Case_3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# MSDS 7331 - Case Study 3 - Clasification of e-mail as ham or spam\n",
    "Daniel Crouthamel\n",
    "\n",
    "Sophia Wu\n",
    "\n",
    "Fabio Savorgnan\n",
    "\n",
    "Bo Yun\n",
    "\n",
    "# Introduction\n",
    "\n",
    "In this study, we will be building a classifier to predict email that are spam or ham. "
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Business Understanding"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "The business is receiving multiples emails that are spam. This causes that the member of the business miss important email because the member erase multiples e mail that are spam and among those the member erase ham email which can be important.\n",
    "\n",
    "An internal IT department has raised an alarm that the staff is receiving a large amount of spam email. They have requested the rock star data scicence team at SMU to build a solution that not only minimizes the number of spam emails, but also keeps important emails from getting flagged as spam and missed.\n",
    "\n",
    "With that in mind, Precision will be and important classification metric to use in our analysis.\n",
    "\n",
    "Our team will build a Naive Bayes model for classification, and leverage the insights from EDA to incorporate clustering as part of the model features. "
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Data Evaluation / Engineering"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "#importing libraries and reading in file\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "#general sklearn libraries\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import RobustScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "#NB\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "\n",
    "# Files\n",
    "from os import listdir, getcwd, chdir\n",
    "from os.path import isfile, join, dirname, realpath\n",
    "import email\n",
    "import email.parser\n",
    "from email.parser import Parser\n",
    "\n",
    "#NLP\n",
    "import nltk\n",
    "import re \n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.metrics import silhouette_samples, silhouette_score"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "source": [
    "def get_cwd():\n",
    "    try:\n",
    "        chdir(dirname(realpath(__file__)))\n",
    "    except:\n",
    "        ##chdir('/Users/fabiosavorgnan/Desktop/QTW/Week_1/Data/MSDS-7333-QTW')\n",
    "        chdir('/Users/boyun/Desktop/SMU/DS7333-QuantifyingTheWorld/MSDS-7333-QTW/')\n",
    "        # chdir('D:\\Projects\\MSDS-7333-QTW')\n",
    "\n",
    "    active_dir = getcwd()\n",
    "       \n",
    "    return active_dir\n",
    "\n",
    "def main():\n",
    "    \n",
    "    get_cwd()\n",
    "    \n",
    "    directories = [\n",
    "            'easy_ham',\n",
    "            'easy_ham_2',\n",
    "            'hard_ham',\n",
    "            'spam',\n",
    "            'spam_2'\n",
    "            ] \n",
    "    \n",
    "    res_frame = pd.DataFrame()\n",
    "\n",
    "    # *dc - Added to keep a collection of email text\n",
    "    emails = []\n",
    "\n",
    "    bodies = []\n",
    "        \n",
    "    for d in directories:\n",
    "        mypath = getcwd() + '/data/' + d + '/'\n",
    "        onlyfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]\n",
    "    \n",
    "        try:\n",
    "            onlyfiles.remove('.DS_Store')\n",
    "        except:\n",
    "            pass\n",
    "        \n",
    "        for file in onlyfiles:\n",
    "            with open(mypath + file, encoding='latin1') as f:\n",
    "                lines = f.readlines()\n",
    "                f.close()\n",
    "            \n",
    "            with open(mypath + file, encoding='latin1') as f:\n",
    "                body = f.read()\n",
    "                f.close()\n",
    "                \n",
    "            msg = email.message_from_string(str(body))\n",
    "            tmpStr = ''\n",
    "            \n",
    "            if msg.is_multipart():\n",
    "                for payload in msg.get_payload():\n",
    "                    #tmp.append(str(payload.get_payload()))\n",
    "                    \n",
    "                    tmpStr = ' '.join(str(payload.get_payload()))\n",
    "                bodies.append(tmpStr)\n",
    "            else:\n",
    "                bodies.append(str(msg.get_payload()))\n",
    "                \n",
    "            in_reply_count = 0\n",
    "            sub_line_all_caps = 0\n",
    "            attachments = 0\n",
    "            subject_line = []\n",
    "            n_lines = 0\n",
    "            blank_lines = []\n",
    "            \n",
    "            for line in lines:\n",
    "\n",
    "                n_lines += 1\n",
    "                if \"Subject: Re: \" in line:\n",
    "                   in_reply_count += 1\n",
    "                if \"Subject: \" in line:\n",
    "                   s_line = line.strip().replace('Subject: ','')\n",
    "                   s_line = ''.join(e for e in s_line if e.isalnum())\n",
    "                   num_upper = sum(1 for c in s_line if c.isupper())\n",
    "                   ttl_chars = len(s_line)\n",
    "                   if num_upper == ttl_chars:\n",
    "                       sub_line_all_caps += 1\n",
    "                   subject_line.append(s_line)\n",
    "                if \"content-type: multipart\" in line.lower():\n",
    "                   attachments += 1\n",
    "                if line == \"\\n\":\n",
    "                   blank_lines.append(n_lines)\n",
    "        \n",
    "            temp_frame = pd.DataFrame({\n",
    "                        'directory':d,    \n",
    "                        'filename':file,\n",
    "                        'is_spam':['Y' if 'spam' in d else 'N'],\n",
    "                        'in_reply': ['Y' if in_reply_count > 0 else 'N'], \n",
    "                        'subj_caps': ['Y' if sub_line_all_caps > 0 else 'N'], \n",
    "                        'attachments': ['Y' if attachments > 0 else 'N'],\n",
    "                        ## *dc+3 \n",
    "                        #'body_lines': n_lines - min(blank_lines)\n",
    "                        'body_lines': [0 if len(blank_lines) == 0 else min(blank_lines)]\n",
    "                        }, index=[0])\n",
    "           \n",
    "            res_frame = res_frame.append(temp_frame, ignore_index=True)\n",
    "\n",
    "            ## *dc+2\n",
    "            #append body of email to collection\n",
    "            text = ' '.join(lines)\n",
    "            emails.append(text)\n",
    "            \n",
    "    #res_frame.to_csv('output_file.csv', index=False)\n",
    "    \n",
    "    ## *dc - add emails\n",
    "    return res_frame, emails, bodies\n",
    "\n",
    "df, emails, bodies = main()"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Above we loaded the data set provided to us, along with creating data sets that represent the entire file (emails) and the body of the email (bodies). To pull the body of the email we used an EmailParser libary. We'll use this information later during clustering.\n",
    "\n",
    "Below we do a quick data set summary, outputing our features, names and a few sampled rows. We created the directory attribute, in case it comes in handy. There are no missing values in the data set. There is only one numberic feature, the number of lines in the email, along 3 categorical features and 1 target feature. For modeling later, filename and directory will be removed."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "source": [
    "df.info()\n",
    "df.sample(5)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 9353 entries, 0 to 9352\n",
      "Data columns (total 7 columns):\n",
      " #   Column       Non-Null Count  Dtype \n",
      "---  ------       --------------  ----- \n",
      " 0   directory    9353 non-null   object\n",
      " 1   filename     9353 non-null   object\n",
      " 2   is_spam      9353 non-null   object\n",
      " 3   in_reply     9353 non-null   object\n",
      " 4   subj_caps    9353 non-null   object\n",
      " 5   attachments  9353 non-null   object\n",
      " 6   body_lines   9353 non-null   int64 \n",
      "dtypes: int64(1), object(6)\n",
      "memory usage: 511.6+ KB\n"
     ]
    },
    {
     "output_type": "execute_result",
     "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>directory</th>\n",
       "      <th>filename</th>\n",
       "      <th>is_spam</th>\n",
       "      <th>in_reply</th>\n",
       "      <th>subj_caps</th>\n",
       "      <th>attachments</th>\n",
       "      <th>body_lines</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3651</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>00187.f2e1e617b73fa1c5137d78383372886f</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5870</th>\n",
       "      <td>easy_ham_2</td>\n",
       "      <td>01163.ce66ad7634e93c6c76e449b74f5139c4</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5320</th>\n",
       "      <td>easy_ham_2</td>\n",
       "      <td>01325.d94c9e1cca235f9f1bcecf469954490f</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2531</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>00630.d4dad4b8734a30afbf8c80499e652c2c</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3036</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>1552.397ac9c4693e0e39de3c115fe7ded51f</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       directory                                filename is_spam in_reply  \\\n",
       "3651    easy_ham  00187.f2e1e617b73fa1c5137d78383372886f       N        N   \n",
       "5870  easy_ham_2  01163.ce66ad7634e93c6c76e449b74f5139c4       N        Y   \n",
       "5320  easy_ham_2  01325.d94c9e1cca235f9f1bcecf469954490f       N        N   \n",
       "2531    easy_ham  00630.d4dad4b8734a30afbf8c80499e652c2c       N        Y   \n",
       "3036    easy_ham   1552.397ac9c4693e0e39de3c115fe7ded51f       N        Y   \n",
       "\n",
       "     subj_caps attachments  body_lines  \n",
       "3651         N           N          46  \n",
       "5870         N           N          50  \n",
       "5320         N           Y          83  \n",
       "2531         N           N          44  \n",
       "3036         N           N          71  "
      ]
     },
     "metadata": {},
     "execution_count": 3
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Below we will replace the 'Y' and 'N' values with 1 and 0, respectively."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "source": [
    "df = df.replace(['Y','N'],[1,0])\n",
    "df.sample(5)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>directory</th>\n",
       "      <th>filename</th>\n",
       "      <th>is_spam</th>\n",
       "      <th>in_reply</th>\n",
       "      <th>subj_caps</th>\n",
       "      <th>attachments</th>\n",
       "      <th>body_lines</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2697</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>0726.b9cf1bbdedddc254ac0a78e9fdb22a8e</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4564</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>01537.ded3233d7649ef141c30e4cc1349f790</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4471</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>0966.20a99be33fcc43b63c8b082d9348ee02</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9127</th>\n",
       "      <td>spam_2</td>\n",
       "      <td>00010.2558d935f6439cb40d3acb8b8569aa9b</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1987</th>\n",
       "      <td>easy_ham</td>\n",
       "      <td>00936.e8fd8c240b680e948f85f2326cc87250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     directory                                filename  is_spam  in_reply  \\\n",
       "2697  easy_ham   0726.b9cf1bbdedddc254ac0a78e9fdb22a8e        0         1   \n",
       "4564  easy_ham  01537.ded3233d7649ef141c30e4cc1349f790        0         1   \n",
       "4471  easy_ham   0966.20a99be33fcc43b63c8b082d9348ee02        0         0   \n",
       "9127    spam_2  00010.2558d935f6439cb40d3acb8b8569aa9b        1         0   \n",
       "1987  easy_ham  00936.e8fd8c240b680e948f85f2326cc87250        0         0   \n",
       "\n",
       "      subj_caps  attachments  body_lines  \n",
       "2697          0            0          46  \n",
       "4564          0            0          33  \n",
       "4471          0            0          54  \n",
       "9127          0            0          56  \n",
       "1987          0            0          45  "
      ]
     },
     "metadata": {},
     "execution_count": 4
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Below we lift some code I (Dan) used in a recent NLP class. Here we use nltk to remove stop words, etc, from the bodies in our emails. This will be used in our EDA and modeling later."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "stop_words = nltk.corpus.stopwords.words('english')\n",
    "\n",
    "def normalize_document(doc):\n",
    "    # lowercase and remove special characters to form a normalized document\n",
    "    doc = re.sub(r'[^a-zA-Z0-9\\s]', ' ', doc, re.I|re.A)\n",
    "    doc = doc.lower()\n",
    "    doc = doc.strip()\n",
    "\n",
    "    # tokenize document\n",
    "    tokens = nltk.word_tokenize(doc)\n",
    "    \n",
    "    # filter out stop words\n",
    "    filtered_tokens = [token for token in tokens if token not in stop_words]\n",
    "\n",
    "    # Remove numbers\n",
    "    filtered_tokens = [token for token in filtered_tokens if not token.isdigit()]\n",
    "\n",
    "    # Remove short tokens\n",
    "    filtered_tokens = [token for token in filtered_tokens if len(token) > 2]\n",
    "\n",
    "    # stem tokens - Skipping for now\n",
    "    #filtered_tokens = [stemming.stem(token) for token in filtered_tokens]\n",
    "\n",
    "    # re-create a normalized document\n",
    "    doc = ' '.join(filtered_tokens)\n",
    "    return doc\n",
    "\n",
    "normalize_text = np.vectorize(normalize_document)\n",
    "norm_text = normalize_text(emails)\n",
    "\n",
    "print(type(norm_text),len(norm_text))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "<class 'numpy.ndarray'> 9353\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Using processed norm_text and directory to convert it to a dataframe"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "source": [
    "norm_text_2 = pd.DataFrame(norm_text)\n",
    "directory = pd.DataFrame(df['directory'])\n",
    "combined_df = pd.concat([norm_text_2,directory],axis=1)\n",
    "combined_df.columns = ['text','directory']\n",
    "combined_df['text'] = combined_df['text'].apply(lambda x: x.split())\n",
    "combined_df"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>text</th>\n",
       "      <th>directory</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[rssfeeds, jmason, org, mon, sep, return, path...</td>\n",
       "      <td>easy_ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[fork, admin, xent, com, tue, sep, return, pat...</td>\n",
       "      <td>easy_ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[fork, admin, xent, com, tue, sep, return, pat...</td>\n",
       "      <td>easy_ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[rpm, list, admin, freshrpms, net, mon, sep, r...</td>\n",
       "      <td>easy_ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[secprog, return, jmason, org, securityfocus, ...</td>\n",
       "      <td>easy_ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9348</th>\n",
       "      <td>[687ifsuy, bol, com, tue, aug, return, path, 6...</td>\n",
       "      <td>spam_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9349</th>\n",
       "      <td>[mraimecoilcipc, msn, com, mon, jul, return, p...</td>\n",
       "      <td>spam_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9350</th>\n",
       "      <td>[fork, admin, xent, com, thu, aug, return, pat...</td>\n",
       "      <td>spam_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9351</th>\n",
       "      <td>[niddeel, hotmail, com, tue, aug, return, path...</td>\n",
       "      <td>spam_2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9352</th>\n",
       "      <td>[received, actioncouriers, com, linux, midrang...</td>\n",
       "      <td>spam_2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9353 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   text directory\n",
       "0     [rssfeeds, jmason, org, mon, sep, return, path...  easy_ham\n",
       "1     [fork, admin, xent, com, tue, sep, return, pat...  easy_ham\n",
       "2     [fork, admin, xent, com, tue, sep, return, pat...  easy_ham\n",
       "3     [rpm, list, admin, freshrpms, net, mon, sep, r...  easy_ham\n",
       "4     [secprog, return, jmason, org, securityfocus, ...  easy_ham\n",
       "...                                                 ...       ...\n",
       "9348  [687ifsuy, bol, com, tue, aug, return, path, 6...    spam_2\n",
       "9349  [mraimecoilcipc, msn, com, mon, jul, return, p...    spam_2\n",
       "9350  [fork, admin, xent, com, thu, aug, return, pat...    spam_2\n",
       "9351  [niddeel, hotmail, com, tue, aug, return, path...    spam_2\n",
       "9352  [received, actioncouriers, com, linux, midrang...    spam_2\n",
       "\n",
       "[9353 rows x 2 columns]"
      ]
     },
     "metadata": {},
     "execution_count": 6
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Data engineering"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Feature Set:\n",
    "\n",
    "1-\twe will create a variable which states the replay or not replay (n_reply_count)\n",
    "\n",
    "2-\twe will create a variable which states if all the subject lines are capital or not (sub_line_all_caps).\n",
    "\n",
    "3-\twe will create a variable which states if the email has attachment or not (attachments).\n",
    "\n",
    "4-\twe will create a variable which states the number of lines in the subject (subject_line).\n",
    "\n",
    "5-\twe will create a variable which states the number of lines if it is more than 0 (n_lines) in the e mail.\n",
    "\n",
    "6-\twe will create a variable which states the number of black lines in the email (blank_lines).\n",
    "\n",
    "7-\tWe will extract the body from the email, then we will run the TDIDF in the body; furthermore , we will run a k-mean  to cluster the TDIDF that we extracted from the body.\n",
    "\n",
    "We will calculate the bag of words. A bag-of-words model, or BoW for short, is a way of extracting features from text for use in modeling, such as with machine learning algorithms. The approach is very simple and flexible and can be used in a myriad of ways for extracting features from documents. A bag-of-words is a representation of text that describes the occurrence of words within a document. It involves two things:\n",
    "\n",
    "\tA vocabulary of known words.\n",
    "\tA measure of the presence of known words.\n",
    "\n",
    "It is called a “bag” of words because any information about the order or structure of words in the document is discarded. The model is only concerned with whether known words occur in the document, not where in the document.\n",
    "\n",
    "TF-IDF (term frequency-inverse document frequency) is a statistical measure that evaluates how relevant a word is to a document in a collection of documents.\n",
    "\n",
    "This is done by multiplying two metrics: how many times a word appears in a document, and the inverse document frequency of the word across a set of documents.\n",
    "\n",
    "It has many uses, most importantly in automated text analysis, and is very useful for scoring words in machine learning algorithms for Natural Language Processing (NLP).\n",
    "\n",
    "TF-IDF was invented for document search and information retrieval. It works by increasing proportionally to the number of times a word appears in a document but is offset by the number of documents that contain the word. So, words that are common in every document, such as this, what, and if, rank low even though they may appear many times, since they don’t mean much to that document in particular.\n",
    "\n",
    "However, if the word Bug appears many times in a document, while not appearing many times in others, it probably means that it’s very relevant. For example, if what we’re doing is trying to find out which topics some NPS responses belong to, the word Bug would probably end up being tied to the topic Reliability, since most responses containing that word would be about that topic."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "The code below is used to create a TfIDF matrix that will be used later in clustering. In this case, we arbitarly chose grams up to 3 words, shoudl appear in at least 5 emails, and at in no more than 80% of all emails."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "tf = TfidfVectorizer(ngram_range=(1,3), min_df=5, max_df=.8, stop_words=stop_words, norm='l2')\n",
    "tf_matrix = tf.fit_transform(norm_text)\n",
    "\n",
    "print(tf_matrix.shape)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "(9353, 166742)\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## EDA\n",
    "### `Latent Dirichlet Allocation (LDA) Topic Modeling`\n",
    "- Topic modeling using LDA will be utilized for the EDA to extract main topics of the given email content and see how certain lemmatized texts correlate with each email directory. \n",
    "- LDA is a type of statistical modeling used to classify text in a document to a particular topic. Dirichlet distribution is used here to calculate a topic per document model and words per topic model. \n",
    "- Due to the newest version of Gensim package removing one of key LDA wrappers, Mallet's LDA result cannot be transferred to a Gensim model for PCA visualization. Gensim will be separately modeled and the results will be compared against Mallet's version to verify both results and enhance the insights."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Training Mallet's topic model"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "source": [
    "from gensim.corpora import Dictionary\n",
    "from gensim.models.coherencemodel import CoherenceModel\n",
    "from gensim.models.ldamodel import LdaModel\n",
    "import little_mallet_wrapper as lmw\n",
    "\n",
    "combined_df['text_2'] = combined_df['text'].apply(lambda x: ' '.join(x))\n",
    "\n",
    "# Download File: http://mallet.cs.umass.edu/dist/mallet-2.0.8.zip\n",
    "\n",
    "path_to_mallet = \"/Users/boyun/Desktop/SMU/DS7333-QuantifyingTheWorld/CaseStudy_Bo/CS3/mallet-2.0.8/bin/mallet\" # update this path\n",
    "# path_to_mallet = \"D:/Projects/MSDS-7333-QTW/mallet-2.0.8/bin/mallet\" # update this path\n",
    "\n",
    "training_data = [lmw.process_string(t) for t in combined_df['text_2']]\n",
    "training_data = [d for d in training_data if d.strip()]\n",
    "\n",
    "\n",
    "# Number of topics that user specifies\n",
    "num_topics = 15  # CHANGE THIS TO YOUR PREFERRED NUMBER OF TOPICS\n",
    "\n",
    "# output_directory_path = 'D:/Projects/MSDS-7333-QTW' # CHANGE THIS TO YOUR OUTPUT DIRECTORY\n",
    "output_directory_path = '/Users/boyun/Desktop/SMU/DS7333-QuantifyingTheWorld/CaseStudy_Bo/CS3' # CHANGE THIS TO YOUR OUTPUT DIRECTORY\n",
    "\n",
    "topic_keys, topic_distributions = lmw.quick_train_topic_model(path_to_mallet, \n",
    "                                                              output_directory_path, \n",
    "                                                              num_topics, \n",
    "                                                              training_data)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Importing data...\n",
      "Complete\n",
      "Training topic model...\n",
      "Complete\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "source": [
    "assert(len(topic_distributions) == len(training_data))\n",
    "\n",
    "for i, t in enumerate(topic_keys):\n",
    "    print(i, '\\t', ' '.join(t[:10]))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "0 \t NUMd font NUM size face color NUMdNUM nbsp width align\n",
      "1 \t NUM net sourceforge com spamassassin example users exmh list received\n",
      "2 \t NUM width http height src img gif font www border\n",
      "3 \t NUM money blockquote business people government mail free report states\n",
      "4 \t com received net localhost org netnoteinc http esmtp content zzzz\n",
      "5 \t NUM NUMe net NUMc pro ssz cypherpunks jul iiu minder\n",
      "6 \t org localhost oct com received dogma slashnull jmason http taint\n",
      "7 \t com xent fork NUM localhost list received org sep mailto\n",
      "8 \t font NUM http www com size href color verdana table\n",
      "9 \t com bNUM dNUM yahoo aNUM cNUM NUM yahoogroups zzzzteana received\n",
      "10 \t one would com like people time net get could date\n",
      "11 \t NUM http code securityfocus com software java web xml www\n",
      "12 \t http www nbsp href com lockergnome html font NUM table\n",
      "13 \t linux ilug NUM org localhost received lugh aug tuatha com\n",
      "14 \t net NUM freshrpms rpm list zzzlist egwn localhost received http\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "source": [
    "# Organizing the result into a dictionary\n",
    "\n",
    "topic_word_probability_dict = lmw.load_topic_word_distributions(output_directory_path + '/mallet.word_weights.' + str(num_topics))\n",
    "len(topic_word_probability_dict)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "metadata": {},
     "execution_count": 10
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Each topic and its weight distribution of words "
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "source": [
    "for _topic, _word_probability_dict in topic_word_probability_dict.items():\n",
    "    print('Topic', _topic)\n",
    "    for _word, _probability in sorted(_word_probability_dict.items(), key=lambda x: x[1], reverse=True)[:15]:\n",
    "        print(round(_probability, 4), '\\t', _word)\n",
    "    print()"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Topic 0\n",
      "0.1403 \t NUMd\n",
      "0.0904 \t font\n",
      "0.0521 \t NUM\n",
      "0.0336 \t size\n",
      "0.0243 \t face\n",
      "0.0237 \t color\n",
      "0.0226 \t NUMdNUM\n",
      "0.0211 \t nbsp\n",
      "0.0209 \t width\n",
      "0.0176 \t align\n",
      "0.0147 \t arial\n",
      "0.0122 \t center\n",
      "0.0111 \t table\n",
      "0.0096 \t height\n",
      "0.0092 \t http\n",
      "\n",
      "Topic 1\n",
      "0.0817 \t NUM\n",
      "0.0354 \t net\n",
      "0.0329 \t sourceforge\n",
      "0.0301 \t com\n",
      "0.0246 \t spamassassin\n",
      "0.0216 \t example\n",
      "0.021 \t users\n",
      "0.0199 \t exmh\n",
      "0.0184 \t list\n",
      "0.0172 \t received\n",
      "0.0158 \t org\n",
      "0.0148 \t localhost\n",
      "0.0118 \t talk\n",
      "0.0116 \t razor\n",
      "0.0113 \t aug\n",
      "\n",
      "Topic 2\n",
      "0.0986 \t NUM\n",
      "0.0626 \t width\n",
      "0.0554 \t http\n",
      "0.038 \t height\n",
      "0.0377 \t src\n",
      "0.037 \t img\n",
      "0.0356 \t gif\n",
      "0.0303 \t font\n",
      "0.0271 \t www\n",
      "0.0238 \t border\n",
      "0.0221 \t cnet\n",
      "0.0184 \t href\n",
      "0.0172 \t bgcolor\n",
      "0.0172 \t com/b\n",
      "0.0166 \t table\n",
      "\n",
      "Topic 3\n",
      "0.0233 \t NUM\n",
      "0.0075 \t money\n",
      "0.0073 \t blockquote\n",
      "0.0069 \t business\n",
      "0.0057 \t people\n",
      "0.005 \t government\n",
      "0.0048 \t mail\n",
      "0.0046 \t free\n",
      "0.0043 \t report\n",
      "0.0038 \t states\n",
      "0.0037 \t make\n",
      "0.0036 \t new\n",
      "0.0035 \t order\n",
      "0.0035 \t send\n",
      "0.0034 \t one\n",
      "\n",
      "Topic 4\n",
      "0.0671 \t com\n",
      "0.0237 \t received\n",
      "0.0217 \t net\n",
      "0.0199 \t localhost\n",
      "0.0125 \t org\n",
      "0.0124 \t netnoteinc\n",
      "0.0112 \t http\n",
      "0.0108 \t esmtp\n",
      "0.0107 \t content\n",
      "0.0101 \t zzzz\n",
      "0.0101 \t aug\n",
      "0.0093 \t mail\n",
      "0.0084 \t jul\n",
      "0.0075 \t subject\n",
      "0.0075 \t mon\n",
      "\n",
      "Topic 5\n",
      "0.2378 \t NUM\n",
      "0.0328 \t NUMe\n",
      "0.0204 \t net\n",
      "0.0161 \t NUMc\n",
      "0.0119 \t pro\n",
      "0.0103 \t ssz\n",
      "0.0098 \t cypherpunks\n",
      "0.0095 \t jul\n",
      "0.0093 \t iiu\n",
      "0.0092 \t minder\n",
      "0.0087 \t cpunks\n",
      "0.0078 \t einstein\n",
      "0.006 \t NUMcfont\n",
      "0.006 \t NUMffont\n",
      "0.0052 \t sightings\n",
      "\n",
      "Topic 6\n",
      "0.0708 \t org\n",
      "0.048 \t localhost\n",
      "0.0392 \t oct\n",
      "0.0304 \t com\n",
      "0.0265 \t received\n",
      "0.0231 \t dogma\n",
      "0.023 \t slashnull\n",
      "0.022 \t jmason\n",
      "0.018 \t http\n",
      "0.0179 \t taint\n",
      "0.0172 \t sep\n",
      "0.0171 \t rssfeeds\n",
      "0.017 \t jalapeno\n",
      "0.0165 \t spamassassin\n",
      "0.016 \t ist\n",
      "\n",
      "Topic 7\n",
      "0.0713 \t com\n",
      "0.0583 \t xent\n",
      "0.0494 \t fork\n",
      "0.0336 \t NUM\n",
      "0.0241 \t localhost\n",
      "0.0238 \t list\n",
      "0.0221 \t received\n",
      "0.0218 \t org\n",
      "0.0156 \t sep\n",
      "0.0156 \t mailto\n",
      "0.0149 \t admin\n",
      "0.0145 \t esmtp\n",
      "0.0142 \t http\n",
      "0.0139 \t subject\n",
      "0.0116 \t postfix\n",
      "\n",
      "Topic 8\n",
      "0.0349 \t font\n",
      "0.0343 \t NUM\n",
      "0.0275 \t http\n",
      "0.0194 \t www\n",
      "0.0148 \t com\n",
      "0.0131 \t size\n",
      "0.013 \t href\n",
      "0.0102 \t color\n",
      "0.0097 \t verdana\n",
      "0.0091 \t table\n",
      "0.0081 \t bgcolor\n",
      "0.0078 \t width\n",
      "0.0075 \t arial\n",
      "0.0075 \t face\n",
      "0.0075 \t html\n",
      "\n",
      "Topic 9\n",
      "0.031 \t com\n",
      "0.0272 \t bNUM\n",
      "0.0265 \t dNUM\n",
      "0.0246 \t yahoo\n",
      "0.024 \t aNUM\n",
      "0.024 \t cNUM\n",
      "0.0221 \t NUM\n",
      "0.0156 \t yahoogroups\n",
      "0.0142 \t zzzzteana\n",
      "0.0118 \t received\n",
      "0.0102 \t grp\n",
      "0.0102 \t scd\n",
      "0.0089 \t eNUM\n",
      "0.0084 \t aug\n",
      "0.0083 \t groups\n",
      "\n",
      "Topic 10\n",
      "0.01 \t one\n",
      "0.0064 \t would\n",
      "0.0063 \t com\n",
      "0.006 \t like\n",
      "0.0054 \t people\n",
      "0.005 \t time\n",
      "0.0048 \t net\n",
      "0.0044 \t get\n",
      "0.004 \t could\n",
      "0.0039 \t date\n",
      "0.0038 \t message\n",
      "0.0036 \t NUM\n",
      "0.0035 \t also\n",
      "0.0034 \t even\n",
      "0.0033 \t said\n",
      "\n",
      "Topic 11\n",
      "0.0274 \t NUM\n",
      "0.0083 \t http\n",
      "0.0071 \t code\n",
      "0.0066 \t securityfocus\n",
      "0.0066 \t com\n",
      "0.0061 \t software\n",
      "0.0053 \t java\n",
      "0.0051 \t web\n",
      "0.0047 \t xml\n",
      "0.0045 \t www\n",
      "0.0043 \t server\n",
      "0.0043 \t list\n",
      "0.0042 \t technology\n",
      "0.0039 \t secprog\n",
      "0.0034 \t new\n",
      "\n",
      "Topic 12\n",
      "0.0685 \t http\n",
      "0.0445 \t www\n",
      "0.0298 \t nbsp\n",
      "0.0276 \t href\n",
      "0.0258 \t com\n",
      "0.0224 \t lockergnome\n",
      "0.0143 \t html\n",
      "0.0123 \t font\n",
      "0.0123 \t NUM\n",
      "0.0107 \t table\n",
      "0.0097 \t border\n",
      "0.0092 \t target\n",
      "0.0087 \t blank\n",
      "0.0081 \t size\n",
      "0.008 \t class\n",
      "\n",
      "Topic 13\n",
      "0.0584 \t linux\n",
      "0.0409 \t ilug\n",
      "0.0366 \t NUM\n",
      "0.0296 \t org\n",
      "0.0254 \t localhost\n",
      "0.0232 \t received\n",
      "0.0216 \t lugh\n",
      "0.0185 \t aug\n",
      "0.0181 \t tuatha\n",
      "0.0155 \t com\n",
      "0.0152 \t esmtp\n",
      "0.014 \t admin\n",
      "0.0094 \t list\n",
      "0.0083 \t net\n",
      "0.0082 \t irish\n",
      "\n",
      "Topic 14\n",
      "0.0629 \t net\n",
      "0.0623 \t NUM\n",
      "0.0499 \t freshrpms\n",
      "0.0474 \t rpm\n",
      "0.0426 \t list\n",
      "0.0286 \t zzzlist\n",
      "0.019 \t egwn\n",
      "0.0175 \t localhost\n",
      "0.0153 \t received\n",
      "0.0127 \t http\n",
      "0.0113 \t esmtp\n",
      "0.0104 \t mailto\n",
      "0.0102 \t lists\n",
      "0.0102 \t subject\n",
      "0.0097 \t admin\n",
      "\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### `Results from the heatmap and box and whisker plots below suggest that __Topic 0, 3 and 4 are highly correlated with 'Spam'__ directory and __Topic 0, 3, 4, and 5 are highly correlated with 'Spam_2'__ directory. This shows the correlation of certain words with directories and address the need to include clustering of lemmatized texts as a feature in the prediction model.`"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "source": [
    "directory = combined_df['directory'].tolist()\n",
    "\n",
    "target_labels = [\"easy_ham\", \"easy_ham_2\", \"hard_ham\", \"spam\", \"spam_2\"]\n",
    "\n",
    "lmw.plot_categories_by_topics_heatmap(directory,\n",
    "                                      topic_distributions,\n",
    "                                      topic_keys, \n",
    "                                      output_directory_path + '/categories_by_topics.pdf',\n",
    "                                      target_labels=target_labels,\n",
    "                                      dim=(14,6))"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA84AAAGfCAYAAACOWIsjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOydd0DN+//Hn6dSiCQjO/MWJVnJTKUhURGJRCVFQ2ZWRkZxSdfOHlFGZCREQ2TUrZtVKFQo7aV5Oq/fH/3O57bu/S4n173vxz/0OZ/zPq/P+/Ner/f7NXhERGAwGAwGg8FgMBgMBoPRKGLfWwAGg8FgMBgMBoPBYDD+yjDFmcFgMBgMBoPBYDAYjD+BKc4MBoPBYDAYDAaDwWD8CUxxZjAYDAaDwWAwGAwG409gijODwWAwGAwGg8FgMBh/AlOcGQwGg8FgMBgMBoPB+BMkvrcADAaDwWB8C1atWoUrV6784edOTk5wdnb+r8qeM2cOVFRU4Obm9t+Kx2AwGAwG4weGx/I4MxgMBuPvQHFxMcrLywEA79+/x5w5c3Dx4kV07twZANCyZUtIS0v/V2UXFBRAQkICrVq1+mbyMhgMBoPB+HFgJ84MBoPB+FvQunVrtG7dGgCQn58PAJCTk0OHDh3+57JlZWX/5zIYDAaDwWD8uDAfZwaDwWD87SkuLsbmzZsxduxYDBo0CLa2tnj37h33uba2Nk6cOAFzc3OoqqrC3NwcL1++5D6fM2cOtm/fzv197tw56OvrY9CgQTAzM0NsbGyTPg+DwWAwGIymhSnODAaDwfjb4+LigidPnsDb2xsXLlyAlJQUbG1tUVZWxt3zyy+/YMqUKbhy5Qp69eoFGxsbFBQUNCgrMDAQ27dvh52dHa5du4YRI0bA3t6eO+VmMBgMBoPx94MpzgwGg8H4W/PmzRtER0fD09MTw4cPh6KiInbu3InS0lJcv36du8/Q0BCzZ89Gnz594OHhAQkJCdy8ebNBeefOnYO5uTnMzMygoKCA5cuXY+bMmSgqKmrKx2IwGAwGg9GEMMWZwWAwGH9rkpOT0axZM6ioqHDXWrZsiQEDBuDt27fctWHDhnH/l5SUhKKiYp3PhaSkpGDgwIHc3zweDytWrICCgoKInoDBYDAYDMb3hinODAaDwfhbIyUl1eh1IkLtxBLi4uJ1PhcIBA2uAUCzZs3A4/G+rZAMBoPBYDD+0jDFmcFgMBh/a/r06YOqqiq8ePGCu1ZWVoakpCT06tWLu/bq1Svu/xUVFXj9+jUUFRUblNezZ886gcMAYNKkSbh9+7YIpGcwGAwGg/FXgKWjYjAYDMbfmp49e0JPTw9r1qzBhg0bICMjg3379kFcXByTJk3i7jt//jxUVFQwYMAA+Pr6QkJCAhMnTmxQ3rx587Bu3Tr0798fampquHjxInJycjB8+PCmfCwGg8FgMBhNCFOcGQwGg/G3Z9u2bfD09MTChQvB5/MxfPhw+Pn51cnPbGZmhhMnTiAlJQVqamo4efIkWrVq1aCsSZMmITs7Gz4+PsjNzUX//v1x5MgRyMnJNeETMRgMBoPBaEp4VNvBi8FgMBiMfyDa2tqwsbGBpaXl9xaFwWAwGAzGXxDm48xgMBgMBoPBYDAYDMafwBRnBoPBYDAYDAaDwWAw/gRmqs1gMBgMBoPBYDAYDMafwE6cGQwGg8FgMBgMBoPB+BOY4sxgMBgMBoPBYDAYDMafwBRnBoPBYPywVFdXN7jGPJBEh0Ag+N4iMBgMBoPxXWA+zgwGg8H44RAIBODxeODxeACAhw8fgs/nY8SIEWjevPl3lu7vh1BhFhNj++0MBoPB+GfCFGcGg8Fg/DBkZWWhY8eO3N8ZGRlYuXIl3r9/D2lpaSgoKODw4cPfUcK/H0TEbVA8e/YMERERGDRoEBQUFNCzZ08IBAKmUDMYDAbjbw+b6RgMBoPxQxAaGorly5ejoqIC1dXVCAoKwrp16zBq1Cg8ePAA+/fvx+PHj3Hp0qXvLeoPT2lpKZ49ewYA4PF44PP5+OWXX2Bvb4+srCz4+flx/2dKM4PBYDD+CbDZjsFgMBg/BIMHD8apU6eQmpoKcXFxZGZmIjo6GgoKCgCAvn37wtHREXv27EFubu53lvbHxsvLC3v37sWbN28AALm5uYiOjoa/vz+2bNmCI0eOICsrC9u2bUNlZeV3lpbBYDAYDNHDFGcGg8Fg/CURCAR1glG1b98e4eHhmD17Nl69egULCwsMGTIE9+/f5+6xt7eHhIQETp8+/T1E/qEhIi7YmoWFBcrKyvD48WNUVlbi9evXaNasGXr27Inr169DV1cXgwcPhq2tLac4M88vBoPBYPydYYozg8FgMP5yCP1mxcTE8OnTJ7x9+xZlZWUYNmwYtLW1cfjwYbRp0wazZ8/Gy5cvERMTw3139erV8PX1RWpq6nd8gh8LPp8PHo8HcXFxAED//v0xdOhQREZG4t27d+jevTt+/fVXTJs2DXv27IGLiwuOHz+Oq1evYtu2bQDA+UEzGAwGg/F3hCnODAaDwfjLISYmhsrKSqxYsQIzZ87E2rVrsW7dOsjIyMDQ0BDv37/H3bt3oaenh4EDB+LYsWPcd3V1deHj48OZcDP+GD6fDwCQkJBAWVkZLl68iKioKBAR5s+fj6qqKty4cQPdu3eHiYkJcnNzERoaismTJyMnJwfx8fGYOXPmd34KBoPBYDBED4uqzWAwGIzvTv3IzAKBAHv37sWnT5/g4eGBsrIySEtLQ1JSEl+/fsX+/fsRHx8PPz8/xMfHY+XKlXBwcMCMGTO+41P8uLx8+RKzZ8/G0KFDkZKSAjU1NSxZsgTJyck4cuQInJ2d0b9/f1hYWKBHjx7o1KkT7ty5A319faxbtw6SkpLf+xEYDAaDwRApTHFmMBgMxnelurqaMxEuLi5G69atAQALFixAx44d4ejoiAcPHiAiIgIdOnTArFmzIBAIsHnzZowdOxYODg64ePEi1NXV2Snzv0A45QvNqsvKymBtbQ05OTkYGBhgypQp+O2333D58mVkZWXh0KFDcHV1hbS0NNauXYuCggK8evUKSUlJGD9+PFRUVL7n4zAYDAaD0WQwxZnBYDAY35309HRs3boVFRUVkJeXh5eXF27cuAEvLy/07NkTnz59wqRJkxAdHY3q6mpcvXoV+/fvR1xcHA4ePMhOPP8Nam9Q1M7N7OHhgXPnzuHChQsYOHAgeDweoqOjsX//fqxYsQJEBA8PD5iZmWHWrFl1fJmFwdtYSioGg8Fg/N2R+N4CMBgMBuOfRX2z7Pj4eCxevBjTpk2DpqYmnj9/DgAwMjLCsGHDkJeXhwEDBgAA7t+/j4CAAAgEAsyePRuOjo5cObWVQUZDxMXFIRAIsHv3bpSXl6NDhw6wsbGBs7MzQkJC8P79e6iqqgIA+vTpg+LiYhQUFGD8+PEYNGgQOnbsCB6Px9Vz/ffIYDAYDMbfGTbjMRgMBqNJEAgEIKIGytazZ89gZGSEhQsXQk1NDT169MC1a9eQlpaGli1bIiwsDBEREfDz84Obmxs0NDQgJiYGWVlZAOBSKDGl+c9JTk7GxIkT8fbtW/To0QMBAQFYvnw5CgoKsGzZMnh5eXHBwuTl5dG8eXPIyMgAANavXw9dXV0Av9czU5oZDAaD8U+CnTgzGAwGQ+TUVphjY2Nx69YtjBo1ClpaWpCUlERgYCBev36N9+/fo7KyEh07dkTbtm2xcOFCVFVV4fTp0yguLoa3tzdGjhxZp2yh+TGjBiKCQCCAuLh4nVP4qKgoDBo0CDt27ABQE3185cqVuH79OqytrXHu3Dno6enBwsICQUFB6NatG3766ScANUoyO9FnMBgMxj8ZpjgzGAwGQ+TweDyUlJRgw4YNiI6OxogRI7Bt2zakpKTAzs4OP/30E1JSUtChQwdoaWkhISEBGzZsgLy8PJYsWVInaBgRNXpyzfjdDF5cXBwlJSXg8/ncyfybN2/QrFkzADWn9J06dcL48eNx584dODk5YcWKFbC2tkZxcTGWLFmCCRMm1CmbKc0MBoPB+CfDFGcGg8FgfHNqB6ISEhoaiurqajx8+BBiYmJYsWIFHj58CGVlZYwaNQpEhN69e0MgECA0NBRycnKQlpYGAE5pFpbLlLjGEW4m/Pzzz7h79y66du2K7t27Y+XKlWjbti2ysrKQkZGBzp07AwDU1NSwd+9eZGVlYeTIkdDV1cXz58+xdOlSAI2/RwaDwWAw/omw7XoGg8FgfDNqmwkDNemlhJw7dw6jR4+GmJgYLl26hIKCAlRUVCAsLAxVVVU4f/48bG1toaWlhfj4eHh4eEBOTq5O+UyJq0tjiTEOHjyIZ8+e4dixY1i+fDlycnLg7OwMPT09vHz5Enfv3kVeXh4AIDIyEpMmTUKnTp0AAMuWLUNcXBzu3LkDgPkxMxgMBoMhhJ04MxgMBuObwePxwOPx8Pz5c+zatQtEBCUlJaxevRre3t5o3rw5HBwc8OnTJ+zZswehoaHw9/fHkCFD8PPPPyM9PR1FRUVQVlYG0DACN+N3+Hw+JCTqTuOVlZX47bffYG5ujm7duuHVq1fIyMhAZWUllJWVYWNjg0uXLuH69esgIhQUFGDnzp3c93v27AljY2NcunQJenp67GSfwWAwGIz/hynODAaDwfifqK/c3r17Fxs3boS9vT169+6N7OxsAED37t3x4MEDVFVV4fr16wAACQkJtGrVCgEBARg+fDi6d+/OlcPMhBtHqDBLSEigqKgIt27dQo8ePTB48GBUVVXhw4cPaN68OVasWIH79+9j4cKFmDRpEk6fPg1bW1toamrixYsXqKyshIGBQYPy3d3dOV9oBoPBYDAYNTDFmcFgMBj/FULFtrbSLBAI8OzZM1hbW2POnDmorKzE/fv3cfPmTQwePJjzcX716hXCwsK4wFQ6OjoNTk+Z0tw4wnqKi4uDlZUVRo0ahcTERIwYMQJr1qyBjo4OnJycYGlpibCwMEhLS+PEiRMIDw+HpaUlOnbsCG1tba68+ifXTGlmMBgMBqMhTHFmMBgMxn+FULGNiIjAo0ePMGTIEOjq6oLP5+P06dOIjo5GUlISWrRogZYtW6JXr16YN28eZs2ahdWrV6NZs2bYuXMnl/KInTD/e5SUlGDOnDno3r07du/eDV1dXTx9+hRXrlzBxo0bsXHjRvj7+6NXr17Izc1FbGwszpw5AwcHB0hJSTUor/6GBYPBYDAYjIbwqLHIIgwGg8Fg/AsKCwvh5uaG9+/fY9CgQeDz+Vi3bh2kpaURHx+Pd+/eoXfv3tDQ0MDNmzfh5+cHX19ftG7duk5kZ4FAwPlGM/491q5di8DAQFy/fh19+/YFj8dDREQEDh06hG3btuHTp08ICgpCVlYWCgoKsHDhQhgaGn5vsRkMBoPB+GFh28wMBoPB+Jc0FogqKSkJfD4ft2/fBgBUVVWhqKgIfD4fGhoaICKoqKigoqICUVFR6NSpE1eGUGlmp8yNU79eiKjOxoKLiwtu3ryJjx8/ol+/fgCAPn36oLi4GDk5ORg7dizGjh2LT58+oWvXrtz3WLA1BoPB+PawueyfAVOcGQwGg/GHCI2ShArvo0eP0KlTJ/Tq1Qt5eXl49+4dNm7ciKqqKoSEhKBv374QFxfHqVOncPjwYZSWliI9PR2DBg3C9u3b0aJFizrls4VGXYT1LayXjx8/olu3bg1O4+Xl5bFgwQJs3rwZWlpaAICuXbuiRYsWaNOmDXefUGluzB+dwWAwGN8G4ZidnZ2NDh06fGdpGKKCmWozGAwG41/y+PFjuLu7QyAQoHXr1pg+fTpmzZqFkydP4s2bN6iqqoKxsTEEAgE2btyILVu2YNCgQcjOzoZAIECfPn0AsBPPf5eoqCjs3r0blZWVGDp0KBYtWgR5efk691RWVmLy5MkAgBkzZiAgIADKysrw9PRssEHBYDAYjG8HEYGIuPns0aNH8PDwQIsWLWBsbIw5c+awue5vCDtxZjAYDEYd6iu3T548wbZt27Bs2TKMHTsWv/zyC/bs2YOJEyfC2toa2dnZaNGiBVq1aoVnz55BQUEBvXr1QqtWrdCqVSuuTABsIdEIjaXz2rZtG9atW4fOnTujvLy8jtIsNNuWlJTE6tWr4eDggLKyMqxevbpOtGwGg8FgfHuEYzCPx0NJSQmKiopw48YN2Nrags/nY9euXejRowfGjx/PYnf8zWCKM4PBYDAA/K7A1VduY2Ji0K9fPxgYGCAvLw/i4uJo1qwZtm3bhp07d+LatWsIDQ1Fly5dEB4eDltbW3Tp0qVOGUxhbsgfmU9//vwZGhoaGD58OEpLSxESEoLnz59DUVERI0aMqHPv+PHjMWrUKLx+/RpOTk4gIggEAmYCz2D8Azh58iQmT56Mdu3afW9R/lEIlWF3d3fExsZCWloaXbp0gZmZGQAgPj4eFy5cgJKSEhfPg/H3gK1kGAwG4x9O/dPgixcvYsWKFYiMjIRAIICKigqcnJwQHBwMMzMziImJYe3atbh37x4+fPiAiRMnYubMmVBQUMDNmzfh5OT0PR/nh0Go3AYGBmLZsmXYt28fioqKICUlhefPn8POzg5Tp05FYGAgHj58iPXr1+P9+/cNTjBWrFiB8PBwREdHg8fjMaWZwfgH8PPPP8PLywsHDx783qL8YxB6t4aGhiI0NBRFRUXw8vKCmJgY8vPzUVhYCABYvHgxXr9+jQcPHqCysvJ7isz4xrATZwaDwfgHUz8S6O7duxEWFoaffvoJPj4+SE5Ohq2tLXJychAQEIAVK1Zg4sSJuHXrFsrKymBhYYG7d+/CxMSkTpliYmLMRO1fUFhYiKVLlyInJwe6urooLCxERUUFjI2NMXLkSCQlJaFr165QVlZGbGwsDh8+jGbNmjUop3///tDT00NgYCBGjRr1HZ6EwWA0FUIz4UGDBqFr1664desWJk6ciKFDh35v0f521LfgEc5pW7duRVZWFg4ePIhBgwbB09MTFhYWiImJwfjx49GlSxdMnToVp0+fhoqKCvr37/89H4PxDWGKM4PBYPwDEZpli4uLIzc3F97e3ujcuTNyc3Nx/fp1AMDhw4cRHR2N0aNHIz8/H1lZWVBTU0N+fj6uXbuGbdu2oWXLlpCWlq5TLjvxbEhj6bzevn0LIsLVq1cBAEVFRcjIyACfz0ePHj2QkZGBr1+/Ii8vD0eOHIGYmBjat2/faPleXl6QlJQU+XMwvi3104wxGI1RWlqKli1bAvhdeePz+VBSUkKXLl2wc+dO+Pv7f08R/3YI+6a4uDjy8/Px/v17tG/fHj169MChQ4dgYmKCZs2accEvDQwMcPToUQwcOBDy8vJYtGgRHj169L0fg/GNYabaDAaD8Q9EaJb99OlTWFtb4+vXrwgPD0dwcDBev34NABg5ciRkZGQQFBSE4cOHg8/nw83NDdra2mjXrh1MTExgYGDQaLmMugiV5ri4OOTl5QEAWrVqhejoaLi4uMDBwQGjRo3C6tWrMXPmTOTm5uL27dvYtWsXTE1NIScnh3379qF58+aNlv8tlebq6upvVhbjz/n69ev3FoHxF+fUqVMYMmQILl26hJKSEu76sGHDkJCQgBEjRqCiogIXLlz4jlL+/RBuUOzduxcGBgbw9fWFqakpzp49CyUlJUyZMgX79u1DRUUFAGDNmjX4+PEjgoKCUFlZCTExMZw9e5adNv/NYCfODAaD8Q+gth8zEaGqqgq7d+/G8+fPYWRkhAULFuDx48c4d+4cQkNDoaioiIEDB2LkyJG4ceMG4uLicPPmTcTExEBBQQHdu3cHwE7M/l3u3r2LrVu3Ql5eHrm5uTA2Noa5uTmOHDmChIQElJSUYPHixejUqRP09fXx9OlTrF+/Hrm5uRAIBFxe0Pqm9d8SgUDAfKSbkJUrVyItLQ0BAQGsHzEaUFRUBBkZGXz+/BkAcPXqVaSkpGDRokVo3bo12rRpAz09PQCAgYEBTpw4ARMTE2Z58l8iHP94PB7XHx8/foynT5/C398fvXv3xpkzZ3Dv3j3weDxs3LgR6urquHfvHoyMjNC8eXNYWVnh48ePdTaQWQrGvxfsTTIYDMbfHKHPsZiYGDIzM1FQUABJSUloaGggIyMDnz59AgBoaGhAUVERL168wNOnT7lr7dq1Q3h4OCQlJTFmzBh0794d1dXV3EKDURdhABkheXl58Pf3x7JlyxAQEIC9e/fi5s2bOHbsGDQ0NDBnzhzMmzcP/fv3x4cPH6CoqMjlvW7Xrh06dOgAgUAgUjN44eJOmF7FwcEBly9fBsBOoL81fD4fAODg4ICXL1/i6dOn3GKdwaiqqsLRo0excuVKAMCiRYvQtWtXtGnTBgUFBXB1dUVBQQE3nsvIyMDY2Bht2rTB/v37v7P0Pya143IUFhZy85pw3uvduzcAYM6cOVBVVUVUVBRatGiBhQsXYtu2bcjNzQUALFiwAB4eHnXccpjS/PeCvU0Gg8H4myMuLo6vX79i6dKlMDMzg4uLC65fvw5NTU1MmDABhYWFePXqFQBAR0cHrVu3xs2bNzlf26VLl8LNza2OktxYGqV/OkIFs74S9PLlSyQmJsLIyAglJSUICAhARkYG1NTUUF1djcjISJiYmMDe3h62trYYP348fvrppzplN5Ym7FsglFNMTAylpaVwdXXFli1bkJCQgAMHDgAAO4H+xkhISKC4uBhJSUlQUVGBt7c3ALBNKAYAoFmzZmjfvj2KiooQEhKCNm3awMbGBjExMXB1dQURYevWrUhNTYW6ujquXr0KeXl5mJmZ4fjx40hPT//ej/DDIS4uzo1/8+fPh52dHaKjoyEtLY1OnTqhqKiIu3f48OH47bffUFhYCEdHR7Ro0YLbfBaOp2yz8e8LW/UwGAzG35zPnz9jzpw5kJSURGBgIGRlZXHlyhUkJydj2rRpKC0tRVhYGABASUkJAwYMwIcPHzhfZ6FZttDcm9EQIuIUzOPHj8Pd3R0hISGorKxE69atMXDgQHh4eEBXVxdFRUWIiopCZmYm7t69C21tbezatQv6+vq4f/8+bG1tm0ReoK6y5u7uDoFAgKVLl8Le3h7l5eX4+eefAbB3/y25f/8+NDU1ERERgY4dO+K3337DuXPnADS0VmD8MxBaIQj/HT16NFRVVXHp0iWUl5dj9uzZkJWVxYULF+Dt7Q0FBQXMmzcP1dXVaNGiBfed9evXQ15enrWj/5CSkhI4OTlBTEwMPj4+6NevH/r06YMuXbrg48ePdYJ8tWnTBj169EB5eTkA4M6dO1BVVQXw+3jKNhv/vjAfZwaDwfibUD91hpCwsDB06dIFXl5eAIC+ffvi3r17uHz5MlauXIkRI0YgLi4OERERGD9+PExNTWFkZMT51QphJ8x1efbsGWRkZNCzZ0/weDxkZmZi/fr1KCwsRLdu3bBv3z5kZmbC1NQU2dnZ+PDhA3x9fblF1rVr12Bra4tWrVph9OjRXLmNReD+ltT2p7137x6SkpIwdepUfPr0CY6OjujYsSPmzZsHBQUFODo6wsbGBu3atWN+uP8h9eMKCOsuNDQUVlZWcHV1RWVlJcaMGQMvLy+YmppyStD3oqKiAllZWejevTuqqqoaTX/G+HZkZWVh3759GDRoEKZNm8b1+w4dOmDcuHF4/fo1Tp8+jQULFmDt2rVwcHCAoaEhnJycUFZWBl9fX8jJyQEAOnfujOnTp3/Px/lhSUtLQ3FxMbZt24ZOnTph2bJlEBcXh7a2Nn799VecPn0aSUlJ6NevH3bu3InJkydDXl4eQI2SLMrYE4y/FmwVxGAwGH8Dagd2Sk9PR0xMDD5+/AgAkJOTg7a2Nj5//gxra2ukp6djzJgxiIuLQ1xcHExNTVFUVISXL1+Cz+dDVlYWHTp0YKcWf8KHDx9w9OhRTtHx9/fHypUr0bp1a5w/fx47duyApaUlLl++DD6fDwsLC/Tt2xc3b95EbGwsLCwsICkpiZEjR9Ypl4hEqjQDNacinz9/xvHjx3Hs2DEMHDgQRUVFKCsrq5NabOzYsejTpw88PT052Rj/HkKfcTExMRQVFaGqqgpAjWL65csX7h03a9YM06dPh4KCAmey/T1P9+/cuYPly5dzsgHsvYuCrKws5OXloWXLligsLERcXBwyMjIA/H7qPGjQIIwcORJRUVFIT0/HuHHjMG7cOGzevBkA4OLigu3bt3O+0ELY+2qcP+tXubm5ePfuHTp16gSgZrMrKysLd+7cgZGREaZPn46PHz/i3LlzcHV1xZIlSxq4LjH+GTDFmcFgMP4GiImJQSAQwNfXFwYGBjhw4ACMjY05U+BJkyZh7dq1UFJSws6dO2FiYoIXL15g06ZNqKqqgoeHBxwdHesobex0sSHCxVfPnj2xZ88evHjxAq9fv4ampiaKioqQnZ2NkpISiIuLQ11dHT179sS+ffswffp0zJ07Fzk5OfD19YW6ujoCAgK40yIhoqjzxvztHj58iB07dmDQoEEYN24cFBUV0apVK5w7d44zQRQXF4e0tDRu3LiB+Ph4ro0x/jViYmLg8/lwd3fHvHnz4OrqipMnT0JKSgoyMjIoKChAXl4e974HDhyIc+fOISUl5btadgwYMAAAYGxsDHd3dwBsHPjWCAQCzJkzB0eOHEGrVq1gZmaGL1++4O7duwBqfOCJCNLS0tDQ0ICsrCwiIyMBAKtWrcLjx48RFhYGKSkpaGlpQVtbu0757H3VRTj+1e9XtTcY1NXV0aZNGxw+fBhATR1KSUnh6NGjaN26NUxMTLB161b4+flhypQpAJj7yj8VpjgzGAzGD0j9STspKQmurq54/vw57ty5gxMnTmDWrFnw9/fH/fv3kZ2djbdv32LJkiUAgJCQEOjr68PCwgKtW7dGr169Gi2XUYNAIAAR1Vl85eXlYefOnbh27Rq6dOmCqVOnQkxMDC9evAAA9OrVC0ZGRoiPj0dYWBjU1dWxfft2HDhwgHsPTRFERnga8vTpU6SlpaGqqgrTp0/H4MGD8e7dO5SVlQGo8XGOiYnBzp078ebNGwQEBKBbt26YNGkS5+vMzPX/GOFCnIhQWVmJdevWIScnB3v37sXEiRNx+/ZteHt7Y+7cuYiKikJYWBh3uiglJQV5eXlERUU1qczCdi2kuLgYCQkJXFwEgAU6+pZUVVVBTEwMCxYsQFRUFJ49e4axY8eiX79+iI2NxcuXLwH8Pg4PHDgQBQUFnMVCz5494ePjg+HDh3+3Z/jREI5/Fy5cwIYNG7h82LU3GKSkpODs7IxffvkFt2/fRnFxMYKDg9GlSxd07twZALg0X3+kiP+TuXjxIucHXlVV9Ze3evhf5GNvncFgMH4giIhLnVGb/Px8JCYmoqCgAF27dgUAODo6om3btoiJiQGfz0e7du2wYMEC6Orq4tGjR1i0aBFmzpyJli1bcuWwxUDjCFOVxMTEYO/evYiMjIScnBwcHByQkJCAR48ewdzcHDweD1FRUcjJyYGYmBhUVVWhpKTELSrExcXRrFkzTmFpChO/+Ph4GBoawtPTE46OjnByckJGRgY2btyIyMhIvH37FkBNYDh3d3eUlpZi2bJlOHPmDKysrGBsbIzy8nLO9J/RODweD8+ePUNRUREqKirw7t07rF27Fl27dkW7du2QlpaG3377DQMHDsT06dNx5swZ2NraQltbGx8+fMClS5cwb968JpO3dgqy9PR0lJWVoXv37jh8+DA0NDQQFBQEgJmh/q+8ffuWi3QtrMtp06ahbdu2CAgIAJ/Ph7GxMSoqKnDnzp069/F4PC6ysxADAwO0bt36L6+cfC+EsT6EZGRkYPr06bh06RJ4PB6Sk5NRUVHR4HumpqawsbGBv78/LC0tceHCBSxfvrxBrA/WH+ry9etXvHv3DkePHsXTp09RUVHxl7d6eP78Oeca8R/3I2IwGAzGD8fnz59p3bp15OXlRVevXiUiokOHDtHEiROpuLiYu+/06dM0ZcoUIiJKS0ujo0eP0rVr17jPBQIBCQSCphX+B6G6upr7f3FxMbm4uND48ePJ2dmZrK2t6cuXL0REtGjRIlq7di2VlpbS7du3ydLSkoKDg7nv5uTkNLnsQoqLi8nZ2ZlOnDhBRERJSUnk5ORES5cuJT6fT8uWLSNLS0uqqKio8720tDTu/35+fuTq6tqUYv8Q1K+z4uJiMjc3p9OnT9P79+9JXV2dkpOTyd7enkaNGkVXrlyhDx8+0I0bN4iopo6vX79OISEhXBkCgaBOuxM1mZmZZGdnR3p6ejR79my6d+8eEREFBgbSpEmTKCEhgYiI+Hx+k8n0d+LFixekqKhI6urqdObMGUpOTuY+e/r0KWlqalJoaCgR1YzVdnZ2dPLkSSKqeTc2NjY0a9YsysrK+i7y/8hUVlYSEVFwcDA5Ojpy14uLiykjI4MKCwuJiKiqqqrOHFhRUUGvX7/m/mbzY+PUHqeCg4NJTU2NFBUVKTExkav7vyKZmZmkp6dHERERlJubS/n5+f/R99nRAoPBYPxgBAcHY8qUKeDxeJCQkOB8UsePH49+/frh2LFj3L2dOnVCly5duNMkW1tbTJ48GUCNyRmPx/vL7w5/L2qfvr948QKlpaUIDw/Hnj17cPToUc50z8bGBomJiQgNDYWenh6aN2+OqKgoFBQUAADatWsHQHQmrxkZGaisrGz0N9LT0/Hw4UPo6uoCABQVFTFjxgzk5+cjPDwc7u7uiIuLQ3BwMPed7OxsBAUFYfv27di8eTN2796NsWPHAmCBhwAgICAAeXl53PsXwuPxIC8vz0Va79GjByZNmgQ1NTXcv38fJiYm2LNnD54+fQqgJs2bkZERDAwMAPzeH0Vl9VHfDSM3Nxeurq4YMGAAAgMD4erqCi0tLQDAyJEjoaKiwvl8slO2/w5lZWWYmJigsLAQv/76KywsLLBhwwYkJydj+PDhmDBhAvz9/ZGbmwtTU1Po6OjAx8cHCxcuhKmpKXr27ImzZ882OPVk1KV+246MjISpqSmAmjHr/v378PDwgL29PUaOHAknJydYWVkBqPEpF86BRARJSUn89NNPAH7vk4zfacxUPTs7GwoKChgwYACUlJTQrFmzv+RcQUSQl5fH+PHjsXr1asyePRtpaWn/URlMcWYwGIwfjPv372PTpk3w8PDAsmXLMHToUHz48AG9e/eGlpYW/P394enpiYCAAKxbtw7jxo2rk+ZGOKGxxXBd6k/0b968gYODAyoqKvD582d8+fIF27dvx7Zt2zB69GjY2dnBxcUFAwcOxIgRIxAYGIiMjAysX78eK1asgKysbJ3yRFHfpaWl2LFjB/z8/Br9jYqKCigrK3OmokBNtOycnBy8e/cObdq0gYWFBV6+fMk9f4cOHdCnTx+IiYmhvLwcQUFBmDp1KoBvF3jor7io+nd4/Pgxvn79ilatWiE/Px/a2tq4ceMGCgoKIC0tDWlpady+fRsAsHDhQvB4PIwYMQJlZWUICgpCfHw8Jk6c2GjZouqPAoGAM8uuTUlJCSQkJDBkyBAAwLt37+Dp6YnAwEB07twZkydPxtu3b7Fo0SL88ssvIpHt74awrmtjbW0NcXFxWFlZ4eeff0ZaWhoWLFiAZcuWQUNDA6mpqQgLC4OkpCTMzc0RFBQEOzs7BAUFcQHamJ/5n1O/bauoqCAtLQ1BQUGYNGkSVqxYAXFxcXTt2hUXLlzAqlWrkJ6eXic/M9BwfGNzZEOEdRIcHIzt27cjPDwcZmZmXGq07du3f2cJ61K/TwoEAqSmpqK8vBw6Ojpcesh/F5bHmcFgMP5iCJWKxpQUIuKi9f72228IDQ1FdXU1iAizZs3C3LlzER4ejtu3b2PixInYvXs3Ro0aVacMtoPeOPXrpaioCFlZWQgICMC0adOQnp6O9+/fo7KyEu7u7qisrMShQ4fw66+/Ytq0acjPz4eYmBi6d+8OAI0qK9+a5s2bY+zYsTh8+DD8/Pzg4+NTZyHQrVs3iIuL49GjR1BSUuKUeXl5ee4Ua926ddz99P/5hg0NDWFoaPjN5aU/yDX+V0f4LjU0NKCmpoaDBw9i2rRpsLKywvXr1/HgwQN4eXlBX18ffn5+yM3Nhba2NhYvXox9+/ahqKgI5eXl2Lx5MzQ0NJpUdmEbTEhIQGBgIBQVFdG/f390794dbdu2xdatWyElJYWCggKMHj0a27dvR1FREaytrbF48WJcvHgRgwcPblKZf0Rq9/e3b99yVj6KioqYMmUKPDw8cOXKFYwZMwYvXryAj48Pjh49io8fP+LkyZNQVlbGgAEDoKCgAAUFBQDg4lmIqr/8qP2RiOoEa8zNzcWRI0dgZGQEFRUVtGvXDsuXL4enpycmTpyIOXPmIDs7Gy1atECrVq0QFhaGwYMHcyfLjH+fzMxMLF26FGVlZejVqxfCw8MxePBgtGnTBrq6uvDz84OxsTGUlJS+ey742vm1k5KSkJOTgwEDBuDQoUO4cuUKTp48iQkTJkBNTe3fLpNHP+q2L4PBYPwLmkJx+dbUHuiFSkz9/+fk5CAwMBCFhYVQUFCAqakpjhw5gqCgIISEhODBgwc4cOAA7OzsoKuri6qqKoiLi4u0Ln7Eugbq1mt5eTmuXr0KFRUVKCsro6ysDJcvX0ZAQAAOHDiA7t2748uXL2jbti0kJSWRnJyMzZs3Y8uWLZyy3BTUbiMA4OnpiVOnTkFZWRmBgYHcMwnfyeXLl3Hr1i1UVFTA2dkZJ06cwMePH+Hr64tOnTpx99cvV5Tk5+cjMDAQXbp0Qe/evaGkpNQkv/uf0lid/Pbbb9i9ezeUlZWxcuVKvHr1Co6Ojpg6dSpatmyJpKQkuLu7Q0ZGhivjw4cP6NOnD4A/3xj71ggVjAMHDsDPzw+mpqbIyMjA/fv3sXv3bqioqODTp0/4+vUrl1N89erVUFNTg7m5ucjl+7tRVFSENWvW4NGjR/jpp58gJSWFkydPorCwENra2li2bBlmzZoFoKZdfP78GSdPnsSdO3ewf//+//j0679B2N9rj31FRUVIT09Hly5d0LZtW5HL8L9Qu0/y+XxISEggKSkJP//8M3r16sVtBFZVVcHExAQjR47EunXrcODAAdy4cQPdunVDfHw8VqxYgRkzZnzPR/nL09j4FxISgps3b2Lv3r0AaqyaSktLISMjg8LCQmzcuBEtWrTA1q1bUVVVVcfarSkoKirC69evucjzX79+xerVqxEXFwcFBQWUlpbiypUrAABzc3MMHDgQixYtgpyc3L+1jvnxVjkMBoPxL/gR00UITYmEk9SBAwewYcMGxMTE1FnoAED79u0xf/58TJs2Debm5pCUlEReXh709PQgISEBVVVVaGho4MSJE6ioqECzZs1EVhc/Yl0DvysUtZWXjx8/4vjx43j8+DHKysrQokULaGhooFevXti/fz8AIDAwEIsXL8bq1asxffp0qKqqolu3bty7EWU6r/om9u/evQMAmJmZwdPTEwKBAPHx8ZwSLHy2qVOnYunSpejcuTOOHz+Odu3a4eLFi1ykXuF9olKa65uZXrp0Cfr6+khMTMS5c+fg7OyMiIiIOs/4V0FYJxcvXkRQUBDS09OhpqYGY2NjPH/+HBERERgwYAB27tyJvLw8hIWF4fr168jKygLw+8JTqDSLOq5A/brm8XgoKChAdHQ0Tp8+DTc3N/j4+GDGjBk4evQoCgsLISMjg48fP+LLly/Ys2cPnjx5gv79+/9puYyGlJSUYPny5ejYsSOio6NhbW2NxMREnD17Fm3atIGjoyN8fHy4iM7V1dXo3r073N3dERkZ2SRKMwAuBZqQAwcOQFdXFz4+PjAyMuLSMf1VEfbJ3bt3cybvrVu3hoGBAVJSUrixpFmzZnB0dMSlS5fw+fNnmJubY9GiRRg7diwePnzIlOY/of5c8/r1ay6WSl5eHl68eAEvLy8sWbIEGhoaWLhwIezs7CAnJwdzc3M8e/YMKioq3LtoSnx8fJCdnc39fe3aNUhLSyMsLAxnz57F5cuXufHM1dUVYWFh3Fxav280BjPVZjAYfzuEg31oaChiYmIwdOhQqKiocGma/krUNzmrrq7GkiVLkJOTgzZt2mDXrl2YNWsWpkyZUkfRy8rKwrp16yAlJYX379+jS5cu2LlzJwBATk4O48aNQ0REBMLCwv7Qp/JbULuu4+LiMHToUAwcOBDy8vIi+83/hadPn+LXX3/FwoULAQBRUVEoLi6GiooK+vbtC3Nzczx8+BDKysqc0qyjo4N9+/YhJSUF+vr6kJGR4dIHCRUiIaLcQBC+++DgYBw8eBDV1dVQVVWFm5sbxo0bh5iYGBw6dAi+vr4QFxevo4QqKSnBy8sLFRUVkJKSAtD4acK3RNi2hb9RVFSE5s2bIzQ0FB4eHjAwMEBJSQnOnj2LVatWITIykpPtr8LLly+xfPlytGvXDlJSUggMDISOjg5mz56N2NhYXLt2DSNHjsTQoUOhoKCA48ePIz4+Hh8+fEDfvn0b1K+o6js5ORkdO3bkTrmTk5PRvXt3SElJ4e3btygoKIC4uDh3orJ06VJoa2sjJSUFrVq1QkhICE6dOoXWrVvjyJEjDdr1j2bK+z14//498vLysGvXLkhJSSE7Oxu9e/fGwYMHYWJiAhsbGwQGBmLTpk3Ytm1bHRNWMTExkffHkpISbNmyBS1btsT69evB4/Hw/PlzREdH49SpU1BSUsKlS5fg4+ODvLw8WFhYiEyW/4Xy8nIsXboUeXl5nCtBbm4uNDQ08Pz5cwQHB2PUqFGQlJSEQCBAeXk5li9fjnPnzsHIyIgrR3hazWiIcK65ffs2du7cibZt26KqqgqzZs3CyJEjUVlZiaSkJEhKSuLIkSPIysrC9u3bERMTg9GjR2Pfvn2QlJRsMkuslJQUnDx5ErNnz8b69euRlpaGGzduYNKkSYiOjkZFRQUePXqEpKQk3L17F+3atYOhoSGmTJmCIUOGwNPTE6WlpVi9ejXGjRv35z/2H8XgZjAYjB+Ar1+/0qpVq0hHR4c8PT3JwcGBDAwMKCMj43uLVofa6RwSExNp6dKl5OvrSxs3biQioqKiItqxYwc5ODhQbm5une8IBAJKTk6m4OBgevDgAVeOMG1MeXk5ffr0SSRy105NI0x3pKWlRe7u7jR//nx6+PChSH73WxAaGsqlzFi7di2NHz+ebGxsaOLEiXT+/Hmqqqoia2tr2rVrF5cC5s6dOzR06FCaPn16g/L4fL5I0wfVT4USEhJCEydOpLi4OEpOTqaUlBTuswcPHpCpqWmdVFh/VGZTpjzKyMggS0tL2rt3L0VERNC4ceMoMzOzThoYbW1tOn78OCff96CxlEve3t60YcMG7u+jR4+SoqIiZWdn0507d8jGxobOnz/PfS4QCOjVq1dNIS5HYmIiDR48mO7fv0/v37+nGTNm0KhRo8jW1pbCwsIoIyODhg8fTrGxsUT0exotc3Nz8vb25spJT0/n/t+U7eNH4s/a5qtXrygyMpJSU1Np5syZNG/ePIqOjiZDQ0PasWMHEdWMJYaGhlRWVtZUInNUVFTQ6dOnydjYmN6+fUtERDt37iQbGxsiqmlHlpaWZGhoSM+ePWty+eojEAga7ZMvX76kGTNmUEFBQYPPgoKCyNLSkjZt2kSvXr0iZ2dnCg8P5+Yk4ftjKabq0lhayoSEBDI1NaWIiAjKy8ujCxcu0IgRI7h0erm5uVReXk5ERGFhYTR//vwG74TP5zdJXX/8+JGMjY3p4MGDVF1dTV5eXjR9+nRKTU2lFy9ekKmpKVlaWpKxsTGdPHmSFi9eTAYGBlReXk75+fl08eJFioiI+Ld+i221MBiMH5rGdo0/fPiA5ORkhIaGcjunSkpKOHv2LBYtWtTkPjd/hJiYGPh8Pm7fvo09e/ZgyJAhCAgIgISEBIqKiiAjIwN1dXWkpKTg4sWLsLe3r3Oi2adPnzonQ7VPLKSkpNClS5dvKq+wrsXFxVFZWQkiQlpaGjIyMhAWFgag5lSjVatW3/R3vyUTJkyAjo4OXFxcMHr0aISHhwMAzpw5g4sXL+Knn37C3LlzcfDgQUhKSsLS0hL379+Hk5NTg6BfogyqI3yX9U16v3z5gl69eqFPnz748uULbty4gebNm0NVVRWjR4/GgwcPsGfPHty8eRP29vYYOHBgg7JFaSpc30dM6As3evRoODk5oaSkBHl5eUhLS4O8vDzKy8shKSmJ/v3749OnT5x8TQn9vyWHuLg4KioqkJGRgS5dukBcXBwPHz7E+vXrUVFRgXXr1uHRo0dYunQpmjVrBi0tLcTFxSEwMBCampqQl5cHj8fjzJypnivAt2bv3r3o0aMHjI2NoaWlhQsXLqBNmzaYOHEi9PX1sWvXLpw9exa//PILZzXh5uYGJSUlzpRRT0+PK69bt24ARG+J8CPyR/2x9jvu378/+vfvj61bt6Jv377YvHkz+Hw++Hw+jh07hmHDhkFXV5dLDdcU1J4fJSUlMWbMGMTFxeHgwYPYtWsXOnToAFlZWSxZsgSPHj3C7NmzYW9vjx07dgBAo+NHU1C7T5aXl+Pdu3fo3Lkzd/L59u3bOu4x+fn5uHXrFsaNGweBQAA/Pz/cv38fkydPxvjx47lyhe+KBcisgepZvNUmPj4e0tLS0NTUBBFh+vTpeP36Nc6fP49Bgwbh8uXLePz4MeTk5BAdHY1ly5ahTZs2dcpoinFEIBCga9eumDZtGu7cuQN1dXUsWrQIzs7OCAwMhKOjIy5fvowPHz6gZ8+eAAA1NTXs2bMHfD4fsrKyMDMz48r7V+P2j+WUxmAwGPUQLgoeP37MLbxTU1PRvn17VFVV4cCBAxgzZgymTp2KKVOmfFfTrPp+gnl5eVi/fj1OnTqFlStXwtPTE4sXL0afPn0QEhICABgxYgQGDx6Mp0+f4uXLlwAaDuxETZNeSlh3YWFh0NHRwdOnTxEbG4vmzZujqKgIANCqVSuUlJTA399fpP6+/wvLli1DTk5OHT8+fX19DB48GJcvX4ampiZGjRqF+/fvY+LEiSgtLcXMmTOho6MD4HdzbFGaZQvfpZ+fHzw9PXHx4kWUlZWhY8eOSEtLw/z58zF37lw8ffoUr1+/hpOTE3Jzc2Fvbw8tLS00a9asgbmtKBG+69p1IhAIICUlhfT0dPTq1QtAzeLdzMyMS7PTvHlziImJIT8//1+byIkIYV86fPgwtLW14ebmBhsbG1RUVEBSUhLr16/HhAkTUFFRgZs3b0JTUxM2NjYoKSnBuHHjYGpqCjk5uT8sV1R8/PgR/v7+yM/Px9KlS5GQkICoqChMmjQJnTt3hqWlJZo3b46TJ09iw4YNaNasGVxdXbFy5UoYGhpCSUmp0ajCTeXr/iPRWH8sLi5u8I7z8vIQHR3Npfg6dOgQtLS04OnpWSea+r/jS/m/IJwThGP269evAQC9evWCkZERXr9+jdjYWPTv3x+xsbHIyclBZGQknJ2dUVBQgMjISLRu3VqkMjbGnj17cPXqVa5ejx07Bh0dHWzfvh0mJiYICgqCQCDAyJEj4e/vD6BmzJGRkcGePXuQkZEBU1NT+Pr6Ijg4GIsXLwYg2tgJf9V57l8h3LAXcvLkSWzZsgWxsbEAatzBOnbsWKetLly4EDExMRATE4OhoSFMTU0xYMAAREZGYvbs2U3+DMDv4+zMmTPRokULbjPZxMQEMTExePDgASoqKuDr64szZ85g9+7dmD9/PjQ0NNCyZUuubQj//VfjNjtxZjAYPzRPnz7FmjVruLyqTk5OkJOTw/Pnz2FkZAQZGRns27cPampqmDt3LtTV1eHo6NjkctY+nczLy4OcnBzk5OQwaNAghIWFcXl2x48fjxcvXiA6OhoaGhpQUFCAuro6YmJiEB8fD2Vl5QYDu6gW6PUnktLSUtja2kJSUhKbNm3C2LFjUVBQgIyMDJSUlHD+lWlpaQgJCYGKisp3ObH4VxNg7969MXPmzDoBqTp27IgWLVogMzMTQM0CwdzcHFVVVdxJnKhPEGuTkZGBJUuWcIvEEydO4NatW1izZg1OnDiB169fo3PnzujduzdevHiBr1+/orS0FN27d8eKFSuaNFhb7ROL8PBwXL9+HWpqahgxYgR0dHRgYGCAixcvYvbs2ZCUlMTSpUthYWGBGTNmQFVVFVFRUejevTtUVFSaTN767/HevXu4c+cOTp06BR6Ph8+fP6NFixawtbWFo6MjfH19oampCaBmca+oqAhZWVmMHDmSi0bdFHLXrmt3d3dMmzYNFy9exIIFCzBnzhxcvHgRhYWF6NChA9TU1KCuro579+5h/PjxOHDgAJ4+fYqUlBRYW1s3CAAmSn7ElEe1qd8fT58+jatXr8LT05OzRCEiyMnJoV+/fvD394e3tzc6dOiAHTt2oG/fvtw9PB5P5Bu4wvb96NEjrF+/HpKSkujQoQOMjY0xZcoUREdH4+DBgzh27Bh0dXWRkZGBCxcuoH379vj5558xZswYbtxrCqqqqrBv3z5cvnwZ169fBwBER0fj9u3bOHbsGHr37g0/Pz8EBARg8uTJGDx4MG7dugUlJSVoaWkhIiICysrK3EZQ+/btAfyezkuUwfh+tHadkpKCPn36gIhw8eJFqKio4Pz588jNzYWcnBy2bNkCJycnTJo0CTNnzsS8efO4eTw7Oxv9+/dHdXU1+vbty7VrQPR1TX+QOk2YQaJZs2aYMWMGjhw5gnv37sHExAR3797F/fv3oa6ujoEDB+L58+fIzc3F8ePHG6xN/l25WToqBoPxw1DfDLS0tBRubm4YMWIELC0t8ebNG8jIyKBTp06YMWMG2rZti71790JSUhJxcXHw9PTE7t27m3RBUJv09HRs3LgRfD6fC06hra2NFStWoGXLlnBycoK8vDwePHiACxcuoG/fvnBxcQFQc7rUlHLXXhBUVlZCQkICYmJiWL9+Pa5du4bLly+jd+/eAAADAwOMHz8elpaW6NatG+7evYvz58/jyJEjTSZvY3LXDoRVn+LiYujq6mLZsmWYPn06AGD//v3Iy8vjTkOFNHaa+i1pLNfltWvXEBwcDF9fXwA1J1SmpqbQ0dGBra0tXrx4gaKiIgwYMACrVq2CrKwsdu/eDUlJyTpyN5UCXVJSgtOnT+PKlSvQ09PDmzdv8OzZM5w6dQpEBHt7eyxbtgzGxsYAahSR+Ph4vHjxAgMHDhRpALva/FFAIC8vL6Snp2P//v1cvZWXlyM7OxsnT57E06dPoaGhgbdv3yItLQ07duzAsGHDGk3tIwpqt2vhSWerVq1w5swZ+Pv745dffkHPnj1hYmKCSZMmwcrKCq1atcL79++xf/9+lJSU4MCBAw0sAgDRWk7UboOpqanYtWsXevfujZEjR2LEiBEi+93/hcaUoT/qj7q6urCzs0OLFi2471VUVCAlJQWFhYXcpsq/e5r1rcjLy0NUVBRCQkJgaGgIdXV1XLt2Dbt378b58+fB5/OxZcsWzJkzB6amprh48SJiY2Px+fNnTJs2DSYmJk0iJ/B7G9m4cSMqKiogEAhgZmaGBw8eICkpiatzAFi/fj34fD6srKwQEhKCGzduoGPHjvj8+TNWr14NAwODJpG5dn9PTU1FYGAg1NTUMGbMmDpj8F+N1NRUmJqaYvPmzZg0aRJcXFyQkZEBRUVFbNmyBRUVFThw4ABevnyJAwcOYP369fj8+TPGjRsHbW1tuLu7o0uXLvDy8qrTR5pyU/nTp0/IyMhA7969IScn16C/rlq1CpWVlVi3bh3evXuHzZs3Y9asWQ3S6/3X49//4IvNYDAYIkUYVKJ+gBDh9cTERFJRUakTBKuqqopycnLo/v37ZGVlRebm5rR06VIaMmQIHTp0qMlkrx9cJyUlhSZNmkTe3t70+vVr2rhxI2lra9Pjx48pNjaWrK2t6dy5c9z9Xl5etGjRogYBzZoyaA+fz6fNmzeTvb09OTk5EVFNQJDhw4fXkTU2NpYsLCxo6tSptGzZMhoyZAgFBAQQkeiDsDQWPIbP59OGDRvI3t6eVq1aRS9fvuTqrXb9nT9/npSVlWnNmjW0fv16Gjx4MEVHR4tUXiHV1dWUnJzM/V1VVcUF7CEi2rhxI61YsYKqqqq4YE5BQUGkpaVFubm55OPjQ9OnTycdHR3y8vJqEpn/iLdv35KTkxMZGhrSy5cvuevz588nV1dXysvLo71795Kuri73WWPtuLF3KQr4fD75+flRSEgIFxBr8+bNtGXLFiopKeHue/LkCS1ZsoSIiO7evUv79+/nAph9D/h8Prm7u5ORkRHZ2tpSamoqEdUE+RIGMAsKCiJ9fX2Kj4/nvnf9+nUuYFzt4IKipP77TU1NpcmTJ9OqVavI0dGRTE1NKS0tTaQy/KfUlrm6upqys7O5v/+oP44fP/5fBvoSdbturPwHDx7Q6NGjSVtbu458rq6uZGVlRUREhw4dImNjY+45Kysr/2W5osTR0ZEUFRXJ0dGRiIg8PDxo8+bNdfpkTEwMDRkyhPLz84mIKDMzk54+fdpkMtavkydPntDgwYPJ3t6e1NXV6fTp0/T169cmk+ffpXbwr3Xr1pGNjQ0VFRVRamoqjRgxglxdXbl28uTJE7K3tydfX1+qrKykQ4cOka2tLZmamjb5XFO/vvfv30/9+/cnExMTsrCw4IKTVVdXc/335cuXZGFhQadOnSKimrE9NDT0T8v9T2A+zgwG4y+L0AdVuJt49uxZLFiwAPv370d2djbk5eWhoqKC58+fA6g5KZCQkICLiwtkZWXh4+MDe3t7qKmp4fbt27C3txe5zAKBoNFgGzExMejWrRuWLFmCn376CWvXroWRkRHWr1+PoUOH4qeffkJsbCxevXoFAJgzZw62b9/O5doV0lQniBkZGZgyZQoyMzMxd+5c9OrVC1+/foWcnBwcHBzg4+ODyspKAMDQoUPh5eWFhQsXYsCAAbh16xa3uyvqXejw8HC8ffsWQM0pUElJCebPn4/Pnz9j4cKFkJCQwP79+3H69OkG8piZmaFfv35ISEjAoEGDEBIS0mRmt9u2bcPatWuRk5ODwMBAjBs3Do6Ojli/fj0KCgrQvn17pKeno7y8nDuNHjJkCPLz81FUVITFixfj8OHDCAoKgpubGwDR+5DSHxio9e3bF7169UJubi4KCgq46+7u7ggNDUVeXh5mzZqF0tJSbN++HUBDX2hANL619evkwYMH0NLSQmRkJM6ePYvVq1cjLCwMmpqaePz4MZKSkrh7U1NTueBOOjo6WLRoEaytrQE0nY+q8N93795h6tSpyMrKwrZt2/Dp0yfs2rUL1dXVcHV1RWRkJB4+fAhjY2N07twZZ86cQU5ODgBg0qRJMDQ0BPB7vYu6Xwp/58yZM3B2dsaFCxcwdepUeHp6wtnZGT169MDBgwdFKsN/ilDms2fPYvLkyVi2bBlWr17NWfs01h+LioqQmpr6p+WK2pRXWH5YWBgiIyPx5csXjB49GjNnzkTnzp252B8AsHjxYiQmJiIzMxPa2tpo1aoVoqOjAYB7LmGfaSoT5OTkZFhaWqJnz57Q1tZGp06dUFlZCVVVVTx9+hSJiYncvTweD/369eP6n7y8PIYPHw5AtH2y/hh1+fJlHDx4EE+ePMHhw4dx6NAhWFlZ4f79+0hISBCZHP8pwvVI7YCQixcv5k7Je/TogWnTpiEnJ4dzUxo2bBg0NDQQHR2NZ8+ewd7eHj4+PvDz82vyuUZY30+fPkVSUhI+ffqE8PBweHl5gc/nY9u2bdx3hP13wIABUFZWxu3bt/HlyxesXbsWEyZMqFP+/9K2meLMYDD+ciQmJmLMmDE4fvw4gBoTZysrK5w/fx6qqqq4ffs2vLy80LZtW3Tv3h1RUVHIz8/nBkMJCQkUFxejbdu20NLSwpw5c9C+fXtUV1eLNEgIAM7H58mTJ1izZg3Onj2L1NRUFBYWcgtZoYJvbGwMHo+HV69ewcDAAJ8+feIW7l26dEGrVq1EGniEz+dzmxP16+XZs2cYMGAA9u3bh5EjR8LQ0BCPHj3C58+fMXv2bHTq1Am7du0CUDM59+jRAxMmTICNjQ06dOjQJHWdkpKCoKAgPHz4EK9fv0ZlZSVSU1NRWVmJw4cPY9CgQdDW1sbTp0+RnJyMyspKzqwWqHlX27Ztg6enJ6ZOnQp5eXmRLQiICBkZGVi7di2+fPkCOzs75OXl4dy5c7h37x6OHTsGV1dXpKen49SpU5g/fz6Sk5Nx9epVrt2Eh4dj7NixUFBQAAC0adMGrVq14upaVAtdIkJ1dXWjypawviZNmoQhQ4bgzZs33GKtR48e6N69O5KSkiAnJ4etW7di8uTJDcoQ1WZQbX+4srIyAEBISAgsLCxw+PBhnDlzBn369IGTkxMXT+Dw4cPw8/NDXFwcTpw4AVVV1Tqm3fT/fsai8lGtX9fCf+/cuYO+ffvi0KFDGDhwIAwMDPDrr78iJCQEGhoaGDJkCPz9/VFZWQlHR0eUl5dzbgrfI4JwZmYmduzYgRs3bkBeXh5nz57lNjh79eoFXV1dPH/+HI8fP25y2YQ0Nj6dOXMG58+fh5eXF2xsbFBZWYn58+fD2Ni40f44evRoKCoqNrXodXjz5g2MjY2xZ88e+Pr6YtGiRbh8+TKsrKxQXV2N3377DeXl5QBqovL369cP0tLSUFBQwN69ezFlypQ65YliHBHWdWPj67t37yAtLY3ly5fD3t4eCQkJuHHjBoyNjdGzZ08cPHgQR44cwaNHj7BmzRoMGTKE82GujSj9xoVjVFFREZydnXHw4EFER0dj//793GbhvHnzUF1djfv373Nt5Hsi3MDn8Xj49ddfcfr0aSQmJqJ9+/awsLBAYGAgUlNTsWjRImRlZeHu3bv4+vUrxMTEoKGhgRYtWiAmJgYAIC0tjZYtW3Jju6g3VYRjVlpaGqZOnYpFixbByckJHz58gLy8PHr27AkHBwfcvHkTKSkpXB504XrJyckJmzdv5jIdfNO1yH99Vs1gMBgioKSkhJYsWUK7du3izIsCAwNp0aJF3D179+4lZWVlevnyJcXHx9P8+fNp+vTpFB0dTXZ2dmRubk6FhYV1ym2KXILCvJO7d++mESNGkJeXFzk7O5Ofnx89ePCAtLW1uVyqRDUmRXp6evT+/XsiIkpKShK5jEIqKirI29u7Qc5loQnTkydPOLO5KVOmkJqaGpmYmJCpqSm9f/+eIiIiSFFRkT5+/Nig7KbMkenl5UXDhw8nLS0tys7Opnv37pGenh5lZ2eTvb09jRo1igIDAyk1NZV+++23PyynsTyW35qysjIaOXIk7d+/n4iIfH19SVlZmdzc3Lh7AgICaPbs2fTmzRsKDQ0lMzMzmjFjBtnZ2ZGGhgbdvXtXpDL+GZmZmeTn50cPHz7k8orXxtfXl5YsWcKZBb98+ZJmzpzJ5cMW0pTtIz8/nxYuXEheXl6UlpZGRkZGlJ2dTRkZGbRgwQLS0tIif39/qqiooMLCQjp8+DAtWLCATE1Nyc/Pr8nkrI+wrqOioig3N5fOnz9PV65coS9fvpCDgwOtX7+enJycyNTUlIqKiig9PZ2UlZXp+vXrTS5rfbPH+Ph4cnZ2pilTplBxcTEREZ05c4bMzc0pISGBiIg+ffpEHh4eNHfu3KYWl4gamiUTEZWXl5OjoyOdPHmyznVhG4mIiKBp06Z91/7YmIuDl5cXZ0JbWFhIZ8+eJT09PUpJSaHz58/TjBkzaPv27fT+/XuytramVatWNShHVH1SIBBQfn4+rVu3jruWk5ND4eHh3N+JiYk0atQoysvLIyKiDRs20MKFCyk9PZ3y8vLoxIkT5ODgQNOnT2/SPlm/XZ85c4YsLCzI3d2diGrWKqampnT06FHOdPzatWtkaWnJjYHfG+H4p66uTtbW1jRixAhKTEwkIqKpU6fStm3biIjI39+fjI2Nuf5JRE3uSlG7TVZWVtKFCxfI3t6ezpw5Q0REBw4cIA0NDc61ID8/n1xdXcnW1vYPyxRFu2aKM4PB+MsgHOTWrFlDo0ePJh0dHQoLC6O1a9fSjRs3iIho27Zt5OzsTPb29jR16lQiqllkuru7k7OzM23evLlJ/YDrk5OTQ+bm5py/qtBvKCsri7Zs2UIGBgb06tUrKi8vpy1btpCjoyO3uBTSVIqFs7MzLV68mPbu3UsRERENfj8sLIwuXLhA586do4qKCsrLy6MxY8bQkydPiIjo8uXLTSKnkNoLmerqaiorKyM3NzfS1NTkFgBv376lyZMnk6KiIu3atYuqqqqIiMjIyIjzeWpKxU0os7BNnj9/nvT19enVq1ckEAjI1NSUPDw8uDbw4cMHWrNmDbe4zcjIoBs3blBAQECTtuv6i6aTJ0+SqqoqLV68mLS0tCgkJKTBs2VmZpKzszOpqamRm5sbjRo1inbv3k1ETVvnQqKjo8nc3JxWr15NRUVFRESkqalJM2fOpJEjR9LatWuJz+fTgwcPyNzcnHsH9ftjU9W7sI6Ede3i4kJaWlp069YtKisro8zMTDI1NaWNGzcSUU1sAUVFRXJ3d6eKigp68uRJnXoWtY9q/XrJzMwkopoxb/fu3aSurs4t0pOSkmj58uW0YsUK7v6wsDAyNTWl58+fi1TO2hw6dIji4uK4v0+fPk0XL17k/GjHjRvHzTVCX2Y/Pz+aPHkyCQQCSktLo+Dg4Cbvj3/0W8KYE5GRkdy1rKwsWr16Nbm7uxOfz6cFCxaQjo4OzZ07l7Zt29akvsvC3xo/fjx5eXnRvXv3yN/fnwYMGEBHjhzh+pqFhQWdPXuWiIjev39Ps2fPJl9fX+691PcbFmXd1y9b+AwfP34kTU1Nsra25saT06dP0/Tp0+nRo0fc/ba2tuTm5sZtBDQV9d8rn8+nHTt20IYNG7hnMjQ0pDVr1hARUXh4OGlpaVFMTAwREenr65OHh0cDn31Rt/PG2mNeXh7t3r2bFBUV6cGDB0RUMyc5ODjQsmXLiKhmvExISCBFRcUGPsyihCnODAbjLwOfz6eKigoaOnQo9e/fnzuZy83NpdzcXDI3NydbW1vKzMyk4OBgUlRUpIMHD3Lfr32K0NSBTYQkJCSQkZERd7pZXV1NAoGATp48Se/evSMXFxeysLAgXV1dsrKyahD8S5Tw+XwSCAScMunv709KSkqkr69PX7584e4TLr4FAgG9ffuWcnJyiKgmGI65uTm9e/euyWSuLxNRjUKUkJDAWRXcuHGD7Ozs6M6dO1RVVUU7d+4kMzMzbmf60qVLNHny5DrBuJpC3toLDqFSQUQ0a9YsWrt2LRERBQcHk56eHrd4ISK6efPmH56uCN+dqMjMzCQDAwOaNWsWd1JcVFRElpaW3AJGGJSqNsL3ExISQlZWVrRp0ybuFOZ7IBAI6MqVK6SmpkZHjx7lrgcEBJCioiJ9+PCBu+bi4kJ79+5tUEZTK55ENUp7/boW1m1CQgLNmDGDu3fXrl1kY2NDvr6+dRSLpt6kuHz5Mk2aNInmzJlDK1eupGfPnlFmZibZ29uTp6cnd9+NGzdo1qxZ3Kl4cXFxg00KUePo6EhGRkb06dMnMjAwIDMzMxo3bhwtWbKEsrOzaceOHWRkZFTnO97e3rR06dJG35eo+2N9rly5Qs7OzrR37156/vw5CQQCsrW15dqvsM16eHiQs7MzEdWMMbNmzaKgoKAmkzsnJ4fs7e0pJCSEiGpOkZWUlMjU1JSIajYQ582bR0uWLKGSkhLavHkznT17lpN/z549NGPGDG7jRUhTzuuBgYE0d+5cWrVqFV29epWIiE6dOkVTp06tYx1mbW1NW7du5YKUpqSkcHNmU1G7bQqtgYqLi0lbW5vbKAoMDCRra2saOXIk3b59m4iInJycyMbGhgQCAb1586bOOkDU1LfyCgoKIg8PD7p16xbl5+dTZmYmmZmZ0ZYtW4iops2GhYWRjo4Ot3lfUlJC165da2DVJEqYjzODwfjLwOPxICkpiTt37sDd3R3+/v748uUL5OTk8Pr1a3Ts2BFHjx6FvLw8Pn78iNGjR+PevXv4+vUrgJrgJvT/Poii8tP6Vz6wLVu2hJiYGLKysri8huXl5QgPD8erV6/g7e2N/fv348CBAzh16hQ6deokUj9moMYvC6jxXRPmPAQASUlJ6OrqonPnznV8gGr7VwYHB2PatGmYPHky9u3bB2dnZ/Tq1Uuk8jYGj8fDu3fvMG3aNHh4eGDr1q2YN28eIiIioKurC1lZWYSFhaGwsBD29vbo27cvbGxsYGZmhiNHjsDNzQ19+vRpUnnFxMTw7NkzTJ8+Hc7OzlxgrKVLl+LBgweIjIyEoaEhunXrhsuXL3N+cSNGjIChoSEX9IZqBYoSdf5XoMYP8uXLl1zQoLi4OHz69AmDBg0Cn89Hjx49wOfzERsbi4yMjDoyamlpQUFBAV+/fkVFRQWAmnRbouCP+iP9fzAcTU1NTJo0CTdu3OA+MzU1hbq6OtauXQtPT09MmzYN7969azSNjSj9+IRjAwAuyB5QE0Swfl1XV1cjISGB8xEWpvV69OgR1q5diwULFqBly5ZcGaLMo0r1fAWvXLmCI0eOYMOGDVi4cCGaNWvGpWcaO3YsEhMT8ejRIwCAmpoaevTogbi4OABAq1atRB7Hgf4/96uQTZs2ITs7G+7u7lzeax8fH3z+/Bm3b9/G1KlTUVhYiE2bNiExMRFv3rxBeHg4hg0b1sAXv6n6I1DTXjw8PHDgwAGMHDkSX758wZYtWxAQEIDhw4cjJiYG79+/59osn8/nxmlDQ0N07twZMTExSE5OBiA6f2Bh+5CTk8O2bdswatQoVFZWok2bNujduze6dOkCAJg6dSo8PDwQFxeHM2fOIDExEenp6Zz8dnZ28PDwgJKSUp3ymyJgGRFh+/bt8PX1xcyZMyErKws/Pz9s374dVlZWEAgECAkJQWFhIQBg1qxZuHnzJhckrnfv3mjXrp3I5/XaCOcaMzMzLF68mIulcfDgQXTo0AEzZ85EQEAAvL29oaamhlOnTnG5yQEgPz8f/fr1Q8eOHUUut7BPCgOWFRQUYO3atThw4AAEAgEOHjwIBwcHyMrKwtzcHI8fP8aLFy8gISEBVVVVqKurY+PGjQBqfK8nT56MDh06iDymSu0HYDAYjCblz3a7a+9Aampq0vbt24moJqWKsrIyBQQEkKOjI2lpaTX5CaKQ4uJiSk5OptLS0gafEdWYmjs5OXH+w2lpaWRhYcGlvqmNqHfQg4ODydbWlvh8PlVWVpKrqys5OjrSsWPHuHv09fVp//793KlVfXPP5OTkBr7Q34Nt27Zx7SE/P5+8vb3J0tKSXr9+TU+fPiVra2sKDAzk7i8oKKhzkitq6reDc+fO0bhx4+jw4cMUEhJCKioqdOXKFSIicnNzI1tbWyorK6NXr17RmDFj6Nq1a01+glUb4amFp6cn6erqkqmpKWVmZtKXL1+of//+9OLFCyKq6b8fP34kGxsbevz4cYPvR0dHk62tLWemLWrKysoaxDQQ8uTJE5oxY0ad0/vi4mK6desW7d69m86fP98kMjZGQUEBLV++nJYuXUru7u70+fNnys7OJmVlZc58uaqqijNRjIiIoEePHtHGjRs5vz8hTWlOWbudL126tMFpvbGxMXl6elJxcTGtWrWKli9fzn2nqS1shFRWVnKm16dPn25g3vnzzz+Tvb09vXnzhp49e0ZWVlY0a9Ys0tbWpsOHDzeZzESNz4/p6ek0bdo0zoqjoqKChg0bRqtWraLo6GhatWoVjRkzhvz8/MjZ2Zk0NTUpISGBaxe//vor6erq0qVLl0RmkVC7vouLi+nXX38lJycnWrp0KRERZWdnU//+/eu4BcXExJC3tzeNHj2aTE1NOQuEprKaaKyuMzIyaPr06XXWFzdu3CALCwuKiYmh8PBwMjAwqJMGq7ap9vfg+fPnpKurS6dOnaKEhATat28f11bOnj3LWR8Q1fTZkSNHkqenZ5O7tdVuIxUVFRQSEkJz584lS0tLrn8WFxfT6NGj6cCBA5Sbm0tLly4lV1dX7nsJCQmcJcP3cAFiJ84MBqPJEe52P3z4EGlpaXU+4/F43CnShg0bcPr0aaSkpMDIyAgLFixAVFQU2rRpg5s3b3IniKJODSOUCwAOHDgAAwMDbN68GdOnT8fjx4+5HVqhHK6urpCRkcG6deuwfPlymJqaYtCgQejSpUuDXVFR7aAL61BCQgI8Hg87d+7EsmXLICkpCWVlZezYsQOBgYEAACsrKwQHB3NpnYTPKoxK3KdPH4waNarOMzYVN2/eBADk5uYiMDCQk0NWVhYmJibo3Lkzrl27huHDh0NJSQknTpyAmpoanjx5gjZt2mDYsGEARJs+Q1h27ZO+kpISXL58Gbt374adnR0nX2BgIHJycrBkyRKkpqYiKCgI/fv3h6amJr5+/doggrMoOX78OE6ePImsrCzu98rKyiAtLY1ly5ahtLQUp06d4ura3d0dQE2b6tSpEwoLC+tEtxWeyI0cORKKioro0aPHN3+G+u9x9+7dMDIywqJFi/Dy5UvuurBPDhgwAFpaWrhy5QoX/bZly5bQ19eHq6srZsyY0Wi535ozZ84gMjKSk+3du3eYMWMGxMTEYGZmhsTERGzatAm//fYb5syZg3Xr1gGoqesuXbrgy5cv6NatGzQ0NLBhwwZYWlrWkVvUaeqE49ShQ4ewbt06LpXU27dvISsrC+D3k3M7OzuEhoZCSkoKw4cPR2ZmJl68eAEAXHo9Udd3bZn37duH+fPnw9PTE2VlZZgzZw4UFBQQHx/P3Tt37lyUlZUhJCQEioqKOHXqFDw9PRESEgI7OzsAaJKTOOD3+TE5OZm7VlVVhYqKChQXF+PAgQPQ1NSEgYEBFi1ahN69e2PLli2YPXs23r9/D3l5edy6dQuqqqpcuxgyZAjc3NwwZcoUkVkkCOv73r17MDAwwKtXr6Cnp4d3797hyZMnaN++PWbMmMFZ3gA1qY8WLlyI8ePHg8fjoaSkBEDTRYIX1nVAQADOnj2Lly9foqqqCp8/f+ZkAQANDQ106tQJcXFxGD9+PFq0aIGLFy9y92hoaAAQ/Zj9R/0mPj4eioqKsLKygqqqKhwdHSErK4uMjAx8+fIFOTk5KCgowL59+1BcXAx3d3csW7aMax9N0R+B39vIhQsXMHPmTMjLy0NcXBzZ2dmcxVWrVq2wZs0anDx5ErKyspgyZQp+/fVXbi2gqqrKWQh9j4wB7MSZwWCInPq7gg8fPiQtLS0yNzcnPT092rhxIz179qzOvcJ/58yZQ3PmzOG+K9yVJBL9aW19WSIjI2nmzJn05s0bIiI6duwYjR8/vtHAWsXFxRQXF0cXLlxo0pNxoU+1kMrKStq3bx8ZGhpyfrVERCdOnCB1dXWuDmfMmEGbNm2iy5cvc/5DTU393e8vX76QoqIiXbt2jQQCAc2bN49Onz5d597ly5fT8uXLiajmBO/+/fuUkpLStIL/PxEREXTp0iVKTEykzMxM2rhxI2VnZ9P169fJ0tKSDh48SKNHj6Z9+/YREZGPjw9paWlRbm5uk++ch4aGkqKiIg0bNowWLlxY5+Rl8+bN5OPjQwkJCaSurk4JCQmUm5tL2traNGfOHNqxYwdpaWmRm5vbHwaSaSxq8f9CfX+47Oxs+vXXX8nR0ZGLrG9paVnHd1lIYmIizZw5k9avX99ouU2Bu7s7jRkzhvv7xIkTdQJlff78mbZs2UILFy6kjx8/kp6e3p/WdVO3l5ycHLKwsKCZM2fSsWPHaP78+fTmzRvau3cvaWtr17n36NGjtGDBAiKqsQxpLAK7KKjfRpKTk8nMzIxmzpxJAQEBpKmpyZ1+hoaG0oABA+pYAZ04caJBZGGipo+XIQzaZGRkRNbW1hQcHMxZLamrq9Ps2bM53197e3su2BNRQ2uh+te+NfXnGisrK5o5cybdunWLiGpOb93d3Wn+/PlEVDMvDh48mM6dO8f5whPV+Krq6ek1efyMxMRE0tPTIwsLC7K2tqb169dTdHQ0LVy4kA4cOFDnGc3Nzbno5cnJyY1akImS2nUdExND8fHxnG/vpk2buAj1wrH81atX5OfnR6dPn+asEKZNm1ZnfmzqE+ecnByys7MjKysrio6OJqKamB6Wlpac9ZJAIKCkpCSuPeTk5NCFCxea1Frlz2CKM4PBECn1TaEEAgG5uLhwpobZ2dlkY2NDdnZ2XKRKot8n/ZSUFNLX16+z+KofeOlbU3uhlJmZyaVc2rx5M6ekvXnzhubNm0f6+voNApg0RnV1dZOaUxYVFdGOHTvoyZMnlJGRQebm5mRjY8N9XlFRQXp6elzgjV9//ZXmz59Purq6DRaOoqb2gkBoGixEuDAvLy+nzZs30+LFi+tsRGzYsKFRk+CmXOzm5uaSra0taWpq0oIFC8jKyopbAN64cYN0dHS4heSiRYto7NixdPHiRSKqqffaiDItTF5eHq1atYqIatqjs7MzOTg4kLa2Nq1cuZIza3/16hVZW1sTEZGNjQ25uLgQUY3LgTCAS1NGMa3N+/fvydTUlNt0E9ZjUVER6evr04kTJ6i8vJyIfl8UVlRU0K1bt5rUbL8+xcXFNHbsWPL19SWimrFk3rx5RPT7O7916xZZWlpSbm4upaam0tWrV79rXdfm3r17ZGlp2eC6MMjWsmXLKCEhgVJTU2nq1KnccwppSkVfaLZ/5MgRWrJkCXd9z549pKioSPHx8URU07YdHBzqyCh0R/geZGZm0pUrV2j+/Pl0584dSk9P51Ivvn37ljw8PMjGxoZevXpFREQvXrygadOm1UlxKKQp6/vTp0+cC8qJEydo0KBB3N9ERA8ePKCpU6dyffXKlSukrq5OWlpanPJZXFxMs2fPFpnJ8+fPn+tsugvZv38/NyaWl5dzY8cvv/xCDg4OnCtHUlISzZw5kxtD6m+qiwrhXCj8nU+fPtGsWbNIR0eHLC0tycDAgEJCQujFixc0cODAOnPj9evXaebMmVRSUkJlZWV1AprV32BvKrKyssjBwYF0dHTqXHdxcaHNmzdzLipBQUFkb2/f5PL9OzBTbQaDIRKEJr0SEhKorKyEn58fkpOTkZKSgqioKFhaWqKyshL79+9HfHw89PT00KxZM+774uLiqK6uRu/evXHr1i3IyclxnwkDL4kKcXFxVFRU4Pbt29i0aRP27dsHoCb4WPv27eHm5gZLS0uoqanhypUrCAwMxJcvX/6wPCKCmJiYSGSubT4olHvDhg1wd3fHlStXcOHCBXTq1AnTpk1DZWUloqKiANQEBlu9ejXOnDmDDx8+YMiQIdi+fTvu3LkDVVXVby5nY9Q3cc7NzYW7uzsOHDjA3WNraws+nw9/f3/MmjULYmJisLe3x+3bt7Fq1SqEh4dDR0enQdmiNoGvTWxsLPh8PiIiIuDr64vt27dzgXlSUlIwZswY6OvrIycnBzweD2pqapyJ85AhQ+qUJSrTMx6PBxkZGVy5cgUHDhyAmJgYTExMUFlZiTlz5kBSUhL79+/H48ePOVNsAFi1ahWioqIQHByMbt26wdjYGO7u7pgwYcIf1se3hohQVVWFq1evIjg4GJqamnByckJqaioqKytRWVmJ1q1bY+bMmbhy5QoXAElMTAxEBElJSejr63Nm+6KksrISSUlJdQK6ATXmhy4uLjh48CCKiorQq1cvSElJ4bfffuPeeadOnfDq1SuUlpaiR48emDJlSpPXdW2oltlpXl4evn79ygViJCJkZmYiPDwcx48fR05ODnbs2AFra2uMHTsWCxYsqFOWqNp1ffPpx48fw9bWFvn5+cjOzoaenh7KysqwevVqfP78GaNHj4aHhwcAcOOHMHgZj8eDsrKySOT8d3jx4gV27NiBjx8/QkdHB926dYOTkxNGjhzJBUxSVFSEnZ0dnJycMG/ePOjo6GDo0KENympKE9bHjx9j69atePbsGebNm4c+ffrg9evXXFBKZWVlaGtrw9/fHyUlJTAxMcG5c+cQFhaGbt26oaqqCrt370Z6evo3DzwpEAiwadMmGBoacnO0sF1//foVMTEx6Nu3L4CaOaNZs2bIzs5Ghw4dMH78eHh6emLBggWwtLTEuHHjuDGkdgBNUZGdnY1x48YhKiqK+53AwED07dsXd+/exeHDhzFp0iQEBATg48ePmDFjBlxcXBAXF4fS0lLcv38fAwcORIsWLdC8eXMoKioC+D0woahk/7NxqkOHDrC1tUVmZibX74AaV4nHjx9j2bJlWLVqFbZs2QJdXV0AaDCWfm+Y4sxgML4ptRVmoMaPcsyYMdiyZQuaNWsGWVlZDBw4EG5ubtDW1kZOTg7CwsIgISGB48eP14kwW1v5EaVvbf2BPigoCOPGjcPt27eRlJSEX3/9FUlJSRgyZAjOnDmD/Px83Lt3D4sXL8anT58QERHxp5OFqCaou3fvYuvWrcjPzwdQ41c7f/585Ofnw97eHiNHjkRycjKuXbuGiRMnolu3brh69SpXl+PHj8fw4cM5ZVq4OSFqP2aqF/n8xYsX8PT0RLt27TBmzBikp6cjJiYGANCiRQusWrUKPj4+kJWVxbZt26Cjo4P79++Dx+Ph6tWrGDhwoEjlBWoWYLVlrv2+Q0JC0L17dwA1ilOnTp2Qn5+PyMhIlJSU4NmzZ1izZg309fXRr18//Pzzzxg/frzIZa4tu9BffceOHThy5Aiys7Ohra2N9u3bIyUlBcbGxhg/fjw2b96Mly9fIiYmBqmpqejXrx/09PTw+fPnOu1YqLCIYoOiurq6QZT3z58/w83NDaGhoVi4cCHGjBkDBwcHBAQEcD7a8+bNQ/PmzREQEMBFvG1KBSIjIwMGBgbw8PDA+/fvG/y+mZkZevbsiT179kBbWxuSkpI4c+YMt6CPiIiAjo4OOnbsWKdcUdZ1Y4SHhzeQXUxMDB06dMCzZ8+4z6qqqnDo0CGIiYnhxIkT2L59O65fvw5XV9c6couS+puRP/30Ez58+IDS0lK4ublh8ODBmDVrFiorK7F27VrMmDEDr169wtGjR9GvXz/4+Pg02UYh8Od1oqOjAyMjI3Tp0gUfP37kri9evBgPHz6EmJgYVq5ciSNHjsDExAT37t3DwoULAYheqahffkREBPf/qVOnQlVVFf7+/gCARYsW4e7du0hMTARQE49i5MiRaNasGR4/fgwAXHySqqoqNGvWDDo6Orh+/Trk5eW/qczv37/Hu3fvsGfPngbKrrS0NKSkpLjYCDweD0SEK1euICkpCTNmzEBQUBBsbW0RFhbWZHUtRBgN++effwZQo+ifP3+eU95btGgBExMTKCgo4NGjR1ixYgUUFBSwd+9eGBsbo6ysDC4uLg36iKjGkfrzutBnuT6qqqowNzev4+s+ZMgQ6OjooF27dujevTsePnyIadOmAWiaTYr/BKY4MxiM/xkiwocPH+Dm5sad9ghPAm/duoXFixdj9OjR6Ny5M6qqqtCuXTs8evQI3t7e2Lt3L2RlZXHt2jW0bNkSkpKSjf6GKNJn0P+nRRAO9IWFhVyqiRUrVsDb2xu7d+/GiBEjcODAAejq6kJTUxMdOnTA/fv3ERERAXt7e2hra6Nz587fXL5/Rbdu3eDg4IBmzZqhuroaOTk5KC8vx9q1a9G/f39s2LAB+vr6uHjxIgBAX18f+fn53N8AcOzYMcyZM6dOuaJOsSJMQxEZGYn58+dj3759CA4ORlxcHIyMjCAnJ4fLly9z9//000/g8/nYsmULmjdvjtWrV2PTpk3w9PSErKysyE/ihBYDQpnnzp0Ld3d3nDx5EkDNQuDp06cAwFlN3Lt3D/fu3YORkRGsrKxARDhx4gQWL14MKSkpkSsVL1++xNGjR+vIDwBTpkxB7969sWPHDgA1yubbt2/x4sULzJ07FyYmJoiOjkZ6ejoXPMnT05MLkCREVBYfwoUXj8fDixcvkJKSgpycHCgoKGD+/PnIzc1FeXk5gJogfCUlJbh27Rp3zdbWFqWlpU2WJqg2fD4f5eXlyMnJQUxMDMrKyrhnErJ27Vr4+fmhrKwMixYtwsePH+Ho6IiJEyciIiICdnZ2DcZAUVrX1G+HHz9+hLu7O/fuhX1LW1sbQM1mXVJSEgDg/v37GD58OOTl5cHj8dC1a1e0bNmS2/gQdcAyoCY4mY+PD6doVldXY8yYMXjx4gXExMSQnJyMbt26YdeuXWjVqhVev34NdXV1JCYmorS0FAYGBpCWlm4SxbN2CrL6CN/D1KlTUVZWhmfPnnGbyNnZ2ejfvz/XLvr3748JEyZARkaGq2tRpiCrrq6uU35KSgocHBzqpHlzdHTEkydPcPfuXejo6KBfv364fPkytymkoqKCQ4cOcdYTQoT9dNSoUZCRkfmmsvN4PHTv3h2ZmZlYuXIl1q1bh8rKSlRXV3P17ejoiJs3b+Lx48cQExODuLg4Pn78iA4dOoDH40FBQQEjRoxA69atRV7XjbF48WJkZmbi7NmzkJaWhoqKCtfWiQjdunWDhIQEMjMzISUlBR8fH3h7e+PEiRPYu3evyNO91ab2vD5jxgwsWLAAv/zyCzc2C5GUlMSsWbNQUlLCbbYAwIwZM7iUpI1tUP9VYIozg8H4n+HxeOjQoQPmzJmDvn37ciZQc+fOxYULFzBo0CB8/PgRPB4PnTt3hq6uLtTU1ODn54fIyEhYWVkhKyuLW5w1pdxiYmLIy8vDwoULsWfPHsTFxeHt27cYMWIEiAiDBg2CgYEBcnJycO/ePXh6ekJBQQE3b97EwYMHsWjRIqxevbrJJtPaE4mSkhJiY2Ph6OiId+/e4evXr0hOTuZOrGRkZKChoYGvX78iMDAQY8eORadOneosyiQlJRvkOhW13EBNtOzVq1fDyMgIOjo66NSpE06dOoVWrVpBR0cHX7584RT8/Px8mJmZ4bfffuMiIwsXXLU3Pr4lJSUlWL9+PYDfd7pPnz4Nd3d3TJo0CX379sW1a9ewevVqjBgxAlVVVTh37hx3L5/PR2lpKVRVVWFiYgJPT0+oqqpyp7+iUiqEprTJycnYuXMnUlNTObcHoXKwYcMG3LhxA8+ePUP//v0xYsQIPHjwAC9fvoSdnR3Xj/v371+n7KZYgPF4PGRnZ8PGxgbOzs7YtGkTZs+ejbi4OCxZsgSlpaW4evUqd7+rqytOnTqFN2/eAAD09PSwa9cuSEtLi1zW+m26e/fuGDFiBAQCAR4+fMjJVHtsGDZsGCZOnIgNGzZASUkJx44dg5eXFzw8PHDlyhX069evSU0SxcTE8OXLF27jR0JCAiNGjEBxcTGAmtMpgUAAWVlZzJs3DwUFBXB2dsa8efNw6NAhGBsbNyhTuPHxLcnOzuZO8Wvz/v17xMXFwcXFBTk5OejQoQNycnI4M+EPHz7g5cuX2L9/P6ysrHD79m1s2bIFu3btapLc17XLFxcXR3p6Ovbu3Ytbt25x7UM4HhARlJSUMHz4cPj7++PAgQNISUnB0aNH0atXL7Ru3bpBuaKo6z+S++zZs4iJiUGrVq3g6uqKvXv3cvcNGTIEI0aMwLFjx1BeXo7FixcjODgYr169AlAzz8jKyjbIBf6tZa9vLXXz5k0UFRVBTk4OS5cu5ZQyYX0PHDgQdnZ22LhxI5YvXw5bW1s8fPgQWlpaDcoWZV3/0djatm1bODs7Y8+ePSgtLYWioiJevnyJly9f1jk579atG4Caem7bti26devGzetNufF2+fJlbNiwAQsWLMCKFSugoaGB5s2bN7i/Z8+emDlzJjZt2sStRbp27Qo9PT1ERERwVi9NZWXzHyEa12kGg/FPoH4AppSUFDI3N6fr16/Xub59+3YuqJaQtLQ0LtjTrl27RC6rkPoBuiIjI2nixIm0bt06LliGuro6BQcHc/dkZWXRlClTyNLSksuNWDuQWWPlilpuYdC19+/f08yZM8nX15eKi4vJyMiIvL29uftycnJo9OjRZGxsTGlpaVRYWPhdgoII+fz5MxHV5GTetm0bdz06OpocHR3J39+fqqqqaMeOHTRs2DCyt7cnVVXVJg/sVFpaSsrKylwE7KKiIrK2tqb79+9z90RERJCdnR2dPHmSQkNDSU1NjVavXk1r166lYcOGcQHBhO9O1G3k0aNHZG1tzQWIcXJyIjs7uzr3CN+9i4sLF+wpNzeX5syZQ5s3b6acnByRylif+mNIdXU1eXt7k7u7OwkEAioqKqJ169bRggUL6MOHD3T58mVSV1fnovESEc2cOZPu3bv3p+V+a3x8fGj37t2cHFVVVVRZWUmnTp2i48ePk5mZGe3evZsKCgqIqG4Qoc+fP5OiomKdMaap5K7f98vKymjZsmV18ho7OjrSkSNHGpWnrKyMfv31V65tNwUJCQmkqKhIa9as4dpn/Xyws2bNIicnJ0pISKCLFy9ygdeIavI2L168mHbu3Fmn3Kaqa2G/9/f3J3V1dVq5ciXZ29uTnp4excXFcfcL7ysoKCArKyvS19enOXPmNOn8WBuh/IcOHSJVVVVydXUlLS0tCg8Ppw8fPpChoSE3PhLVRCpXUlKiY8eOEVFNULDvJXdsbCwXDKy8vJwWLVpEW7du5eaf+hHYY2Ji6MSJE3TixIkmlbV+G0xLSyOiuv20urqaDAwMyNvbmz59+kRubm40fvx4unHjBm3YsIFGjRrV5Jkw/mgu8/b2pv3791N1dTXl5ubShQsX6OrVq1xgstrP9eXLFwoMDCSi39czFRUVZGdn993azr8DO3FmMBj/MVTPlyU9PR0A0Lt3b3Ts2BFPnz6tk585Ly+PC0wBAMHBwcjKysK6deuwc+dOLF26FIDoc+3W330lIkhJSSEzMxNycnLg8XgQCASYM2cOtm3bxu1ed+jQAXJycqiqqoKfnx8AcCdaTZVHVVj+uXPnYGlpiTVr1uDWrVvo2bMnTExMEBUVhYSEBLi4uMDX1xfBwcEoKSlBeHg4VFRU0KdPH/j7+0NGRqZOrmxRkZ6eDldXVy5XdEFBAfT19TkT0KSkJFRVVXH3Dx48GEpKSggJCUF+fj6WL1+OTZs2QU1NDXfu3OH8upoqj3SLFi2wfft2HD58GHl5eWjZsiXev3/P+c4CgLq6Onr06IHU1FRMmDABPj4+UFJSQqtWrXDjxg3o6+sD+P3didK8Gag5LWzRogVn5m5ra4uEhAQufzCfz+dOKRYuXIgPHz4gNTUVcnJyMDAwQLt27eqcaomyjdQfQ4R++qWlpQgODsbw4cPB4/HQunVrzJ8/H1JSUrh69SpMTU0hKyuLPXv2cKcXZ86caWCtIsqTirt37+LgwYM4ceIE1q5di7KyMkhISKBZs2Z4//49xMTE4OLigvDwcDx//hxA3ZO1zp07Y+/evVzu16aSu7YcwjbRvHlzKCsro0uXLrh16xaCgoIwe/ZsXL9+vVF5mjdvjiFDhnBtuyn6o9CK4t27d4iLi6sjV3V1NSQlJeHl5QUVFRXMnz8fnz9/RqdOnZCXlwcAmDNnDpfHXvidxp7tW1LbvFlMTAwVFRW4c+cOtm/fju3bt+PQoUNo2bIlNm7cyOUCFhMTg0AgQJs2bWBiYoIOHTrAwsKiSeZHoOHpIY/HQ15eHiIjI3Hq1Cns3r0bx48fx/jx46GgoABra2ucOHECKSkpAGrG/FGjRqF58+YQCAQYPXo0ANH6BNeX+cqVKxg7dix++eUX6Ovr48iRI5CSksL06dORkJDAzT9Cs2KhbMOGDcO8efMwb948AE03zwjbYGxsLNavX49Tp06htLS0jmxiYmJYs2YNjhw5gurqanh6esLQ0BBPnjxBcXExLl++DHV19SaRV1jftdcjDg4OuHz5MjefX716FVZWVpg4cSJOnz6Nc+fOYfv27cjKyqozDnbs2BFTp04FUDN3Cfvyvn37uLbzV4QpzgwG4z9GOOnExMRg+vTpWLhwIZYvX47S0lLMmzcPb968wePHj7nJJy0tDWpqanj16hX09fVx6NAhyMrKAqgZMOsvor8lGRkZAMCZZwknqLNnzyI+Ph4jRozA9OnTce3aNQA1E8L8+fPRrVs3zJ49G56enjA1NeUCV6WkpKCgoICbOJrSlGj37t3w8/ODnZ0d+vbti+vXr2PTpk2YMWMGmjdvjvv372Ps2LFYsWIFLl68iClTpuDYsWNwcnJC586dkZ2dzfldijJACFBjMiYjI4PIyEgUFBSgsLAQ0tLSGDNmDICaQElXr17lFsXNmzcHESExMRF+fn7g8XgwNDSEg4MD5OXluUXjt/ZdFQgEnF9+/cXSpEmT0L9/f2zbtg1EBDU1NTx//py7r0WLFigsLOSCU2lqasLKygqrVq3iZBblolG44BYuRoYNG4aRI0fixYsXePLkCdTU1DBlyhTOn1kY4R6oMXHt27cv1w9nzZqFhQsX1vGvFWXbFo4hsbGxMDU1hYuLCzZu3Ii3b9+iX79+KC0tBVDTnhQUFCAtLc2Zfa5atQopKSncs0hISHDB20QBEeHz589YvXo1CgoKMGrUKEybNg3a2tpISkrCzp07cffuXQDAxIkTcfv2bYwdOxbdunXDnTt3kJmZyZUjRFdXF3Jyck0e1Ck9PR1Xr16Fq6srNmzYgNLSUgwfPhxVVVUwMjLCrl27kJCQACUlJU4h+jNE4Usu7F/CPj906FAMGjQIhYWFiIqKQmpqKnevsI12794d9vb2sLKyQmBgIGJjY9G2bds6copyngFqFPtz584BqJlHkpKSsH37duTl5eHt27coLi7G8OHD8fjxY0yZMgU8Hg+rVq3ifEAFAgHXl01NTdG2bVvExMTgw4cPdZ5VFNT2v66toD98+BC5ubkYMGAAqqqq0LNnT1RWViI2NhbDhw/HqFGjsGDBAmhpaeH27dvw8PDgsh8IEYV5MzVihvzlyxdcvnwZW7duxenTp7Fjxw5cunQJ+/bt45T9qKgozuSf/sBfmYhEFiOB6pmqJyYmwsjICEeOHMHjx48RGRnJBcasLdvYsWOhpaWF5cuXg8fjYcWKFdiwYQN27dpVZ34UJfXr+/Dhw7h48SK6du0KPz8/7NmzB0uWLIG3tzemTp2KY8eO4fr167CwsICkpCRatGjxp+UL2/cfxbn5q8AUZwaD8V8RHR0NNzc3zJs3Dz4+Ppg+fTpatmyJIUOGQFVVFREREXjz5g0KCwuRkpKCTZs2Yf78+bC0tMT169e5qJrA74vob0lZWRns7Owwb948fP36FXw+H7t27YKLiwukpaXx5MkT2NnZ4eHDh5gxYwakpKS4YErNmzfHwYMHYW1tja9fv2LatGnYuXMnOnfujI8fP9bx2REFtU9ihRQUFCA+Ph5eXl7Q1NSEmZkZ0tPTERsbi6qqKkyfPh3Pnj1DZGQkbG1t4evriz179iAkJAQqKiqQkJBA8+bN/+Xk9b9Q+0RTXl4eurq6+Pr1K+7cuYN3796hR48eXAAYLS0tKCkpwcnJCXFxccjLy0NKSgq0tLTw/PlzxMbGAvg9wrKoFo0JCQkwNzcH8LsSUHths2bNGty4cQPv3r2DpqYmkpOTcfz4cQA1fpfZ2dkwNDRsUK7Q91oUi8bq6mp4e3tj8uTJuHDhArc5BAAaGhro0qULAgMDIRAIYGtri7KyMhw6dAhAzaIkOzsbV65cgYqKCtq0adNA7qYiNjYWq1atgoWFBZydndGlSxf89NNPkJGRwW+//Ya3b99y9deuXTvOb09LSwsBAQF1+qEo06vweDy0b98e169fR2BgIFq2bAktLS2Ii4vDyMgI8vLy+PnnnxEXFwcZGRkoKSnh69evsLGxQUxMDKKiohoEWKpdtigRli9s03fu3EFQUBCCgoKQmpqKXbt2oWvXrlBRUYGYmBhWrFgBPz8/hIWF1UkP2BQUFRVh48aN8PPzQ2VlJeebX15ejnHjxsHU1BTx8fF4+vRpnewLtZ/PyckJ7u7u4PF4iI6OrnOPKOYZAJw/eHR0NA4dOoQ3b96Ax+MhMDAQeXl5kJOTQ48ePfD69WvMmzcPK1euhImJCS5fvozk5GQsWbIEwO9tWKgEmZub48GDB3jx4sU3l7k+4uLiyMnJwbJly7Bp0yZ4e3tzm4VpaWn4/PkzF4Ty06dPOHDgAEpKSvDzzz/D3d0d69atQ0BAALp27QpA9ONI7fgk7u7uePr0KZ49e4bMzExoamriy5cvOH/+PAoLC7kUY3PnzsWbN29w7969OnNVY2WLAuHGiDAiPQD4+/tjzJgx8PX1xf79+9G9e3dERkZym22163Hp0qX4/PkzF3BNOCeKKs5HbbmBmvaZm5uLpUuXYteuXXjz5g1OnjwJd3d32NjYICIiAjExMVBWVoa8vDzatGmDwsJCBAcHo02bNnViCvzIMMWZwWD8KUITyvpkZGRwJ1xycnJ48+YNjh49imfPnsHJyQlFRUXcBNCpUyeMGTMGUVFRXARnUe6QCgQCREVFoby8HCdPnkRFRQUqKysRExMDf39/uLm5Yc+ePZg8eTJOnDgBIsLcuXNx6tQp7hSvTZs2MDAwgKWlJSwtLQEA8fHxGDRokMgU51u3buHLly/cgjUyMpI79ZGSksK7d+8AAPv378fEiROhrq4OX19fPH/+HHp6epCVlUVoaCjy8vJQUFCAoKAghIaGwtPTE+fPnxd5CiSh4hkWFob3799jyJAhGDx4MEJDQ7F161YkJydj7969yM3NRatWrbB37160adMGe/fuhYmJCbp16wY7Ozt07tyZM/EWdfCbwYMHY+DAgdDX14eVlRWqqqrq/J6qqiqmTJmCTZs2wcjICJMnT8ahQ4fg4OCASZMmoXfv3pg4cWKDckWdZ5zP5+PLly/Ys2cP5s2bh127dqGoqAh9+/aFjo4OsrOzcfXqVXTq1Alr1qzBoUOH4OLignXr1kFfXx9dunThFuyilPvr168NTj2Ff8fFxUFNTQ0zZsyAuro6FixYAGlpaUyYMAEVFRVwc3NDZGQkduzYgaCgIOjp6dUxYWwqc0o+nw9JSUkuIrbQNF9OTg45OTkYN24czMzM4OPjg5s3b+LFixeoqKjA0KFDMWTIkDpRYpsCYf0IF7wXL17kTkJnzZqFt2/fgoiwevVqFBYWYvHixRg6dCjevXuHKVOmYOHChaiuruZccJqKgIAABAQE4MCBA/Dy8gJQ09ZlZGQQFxfH9cUbN25wViJCareLESNGYMiQIZx1jShJSkrCvHnzuJy76urq3CZVbGwsF2RKRkYG1tbWSExMREhICGxsbADUbGTMmjWrTpnCtjJ69Gi4ublh0qRJIn+O5ORkWFhYoHnz5tDR0UFMTAxWrVqFzMxMmJqaYu3atZxsHTp0QFFREWRkZCAlJYXx48dDR0cHgGhdl+q366CgIOzYsQMSEhJQVFQEj8eDkpIS3N3dMWXKFLRo0QIRERF4+fIl7t27B2VlZairq6N9+/bfJeq+sE52794NJycnFBQU4Pbt29DU1AQA9OvXD7Nnz0Zqaiq36VO7Hvv06YP79+83SNnVVG5iDx8+hKurK8TExPD27VvcuHGDs/5QU1ODmpoat6l88uRJLF++HIaGhujQoQM8PT3/moG+/guavuUwGIwfgvT0dMyaNQvKysrcQqA2fD4fr1+/hoODAzIzM9GqVSv06NED586dw7lz56CtrY07d+5AXV0dfn5+nO9kdXU1xMXFRTqIiomJQUFBAUlJSZg+fTomT56MYcOGcUqxUIYVK1ZgwoQJSE1NhampKfz8/LB+/Xp4e3tz/mYbNmyAjIwMkpKS0LNnT+zatUskMsfHx3PRgbW0tODi4sKZ1jo5OWH69OkYMmQIzM3NMXbsWAQEBKBXr17YsmULPn78iKFDh2LFihWQkZHh8jHzeDzcuXOHS/vQs2dPkcguJDw8HO7u7ujRoweSk5Oxbt06jB49Gm/evEFGRgacnJxw/fp13Lp1CzNmzICuri58fHy4fI/t27cHAAwaNAiXLl1CbGws59v8rai96w8Aubm5eP78OcrKyrB48WI0a9YMfD6/zsJq+fLlmDhxIoKDg2FiYoIhQ4YgOzsbnTt3RpcuXQD8sdnft0ZoLmdjY4Pnz59jyJAhkJaWRlhYGG7duoXRo0fD0NAQAwYMwJ07dzBmzBhMmDABe/fuxZcvX5CWloagoCD06NGjTnmiIDo6Go8fP+Z8NIUI6+n169ec8ltVVQUJCQk8f/4cWVlZmD9/Pq5du4Zr167h69evOHv2LBQUFOp8X1SL3wcPHiApKQnz58+HQCDgfsfCwgLnz5/H8ePHsWnTJhgbG2PPnj0ICwvDokWL0KFDB4SGhnKWH6ampti8eXOTpGUCfjcDre1TX11djZ07d6KoqAgdOnSArq4udHR08PTpU8yYMQNbt27FrFmzcPz4cc5PcurUqZgyZUqjUZy/NbGxsbh16xbWrVuHiRMn4vnz5xAXF8edO3cgIyMDLS0tDBo0CKNGjcK1a9ewYcMG3Lp1C/fv30ePHj3QqlUrru/Vbhfx8fGNRvz+Vjx+/BgVFRXQ1NTEyZMnUVBQgB49emDSpEnYuXMnTp06BVlZWe7kEACWLFmC8PBwODs7o2vXroiKioKSkhJGjRrVoHzhMwkV0m9FfR9VIZGRkVBXV8fWrVsBAM+fP+fahIuLC6ZPnw4bGxsoKSnhxo0b0NHRgby8fINxT1TzemVlJWfCKyYmhqqqKuzYsYPLhd6mTRvIyckhNTUVHz58wIULF7jx4ubNm1yu7hUrVjRpfxRu6PB4PLx9+xZnzpzBp0+f4OrqCllZWQwaNAixsbEYOXIkgJpc3sIYJWpqaujdu3edOhZaJDTVCTMRgc/nw8/PD0FBQRg1ahTc3Nzw5s0btG7dGv7+/lBVVUW3bt2gr68Pb29v3Lx5E0eOHEFqaiqaN2/OKfqilrvJ+LaxxhgMxt+FtLQ0UlRUJEVFxToRDoXRFCsqKigtLY1CQ0Ppt99+IyKiV69ekbW1NX38+JEqKiooMTGxzvdEFVVYIBA0iE7p4eFBKioqZGZmRkQ10afV1dXp2bNnnPxERKamprRnzx4iInry5Ak9ffqUiH6P8vjx40eKj48XeVTnqqoqOnLkCC1YsIAWLFhA4eHhRER08OBBGjp0KBUXF9O1a9fI1NSU7ty5Q0REiYmJNGXKFLpy5QoR/R6xUvhstZ/jW1M/Om95eTkXFZuIKDk5mYqLi4mI6OzZs2RkZEQpKSlEROTn50eTJk0iW1tbqqio4MqqrKwkIqK8vDz6P/a+PKCmtW3/ajLPw4lzOA6Os80iMiRKKaU0SaHSpFGliZAjSgghHPNYpuIYkjKnyFCGEiUKRZNK87y7f3/0rufsXbzf+/0+e3vP+3b9U3uvtZ59r2fdz7Oe4bqv+927d9/cZkH/KyoqYnXz8uVLWrNmDS1cuPCr1+7evZt4PF4LH25oaBCpWnZqaipVVFQIfcf93pEjR8jKyoopqt67d4/mzp1LmpqapK6uTurq6kwduTlEbTcRUXp6OvF4PAoLC6OzZ8+y++B+NzY2lkaNGiWkkH3mzBlauHAhO6e6ulrIZlGiuLiYiIjOnz9PPB6P2dXY2Mh85f79+zRhwgR6/PgxERHt27eP7Ozs6MGDB0TUpAC9aNEiyszMZNc2V/AVNbKzsyk8PJz1vwcOHCAVFRVSV1enmJgY8vDwoNDQUHZ+RkYGOTs7k76+vlDfwefzRW53cnIy8Xg8iomJISKio0ePkre3N23cuJG2bdtGWlpaVFBQQDExMbRx40YialKn1tbWZtcIora2lg4ePEgqKipMofhbo7y8nPz9/Wn27NlUU1ND79+/p2nTptHp06eJiGjjxo00adIkUldXpwkTJtCSJUtYponCwkK6c+cO7dq1S+zKwYLPMiMjg5KSkig3N5eIiPz8/Mjb25vevn1LBgYGNHPmTLpz5w69evWK6urq6O3bt3T+/Hny9fX9Yr2LAjk5OUIZIl6+fElnzpyh58+fExFRVFQUKSoqCtmzbds2cnZ2pu3bt1N8fDzp6emRtbU1exdxELVff6mvunz5MqmoqJCrqysREVVWVtK+ffvI1NSUXr9+zc5zdXUlLS0tOnbsmMj7vOYQrJe8vDyqrKwkIqLnz5+Tvr4+OTg4EFFT33D+/HkyNjamGzduEFGTb/v7+5O7u7vQu0WUY7/vgdaJcyta0YoW4AZM69evJ01NTdLW1hYaUBH91cEmJiZSfHw8lZSUkIuLCy1atIhKS0vFZqtgR19ZWUmvXr1i95CXl0czZsygsLAwIiKys7MjOzs7lhrhw4cPZGxsLDTB/x7g7iE1NZUcHR1JQUFBqA7nzJlD69atIyKisLAwmjp1KllZWZGioiLt27dPrLYKvsgF6z4pKYnGjBlDnz59YscaGhooKyuLHj9+TKtXryZ3d3fmR/n5+WK1m0NFRQW5u7vT3LlzacmSJWwC8erVK5oyZQpbhGg+YKmrq6NLly4RkegHXRxKS0tp3LhxFB8fT/X19Wzwxw1C6uvrydbWlnx9fenDhw9E1DSof/ToEbm5uRGPxyMzM7MW7VGUgxiu3rjnrKysTDwej06cONHi3KqqKrKzs6N58+bRs2fPqLy8nJydnYUGy+KwmagpHZq6ujqrR3Nzc3JyciKiv54395frR4iaFhidnZ3Jy8uLTbzFieYLY2fPnqXp06fT8uXLWSqyyMhIOnDgAK1du5YCAwNJT0+PFi1aJHSdOPtsDtwz9fHxIRMTE6qsrKTi4mLy8vKi1atXU15eHnl4eJCXlxf5+PgIpVazt7f/6mKmOO6FWyDh+t/g4GCaN28eVVZWUkpKCqmpqZGfnx8VFRXR5s2bSV5ennR0dCgpKalFWeKcHFVVVZGrqyspKiqSsbExmZmZERHRsWPHSE1NjeTk5Ojo0aNERPT582caP348W2xuDlG3yYSEBOLxeHT79m06ffo0jRo1ihYuXEhjxoxh9WhhYUGrVq1i753y8nKKiYkhT09PsrCwEPv7URBlZWV06NAhunjxIls0DgoKIj09PXr79i0RNb3zV6xYQdOmTaPz58/T0qVLycXFhVauXEnOzs4tJvziQFVVFTk7O9OMGTNo0aJFtH37diIiCgkJIVNTU7ZQm52dTevXrycbGxvWD4k7neH3QOvEuRWtaAUdOnSIIiMj2Y5QY2MjlZaWkpeXF927d4+UlJS++gJav349mZiYkLKyMrm5uVFVVZU4TWfYu3cvKSoqkrm5Oc2ePZvt2G7fvp0MDAwoPz+fCgsLycrKimbNmkVeXl4kLy9P/v7+Yh24/E+DjfPnz5O+vj5bxSVq2gkfMWIEm+Dn5+dTUlKS0G6cOFZ0BSeMBw4cIB8fH9q7dy8VFhZSYWEhGRgYsNy09fX1VFVVRQsWLKCkpCS6dOkSLVy4kJ4+fSpUpijrvvkEt6qqihwdHcnDw4NycnLo/PnzpKenRzt27CAiol27dpGKiorI7PnfgKuXrVu30vjx40ldXZ1SUlJaHL9x4wYZGxvTxYsXW5SRkpJCNTU1YrG3eV1XVlZSUVERrV27lkaMGPHVXaqKigpavHgxWVhY0IwZM8jd3V1sNgsiNTWVbG1tae3atUTUtLs1dOhQxkDh8/mszlNTU2ns2LFskBgWFkanT5+mhoYGVg/i3inibAwLCyNvb29KT08nfX192rBhAxuQ5+fnU1RUFKmqqtKoUaPYIqMgRMVQIWpimnD5tgV3ocrLy2nSpEkUEhJCRETR0dFkaWlJly5dotraWgoJCSFDQ0Pi8Xgs7/H3zEVP1MSECAkJIW1tbcrJyaGsrCwyNTVlk4w//viD5s6dS0VFRURE9OTJEwoJCRFagP4e93Dw4EFatWoVs4Nj+WRmZpKdnR39/vvv7NzAwEBydHQUes8QieddQ9S0+BYYGEhTp06l1atXs5zoDg4OZG9vT7W1tZSUlETKysp08+ZNoTYn2F6JxNceuWd66dIlUlBQoKVLl5KpqSmpqqpSTEwMvX37luzt7Wnz5s1C1+3cuZM8PDzIw8ODiIjS0tJo1KhRIp+INn+W0dHRtG/fPnJ3d6dPnz7R8ePHSU1NjXbv3k2lpaXk7u5O3t7e7PyrV6+SiYkJ6yc5fI/+T1xonTi3ohX/5eBoiUpKSmzQyGHFihW0c+dOunPnDsnJyVFBQQE7JrjrVVJSwihfRKLtNL9EH7x37x6ZmJiwneSjR4+SoqIi3bt3j+rr60lLS4t27txJRE0U3fj4eDp+/Dilp6eLzM7maGxs/OqAQ/BYbm4u+fr6kouLixBFd/HixTR79uwW14r6BZWUlMQmZQ0NDZSfn08mJiZkbGxMISEhpKenR5aWlnTlyhXasmUL2dnZCU185s+fT0lJSVRVVcUGkaLGl6j7RE308blz57JV/GvXrpGcnBy5uroyhoK6ujotXLiQli1b1oIiLWoITibq6uqosbGRrKysaMSIEeTr6/vV61auXEleXl7M/7kyvsckLjo6mrS1tWnBggVscHjgwAGaPn16i0Gg4O50YWGhUB8i6sE5NwgXtOXSpUukra3NdjJXrlxJ2traLeyNiooiGxub77JL25z6nZeXR1paWhQdHc1s4+iUmZmZZGtrSzt37iQ5OTm6c+cOERGdPn2anJ2dqaysTCw2c+3+999/p0mTJgkd4ybqBw4cIDU1NcrLyyM+n0/r1q0jNzc3Rnn/888/aenSpULPTdQ+8j9RTDMyMmjJkiW0evVqImqq19mzZ9O7d+8oLy+PjI2NhSYZ/w4wNzcnV1dXevHiBR05coTMzMzI3d2dEhISKD4+noyNjWnhwoWkrq5O8+bNY/X/vfDmzRvS1NQUCqPJzs4mFRUVCg8PJyKi1atX08KFC79IzRfHJP9L4Tv29vZ04cIFImpaGDpy5Ahjkh0/fpwsLCyEGBPc+4fD0aNHRb6IKDiOiomJodevX5OamhqNHDmSLXAREV2/fp1GjBhB5eXldO3aNbK0tKRz586xe/v8+bPIbPx3RKuqdita8V+I4uJirFu3DgCgq6uLMWPGYMCAAXj8+DGWL1+Ojx8/AgC0tLTw8eNHKCkpYeDAgdi+fTsrgxPYkJaWRteuXdGnTx80NjaKNDUCl2tSQkJCKB1JXFwcZGVlwePxkJGRgbi4OLRv3x6dOnWCtLQ0bGxs8Oeff+Lhw4fo0aMHJk+eDDMzMwwZMoTZLCp7gb8EqSQlJZGRkQEnJyds3ryZKUdzxwCgT58+UFRURFVVFS5cuMDK8vPzg5eXV4vfEKXYRk1NDW7fvo2dO3eiuLgYUlJSSExMRPv27XH69GmYmpoiPDwcMjIySElJwbBhw9DY2IiFCxfi4sWLMDU1haSkJAYOHIj27duLJW8tl/ZHSkoK7969w5kzZ5CcnAygKX1Ubm4uPn36BFNTU6xfvx6bN2+Gs7MzoqOjISsri23btuGHH36AoqIiOnbsKFJbBcGJdElKSqKgoABlZWWQkJDAoUOHsGXLFkRERLB8roLXAMDChQvx7NkzvHz5UkgcihOUEZcgS3R0NAIDA7F06VJYWFjghx9+QF1dHWxsbMDn83H8+HGh8wXzdvbs2VOoDxGVgE9dXR0CAgKgrq6OvXv3oqCggPVZcnJyGDlyJEtLt2LFCuTl5eHo0aOorq6GlJQUMjMzcfz4cYwePZqlVuMgat8G/kqpxP1WeXk5SkpKsH//fpw9exaqqqp4+vQpXrx4gYEDB2Lx4sUgIlRXVyMqKgpAU5qj4OBgkYt/NTY2IiwsDDY2NgCAVatWQUZGholNCr4rOCG28PBwSEpKQkNDA6WlpQgPDwfQlNd427ZtQunTRCnyxLUjSUlJlJaWora2VugYAAwcOBCampp48uQJEhMTMXv2bAwZMgTbt2+HrKwsrKysoK+v36JcUYJ75zT/He6ztbU1kpOTsWHDBpw+fRqKioooKCjA2rVrMXnyZOzbtw9r1qzB+vXrcebMGQwcOFBsNn8Jv/zyC0xMTJCamsrO69evH3R0dHDu3Dnk5eVhyZIlrM9vDlH5yPv373Hx4sUv/sbz58/x4sULjBs3Dnw+Hx06dICFhQW6deuGP//8Ezo6OujatSuOHj3KrqmtrcX27dsxf/58zJ8/H4cPH4ahoSHatm0rEvuBpr6ksLAQmzZtwooVKyAtLY1169aha9euQmkrp02bhhEjRuDq1auYNGkSevfujeTkZNTX16NTp07o1q2bWPq+fxt8n/l6K1rRiu+J7Oxs4vF4TMzkwoULpK2tTVeuXKFFixaRp6cnvXz5khISEsjd3Z2IiB4/fixElfteqK2tpZUrV5KTkxN5enpSaWkpbd68mTZs2EDe3t6koKBAQUFBVF1dTf7+/mwV19LSku7duydUliipcoWFhUIr4A0NDRQREUH6+vrk5+dH27dvp1GjRtGRI0fYzhW3o1VeXk5BQUE0d+5ckQnc/DMIrqCnpqaSnZ0drV+/noiaKGVWVlZUU1PD7I2OjiZdXV3Ky8ujT58+UVBQEHl6erJd/u9hf2hoKI0YMYJMTU1p5MiRdO/ePWpoaCBtbW3i8XiMFkpEZG1t/cW4WnGjvr6eVq5cSSoqKmRpaUl2dnZst05PT4+WLVvWYndDUGtAXPjaDrafnx8TcOLAUUFjYmJoxIgR9OjRI7p58yYTnfke2LFjB02cOJHk5ORIX1+fvL29Wcz9/fv3ydDQkM6ePUtETfHBKioqZGFhQe7u7jRu3DhGx/1eCAsLI0tLSyJq6g/NzMxoxYoVZGtrS2FhYbRjxw6httfQ0EA2NjYsRp/zIVEwEZr3qXfv3mXsFKKmGOzRo0cLsU+4XefQ0FCaNWsW+/6PP/6g69evi9xmQQiWX19fT56enqSurk52dnZCjAjBHf81a9awmPeoqCjS09Ojjx8/itTO5hDsFxoaGlowKoj+srmkpITevHnDvk9ISKDFixd/kWEjrlAaPp//1TCBwsJCMjExYTv7RE39ysSJE2nXrl0is+9rqKyspDFjxtCIESPYeEiwnrjjXHgKR3V3c3Nj93DhwgWKiooSKjcvL49u3LjBNFm+NZrXb2FhIfn5+dGECRPoxYsX7HuOss/5Q11dHRkaGrL7+R5jkn8ntO44t6IV/0Xgdqj69esHLy8vllpJV1cXXbt2xbNnz1iqDC8vLzQ2NuLp06f48OEDxo0bB2VlZSQkJIjVXhJYyfz06RPmz5+PkpISmJubo7q6GvX19fjll19w5swZfPr0CdeuXYObmxvy8vIQExPDcnkeOHCgRdqPb50+SNBWX19fBAcHA2hK7bVnzx4EBQVhxowZ8PHxgaurK3x8fBAdHY3nz58DaNqBIyJ06tQJioqK0NfXxw8//PBNbfxn4HbxBVfQBw0ahJkzZ+LevXt4//49evfujYaGBnz48IGt8MvLy+Pdu3fIyclBr1694Obmhk2bNmHJkiUARJuzuznevHmDRYsWISUlBdevX0dISAg0NTVx6NAhfPz4EQ4ODpCWloaCggIqKytx/PhxvH//Hurq6kLliIqF8M/K37t3L4qKinD27Fls2bIFjY2NcHR0RH19PQICAnDp0iWkp6cD+OtZcT4nLy8v9FmU4J57bGwsUlJSWJq31NRU5jv19fUgIiQkJCAkJATTp0/HnDlz4OzsjPj4eJZaRpzg6lxfXx+TJ0+Gnp4erKys8P79e8yfPx+rVq1CWVkZlJWVERYWhtraWmhpaWHr1q0wMDDAsGHDcOPGDbi6ugqVJ25MnDgR79+/x6ZNm1BdXQ01NTV8+vQJTk5OCAgIQFJSEmtz9fX1kJKSwr59+6CjowPgr/YtCiaC4M4s0JRabsaMGTh37hyKi4thaGiIIUOGICAgAIBwOrTPnz9j/PjxzHZ7e3uoqamJ3Gbgr3bDlf/p0yfcvHkTRIRt27bh/fv32LFjB0udx0FWVhYaGhp48+YNwsPDMWvWLJw6dYqlqRMXuPo5evQo9PT04O7ujuDgYJZnt6Ghgb3vJCUlcfLkSURERGDfvn1wcnLClClTvsiwESVbhbNn//79mD9/Pjw9PREbG8ve2Zwf9OzZEzY2Nrh69Spev34NAJCRkcEff/yBRYsWsfLE8Z4hIkhLS2Py5MkYMGAA2zXm3t0A0KFDB5iYmGDt2rVobGxEu3btAACFhYVQVFQEAMyZMwezZs0SKltWVhaqqqowMjL6pjYnJyejsLCQpdVLSkpCVVUVevbsiWnTpqG+vh4vXrxg57u7uyM2NhZbtmxBRkYGTp06hfr6epbeq3///gC+X//3vdE6cW5FK/7D8erVK0Z3E8S8efPQs2dP+Pv7A2jKMXn9+nUUFhZi6dKlGD9+PI4fP47S0lJ8+PABALB7927Y2tqKxW5uQCUhIYHHjx+joqICr169QqdOnbB7924oKChg165d6NmzJ+bMmYPx48ejb9++iI2NRWxsLBYvXozp06ezzl5KSkrkHX1SUhKbBPv4+CAyMhJTpkzB48ePMXz4cMjIyKCkpISdP2/ePABgVGKO0g0ACgoKWLBggUipWhzy8/Oxdu1aJCYmAgDu378PPz8/REZGoqCgAJqamhgyZAj27NkDfX19FBcXIyoqCm/fvgUA3Lx5E1OmTMHw4cNZmVwOSCISyeDr7du3+Pz5c4vvJSUlUV5ejrS0NPTt2xcAsHLlSrx//x537tyBlpYWrK2tsWnTJpiZmeHChQvYsmULRowY0aIcUYELORBEXV0dHj58CF1dXfTo0QOZmZl4+/YtOnbsiKKiIgwbNgxqampYunQpjIyM8Pjx4y/aKY5c0snJydDU1MT27duxcuVKODo64uHDh1i4cCHCw8NBRJCRkYGEhAQePnyIhw8fAmhqE5cuXYKPj4/IcjALgvM/Dlxd9e/fH+PHj0dubi46d+6MkydPIjAwEIWFhVizZg1u3bqF169fM2r52LFjoaOjAxsbG3Tv3h18Pl+IEi8Ku782CWhsbMTPP/+MwMBAlJWVYdWqVRg3bhzat2+PIUOGYNWqVXj27BmjkcrIyLB7b74Q+S1RVlYGGxsb6OvrIzQ0FDU1NQCATp06YerUqejRowejaP/++++4fPky7t+/z9rCvXv3cO3aNSgoKLD+QpCSLkqQQJ7c+/fvQ0NDA+7u7ggKCoKJiQmGDx+OwMBAPHz4EE+ePGH9NPc+GT58OJycnFjO+Xbt2ol0EkdEyM7Oxu+//47MzEz2/bFjx3DhwgUEBATAwsIChYWFcHR0BCCc87xz587o2rUrrl+/jrt372LPnj2wsLAQmb1fQ1paGo4dO4aYmBg4OTmhQ4cO2LNnD44dOwZAeNI+adIkyMvLY9myZey7cePGoVOnTuw5iGqSzy0McmFAMjIyqK+vx9ixY9HQ0ICwsDAAwhNJFxcXdO3aFXPnzoW/vz9mz54N4K8FTnH000BTGJu3tzcSExNZv+3k5ARHR0dER0dj2rRpMDAwwNmzZ5nP/vbbb9DT00NYWBh27NiB6OhorFu3Dr/88otQ2eLKh/1vB7Hub7eiFa0QGzhRif3799P06dMZvUaQUnT16lUaOnQo5eTkEFGTGNiiRYuoqKiIampq6ObNm6SiosJSKXAQpeCGIH2rvLycAgICGJXozz//JGVlZaHzMzMz6dy5c5Senk6HDx8mR0dHMjIyEhnd6WsoLCwkS0tLCg0NpaysLNq1axdNnjyZFBUViaip3oODg8nV1ZWys7PZdWvWrGFiPl+COJRXKyoqyMjIiAIDA2nbtm2kpqZG69atowULFpCamhpL8cEJJyUkJJCVlRVpamqSpaUlTZ48mVEqxYHs7GySl5en8+fPM38WTBl0/vx5GjFihJBoycGDB0lfX5/ly6yvrxeiK4oj164gVe7Dhw+0detWunjxIhUXF1NZWRnNnz+fzp07RzY2NqSoqEjnz5+nvLw8CgwMJKImyt+BAwdYyqzvgYaGBlq+fDlT2c/KyqKNGzcysRtzc3Oys7Oj1NRUKiwspEWLFgnlDCb650J5osCXlIxLSkpo+fLl5O3tLdQeHz16ROvXrycej0empqYtsgSI2kcE6yUvL4+eP3/OaM3Nf/vDhw+0ePFi0tbWJkdHR0pPT6f6+nqKjo5mYmGiRnFxMaWmphKfz6dly5bRhAkTaMSIEWRubk4bN25kAmTnzp0jbW1tlkZo69atpK6uTo6OjmRtbU2TJ09m9PjvgU+fPlFkZCStWbOGTp06ReHh4cTj8Sg+Pp71MT4+PmRlZSXkL98LFRUVpKCgQEeOHGFihy4uLkww68OHD2Rubk7jx49n7/7mba55zm5xpKnjUFJSQnZ2djRy5Eihd8eBAwfI2tqaZV8Q9PnU1FShbBOiRnl5Odnb25O+vj5Ld8Xh0KFDZG1tTfv27SMzMzP2rhFU1i8oKKCoqCgKDAxkYRLiguCzdHd3pxUrVpCzszNFRkbSx48fydfXl+bOnUtlZWWUmppKxsbGdODAAXZNZWUlaWhoUHBwMPvuP1kp+3+D1olzK1rxH4gbN27QsmXLKC8vj/Ly8sjZ2ZlWrlzZ4ry6ujpavHgxy1daVFREqqqqtHfv3hYpKMSN3Nxcsre3Jy0tLTZwvH//Pi1cuFAo9c779+9p3LhxbHDQPE5L1J294IRr3759NGnSJNLS0qIbN25QTU0NjRo1ik6dOkVETXHi7u7u5OfnR/X19dTQ0EBWVlZ09epVkdr4z8DVT2RkJBkbG5OxsTG9e/eOiJoGVkZGRuTl5UU1NTXk7+9PixYtosbGRqqsrKSYmJgvpkESpa1cXfv5+ZGdnV2LRR2ipkGZtbU1ubi4CH2vrKxMmzZtauHb4h4QxMXFkZycHDk4OJCioiL5+fkRUdPAnMfjCcUJHz58mBYtWtQij7oo8fHjRxab3Lxu3r17RyNHjhRSpH/+/Dm5uLjQoUOHKC8vj8zNzcnc3JyUlZXZvYkLO3fuFGpPoaGh5OHhITSQ5Hzo+vXrtHjxYqF4dw6cQrk48KVcrX/88QeNHDmSFixYQFOnTqVHjx6xZyJ4L6WlpbRkyRLi8XhM6VZcaGxspCVLltCSJUuouLiYkpKSyMnJifz9/SkyMpI0NDRIRUWFdu7cSZGRkbR69WpydnZm1yckJNCZM2e+uLAiSnypvV+5coV4PJ7QIqajo6OQzgD3fty3b98X1Y7FnV5q48aNNG/ePHrx4gU1NDSQiooK3bhxg6WwCwwMpKKiIjpz5sw/tU2c/V98fDy9f/+eiIju3LlDI0eOFNJneP36NVlbW9OtW7fEZtPX8O7dO5o2bRrxeDzy8fGhy5cvs2PXr1+nnTt3UlxcHFlbW7MJ5v/kA+KM0ef+T0xMpNmzZ5OKigrra1JTU8nKyoplPjhy5AjNmTNHKDY/NDSU9PT0KCkpSawLnf/u+C/dZ29FK/4zQf+gtTU0NKCgoAA3btyArKwsNDU1kZycjAcPHrDjQBONb+HChcjLy0NOTg569OgBAwMDVFRUCFG7xBmjmpiYiMOHD6NPnz4YNWoUioqK8PTpUwBNKqZjxoxBSEgIUlNTAQC3b9+GsrIyZGVlAYDFaXE2izJGq76+nlGuiAgfP36EhIQExo0bhylTpqBt27ZwdHTEzp07UVRUhLFjx2LcuHEIDw+HnZ0djIyMICkpicmTJ4vMxuagr9BXtbS0MHjwYGRnZ6O0tBRAk9rxhg0bcOnSJVRWVsLQ0BCfPn3C7t270aFDBxa3CvzlU6KElJQUJCQkcO3aNXTs2BFPnjzBnTt3WsT7du3aFTY2Nnj48CGjnwPAzp07YWNjw2LOBMsVB0pKSjBt2jRERETgwIED+OOPP+Dh4YH379/jypUr8PDwgKysLHr27Im0tDScO3cOhw4dgqGhoVBMMImQvsrn87Fu3TqmoN+cUigtLQ05OTmkpaUxW0aOHImKigp8+vQJsrKyOHDgADZu3Ihz587Bx8cHgOjj4Tj/+/jxIzZt2sToldevX4eGhoYQrZC7JzU1NQwaNAhPnjxhIRZcv8Hj8YQ+iwq3bt0SUtdNS0uDv78/kpKScOXKFRw+fBiqqqoICgrCkydPAPzVZhsbG9GlSxcsW7YMq1atwvTp00VqKwciYn3fokWLUFZWhtu3b2P06NEYPHgwcnNz8dtvvyEyMhLu7u54+vQpVq5cidTUVDx8+BA3b94EAIwfPx7z5s3DwoULAfz1DEVJYyWB0JHExERkZWWhuroampqamD59OsrLy1l/smLFCjx+/BgPHjxAfX09evToAT09PZSXl3+xzxAX/fbs2bOYMWMGPn/+jOfPn+PMmTOQkpJiNNwPHz7g/Pnz8PLyQnR0NP7880/U1tZ+tQ2Ko/87e/Yspk2bhu3bt8PU1BSpqamYOHEitLW1mYo9EeHXX39FZmamUEiTOPH+/Xt4enoiKioKAwYMYBTyHj16ICQkhGW6kJGRwf379zF16lRMnToV9+/fR3p6OiQkJL7YZ1CzGHpRgSs/KioK+/fvR3JyMuTl5TFr1izIysoyav+gQYOgpaWF2NhYvHnzBpqamujcuTPi4+NZWQsXLkRdXR3CwsJY6EUrWmOcW9GK/whwYhpc56yhoYFhw4bh4cOHSE9Ph5KSEuTk5NgLSlpamg0OPnz4gG7duqF79+4AAEdHR3h4eAhNnEUtyCKI+Ph43L59G48fP8bcuXMhJyeHxMRE1NbWQlZWFgsWLMDPP/8Mb29v6OrqsrRIzcWGRGVzVlYWTpw4AQAsDdPNmzdRUlKCtWvXYt26dXj37h0bHNrb26Nt27Y4cuQIJCQkoKCgAF1dXQDA+vXrceDAAZGnhREEl86moKAAdXV1Qi96Y2Nj9O/fHx8/fmTfde/eHUOGDMHr168xYMAA2NnZMcEeQYgjZrWurg4eHh4ICAjAjz/+iCFDhiAkJIQJZgkOXOXk5KCioiIUEzdy5Ej06NHju4iaEBG6deuG8ePH4+LFiyzdh7a2Nvr164erV69CQkICW7Zswfv37xEYGIjTp08jMDCQiTpxEOUAXUpKCubm5ggLC4OysrLQQApoWpjq168f7t27h0+fPjFbOnbsiJ9++glA04JL37590aNHD5HHAwNNE0jO/9avX4+amhqcOXMG+fn5qKqqgpycXIvBLOcD6urqyMvLw8uXL9n9C0IU/Qifz8enT58AAF26dMGuXbtw7tw5XLlyBYmJibh58yYaGhrQv39/tG3bFqtXr4a0tDQePHggtPAlGLNtZmaGnj17ii0lloyMDB4/foyYmBh8/vwZFy9eRH5+PgwNDcHn83HhwgXU1dVBW1sbhw4dwoYNGzBw4ECUlpYiLi6uRZn0D9ElcdielpYGPT09rF27FkuXLoWdnR3ev38PHx8fJCQkMP2Afv36Yd68eQgODkZeXh4AYMmSJS3ej6JC80VOoOn9Ex4ejt9//x2+vr7w8fHBnTt3cPfuXRgYGKBXr16Qk5NDly5d8OLFC4SGhkJNTQ3t2rUTS0zql9I7vn79GpcuXcLu3btx5swZREVFYdiwYWjbti3Mzc3x6tUrHDp0CJWVlUhMTISsrCyLBRY3pKWlQUSIi4tDRUUFNDQ0MHz4cHz+/Bmurq7YuXMnTp06hcmTJ6Nt27Z4+fIlVFVV0a1bN+zatQvAl/sMcS2qZGVlwdDQEAcOHMCjR49w9uxZEBEMDAwgKSmJu3fvoqysDG3atIGCggJ++eUXbNiwAbKysggODsbcuXMB/LVgGBAQABMTE3To0EEs9v8tIMbd7Va0ohXfEI2NjdTQ0EDr1q0jQ0NDun37tlCKj2fPnpGdnR1t376dGhsbKSEhgfT09OjkyZPsnIKCArKzs2PxioIQFzWnrKyMLl26xChOubm55O7uTmvWrKG6ujo6ffo0LV68mK5du8auaWxspNzcXEpISBCLjRwaGhro5MmTpKioSM+ePaPt27fT6NGjSV9fnxQVFRlN3M3Njby9vRnd8/r16yQnJ0dz586lQ4cOUXh4OFlaWrJ0FKKm4TZ/lnFxcWRhYUHFxcUtqGX+/v7k4eFB8fHxRESUnJxMCxYsEIoXFge+RGl78+YN6enpsfRdRETm5uYUGBjI0q8I3s+bN2/oypUrojdWAF+j4nHfl5aW0ogRIygiIoIde/ToEdnb29Mff/zBvisoKGD/izr+urnNO3fuJB6PRxoaGkLfczbExMSQk5MTaWtr0/Xr18nBwYE0NDTo7du3IrPxf0JZWRnZ2NhQWFgYnT9/nhQUFOjo0aM0dOhQunnzJktLR9SyPYiTlt3Y2EhHjhwRooerqKgQj8ejK1euUFVVFYsZF2xzR48eJS0trf+xbHHh/PnzJCcnR0ePHqUdO3aQgoICe4+cOHGCFi9e/EXdA8G0N+JAc9+uqqoiDw8P2rNnDxE10XFdXV3JxcWFqqqqaN26daSvr8/68sbGRtLR0WE+wtWxqN+PX+tHIiMjadKkSULfeXt7k6urK1VWVlJ8fDwtWrSITE1NSVVVlY4cOSJSOzk0Dzeora1l93Dt2jUaM2YMJSYm0o0bN8jFxYVsbW1p//799PnzZ9q9ezfxeDxyd3cndXX1L4ZOiAqCuhPcs42KiqLFixezEKurV6/SuHHjqKioiGJiYsjKyors7OzIz8+PpaOKjIxk6ZrEhS/5yMmTJ2n58uXsc11dHXs2YWFhZGpqSrGxsex4bGxsi7jxVlr2P0frjnMrWvE3hYSEBKSkpFBeXo6UlBRs2LABZmZmOHnyJOrq6jBmzBhMmTIFKSkpuHfvHsaPHw9jY2OsX78evr6+WLt2LdTV1dGnTx9YW1u3KF8Uq9NfojC9ePECXl5ebMenT58+UFJSQnZ2Nm7cuAEdHR10794dcXFxKCgoYNf16dOHqZiKi0ouJSWFqVOnQkVFBWvXrkVpaSkSEhJw5swZdO3aFX5+fgAAMzMzZGZm4uXLl+Dz+VBTU8OWLVugpKQEKysraGpq4rfffkNISAiKi4tFnpqn+bMcPHgwHj9+jJqaGqZcy+0SWFlZIS8vD15eXli1ahWsrKygqKiIbt26iWWnlppR2jIyMtgxCQkJ5OfnIycnh33n7OyMiIgIpKSksHO4MgYPHgxNTU2R2wyghbIrRxXmICUlhYaGBnTp0gW2trYICgpCcXExAGDChAkYMGAAkpKSWPqY3r17A/hLyVUUOxbN6/rFixfg8/lYsGABjh07hpKSEty/f1/IDgCYPn06fH19MXHiRFy7dg0//vgjIiIiWqiuigqc3dzf+Ph4eHh4oFevXlBVVYWenh5+/vlnbN++HT/99BNOnjyJhQsXIjo6GnV1daw9cNdztGwS025teXk57t27h02bNsHf3x8zZsyAhIQEevXqhfbt22PatGno3r07Ll26xK7r0KEDhg4d+k/7OnHtatXX1+P27dtYvnw5Fi1aBBcXFzg6OuLp06d4+vQp5syZg+7duyM2Npb12RwNm1PeF3Wf3bw9lpWVAWjKIBATE4MZM2YAAAYMGICFCxeisrIS165dw/Lly5GRkYFz584BaKrTixcvMh8RTOkkSnB2HzhwAFu2bMHZs2dRX1+PPn364Ndff2XhSwBgZ2eHmJgYREdHY/LkyTh69Cj8/f1x9epVppYtyr67qKgI2trauHz5MgBg06ZNsLW1xbJly5CSkgJVVVWMHDkS27ZtQ2BgICQlJaGgoICtW7fizZs3MDAwwJgxYyAhIYGrV6/C1NRUZLYKIjo6Glu3bmVsAq6Opk2bhkGDBuHevXvIysqCuro6Jk2aBG9vb0yfPh3r1q1DXl4eQkND2ftJS0tLrKESwF8+8vHjR3YsKysL6enp2LNnDzw8PKCoqAgHBwf4+fnByMgIHTt2xJUrV5CbmwsAUFJSgqqqqlD5/7Vq2f8iWmunFa34m4IbeHh6euLHH3+Evr4+lJSUcOrUKRgaGuKPP/6AgoICunXrhtu3b6OkpAQmJibYsGEDBgwYAElJSZw7dw6+vr5iSdVEArFlT548wbt371BWVoZJkyZBRUUFQUFBjD6upaWFHj16ICoqCo2NjZg5cyZevXr1RUouIL4YVaCJFqmsrIz8/HzU1taiTZs2kJGRwYYNG3DhwgU8efKExTKfOXMGU6ZMwZUrV6CqqgoXFxcATbTWiRMnQlFRkaWL+ZZo/ixTU1Ph7e3N6KG9e/fG9OnT2aRIQkKCpavp27cv1NXV0b17d4wZMwZnz55lKU3E8ULlnm10dDTU1dWxcuVKWFtb4/Lly5CRkcHYsWMRGxvL7nPs2LHg8/kIDQ1Fdna2UBniBFc3ly9fhpGREZYvX45Dhw6xlF18Pp/5qYuLCxoaGnDy5El2vaWlJdavX8/Sp3EQRx7ViIgIaGhoYNmyZbC3t0dJSQkmTpyI2bNnY8OGDS3sICL06tULPj4+CAgIgI+PD2RkZEQ+GeLz+UIp27i/SUlJePjwIeTk5NCjRw8AgJ+fH6qrq+Hl5YXdu3dDVVUVwcHB0NHRYblgm/uJqPyGo6tz7XLEiBGIi4tDREQEjIyM4OPjAwsLC6xatQpVVVWQl5fH0KFDceLECezZsweRkZHYtm2bUKomUePu3btYsWKF0MIVBxkZGeTk5LAJBwDMnz8ftbW1iIyMRIcOHaCiooL09HTWZzenNov6Prj2eOnSJWhra8PJyQm7du1Cbm4uRo0axSYOQNPCVUVFBd68eYM2bdrAxsYGL1++ZM/ra3GrokRmZiZ0dHQQFxcHSUlJ7NmzB97e3qisrETv3r1x7do1NoHi8/lo3749Ll26xPrAAQMGQEpKitktyr67c+fOmDNnDnbu3Iljx47hzZs3cHJyQkVFBbZs2YLLly8jNDQUmzdvxqlTp7Bt2zZYW1tj0qRJbDFg/vz5uHfvHltMFKVmBhfny6UA5LQDuHEQ57+NjY2IiooC0LQ48eLFC1y/fh0//fQT1q9fD3d3d0ydOlVkdn4NXD915coVaGpqYtmyZfD09MSzZ88wf/58TJw4ESkpKWhsbMSOHTswZcoUPHjwAIWFhdDT00PPnj3RtWtXsdv9H4Pvss/dila04n+NV69etfiOo+oEBQWRmZkZvXv3jqqrq+n8+fOkoqJCxsbGpKKiQkZGRl9NYdPQ0CA2as7bt29p3rx5pK6uTvPmzSNDQ0N69eoVFRcX07Bhw+j69euMLhUUFETjx4+n3bt3E1FT2ilxgKPFfqlOONs+f/5Mfn5+ZGBgIHTc2dmZTE1NiahJ3fvWrVt07949oXOap0/61vgSfSsxMZEpZnOUQycnJ6bCy90rZ1NdXR29fPlSqExR0kCbl52cnMzCDyoqKig6OpqGDh1Kd+/epS1bttDixYtZeo8HDx7QokWLyM3NrUXKEFEiIyODNm7cKESzDwsLo5kzZ9KtW7fo6tWrtHr1apo7d67QdRw1MCIigng8Hn348EHouKjbYvO6vnv3Lunq6tKzZ88oPz+fpaYjamqvKioqdPr0aSHbv1SmKP2jeZhAamoq7du3j27evEn5+flUUVFBTk5OZG9vL3TekiVLyNLSkn0uKSmhR48eiczO5iguLhb6XFlZSURNde7v70/W1tYsJIKIaMKECYxCnJWVRdbW1qShoUFeXl5iVRkuLy8nS0tL4vF4dOrUKZaSi3vG9fX1tHPnTrK2tha6xxUrVpCCggJdvnyZGhsbxdZnfw3nz58nVVVVunv3Ll26dInOnz9PBQUFZG1tTRs3bhQKh1iyZEkLdW9x4Uvt6tixY+Tm5sY+FxcXk7q6Op08eZKOHz9OCxcuJDs7O7p58yYtWLCAAgMDKTk5+bvZnJWVRYaGhjR16lT27qiurqagoCBycHCgN2/e0J07d2jt2rUUGxtLBgYGZGJiwsLLPn/+TKamprRo0SKR2v3gwQPi8Xgsc8Ty5cvJy8uLpSYU7H+Dg4Np6dKllJeXRw0NDRQQEEDjx49vUaY4QiSa13d8fDzp6OhQbGwsZWVl0ZEjR2jo0KEsu4hgaMrFixfJycmpNZ3UN0LrjnMrWvE3QF5eHgwMDNhqcn19PYC/Vh5dXFzw+fNn/Pnnn2hoaICenh5Onz4NW1tb9OnTB8nJybhz5w4qKyuFym1sbISUlJRYaNn19fU4fPgwFBQUcPXqVQQHB2PgwIHYsWMHgCZRso0bN+L58+eoqKjA58+fYWRkxOjYAwcOBCBaOiUnlsXtwDY0NLBdcMGdrm7dukFTUxNt2rTBoUOH2PV+fn5ISEhAZGQkOnbsCBUVFUyZMkXIbm6nRVS7W1JSUqioqMCaNWuY+Ji8vDxCQkIgKyuLDRs24OnTp5CTk8OVK1cA/LUbwdkkIyODYcOGMbs5NWtRora2FhcvXgQAJCcno0OHDlBWVkZjYyNTxu7QoQMWLlyIsWPHwtfXF5aWlrCxscH8+fMRFBSEXr16idRGDpzy9JEjR1gd8vl8PHr0CBYWFlBRUYGSkhLy8/ORkZHBFKiBv3betLW1sWPHDiaoxUFUO0NcexSksgNNfYuUlBRkZWWRl5eHQ4cOISAgAJcuXcIvv/wCExMTBAYGwt7enu30N4eoqOR1dXVYunQpXF1dUVlZCT6fj82bN2PRokV4//49du/ejUWLFqG2thZ6enqorq5mPgQA/v7+iI+PZ8+oa9eumDBhwje380t49OgRTExM2A6am5sb7OzsEBAQAEVFRaxatQrS0tKIjo5GVlYWAMDLywv79+/HkydPkJmZCSUlJfzyyy/Q0tKCiooKGhoaxBIu0alTJ/z8889o06YNEhISWAYD7hlLS0tDXl4eEhISCAwMRG1tLcrKylBTU4Px48ejT58+kJCQEEufzb0LBcH9XkJCAubMmQNFRUXo6OhAT08PvXv3xrRp05CTkwMXFxckJCTAw8MDb968gZKSktD14tpllpaWRmVlJe7cuYP09HQ0NjYiKSmJib3V1dWhe/fuMDIyQnh4OObPn4+VK1fihx9+QGhoKCZNmgQvLy+MGjVK5LZy/ictLY2qqiokJiaiuLgY/fv3x/z581FUVISePXsCANq1a4dp06ahtrYW9fX16N69O3Jzc7F//36oqanh1KlTjCHSrVs3uLi4YPbs2SK1f+LEiRg1ahSCg4MBAKampnj//j0ePHjAwji43e7x48fj8ePHaNu2LaSkpGBqaorff/9dqB6ISGTvxsrKSsZY4nwkJCQEJSUlSElJwW+//QYlJSV0794dmZmZICLk5OSgrKwMe/fuhaurK5ydneHr6wsVFRUhZuH3EMj8T0HrxLkVrfgboE+fPjAwMMD8+fOhra2N/Px8AE2DbI4CunjxYkRHR7NB+g8//IAZM2bg5MmTOH78ONavX89SNXEQxSC9eWwZNwgpKirC1atXWRyQrKwsFi1aBD6fj8jISCxZsgQ9e/bE2rVroaioCElJSSxduhQKCgpC5X/rlxRnb25uLhYtWsTitHbt2gUTExOsWrUK+fn5LeIihw0bhpkzZyIyMpLF8XXt2hX79u1j8XOitPtryMzMhKGhIUpKSjBy5EhERERg1apVyMvLg7+/P5SUlGBvb4/6+npIS0vj8+fP/7Q8Udkt+OKWkJDAixcvsHz5clRXV6O0tBS//PIL9uzZAzU1NRQXFyMhIQEJCQkoLy+Hg4MDwsPDYWFhgfv370NDQwOAeAa63GLTtGnTICEhge3bt6O4uBhSUlJISUlB586dcebMGSgrK6Nr166IiopCbm4uysvLWRmcD3F2iwNcezxy5Ah8fHywf/9+FBUVYciQISAiuLi4wMnJCenp6airq8OyZcuQk5MDW1tb2NjYoEePHpg0aZLY7AWAM2fOICsrC8eOHYOMjAwKCgrw5MkThIWFYf369Th06BAGDRoELy8vTJkyBb/++itu3LjBfJprjyoqKmK1m/vtQYMGYf369Vi3bh3atGkDIyMjnDp1Clu3bgXQRG9+9eoVU3E2MjLCjBkzYG1tjRcvXmDGjBno27cvLl68yFIEiqLPTktLYzH53EKhiooKpk6dirdv3yIuLo6lB+La2IQJE+Dg4IAHDx7Azs4O06ZNQ48ePRAUFNRCEVlUfcirV6/g7OyM27dvA4AQtRoAnj9/ztTr6+vr0dDQgHv37qFdu3aws7PDoEGDcPz4cUhLSyM8PBw///yz0PXiosWfOHECSkpKOH78OObOnYv09HRISUkhJycHdXV1LJxHQUEB2dnZyMvLw/Dhw7Fu3TocOHAAzs7OQvcvSnD+d/LkSSgpKcHX1xd2dnbIyMiAoaEhJk2ahJ07d7LzR40ahaSkJFRUVGDUqFHYvXs3jhw5AgcHBwDCffaECRNgZGT0Te199+4damtrAfzl235+foiMjMSDBw8wcuRIKCgoIC4uDsnJyQD+ev5t27bFgAEDWL3279+fZThovtj8rREREQF5eXlGK4+MjISbmxuSk5PRsWNHvHjxArKysjh69ChUVVVRVVWFBw8esNAIdXV1jBo1CgMHDkRsbCwMDQ2F7G6NY/4/4Htsc7eiFa345+Dz+UKUXj6fT7NnzyYej0dBQUHs++awsrIiHx8fKiws/OI54qTq3Lx5k+zs7Gj16tX08OFDKioqInNzc6ZyzNlmbW1Nq1evJiKioqIiSk5OFqL4iZq6+uTJEyouLqaqqipav349zZ8/n3bu3Ek2NjZ06dIl0tfXJ2Nj4y8qX6enp5OZmRktW7asxTFR07cEfUQQR44cEVLVPHLkCI0cOZLOnz/P6nL37t2kq6tL6urqjIYpDsTHxwspuAuqeqenp9PixYspMzOTEhISiMfjkb6+PqWkpBBRE5Vv5syZTMVUEF+jEH8rfEmROSIigvbu3Uvq6uq0atUqImoKL+DxeLRw4UJKTU0lIqI///yTDAwMqLy8XKyqx83x6dMnWrhwIRkZGdHx48dJX1+fzMzM6P79+1RRUUFJSUn0/v17ImpqhxYWFqzuBSGOe+B+49mzZ8Tj8WjmzJl0+vRpOnDgAM2ZM4eIiGpqaoio6dkMHz6cUlNT6dmzZ7Rw4ULavn37d7Fb8Hf4fD5dvnyZ1NXVydramh2PiYmhkSNH0uvXr4moSRF56dKltGfPHrp9+zYRkVDIQWRkJM2bN6+F8u23QGFhIZmYmNCIESMoMDBQ6FhERATt2bOHLl26RHPmzKE7d+60uEeiJl9JS0ujjIwM9p2o+2yu/I8fP5KXlxctW7asxfuSiOjgwYOkpKQkdO3atWvJ29ubfeb8iEi870cOeXl5ZGpqKpTFgKiJ+jxmzBi6ePEiC1UIDg6mFStWtCiDz+eLzb9fv35NR44cIQcHB3r9+jW9evWKnJycWIhSfHw8KSgo0IULF6i2tpYiIyPJysqqRdiCqG1uaGggX19f4vF4ZGdn1yJEY+nSpaSvr09ETe1t8eLF5Ofnx/rt27dv08yZM1nohDhRXV1Nfn5+5OfnR9nZ2fTixQtSVlYmKysrds7169eJx+ORsbExszk9PZ1mzJgh1BY5iPod+d+E1iWHVrTi3wz0j5ynUlJSyM/Px+vXryEpKYnjx49j9erVCA8PR21trdBKJ7dq6+zsjAsXLjCF6u8hosXn8xEQEAB/f39MmjQJPXv2hIyMDLp06YLOnTsjISEBHz9+ZLb17t0bP/74IwCgR48ebJWUywcpypXRzMxMLFu2DG/evEFNTQ0UFRXRuXNnXLx4EWvWrIGOjg5CQkKQlpYmpHBL/9gxHDhwIMzNzWFgYNCibFHuMDf3kYyMDFRVVQFoot62b98eHz58gJmZGY4dO4YtW7Zg9OjR7JzFixfDx8cHJSUlzFdIDIrCVVVVGDduHADg1KlTcHFxYdTaLl26sJ3Z8ePHQ01NDQMHDkROTg7evHkDe3t7/PrrrxgxYkSLckWRU5XP56Ompga2trZQUVFBeHg4KisrmT/y+Xy8ePEC27Ztw9mzZ5GWloZZs2Zh2LBhUFRUxNChQ5GdnY1Tp05h2rRp6NSpk1hVj5vj2bNnkJaWRlhYGMzMzHDu3Dl07NgR0dHRKC0tRVVVFe7cuYOMjAw4OTlBUlKyhUq2YLjCtwTXf3E+KCEhgcbGRqZs3KlTJxgbG0NOTg4ZGRmorKxE27ZtUVdXh3bt2mHQoEF49eoVhg0bhlmzZn1R3VZUdZ+Tk4P169ezHVuOCi8pKYnx48dDTk6OhdgATYrkY8aMQVBQEICmPO/dunXDhQsXmB/36tWL0UWnT5+OgICAFsq3/xfk5+ejsrISHTp0gKysLPr374+QkBBs27aN0caHDRuGM2fOYNasWfjll18QHR0tJKjFoUePHuDxeBg0aJBY+myuboGmdt+3b198/PgRV69ebXFcT08PHTt2hKenJ9LT05GXl4dXr16x8B+gaVeR/iHcJk6RSQ4PHz7Eu3fvMHnyZPD5fIwaNQrV1dXo2LEjrK2tERoaiiVLlmDRokU4e/YsdHV1W5QhKSkpUtV9QSQkJODYsWOorq7Gr7/+isGDB2PJkiXIzMzE1atXMXnyZKirq2P58uXw8fFBYGAgjI2N0b17d7HYzEFKSgrq6uqQlZVFUlIStm7digULFiAsLAxAk/J3ZmYmzpw5g169emHRokUoKCiAs7MzFi1aBF9fXyxZsgT29vYis/FraNeuHSorKxEREQE9PT1ISEhAQ0ODZY4AADU1NcjLy6N///5obGxEVlYWVq9eDXl5eQwaNEioPBJTjvT/GnynCXsrWtGKfwI+n09r166l8ePHk66uLq1cuZLq6uqIiEhVVZU2btxIRMIr/9z/gjsDokTzvI0cCgoKyNTUlO16V1ZWUkJCAhUUFNCtW7fI0dGRjIyM6MGDB+Tj40MqKipsxVRc2Lp1KxMD8fLyokmTJpGhoSElJSXRkSNHaMyYMZSdnc3O37NnDykpKQnlyf6eu4dEwj6ir69PK1euJD6fT/v27aNZs2aRnJwc23krLy+nUaNGCYnH8Pl88vb2/qpo3Le0UxA3b95kO0SHDh2iCRMmsHyvbm5utG7dOiJq2gX4448/yNbWlvT19Wnz5s0itVMQgjtPpqamNHToUJo3b57Qbk9NTQ3NmTOHPn78SO7u7kyE6tGjR6SlpUVWVlY0bdq0Fjt5ogKfz6cnT56wnYW6ujp6/vw5YxTs3buXLCwsqKqqij2Tmzdvkra2NmVmZtKZM2fI2NiYtLS0aNOmTWKxmagp3+uxY8eE8uZy9mVkZNDLly9p6NCh9PjxYyIiMjY2pqVLl7LrP378SHPnzmViP+KGq6sr8Xg8On78+BeP37t3j+bMmUNnz55l371//554PB7ze040TBzg8/k0efJk1jfExsaSu7s7eXt7k5ubGy1ZsoTl4/by8qLHjx9TcnIy6erq0sWLF7/LzlVDQwNlZmYK7eZv3LiRJk2aRMuWLSM5OTmytbVlO/WC7ff169dkampKpqamNH36dLH2I81t+RKSk5NJU1OTCWo1NjbSu3fvyNjYmHJzc+ndu3d08eJFCg8PF4e5XwX33Gtra8nX15cWL17MGCrV1dW0bds2mjVrFhE15UU3Njammzdvfjd7iZoEOjdv3kza2tpUVlZGgYGBNH78eDIxMaGbN2+Sv78/zZo1i41VKioq6OnTpy2E+ETJoGg+jmhsbKT379+TgYEBqampUXBwMBE19XPTp0+n/fv3s3Pz8vJo1apVZGVlRVpaWmJ71/y3o3Xi3IpWfGdwHSf39969e3To0CFyd3enwsJCCgsLI2NjY/rjjz+IiOjatWs0YsQIKisrI6K/XmjNO3dRTuwiIyOFOvDmkJeXJ11dXXJ2dqbRo0eTtrY2TZo0iWJiYujdu3e0evVqcnV1JVdXV/bSEge4Qcz8+fPJxMSEiIjMzMxITk6O9u7dS0RNFLnFixcz+jgHFRUVNqkTN/4VH5k3bx6FhoZSfn4+GRoa0rZt29j1O3bsIGdnZyG6eUFBAWloaFBCQoJIbP4SFa+xsZHi4uJo9OjR7Hf37NlDFhYWtHHjRgoNDaWgoCAh+nhFRYXQZ1HTKevq6sjAwIA9/2fPntG0adPo4MGDpKysTEFBQfTixQsiIvLx8aH09HTKzc2l0aNHszCE8vJyysjIoNLSUlauqOmre/fuJWNjY3r+/DldunSJFBQUaObMmWRvb0/Z2dkUHh5OpqambMGIqIlmO3r0aEpKSiKipgmc4CROHNTVp0+fko2NDZ06deqrvrh8+XJGq0xPTycFBQWysrIiPz8/UlBQIH9/f6qrq2vRTkQJrm7CwsJoxIgRNHHiRCGFdM6GkpISCg4Oprlz5wrV7bJly1pQykVd31z5oaGhNGPGDDZBDggIIG9vb4qIiKCNGzeSvr4+PXr0iHx8fNgzcXd3JwsLC6GwBXHBx8eHTE1NKTc3l4iaFh709fUZ3f3y5cu0ZMkS2rlzp9B1XJurq6uj7Oxs9s4UPCYqNFeaF+wLuONETYtDnp6eQvTxsrIyMjAwYBNTQYhy4aJ5doWkpCS2SC94/O7du2Rra0sHDx5kx7KysmjKlCnsvSP4rhG1X/+zZ/nmzRtSUlKisLAwImpS5D948CCpqamRrq4u8Xg82r59+xf7DFEvEkVFRZGRkRGj6QveB5/PpxMnTpC9vT3FxMQQUVO7nTRpUotNi6KiIqHvWtWzRYtWqnYrWvGd0NjYCD6f3yIn6enTpxEcHIzhw4ejZ8+e0NTUhLa2NqKiovDhwwfMnDkTY8eOhbGxMaytrZl4RHN6nChpUJ07d8a2bdtw7tw5REREMNEN+ge168SJEzAyMsKgQYNw+PBhREREgMfj4ebNmxgwYADWrVuHjRs3Yvv27ejZs6dYRZ0AwNvbG0lJSXjy5AmOHz8OV1dXXL9+HWlpaejXrx90dXXx9OlTJCQksOuXLVuG9PR0JjQiDhDRv+wjOjo6OH36NNq0aYMlS5bg8ePHWLBgAXR0dHD9+nXY29ujTZs2rOw///wTEhIS6N+/v0hs56h4Dx8+hL+/PyIiIpCZmYmpU6fC0NAQa9euBdBEVbW0tERaWho2b96M9+/fo3379kyQpUOHDuwzCeQCFxVkZGRgYWGBCxcu4NOnTxgzZgzGjx+PzMxMeHh4oKamBitWrEB5eTnevXuHN2/eoE+fPpg3bx5WrlyJuro6dOrUCYMGDUKXLl1YDl9R0FcLCgqwZs0aJCYmYt68eWjbti0uXbqEqKgoHDt2DAEBAZCUlERwcDD09PRQVlaGy5cvs9y8t27dwuTJkzFkyBAAQPv27dGhQwdms6jqmpoW7QEAY8aMQZcuXbBp0yb4+fmhvr6+BdXc09MT7969Q3h4OIYMGYIDBw5AR0cH0tLS2Lt3L1atWgUZGZkW7eRbgxNlBP7qb3v06AFHR0f8+OOPCAwMZMc5G7p27QoVFRV06dJFSDRp06ZNcHV1FSpfFPWdk5MDQLj/W7hwIbp27YrDhw8DAObMmYOSkhK8e/cOy5cvx8iRI3Hp0iVcunQJd+/eBQCsXLkSPj4+kJWV/eY2fg2cH9ja2iIvLw/37t0DACQmJqKurg6//voriAgaGhqYOnUqnj59ilevXrH75Z6RjIwM+vXrh86dO4u0PXLvMa7PlpCQwP379zFnzhy4ubkhKCgIb968YfYBwKBBgxgN19nZGeHh4Zg3bx5+++03FsIkCFFSbpsLXj158gSfPn1CTU2NUJ0pKirit99+Q1JSEp4+fQqgScA0ICAAs2bNAgC0adOmhVDotwb9g2LP2SWYNYTrX3755RdYWlpix44dqK2txdChQ2FtbY2TJ0/C3NwcQ4YMQXJyMhvDCELU9ObExEQkJycjOjoa1dXVkJSURGNjIxoaGiApKYlp06ahbdu2uHPnDsrLy6Gnp4dff/21hbJ39+7d0alTJ5H3261oQuvEuRWtEDO4iZdgjOrly5fx5MkTAE2TukGDBqGmpga1tbXo1KkTJk6ciJ9//hn79+8HAOzcuRNKSkrQ0NDAb7/9Jha7BdMYKCkpoUuXLli1ahXatm3LJmTcC5fH40FPTw8aGhqQl5dHfn4+pKSkhGLL2rVrx8oTVUfPxR0CTfX96dMnmJmZITQ0FJ06dWKTNwsLC9TX1+Py5cuora3F5MmTMXbsWPzxxx/s+lmzZiEkJARt27YVia1fgoSExP/KR/r164ddu3ZBWVkZO3bsgIeHB5YtW4aIiAgMHz5cKGbN0tISUVFRIhkIExEaGhqwdetWeHh4QEJCArdu3YKlpSVev36NhQsXoqqqCiEhIQCAadOmwd/fH8OGDWPx2s0HcaKKiRP0EQ6zZs3CuHHjsGrVKgBNaYIePHiAiooKLFmyBIMGDUJAQADatGmDZ8+eAWiaWGzevFlocQKASFN5denSBa9evcK1a9fQpUsX6OnpITIyEg0NDRg6dCjk5eVhYmKC1NRUPHnyBL6+vnj+/DmcnJxgbW2NLVu2QF9fnykPC6oJi8rmhoYGofRVpaWlqK6uxo8//gg1NTXIyMgwFWEOvXr1grOzMzZs2ID6+nqMHj0aenp68Pb2xtixY1lsrSjA5/NRX18PDw8PTJ8+HSdOnMDnz5+F6urx48fYunUrrl27hgcPHrBrufY2ZMgQTJkyBS9evGD+xh0TpRJycnIyZsyYgY8fP0JSUpK1SwBYsWIFoqKi8OjRI4wYMQLjxo3Ds2fP8PjxY6xevRpTp05laabq6urQo0cPDB48WCxaCBxkZGRQUVGB/v37Q1dXFydPnkROTg5GjRqFjIwM5OXlQUJCAtLS0hg0aBDS09NZXPzXJsai8u38/HxERkay3yguLsb58+dx9uxZzJ8/H/PmzUNSUhI2btyIoqIiSElJsWehpqaGTZs2YdCgQYiNjYWVlRU2bNgglrhUQf+rqalhavUAMG7cONy/f5+1V26iCgA6Ojqoq6vDxYsXmfL39OnTMXToUOYjoo5159JGVlRUYNeuXfD19WWK+oLtc/bs2ejbty82b94MoKlN9+7dGwYGBjh16hQOHTok1vc6hy5dumDIkCGorq5GREQEgKY64557v379MH36dHz48AG3b99Gx44dYWpqipcvX6KsrKzFO1IcaSNbgdYY51a0QpxYt24drV69mkpKSoiIaNeuXSQvL0/Ozs6kqKhI3t7eVFxcTEePHiUrKytGk2toaKCIiAhSVFRktB5x4Uu0n+TkZPLx8SEej8dis5ojLS2NxZ6NHTtW7DTnuLg4cnBwYHHJlZWV5OzsTKtWraKSkhI6c+YMKSgoMNXMyMhIUlNTYzTc6OhoMjY2bqEsLO4Yv/+tj0yZMoXu37/fohxR0re+VHZubi4ZGRnRx48f2XejRo0iT09PKioqokOHDtH06dOFKH3ipoGmp6eTtrY2a1OCzzYpKYmGDRtGDx8+JKImRVt9fX3Kysqi2tpa2rdvHw0dOpQ8PT2/qLguKgiqBxMRXblyhYyNjVm8rLOzM3l4eDBaa2FhIQUGBpKZmRk1NDRQVVUVxcTE0IULF8Rms6DdHEJDQ+ns2bNUVlZGjY2NdO7cOVq0aBHz3eb0y9raWpo5cybFxcUJfS8uBWcHBwemYuvi4iJ0joGBAaWnp5O/vz/p6el9sRxOGVkcOH36NGt3pqamtGTJEiJqGfLh7OxMNjY2RET04cMHcnd3p3Xr1rE+88GDB2Kzmailj6SkpJCWlhalpKRQQ0MDzZ49m3bu3EmfP38mR0dHpitARPT27VvS0dEhKysroXAEceHAgQPk4uJCSUlJlJaWRleuXCEejycUj5+UlESLFy+mrVu3Cl37tdACcffZ7969IyUlJZKXl6eYmBhqbGwkW1tbOnXq1BfLOHLkCF26dElkNv4r8PPzo/nz55O5uTkNHz6crl69yo4J1uv169dp/PjxzDea9xuifq/fuHGDjfs4vZq4uDhSU1Oj9evXk7u7Owv1EAxzqq6uJh8fH7Kzs6OsrCyR2tiKfw2tE+dWtEJMOHz4MKmoqLDPGRkZZGpqyoSx4uPjyd3dnTw9Pam+vp7MzMxo27ZtbBCTnZ1NYWFhQmkdxClQdfHiRdq4cSNFR0dTQUEBETXFns2ZM+er4jYvX76kGzduUE5ODvtOHGmaiIhycnKoqqqKiXwVFRXRzJkz2USIqCnGaNSoUSz2zcbGhvT19UlfX59iYmLELtrTvG4yMzP/zz4iagjanJCQQNnZ2VRTU0NPnjwhU1NTKisrozNnzpCysjLZ29vTixcvqLy8nMrLy0ldXZ1WrlzZokxRD2KePn1K7969o9LSUlq2bBk5ODh88X5+//130tbWZp/19fVpy5YtLOY6Li5ObJN9QdEsIhIaRHl5eZG7uzsVFxdTfHw8zZs3T0j07cmTJ6Sjo8NS2QlC3AtB7969I2VlZTI2NiZFRUXy9PSkwsJCys7OZrGe3H02bxPiXqCwsrIiR0dHImqKlZw+fTrt27ePZs6cSQEBASw1mq+vLz19+pTKy8tJXl6exVN+rVxR4u3bt2RiYsKE7FJTU2no0KFsEiyYxq6goIB4PB6lpaUREdGJEydowYIFLRbevmfMpKGhIa1du5aIiC5dukQaGhr0+PFjysrKIkVFRbKwsCA/Pz+aNm0a7dix46uilaLA5cuXKTo6moiaFoq1tbVp3Lhx5OfnR/X19WRsbEyurq7s/Lq6Otq7dy95e3v/0zSAol4MEsSbN2/I29ubifN5e3uTgoICbdiwgXbt2kWHDh2iAwcOCPUTzWOhxYXmCz+XLl0iExMTysnJofT0dHJzcyN9ff0v9sllZWVkbW0t9vRSZWVlZGFhQTwej9zc3Fg9ExG9evWKAgICKDQ0lBwdHYVSNRL9Vc+3b9+mffv2iV1/ohVfRitVuxWtEBN++eUX8Pl8aGtr48aNG4iOjkZJSQmGDh0KPp+PyZMnQ19fH2lpacjOzoaJiQni4uIY9a9fv34wMjISSusgDlpOeXk5XFxcsGfPHgDA2bNnYWVlhcrKSqxevRpZWVk4f/680DX0D6rWsGHDoKqqir59+7L4G1HbzNGX2rRpg3v37sHFxQW3b98Gn89Hjx49UFNTw2ycNWsW+vTpgx07dgAA/P39YWxsDC8vL0yfPh0dOnQQKZWSA/0jfkxCQgLFxcWMwnf16lV8/vz5/+QjooaEhARev34NXV1drFq1Cra2tti5cyfatm2Lt2/fwsTEBEePHsWaNWuwZ88ehISEYP369ejUqRN8fX0xd+7cFmWKkqKYkZGBwMBAnD17Fl26dIG+vj5yc3OZDws+bxsbG1RVVeHGjRsAmij9J0+eZHGKU6dOhaysrFh8hKMlJiUlQV9fH/b29vj9999RWloKa2trZGVlISoqCpMnTwaPx0NcXBzS09MBNFGFbW1toaGhAUA4zYyo6poE4piBJoq2i4sLNmzYAEdHR5w+fRpBQUHIzs5GdHQ0+vXrB2VlZXz8+BGxsbFCVEQOXNwkiYEuLCUlBXNzc5aia/DgwVBRUcHLly/h7u4OaWlp+Pj4oKioCLm5uUhNTUWnTp1gbW2N1atXC8VbNi9XFODigfv37w9TU1OkpKQgMTERQ4cOhZGREfz8/AA09Y+SkpKoq6tD9+7dMWTIEKSmpgIA5s6di4CAAEyaNEksNjf3kerqatjb27P2BgAeHh6IiYlBbGwsdHR00KdPH5w9exb9+/fH8ePHoauryyjGLi4u6NSpk1jaY01NDd69e4fDhw8jMTERz58/R+fOnfHzzz9j1qxZkJaWhoODA27cuMHowzIyMqiqqkJRURELj/gSREVxbq4l4ubmhpMnT6KiogJLly5FQ0MDdHV1MXbsWAwaNAhPnjxBaGgoSktLIS0tza5vThEWR3vkfu/Vq1dMayA2NhY//fQT+vbtiyFDhiAoKAglJSWIiIhokeKuc+fO2LFjh9jSS2VnZyMlJQWdO3fGlClT0L9/fzx69Ag7d+5EXFwcs+3Vq1eYPXs2RowYgSdPniA5ORmAcIy+srIybG1t0aFDB1Z+axzz90PrxLkVrRATbt++jfz8fMjIyEBNTQ08Hg+NjY0oKChgnWCfPn1QX1+PgoICaGlpYdCgQS1iUEX5kvqSSNerV69QUFCAqKgoLF++HNu2bcPr169x5MgRtGnTBq6urti9ezeePHmCO3fuoL6+vsXkmP4hWCGqSXNzu+/evQs/Pz/07NkTw4cPR1RUFHr37o2uXbvixo0byM/Ph4SEBBoaGtC1a1eEhobi9evXkJWVhbGxMSZPniyWOC3ObgkJCVRUVMDe3h5mZmbw8PBAZWUlE23iYsSB7+8jzQelfD4fISEhUFZWxtWrV7Fv3z7o6+tj+PDhGDduHKSlpXHq1Ck2KcrIyICxsTEAsFhycYCze/DgwVBWVkZGRgYePXoEBQUFTJs2DadOnUJ1dbVQ7GGHDh3Qp08fFpc6Z84c7NixA6NGjRIqW1Q+0vw5XrhwAW5ubtDV1YWPjw9iYmJw6tQp8Hg8TJkyBbdu3cLbt29hbW2N9+/f4+bNmywGXltbG8OHDwcg+gU3QXEkDtLS0hgxYgRiYmLQs2dPAICCggLGjh2L2NhYvHr1CtOmTcPo0aOxfPlyzJgxAwUFBS1sFVWsO7egJohp06ZhxowZLNbdw8MDqampyM3NhZ2dHeTl5eHn5wdpaWk8f/4cAODg4ICgoCB07Njxm9vYHJWVldi7dy9ycnJYXLiUlBTk5OQwZswYHDhwAECTuFpBQQFCQ0MBND3/Nm3a4Pnz5+jVqxfLe92mTRsMGDAAgOgnQ1/yESkpKfTs2ZMtZAJNfYScnByOHTvGYs1jY2MRGRmJAQMGQE9PD/7+/pg6dWoLsShRgOtH2rVrhzFjxuD169cwNTWFrKwsgoODMWzYMFy6dAmfPn3C9OnTMWXKFDg5OeHp06coLS3Fy5cvMW3aNJHZ988gJSWF2tpaVFdXA2haUCsoKMCmTZtQUlKC4OBgpKenY+TIkRgzZgxsbGyQk5ODo0ePoqSk5KuTNXG916urq7F582YsXboUQNO7sGvXrkI6FUZGRjh+/DgTTRW0jWuTol5Yqa6uhq+vL06ePImysjJoaGhg/Pjx+OWXX/DTTz9hxYoVeP36NXg8HqSkpHD37l3o6OigQ4cOQrHOzSGOBaFW/M9onTi3ohXfGCdOnGCJ6rmOv7q6GqqqqtiwYQOICCkpKRg8eDD69euHI0eOsGu7du2KDh06oHfv3gCAwMBAyMvLC5UvipcUN0jiXozv379nxzIyMvDjjz+irq4OQUFBUFdXx8KFC9nulYWFBeTk5GBnZ4cXL160EPYRlc2C4Ox+9+4dAGDo0KGIiYmBrKwsNDU1kZubi+vXr8PDwwMpKSnw9/fH8+fPsXPnTowePRq+vr4YOHAgK0/UO+PN6/vTp09scr9p0yY8fvwYBw4cgLS0NAYPHszUb4Hv5yPNdxs4FBcX4/Lly5g6dSqAph2vQYMGobCwELa2tujZsycWL16M5cuXQ1dXF6NHj8bo0aO/uX1fQ3R0NHJycoTsnj17Ntq0aYOrV6+itrYWs2fPRrt27bB3716ha7nBveBEmbtPUUJwQYVDdXU1wsPDsWbNGlhYWIDH42HgwIGIjY1FcnIyLCwsUFdXh8uXL2PAgAFQUVFBr169vovoDefX+/fvR0BAAE6fPo26ujrY2Nhg0KBBePz4MTvXwsIC1dXVuH79OqSkpJj6cHx8PH744Qex2JudnY0FCxbg6tWrAMAWTiQkJODq6oqXL1/i6tWr6NSpExYsWICzZ88iNzcX69atg5ycHOLj4/Hp0yc2IdHS0hKL3WFhYdi+fTv09PTg4eGBhw8fAgB++uknaGlpIT8/H+fOnUOXLl3g5eWF7du349ixY8jOzkZwcDDs7OwwadIkdOvWrcVEWVx99qFDhxAYGMhEvRwcHFBbW4uDBw+yc93c3PDw4UNcuHABo0aNwrRp01BVVSU0kWtsbGSsDFGAm7QIlp+Xl4eff/4ZQ4YMweTJk9GrVy9MmDABubm5uH37NoAmEba0tDSsW7cO7u7u6Nq1K1s4FDWaTzxTU1OhpKSEixcvgogwcuRISEtLQ0pKCj4+PiAihIWFIT4+HkSEyZMnY+XKlVi6dCm6dOkitp3l+Ph4FBYWsufLTYLbtWsHR0dHZGVl4enTpxg9ejRev34tlP1CWloahYWFOHfu3FftFfXCSvv27aGnp4f8/HzExsbi559/hqKiIkpLSyEvL4958+Zh48aNOHHiBGbNmoXKykr0798fPB4PGRkZjNEkTrtb8b+AOPjgrWjFfwMaGxupsrKSHB0dWc5R7nsOhYWF5O3tzeLmIiMjaerUqeTp6UkHDx6k6dOn05o1a6i2tpZdJ458kxyioqJIXV2d1NXVaePGjVRcXEyRkZGkqKhIKioqtGDBAhYPp6+vT0eOHCGipry1XJytuCBod11dHdna2tLs2bNZPLCnpycdOHCA6urqKDAwkGxsbKi8vJxSUlLI29ubzMzMaN68efTu3bvvZndkZCSpqqqSqakpaWlpUUZGBhER3bx5k2bNmkV37tyhixcvkrKyMnl4eHw3HxHEtWvXaP369RQVFUXZ2dksFjQ0NFTIFkdHR7p69SpVVlbS3bt36fjx4+z+xIUTJ04Qj8cjVVVV8vX1pby8PFZn586dI0tLS7p8+TIREZ0/f54UFBToypUr9PnzZ4qKiiJlZWXy9fWlmpoascfzETUJyuzbt48ePnxIZWVlFBAQQB8/fqTLly/TvHnzKDg4mObNm8eE906dOkWzZ8+m9PR0sdrb3P/y8/PJ2NiYTExMaP/+/WRsbEx2dnaUnZ1NMTExNHz4cCHdg0OHDpGenh49f/5cqBxRx18/f/6cXr16RdXV1fT7778LiU0J3tPWrVuF9Cnmz59Pa9euZfG0CQkJX8y5KwoIxrp/+PCBlixZQjY2NrRw4UJSUVEhBwcHun79OpWWltLevXuF8kcHBweTvb09GRsb06JFi1geZHFB0Cc5MS9TU1PaunUrqampsZzjp0+fpsmTJ7OY9qysLFJVVSVLS0sqKCgQa3/3+PFj8vPzY5+jo6Np8+bNlJSURDU1NVRaWkqOjo5Mq6G2tpbWrFlDnp6elJeXR5WVlXTs2DFSUFAQir8W53u9pqaGiJpyik+aNInU1NTo5MmTlJOTQ3p6euzdXVJSQh4eHsTj8Vgee3H2IzU1NZSQkECGhoYUEhJCRESBgYHk6OjIhLOqqqpo/fr1bIy1atUqMjY2ppCQELp8+TLZ2dnRnj17aOTIkUyLRdT38Pr1a/ZsBZ+rt7c3LVu2jLKzs6mkpIS8vb2ZsOCFCxdIX1+fFBQUWO7rT58+iX0c1Yr/PVonzq1oxTeAoFDDkydPSElJiSIjI1scI2oSFNLT06OIiAj2eefOneTi4sKUccWNt2/fUkREBLm7u9Pt27fp4sWLZGRkRH/88QcREeno6JCbmxs7/86dO2RsbEyfPn0SKudLAleixOfPn9kLa8OGDUz1Nicnh3bv3k27d+8moiZ1WEdHRyFxpOYia+K0++3bt3Tt2jVatmwZRUZG0v79+2no0KGUkpLC7LCzs6Nly5bR58+fKSEhgYKDg8XuI4K+W1VVRcuXLydVVVXasGED2dvb0+zZsykzM5PWrVtHHh4e9PbtW3a+s7Mz/fnnny3K5PP5Yhs01tfXk6GhIRkbG9Ps2bNJW1ubnJyc6OXLl1RbW0s+Pj7k7e3NBOS2b99OOjo6ZGxsTOrq6kz4R9z4/Pkz2dnZkYqKCrm5uZGFhQVTd4+JiaFZs2Yx21auXEnjx49ng6/mKvfi9GtOLTYyMlJoEvr06VPi8XgUGhpKjY2NZGlpSU5OTuw4n89navbiwsePH8nc3JwJOT179oyMjIzo6NGjRCTs+3l5eaSpqUlnz54loqb+b/To0fTo0SOhMkXt118SRwsJCSEXFxe6fPkyZWRk0Nq1a0leXp5sbGzI19eX9PX16eDBg0T010KE4KJFQ0ODWH2E63fDwsLI2dmZfV9WVkYaGhp06NAhys7OJmNjY5o7dy5FR0eTlZUV7d27t4WPiMPutLQ04vF4FBMTQ35+fqSsrEzW1takrq7OBPguXrwopM4fExNDjo6OJCcnR05OTlRcXExTpkyh48ePE9Ff7UTUyMvLIzs7O/Lz86M3b95QdXU1eXt7U1BQEMnJyVFCQgLNnj1bSMyuqKiIAgMDKTExkYhaCnJ9azQ2NlJFRQX5+fnR+fPnic/nk7+/P7m5uVF+fj7dvXuXFi9ezCbSRE3vTxUVFfZ+OX78ONnZ2ZG2tjZbCBU8LkokJyfTuHHj6ObNm6yOuL7j0aNHZG5uTseOHSOiJnHPOXPmMCXye/fukaKiItna2gqNQcS5MNSK/z1aJ86taMX/J5p3blVVVeTu7k4ODg6koqJCWlpa7Jhgp1hSUkLBwcFkYGDwVdVmUXWcXyv34MGDQqvMRE3pNWxsbCg1NZUSEhLIzMyM5s+fTy4uLiQvL89eBuLAl+xOTEwkc3NzCg4OJqKmXR8/Pz/S09Oj4OBgcnNzIzs7OyJqGqhs376dbG1tKT8/X6gcUapTfq2+Dxw4QDwej1atWsW+MzY2plWrVjGF79evX5OSkhKdOXPmi+WI8uUquNNXW1tLVVVVlJGRQYaGhkK/y+PxaN++fRQTE0Pu7u6kqalJcXFxtGzZMtLQ0GC7BBxEOdBtXh/cc/3zzz9JV1eX4uPjKTExkSwtLWnKlCm0du1aWrlyJTk7Owv5ckVFBWMtfK3sb4kv+V9cXBwtWLCAfRbchThw4ACbdBYVFdGSJUto9erVdOPGDZHZ+CU0T4t1+PBh5s+HDx8mb29v+vz5M3l6etKkSZPowIEDlJmZSURNSs88Hq/FxFMcEPRBbsAdExNDRER79+4lPT09KiwsJKK/2kFZWRlZWlpSSEgIu/7OnTtisZfP59PGjRuZUnBFRQWdO3eO7t27xyYdy5cvp+XLlzM/efHiBe3Zs4fU1NSIx+ORvLy8UFo4DqJW5m1e/tOnT8nS0pLevn1LXl5etHbtWuLz+WxB4OjRo6SpqUlETerrzs7ONG/ePLYoJG5w9vv4+JCSkhJTKidqSoPk6upKb968ofLycvL29iY7Ozv2THJzc4UW3sLDw4nH47H+/VujeR+VnZ1N69ato2XLlpGHhwe5ublRZWUl+fj4UHh4OF26dIkcHR1JV1eXNm7c+N0na8nJycTn86mqqorS0tLIysqKqUxv2rSJnJ2d2eJheXk5mZmZkaKiIls4//z5M1Mqz8jIoPnz54uMBcLn82nDhg1sAcrZ2ZlcXV1ZGxXsYwIDA8nZ2ZkyMzOpsrKSduzYQQYGBkxdm+sTW/H3QSthvhWt+P/AuXPnsHTpUmRkZAAA6urq4OPjg+rqaqxZswYLFixAXV0dtm7dCuCv+CugKUZVUVERMjIyLCaNw5fiqERlMwnE/1hbW2Po0KGoqqpiNsyaNQsdOnTAuXPnMHr0aAQFBcHKygoKCgq4ffs2zM3Nv7mN/4rdnH1ycnLQ19fHlStXcPbsWcjIyEBGRgYeHh6QkJBAQUEB4uPj8erVK8jIyMDY2BjBwcEt4iZFpU75z+rbxsYGQ4cORV1dHYtDW758Oa5evYrnz5+Dz+fj119/haqqKqqqqoTiDUXpIxw4peXbt29DVVUViYmJePHiBXr37o2Ghgbs3bsXU6dOhZ6eHlRVVTFlyhSsXbsWkydPxqVLlwAA4eHh+Omnn4TKFVXc5Jfqmnuu+vr66NSpEyIjIyEnJ4fDhw9j165dkJSUxLVr13Dt2jWEh4fjxYsXAJoEZIYOHQrg63Hd3wLcc+TsrKqqYsfi4+PRvXt3EBEaGhrQo0cPlJaW4vLly2jXrh3S0tLg6emJmTNnon///vDx8YGqquo3t/GfgbNbQkICdXV1iIuLw7hx4wA09YcpKSlQV1dHY2MjoqKiYGNjAxMTE9y4cQNDhw7Fli1bMHLkSLHZe+PGDbx9+1bIB2fPno3OnTvjxo0bKC0thaamJnr16oU//viD3Rv3t76+HmPGjGHfcQJPJOK4T0lJSdy9exdr165Feno6ZsyYgWvXrmHFihVYtmwZPn36BGNjYxQUFDBhoeHDh8Pe3h4nTpzA0qVLMW/ePPTq1atF2aJW5m2uPTF8+HAkJSWBz+ejZ8+eyM7ORl1dHdPFmDBhAgoKCpCVlYUBAwYgMDAQoaGhTAxK1OJIjx49QnFxMQBhnYvVq1ejtrYWWVlZqK2tBdAkFFhTU4OoqCh06tQJhoaGyMvLw7BhwxAREYE+ffowHZCGhgbMnTsX7u7uaNu2rUh8huujHj16BKBJOCsrKwszZsyAu7s7ampqsGrVKigrK+P48ePQ0dGBtrY2MjMzkZqaylTZOYhbiGrw4MHw9vbGhg0bwOPxMHHiRCQmJuLly5cwMjJCXV0drl+/DgDo1KkThgwZgu7du+PmzZuoq6vD/fv3YW5ujoCAAMybNw8jRoxo8f75VpCUlMS9e/ewZs0aAIC7uzuSk5Nx7949Jo7KvTsWLFiAjIwM5Ofno0OHDpgyZYrQ2I/TVvmSMGsr/k3xHSftrWjF3xJlZWVkampKI0eOpCNHjlB9fT2Vl5eTgYEB27ng8/l069YtmjBhAluFFFzRra2tpZKSku9ms+BOALerEhsbSyNGjBDabTt37hwtWLCATp8+3aJMcVD8vma3oP3Xrl2j2bNn0/nz50lNTY0SEhKIz+eTn58f8Xi8FjtD4lhZ//+tb09PT5o3bx7bFf8ecbVERJWVlTR//nwyNzenmzdvElFTbJ+ioiKpqamRoaEhPX36lIiIzMzMaMeOHexaQRqiOHJN/rO65v5PSEggZWVlunHjhtBuempqKgUFBdHixYvZLqO4cfv2bTIxMSFnZ2cKDg6m4uJiioiIoKlTpxLRXz5w69YtcnV1pWfPntHNmzdp/fr1lJyczMoR945RVlYWGRgYUGhoKBUVFZGTkxOdPHmSiJq0HGbPnk2bNm1i54eFhZG5uXmLPLvi8PHIyEji8Xg0depU8vHxoaysLOYHV65cIUtLS0ZXjY6OpkmTJtHZs2epqKiI7ty5Q2pqauTl5SXWHMGCSE1NpREjRpCDgwMLAXr8+DGtXr2a6WUEBgaSq6sro+p/rzyvgn7Y2NhI7u7upKamxqi/v//+O+3evZuKiopozJgx9OeffzL69q5du2j58uVfLFOUflJXV0e+vr7E4/Ho3LlzQsc4Pzl27BgpKioK9RP79u0jBwcHio2NJaKmXfLmOg7iYNlwjLZjx44Rj8djFPLDhw+TkZERETXd4/z582nFihWkra3Ndm8fPXpE6enpIrOxOZrnoxfEwYMHyc7OjuLj46mwsJCcnJwoICCAGhsb6eTJk6Surk62trY0Y8YMWr16tVB7zM3NpT///JM2b94s1C9+a3DtimuTXM7zTZs2kbGxMb1584aISIhJoa6uzmj64h77teLbo3XHuRWt+BdBAvkAf/jhB/Tp0weZmZl49uwZKisrUVVVxfLsSUpKYvLkyRg4cCB27twJQHi3rU2bNujatWuLPJbisplblSYiSEtLg4igpKQEeXl57Nixg60+q6urY+zYsRgxYoRQeSTi9FL/k93c6joRYebMmZg/fz5ev36NT58+ISwsDJKSkvDx8cHRo0dbpP4Q5U7t/29919XVAQCWLVuGmpqa777636FDBwwdOhRJSUksPY2GhgZ++ukn/PLLLzh16hTk5OTw7NkzVFdXQ19fn13L7R41NjaKdEfrX6lrKSkpEBHGjx8PBQUFhIeHsxygAMDj8eDm5ob9+/ezFEmisrWqqgre3t7Iyclh3589exbr1q2DiYkJlJWV8ezZMyxduhRTpkxBu3btsGvXLrbDVVZWhoaGBowZMwYzZszAypUrMWrUKDQ2Noo8Bc+XICkpialTp+LatWvQ0tLCw4cPMWTIEBARevbsCWtra7x79w6GhoZYtGgRduzYAVNTU3Tq1EmoHHHkotfS0sLEiRPRq1cvpKSkwMHBAfb29khOToaGhgYGDx6MBw8eICMjAxoaGli8eDEOHToEZ2dnrFu3Dg4ODggMDGxhu7gwdOhQLFq0CLdu3cLAgQNBRBg3bhzU1dVRVlaGpKQk6Onpoby8HBcuXBBiXHDtRFx9iKSkJMrLy1FSUgIJCQkMHjwY2dnZ8PX1xYcPHzBgwABUV1ejR48eWLp0KU6dOoUlS5bA0tISp0+fxpw5c75Ypij9hCu/a9euSEpKQlpaGoC/+msAMDc3R5s2bViKL6DJr6qrq/H06VPU19djwIABGDRokFCucVHbXVxcjA8fPkBCQoKlJbx8+TLOnTuHcePG4ccff8TLly8hIyMDX19fdOjQAa9fv0Zubi6Apl1+rt2KA5zyuSDDhvNNfX19dO7cGZGRkejYsSPU1NTw5s0b3LlzB/Pnz4efnx8GDx4MDw8PrFu3juXsJiL06dMH+vr68PT0bJEu8P+C4uJirF27Fvfu3QPwF4ti6NChMDAwYPnRly5disrKSkRHR6O8vBySkpJo06YNEhMT0b9/f6irqwMQ39ivFaJD68S5Fa34J2hsbERiYiJqa2sZJREAlJSU0LVrVxQVFeHOnTuQlZVFv379EBISwlKStGvXDhISEjh79iweP378xRdo8zyW4rL5wYMHKCsrg4SEBMt9CQABAQG4ffs2bt68CaCJEuXp6ckolYLUxW+N/x+7OcybNw+zZ8+GtLQ0cnJyUFRUBACYNGkSANHnNf6/1vetW7cAAL1798bFixcZxYyrZ1FMippTw5rX0ZIlS9C2bVs8ePCAfbds2TLU19ezXNPW1tZQVVVF//79W5QvKnrz/7auuftyd3dHeno6rl27xq4T9GNRUuUkJCTQoUMH3Lp1i6XZqaurw+3bt7FkyRLo6urCwMAAy5cvBxHh7Nmz2LZtG86ePQsXFxesWLECfn5+UFNTY/XA/ZWUlBTZpPmf+chPP/0ENzc3HDt2DA4ODuDz+fjjjz+wcuVKfP78Gfr6+ggICICbmxvmzp2Lu3fvYubMmSKx82s2Ct6Dubk5ampq4ObmhoCAAEhLS8PJyQmrVq1CXV0d6uvrWb9nZWWFsLAw/P7777hx4wYMDAyEyhIF/qf2aG1tjS5duiAzM5P57a+//oqSkhJUV1djyJAhmDhxIsaOHSvk16LsQ76E1NRUuLi4sAmmiooKTE1NISMjgzNnzuDNmzd49uwZgKZUZBs3bsT8+fOhpaWFuLg4TJkyRSx2cmhoaICUlBQGDx4MCQkJVFZW4u7duywVHXcOAKxduxZnzpxhE+t+/frB29sbLi4uQmkYRT3R51BQUICNGzfC0dERQNPipry8PDp16oS0tDTs378fvXr1Ys/+t99+w4oVKxAUFMT6Eg6itLe5b9+8eZOFswFN9UVE6NGjB2bMmIHc3FxER0dDS0sLvXv3Rnh4OHJycqCgoIBly5axdG98Pl/kdZ2cnIxTp07B2toaTk5OOHz4MLN76dKlyM/PR0hICNq0aQMbGxvEx8djzZo1ePbsGXx9fWFra4spU6awRQ0Oohj7tUI8aJ04t6IVX8HHjx+ho6MDU1NTtmvcpk0bAED79u0xa9YsyMvLIykpCY8ePYKvry+ePHmCjRs34sWLFzh16hT69u0LX19fDBky5N/K5pSUFDYhkpCQgJSUFPh8Pn766SfMnTsXL1++FCpX1Cuj/792cy8eaWlpDB8+HMePH8euXbta7B6K6gUlqvoWR7wTt3L++vVrZhfw17Pu0aMHHB0dERwczAYK8vLyWL9+PRwcHDB27FhcvXoV9vb2IrcV+P+va0lJSfD5fMjKymLOnDmorKxku0iCEOXuODfw/uOPP3DmzBmkpKSgTZs2eP36NVtoA5ri3aZNm4anT59i5MiR2LVrFzQ1NfHjjz8iKioKenp6AP6aBIl6MvQ1H+HA+emsWbMgKSkJTU1NZGdnY968eXBzc0O3bt0wdepU6OjoAPirHkSFc+fO4ezZswBa5kpXVVXFzz//jAsXLuC3337D3r17ceDAARbjfO3aNYSFhSExMRFA06Ihj8cTuk9R+si/2h4DAgKYPX369EH79u0Z02nx4sWYNWuWyGwUxNfeCcOGDYOBgQHi4+Nx8OBBSElJQVJSEh4eHmjfvj3y8/ORkJCAp0+fAgAGDRoEbW1tGBkZARC9j0RERMDe3h537twB8Fc9y8vLY9asWWjfvj2Sk5NZrvHmLKGBAwdiw4YN7BlwPiLKHf2vlf3DDz/A29sbbdq0gY+PD168eAENDQ3weDxoaGggLS0NoaGhSE9PZ+VISUmxiaeo3+vN2yB3H3l5efjll19Y/y0IdXV1DBo0CLGxsSguLsacOXMwbtw4oXd6c30IUWL06NFwdHTEgAED8Ntvv+H48eNYuHAhDh48CBkZGaxZs4a9I3V1dWFtbQ0pKSns3bsXZWVliI6OhpWVlcjtbIX40DpxbkUrmuHdu3f49OkTOnfujNGjR2PAgAE4ceIE9u/fj1evXgFoorLevHkTZmZm6N27Ny5cuIDevXsjMDAQtbW1WLt2LUJCQmBqagoTExN06dLl38rmbt264e7du4w2SgJCKP7+/nB3dxcqX1QTz29ht6B9w4YNQ7du3UQ+8RR1fYtiQEBEQvWSkpKC2bNnw9raGitXrkRKSkqLa0xNTdG7d28EBQUBaBqw/PTTT5g+fTpMTU3Rq1cv8Pl8kQ7AvkVdcxPMpUuXYsmSJWKnNXMD7/Hjx0NRURFBQUEoLy/H9OnTER8fzyYLMjIyKCoqYp9HjhwJfX19ODs7o3fv3iKv63/VRwQHxESEsrIy/Pzzz9DR0cGBAwewadMmmJmZfbEeRAkpKSmsW7cOd+7cwcOHD5md3D15enri6dOnuHHjBurq6jB06FCsXLkSx48fh4uLC/r27ftFQaF/l/bIvW+0tLRw+PBhqKqq4ocffsDgwYMBiIf2ztn9pd/iJjQ6Ojrw9vZGeHg4Xr58ibi4ONTW1sLR0RGjR48GAKGQCUGIykcaGxtRVFSE06dPIyYmBt7e3nj06BELhfj06RNqampgZWWFqqoqxMXFoby8vAVLaO/evfDw8GjhE6JkfXypbI6i3KNHDwQFBaG2thahoaFITU1Fp06dMH78eHh6eqJ9+/ZITU39oo2i9heu/OrqapiYmMDNzQ3Pnz9HQUEB2rVr1+JcbmI/c+ZM5OTk4MGDB5gyZQqsra3Rtm1bdq4o6rq+vp7R1wXbZY8ePTB58mT88MMP6Nu3L6Kjo6Grq4vz589j9uzZeP/+PcrLy3H48GEAwIwZM7B582Zs3rwZQUFB+OGHH0Teb7dCvGidOLeiFQIoLi6GtbU1zp8/jy5dukBTUxPy8vKYOHEiysrKsHLlSlRUVEBJSQlt2rTBq1evMHfuXOTm5uLChQuYMmUKNm7ciODgYFy5cgXy8vIij2X5/7X548ePQqvugnHDgOhj4r6V3YLgPotyJfrvWN/cRF1KSgqlpaUoKCjA3bt3YWFhgT179iA7OxsXL15kA0XBFX1PT08cPXoUHz58aDFgEXWs+7eq6+a7d+LaHRL8Pe57f39/3L9/H4mJiZg6dSrKy8sRGBiI6upqFBYWIjU1FcrKyi3KFYeuwP/GRwQXrcrLy5GdnQ0iQvv27TFu3DiMGzdOLINFrl65HUEJCQnY2dkJLU5JSUmhsbERQ4YMgaqqKqN+chg0aBAcHR0REhKCvn37itzm/0t79PLywvv37wE0hR/s2rULHTt2FLnNHDi7P336hPPnz+PJkyeorKxkx4CmZzJx4kTY2toiIyMDxcXFCAsLAwC4urri8OHDYtkZp3+o0gcFBeHy5cvo2bMnTExMoK6ujh49eiA8PBx79+4F0BTu8ezZM/D5fOjr6+Ply5dM+Zi7Z6Bpl3f06NEi921BlkN+fj78/f0REhIiZBPXDgcMGAA7Ozv88MMPOH/+PKtrTU1NhIWFYfny5SK19WsoKyvD+fPnUV9fDwcHB8jKysLJyQknTpxAx44dUVpays4VXCBQUFDA2rVrGcMGEP3u+IULF3D8+HEAf40fuAXMESNGQEVFBSdOnEBJSQlMTU1x4sQJODo6st387du3C/UpnTt3BvDXLn8rLfs/B60T51a04h9obGxEjx49YGBggLt37yI1NRWTJk1C37590djYCHV1dQwcOBCenp64cuUKxo4di/r6ekyePBk9evTA/fv32Sp6nz59AICtzIuq0/y/2NytWzc8evQIBQUFQmWKIyZOFHaLA3/X+uZ+Y8uWLZg5cyYWLlyIP//8ExoaGhgxYgR0dXXx/v173L59u4UtysrK2LRpE/r16/fVckWBv1tdN08TFhcXx+KZJSUlGT3/hx9+wKJFixAUFMTSBl25cgVLliyBjo4OBg0ahIULF7YoX1y7Q/+qjwjak5OTA0dHR7Rv315o4UDU/gH8Vd8SEhLIysqCgoICJCQkWggacli6dCny8vIQERGBmpqaFnaKI0zi/9IeFRUVsXXrVlhaWmL27Nlisxn4qy6PHj0KLS0txMTEYMWKFfD398erV69YGh7u/vT09KCrq4v27dsjNzcXeXl5AMDimEU9GZKQkIC0tDTevn2L6Oho5ObmQlFRET/++CMGDhyI6dOnIyYmBkFBQSgpKYGenh5evnwJNTU1dOnSBVFRUfj48eNXy/7WyMvLYzvg3OTt7t27MDQ0RElJCR49egQXFxdcu3atxe//+uuvcHNzg4qKCgoKClh6PS5M7HukO0pLS8OaNWvw+PFjTJ8+HStXrsSuXbswatQoxMXFwcLCAjdu3GBx5sBfPsFR4MUhssbh7du38PDwwO7duwH8xX7o0KEDlJSU0LdvX5aqrlu3bjAxMUFwcDC2b98OT09P/Pjjjy3KFDe7qRViwL8qv92KVvwnQjD5PJcioba2loyNjWnLli1UW1tLKSkp5OjoSDt37iQ+n09BQUFkbm5OI0eOpDNnzhAR0fv376moqKjV5la7/y3s5lKgcH8TEhIoPDycnJ2dKScnh7Zu3Ur6+vosfVp5eTmtWLGCvL29KTs7W+hexYm/Y103x/nz58nZ2Zn09fXJyMhIKD2JYGqaSZMm0b59+4iIqKioiNLS0ljqOu58UeJb+AiXqkecadO41DscLly4QB4eHnTu3Dn2zJcsWUILFixoYRuXSubw4cO0YcMGodRporZZ8O+3ao+i9pEvpYEqLS0lU1NTevDgARERZWdnk76+PllaWn61nDdv3gj5trjA+Wd2djbNmTOHjh07RkREd+7cocWLF9Pp06cpIyODzMzMaPXq1bR48WKW7isyMpJ27txJlZWVIreTq2NXV1fy9fUloiYfcHZ2pgULFjC7iYgOHDhAkyZNotLS0i/ea0FBAUvv9T0QFxcn9Nnb25vs7e0pKyuLiJrsNDAwoBMnTtCOHTtIS0uLtLS0KC0tTax2cm2H6xPi4uJoxIgRpKioSC9evBA6hzvv4sWLNGfOHHry5AkREUs31Yr/LrQuhbTivxa3b99mYjYAmIhQmzZtsGDBAsTFxSEhIQEjRozA5MmT8fDhQyQnJ8PNzQ36+vqor6/H27dvAQD9+/dHjx49RE5v/jva3Gq3+OymZvGH3N8LFy5g/fr16NevH/r27Qtzc3P89NNPuHfvHvLz89GpUyeoq6sjJycHly5dYvcqTvzd6hpouYuze/duBAUFQU9PD1OnRdCqxgAAYENJREFUTkVVVRXOnTvH7kdCQoLR/zw8PBAUFIT8/Hz06NEDPB4PsrKyLB5OVPX/LX2E25FpToUXJTgGT2lpKXx9fbFv3z7IyMjg2LFj2Lx5M4CmWObk5OQWO3Pc/5aWlvD29hZSQhYFRN0eRc0K4ny2srKS+frdu3dRUlKCiRMn4unTp1i6dCkaGxthY2ODiooKdq0gBg0aBFlZWZG2Rz6fj4KCAnh7e6OsrAxAk382NjaiX79+UFdXx9WrVxlzZeTIkYiOjkbXrl3h5+eH9u3bIzY2ltW3lpYWlixZwoTXRIW0tDQmEGhqaoqIiAhMmzYNr1+/hoKCAt68eSOUntDGxgbt2rXDxYsXAfxV11xb7N27N7p37y6WHebmv5Gfn8/CBzi4uLggPT0d8fHxqK6uhrS0NGRkZNC7d2+4uLjg9OnT8Pf3ZzvM4oAgLZzrs7p06QJHR0fIyckxgT7B9iUlJYUJEyaAx+MhMDAQAFqIm4mj/2vF90frxLkV/1Woq6vD1q1bQURQUVHBkCFDhPIych3lnDlzmJJtQ0MDlJSU8PPPPyM0NBSNjY3Q09NDeHg4ix0SJQX072hzq93it5v7DSkpKZSVleHy5ctMIdjT0xPDhg2DhIQEysrK0KtXL8yYMQNv3rxhMXPKysoYN24c5OTkRGZfc/xd65obIHF1ffHiRVRUVOD58+dYtWoVZsyYAXd3d8yfPx9v375l8daNjY1sgDt37lzs2LGjRZoSUcfDidJHRGV38wF6TEwMnJ2dkZOTgytXrmDDhg1wcHDAy5cvERMTgwEDBsDU1BQbN25Ebm4ukpOThXJdkxhi3YG/X3sE/qobSUlJVFRUwN3dHa6urti4cSOApvy1OTk5WLx4MZydnaGiooILFy4gKSmJUVy/JkIlqvZYVVUFKSkp/PDDD7hw4QKL8RXE4sWLwefzER0dDWlpaaioqKBjx444evQoBgwYAG9vb8ydO5flpBdM+SYqlJSUYP78+UhJSQERITExERUVFZCUlMTYsWNhYmKC0aNHo6qqCiUlJey6cePG4cOHDwC+Xqei1PkQjLmvqKhAYmIiiouLISsri6VLl+L06dMoLy8HAPTt2xcGBgY4duwYCgsLUVZWhtTUVNbvde7cGWPHjhWZrV8C1yaXL1+O1atXY+vWreDxeHB0dISsrCxu376NN2/eCN0rdy/6+vqwtLQE0HKi3BrH/N+B1olzK/6r8PnzZ1y8eBEBAQEAgDVr1iA8PJylxhDcEbK3t8e1a9dQWFiIAQMGQE5ODvn5+UztdNSoUQBEHzv0d7S51W7x280hJCQEqqqquHr1Ktzd3eHh4YGKigqYmJggKSmJ3Yeenh5kZWVx7do1tsLu6uoq1jyqf9e65gZI0dHRMDc3R2JiIjp16oSUlBSh31dVVYWkpCSuXbuG6upqlq+UG3BpaGiI3NYv4e/iI1xdcZMAToRqwoQJkJGRQVZWFmMojB07FhMmTMCRI0cAAMuXL8cPP/wAFRUVJCQkCE0wxJnb+O9S1xy4uomNjcWaNWsgJSUFDQ0NnDhxAnv27EGPHj2gpqaGly9f4u7du3BycgIAXL16FdOmTROrrUBTH6Krq4ujR48CANatW4d9+/YJTSw55oqJiQmio6Px8eNHjBo1CgoKCrh37x7i4uIgISEBf39/1iZFnfKNz+ezOFkHBwcYGhpCQUEBISEhqK+vx40bNyAtLQ09PT0kJCQgNjaWXZufn4+pU6eKxK5/Bq4v5urk6NGjUFJSwtq1a2Fvb4/s7GwYGxvj559/Zn060NQPvnv3jonycQtH32uHlrOTz+dDS0sLz549w8qVK/H8+XMYGRmhuroa169fF7pX7t4nT54MdXV1AK0T5f9aiJMX3opWfC8IxqpERkaSgoICi7mxt7cnU1NTFq/S0NBA9fX1VFtbS1OnTqVbt24REVFZWZlY4yb/jja32i0+u/l8PrOZi5HLz88nc3NzFn/4+PFjWr16NVlbWxMRkZ2dHfn7+9PHjx+JqCnWb+vWrULxcuKIV/271zWfz6fr16/T9OnTafXq1ew8f39/MjMzE7rWysqKVFRU6Pz582KxVRB/Vx958OABvXr1in2Ojo4mY2NjcnZ2pt27d1NJSQndv3+f5s6dS1FRUey8hw8fkomJCe3fv5+ImmLHxRVb+3et66/5tpqaGjk5ObHzoqOjSU5OjjIzMykpKYnU1NTIy8uLdu7cSSoqKuTo6Ngi7lYcqK+vp127dpGKigqrq9mzZ9PKlSvZOYJ1qKioSOfOnSMiovT0dDpy5Ah9+vTpi+eKAlxcPVfnFhYWNGLECFq3bh07Z+PGjaSurs4+e3t7k7y8PHl6epKWlhZZWVmJNY45KyuLPDw8KCEhgYiI0tLS6NChQ+To6EiZmZn08uVLsrW1ZX59+/ZtkpOTo3v37hER0Z9//kk2Nja0d+9eqq6uZuWK07cFcfr0aVq6dCn7fPToURo9ejSFhoYSEdH27dvJ1taWNm7cSCdPnhSpja34+6F14tyK/1jk5OTQ5s2bmTgPh+LiYrKzsyNbW1siahpcjRo1ii5evCh0XnR0NNnb21NNTY3Q96Ls7P+ONrfaLX67BQcD+fn5VF5eTkRNk1BFRUUi+kv0JDExkXR1denx48cUHx9P+vr6FBYWJlL7voT/hLouKCigz58/ExHR58+fydHRkbS0tNjxR48ekaamJgUGBlJFRQVlZmaSvb09ubq60sqVK7/bwtvfxUeqqqrIwcGB9PX1mZ8cP36cVFRUKCoqik6ePElWVlZkZWVFjY2N5O3tTcuXL6eMjAwiahKuCgoKIkdHR6qrq2O+8SWRq2+Jv2NdE33dt4uLi8nZ2Zk0NTWJ6K82ZmhoSJ6enkRElJqaSiEhIbRy5Uq6du2a2Gz+UhvKyckhXV1dWr9+PRE1ia8NGzaMEhMThc4rLy8nMzMzio+PF4utHPLz8+nOnTvsc1ZWFiUlJbHPERERNH78eHr//j0RNYkbamhosAWgly9fkr6+Pq1Zs0aoHFEjLy+PHj16RPX19ZSZmUk1NTVUV1dH4eHhpKioSHZ2dkTU5NvPnz+niRMn0u3bt4mIaNWqVTR58mTS0tIiNTW1FuJf4uy3s7Oz6fnz52yxYfv27eTs7EwfPnwgExMTUlFRoevXr1N6ejqVl5dTQUEB7dmzh3R0dNj9tKIVHFonzq34j8XBgweJx+PRsGHDaMWKFUI7E48ePaIJEyawTnHXrl2kq6tLQUFB9Pr1a/L29iYFBQU6e/Zsq82tdv/b2U3UpOjp5eVFKioqZGJiQkFBQZSQkECzZs2i169fs/NycnJIW1ubrly5QkREK1eupNjYWCIioYmFqPF3r2tvb29SV1enRYsWkbOzM1VUVFB8fDxpamqy3eT6+nqKj48nLS0tMjIyojFjxlBYWBhFRkbSnDlzKD8/X+x2/518pLS0lHR0dOjq1auUmJhI5eXlZGNjw+wiappEGBsbU2hoKL19+5aMjY3p5MmTbCfve6mp/93qWtDur/n27NmzhRaw0tPTicfjMfXv5uAWB0SFW7duEY/HIycnJ7p79y5TvK6vr2eTT4654uzsTFZWVpSenk5ERDU1NbRs2TKytLRssTMu6kUVHx8fsrW1pbS0NNq6dSuNHj2adHV1ydzcnP22lpYWeXt7M3tCQ0NJTk6O7OzsKD4+noKCgsjOzo5NQDkVbVHbvXjxYnrz5g19+vSJFi5cSCEhIcTn82n16tVkZ2dHHz58IKKmRa/AwEDS1tZmZcTFxQn181y54kJ9fT2tW7eOxowZQ/r6+mRhYUFETZkPZs6cSaNGjWKZDerq6mjChAmMHSIuxf1W/P3QGuPciv840D/iZlRUVKCvr8/yefr4+MDBwQHXr1/HkCFDYGZmhi1btgAAnJycoKurizdv3mDNmjWoq6tDdHQ0DA0NW21utfvfxm5B+zdv3oyysjKcP38eZmZm0NTURJcuXTB06FAW7wcAPXv2hIyMDH766ScAgJ+fH5SUlACIJ9bz717XfD4ffn5+qKiowKlTp2BjY4MPHz7A29sbAwYMwMyZM3H69Gn8v/buO66ps28D+JUwRcBZqG0dFesugiiIG3EPLFqxfdxStW5trdZRrVaq1YpVHHW1oLY4UFy4qUJb96qKq6AUFAtONgTI/f7hS2oEYoJkmev7fPJ5mpOT5MrxNsdf7nFkMhnMzc3h6emJLVu2YNasWThx4gT69euH+vXrQwihWBhMF4ypjRQyNzeHVCrFhAkTEBISgvLly+P69euK+YUA4OTkhNatW+Ps2bOoVasW6tati8jISCQnJwMAKleuDEC36woY47EGXt62vb298euvv0ImkwF4dk3grl274saNG0qv8/xiUdpkZ2eHJk2aIDo6GnPnzsXQoUMRFRWF3NxcdO/eHZ6envj6668BPJvrLITApEmTMHXqVLRv3x5ZWVn44YcfYG9vr/S62pqrWrggXc+ePWFlZYUNGzbg4cOH+PPPP/Hll1/i6dOnikXX5s6di127duHWrVuQSCTw9fXFpEmTUL58ebi7u2PIkCHIyspCeHg4nj59qtXvkudzm5ubIyoqClWrVkX9+vVx/vx5JCcnw8fHRzEXGwDKlSsHPz8/pKSkKFbVbt26Nbp27Qqg6PxobTt79iyCgoKQmZmJ48ePY8qUKXj48CFWrFiBTp06oXbt2ujRowdGjhwJAFizZg3ef/99NGzYEAAUK+7r4/rXZOD0VbET6cK2bdvEyJEjxfHjx8WdO3fEwoULRevWrcUHH3wgAgMDhYuLi9i1a5cQ4tmv5QUFBUpzh7T9C/rrkpm5dSs9PV30799fcc3Rwt6Hhw8fil27donevXuLUaNGiZCQENG5c2cxefJkkZGRoXi+Lq+7+zxjPNbJycniww8/VBremZycLN5//33xxx9/iBs3boiBAweKpUuXCiGe9ajk5OSItWvXiv3794uDBw+Kjh07ioULF2q9l+h5xthGjh8/Lvr16yd69OihyD19+nQxefJkpf2++uorMX78eCHEsyHGhUNc9cUYj7UQmrdtIfSXVYhnf//Xr18vxo0bJ8LCwsTSpUtFt27dRPfu3cXmzZtFZGSkUq9hYmKiOH36tNi8ebPi2ryFr6NrQUFBolmzZmLJkiVCiGc9/REREUq95CNHjhRdunQRffv2LTLMXAghfv75ZzF//nzFVABd+OGHHxQ93bGxscLf31+sXr1aCCHE4sWLxYQJE8Tly5eFEM96aY8eParo5dengIAA0bRpU0Xbff54P3r0SFy/fl0MGDBA9O/fX/To0UN88MEH4vr16/oNTUaBPc5k9Iq7XIT4/16utm3b4q233sKOHTvwxhtvYNq0adi8eTPatWuH06dPIzs7G1999RVkMhnMzMwglUpRqVIlCCEgl8u19gu6MWZmbv23kUJPnjxBZmYmKlWqBOBZT11eXh4iIiKQmZmJwMBANGrUCBcvXsTo0aMRGBiI8uXLK56vzdVAi/uF3tiPdUJCAt566y0Azy6h5eDggHbt2mHXrl2oV68eWrVqhcuXLytWzraysoKNjQ0OHDiAH3/8EZ9++immTZtW5r1Ez/fEFpfbUNvIi8e78H7Lli0RGhqKfv36Yf369UhPT0fXrl2RlJSEpUuXIjc3F/fv38ft27fRvn17AEDVqlVRo0YNra/Qa6x/H8uybav7utrILf5/dfVWrVrB2toaV69exaRJk7Bt2za0b98eW7ZsQWBgINLS0rB06VIAwDvvvAN3d3cMGDAADRs2hFwu1/n3SOG23r17w9XVFenp6ZDJZLC0tISnpyeaNWuGBQsWAACWLFmCvn37YvDgwXBzc1O8RuF1nIcOHYqZM2fC1tZWZ7kLe5137doFJycntGjRAufPn8eVK1fg5+eHjIwM7N69GwUFBbCwsIC3tzfee+89nayYrSr34MGD0bx5czx+/Bi5ubmK4+3m5oa5c+eifv36WL16NQICAjBnzhyEh4crRgYRqaSPap1IGwpXfHxRZGSk+OSTT8SGDRuUtqenp4v169crFuDQxy/pxphZLpcbZW4hjPN4l5R5+PDhYuzYsUqrlI4dO1aEhoYWu782e1nkcnmRuaXGdqxVtWtfX1+lVXqFEGLixIli5cqVQoiS59Vq6zNERUUpFklTlduQ2khxCnO/eJyuXLkihg0bpugtOnr0qGjVqpUYPny48PDwEAsXLtRpzkKvw7F+UWnbti69OC82LCxMDB48WOzbt0+xLTU1VYSHh4uePXuKevXqFek91NX3iao2snnzZjFy5EjF/PbCfevVqyd+//13neRTRVVuf39/ER0dLR49eiQmTpwovvnmG1FQUCA2bdqkGGmhL6U53n/88UeR/fU16o2MC3ucyehERkbi4sWLePDgAQAgIyMD06dPR2RkpNIvkOL/fzn08PBA48aNcfr0afzzzz8Anv2Ca2trC39/f4wYMQKAdn/1P3funOJ6nQUFBcjOzjb4zAAQERGBXbt24fz58wCA7OxszJgxw+Bz7927F7t27cKFCxcAGEcb+e233xRzIYUQyMrKKjZzYY/unDlzcOXKFSxatAhXr17Frl278Pfff6NBgwZKr6vt+Ydbt25FgwYNcODAAQAwira9Y8cO/PTTT9i/f78is6p2PXv2bOzatQvLly9HbGwswsPDERMTA3d3dwAlz6st68+wfft2NG/eHMuWLYOVlRUyMjKKzW1obSQiIgK7d+/GX3/9BQBIT09XaiMvHqe6deuiW7duOHbsGK5fvw5vb2/s2LEDU6ZMQUREBKZNmwYAWu0dOnLkCE6ePIm4uDgAQFZWllEca03Pj6Vt22Wt8BrdhaMofvvtN3zxxReK+bbPZ27dujWcnJxw6NAhZGRkAHg2//mDDz5ASEgITp8+jfr16yu9vja+TzQ9P/bo0QMVK1bEyZMnkZycDIlEgoYNG2LhwoWKtR+e319bND0/9ujRA5UrV8bhw4dRrlw5tGnTBlevXsWNGzcwcOBAdO/eXat5C6l7jlTneBfOZX6etufo0+tBIrT9N5SojNy7dw+ffvopcnNzUaNGDSQmJmLx4sVwdnbGkydPFMPknieEgEQiwcWLF7F69WpUq1YNc+fOLXE/bcjJycGnn36KN998U7EQCAA8fPgQVatWNcjM//zzDyZMmIC8vDy4uLggPDwcixYtQq9evQw69507dzBhwgTIZDJ4eHhg27ZtOHr0KN555x08ePAAb7zxhkHmBoBp06bh4MGDOHDggGLoZEnHuqCgAGZmZoiMjERkZCQSExORmZmJadOmwcPDQ2sZn3f9+nV89tlnMDMzw9SpU9G2bVvFY4Z6rB8+fIhx48bhyZMnaNu2LaKiorBhwwZUr169xGNd+A/3sLAwREdHIyUlBTk5Ofjyyy/RokWLMs9YnDNnzuCbb76BXC5H06ZNkZqaiuXLlys+k6G2kYSEBIwbNw4ymQyurq4IDw/HkSNHUL169RLbSKHC7/eMjAz89NNPSo8VFBRAKpVqpY0kJCRg/PjxKCgogIuLC44cOYJ58+ahS5cuSE9Ph52dXZHnGMKxLs35Ud9tu/Dv+YYNG3D//n3MmjVL8Vh4eDhq166NJk2aFPvc3377Ddu2bUOzZs3wySefKF6r8P8L/0y04VXOj0eOHMHGjRvRrl07fPLJJ8Xuoy2vcn48cuQIQkJC0K1bN3z00UdISkpC9erVdZYd0OwcaQjHm15T2u3QJio7mzZtEl999ZUQ4tniKwEBAcLHx0dcuXJFsY+qSx1s2bKl2OE5Ze3FIWGnTp0STZs2Ff369RN79uwRQijnNITMLwoMDFQaDhkYGKh0mQkhDDP3t99+q8hduGDPhQsXlPYxhNzFDRscO3asqFevnpg2bVqRx17M/OLz//33X6XHtD0sMTs7W4waNUqMHTtWsS0jI0NpmKoQhnGsn3fkyBHFolJCCPHgwYMiw/NedqwTExOVHtPWsZbL5aKgoEAsX75cuLu7i5CQECGEED/++KOYOnVqkWyG1kaEEGLdunWKv4+5ubli4MCBioWECqlqI1FRUeK3337TasYXLV26VMycOVNxPzg4WHz44Yfi6NGjim0vthlDONalOT/qq20/LzMzU/Ts2VP4+PiIixcvFnn8xQyF9zMyMsT8+fPFpEmTdLpYlhCvdn6Uy+WKS5Xp2qucHwtznz59WmmbLmlyjiykz+NNrycO1SajcfbsWcUiGebm5pgxYwbKly+PiIgIPHnyBEDxQ5wKt/Xv3x+tWrXSes7CXzAL3/fQoUPo1KkTmjVrhkOHDiE1NRVSqVTlcCxdZ36eTCbDlStXlBav8fLyAgDFcS6JPnNnZGTgxo0biiFYc+fORUxMDEJDQxEWFoasrCwAqhev0nUbAZ4NTczIyECFChUQFBSE3bt34+TJk8Xme/H5hdsdHR0BPPtsEolE67+iW1tbo3Hjxor3/+677zB48GBMmjQJW7ZsQVpamiLPi/TZRqKjoxX//d1332HUqFGYOHEidu/ejdzcXAAlD7cuzP3OO+8o9tPmsZZIJJBKpWjZsiX++OMPDB48WPG+hb2eqt5b320EAC5fvow7d+4AAL799ltcu3YN+/btQ2RkpOK7/GWLyRV+9+hCfn4+zp49q/TdN2TIEDx69Ai7d+9GYmKiUr5ChnCsz58/r/H5UV9t+3lXr16FpaUl6tSpg61btyq2F2Yq7EV+PrMQAuXLl8ewYcOwZMmSMl8sSxWZTIarV6+W6vxYOC1h8uTJaNasmVZzvigtLQ23bt0q1fnx+dyFQ/gB7U+5ev79NT1HFj5PX8ebXl8snMkghYeHY9asWfjll18QHx8P4Nm8NwsLC6WT0/Dhw/H777/j3r17AP6bo3L//n0Auh2K83zmf/75BxKJBPn5+SgoKMDUqVPRpk0byGQybN++HUDRa3bqI/OLuRMTE2FpaYmWLVvC3d1dcTL6448/UK1aNcUKx4aY29bWFitWrECvXr2QkJAAGxsbrF69Gg4ODggJCUF4eDiA/67PaAhtBHj2j1wbGxvExsbCxcUFH374oWJV2H///RdA8e0aKPoPF20NTXw+9+3btwE8+7sXHR2N4cOHIyUlBTNmzECjRo2wc+dOhIaGQghhMMe6sHh77733YG9vjyVLliAhIQEzZsyAk5MTfv75Z+zYsQNA8e0D0P2x3rx5MxISEuDm5gYLCwtFUZSUlITU1FQAyqvKFvf3UR+5n2/bHTt2RGxsLFxdXRETE4Pvv/8ecrkcgYGBiu9BQ2kjiYmJMDc3x1tvvYXy5csjPT0dwLO5t++99x6ys7Px999/A4BidXR9HGu5XI64uDgsXboUSUlJim3vvfdeqc6P+sydm5uLNWvWoGPHjmjTpg3u37+Pffv2KWV7sT08Pyf+rbfeglQq1doK3yUda09PT3h4eGh8fnz+GsaqfjjXRm4bGxsEBQWV6vyoz9zAs2NZmnOkrnKTaWHhTAZDCIG8vDzMnz8fQUFBqFu3Lvbt24e5c+fi0qVLaNq0KeLj43Hp0iXFczp27Ihy5crhyJEjAJ4t4PLNN99g+fLlxS46o6vM8+bNw/Hjx2Fubo7p06ejcuXKcHZ2hrOzs9KCM/n5+ZDJZDrNrCr37Nmzce7cOXzyySdwd3eHRCKBTCbDsWPHFL2DhflkMhnmzZtnELm//vprREdHw87ODvn5+XjzzTcxd+5ctG7dGlOmTEGLFi1w9epVAM8WcDGUNvL7778DAOLi4pCXl4c33ngD33zzDW7duoVWrVohODgYmZmZyMnJMZg2EhAQgKioKNjY2MDX1xePHj3C3Llz4ebmhvHjx6Ndu3a4desW8vLyDOpYX7x4EfXr18edO3dw4sQJzJw5E25ubpg8eTI8PT1x/vx55Ofn6zxzSbkjIiIwd+5cRRsp/Iefp6cn7t27h8zMTKV/GBrS98jcuXNx8uRJ+Pj4YPbs2fD09MTmzZvh5eWFmTNnonXr1rh06ZJioURDaCNfffUVrl69iu7du+PcuXOYMmUKIiMjMWLECLi5ucHS0lKxmJKuzzPPk0ql+Pvvv7FhwwZcvHgRMpkMUqkUb731Fu7du2dQ58eX5bayssLkyZMxevRoxWWMDh48iKdPnyoKYolEgtu3byMoKEjxOsW9tq4yW1tbw8/PD82bN3+l86M2j/2LufPy8mBubq7omX+V86Mucz9/ub2EhIRXOkdyLjOVmTIb9E1UBnJycsTAgQMVc53+/fdfERgYKLp06SKEEGLChAli9uzZIiEhQfGcwMBAMXr0aMX95+d06Svz0qVLhZeXl2KfvLw8IYQQf/31l5gwYUKRy6noOrMQ6uUWQojY2FjRpUsXxWVJ0tPTxaVLl4QQQsTExOg0sxAl5+7QoUOJz9m2bZvo1q2bYm6iobWR9PR0MWPGDJGQkCBmzpwpmjZtKpo2bSpkMpniNQypjRQe6+vXr4tTp04JIf5r43v27BFt2rQRubm5esn9su+QKVOmCA8PD6Vce/fuFa1bt9Zb5pJyF9eud+3aJUaOHCmSk5OLvIYh5S5s21lZWWLOnDkiLi5O8Zzw8HDh5eWluKSWobSRzp07CyGEOH/+vJg1a5YYNWqUWLZsmRBCiJ9//ll8+umnitfQx7EuNG3aNOHl5SX8/f1FbGysEOLZ37/Ro0eLuXPnGtT58XnF5X7e8ePHxciRI8X69euVtq9atUps3bpVCKH7ebXPZ36+DRcytPNjoedz3759W+W++j4/Pq+k3I8ePRKzZs0yuHMkmR72OJNBuX37NjIzM2FtbQ0hBBwdHTFo0CDIZDJs3LgREyZMwN27dxEaGqqYk3j9+nV07NhR8RqNGzcGAK0N31In88CBA2Fubo5ly5YB+K+3yNnZGe7u7jh9+rSiJ0kfmdXNDQAHDx6Em5sbKleujJUrV6JZs2Y4ffo0ACjmSxlCbjMzM/zwww8AgJMnT2LVqlWQyWQAgCtXruDjjz9WDOcypDayZs0a3L9/Hzt27EC/fv0gk8lw4sQJWFhYYPHixYrXMKQ2IpVKERQUhPr166NWrVpKc4fPnTuHDz/8EJaWlnrJreo7JDg4GGPHjkWtWrXw888/Izk5WZHZx8dHMUzRkI61mZmZ0t/HVq1a4dKlS0hJSQGgPCfRkHKbm5tj1apVyMrKwt9//41jx44pepAuXbqEfv36wcrKSi+5S2ojBQUFWL58OZo2bYo5c+Zg7ty5mDBhAoBna2z4+PgoXkMfxxoALly4gNOnT2Pp0qVITEzEb7/9hoyMDJibmytWOzak86Oq3NnZ2UpZPDw80KRJE5w8eRI3b95UPPfTTz+Fn58fAN32HL6YOTIyskhmQzs/Fpf76NGjitzAs+87Qzo/qpP7ypUr2L59u8GdI8kE6algJypR+/btxebNm4UQ//26HB4eLtq2bSuys7PFyZMnha+vrxg4cKDw8vISAwYMUFrFVB9KytymTRuRmpoqhPhvJda4uDjx888/F9tjpGuqcj958kQIIcSwYcOEn5+f6NKli+jXr5+4du2avuIqqMqdmZkpoqKiROfOncXAgQOFt7e3GDBggEhKStJnZJXtOikpSSxYsEBcv35dsf/FixeL7eHQNVXHurB3pWfPnmLQoEGiQ4cOws/PT2l1Xn142XfIiRMnRPfu3cWHH34oOnXqJPz8/JR66fRFne8RIYQYOXKk8PPz00vG4qg63rm5uWL9+vXCy8tLDBo0SHTs2FF8/PHH4u7du/qM/NJjff/+fTFs2DAxcuRI4eXlJYYNGyYePHigz8hCCCFSU1MVPZqhoaGiZ8+eivtCCHHs2DHRp08fgzo/ClF87r/++kvxeOGqyBcuXBDDhw8XoaGhSs/XdU+zEC/PLIQQw4cPN7jz48tyHzp0SHTt2tWgzo9CqG7bWVlZYvbs2eLGjRuK/Q3lHEmmhYUz6cWLl/N4ftvGjRtFixYtlIbgJCYmiiFDhojjx48LIZ4N27l48aI4ceKEbgKL0mUePny4iIyM1FnG4pQ292+//SaysrJEx44dRbdu3URERITS87X9D5nStpGoqCghhBApKSni7NmzBt9Ghg0bJo4dO6bYVlBQoHRpDV38g7E0uYcOHSqio6OFEELcuXNHHD16VOnyUobaPgq/Q548eSJiY2PF+fPndZb5+YzFbVP3e+T69euKdq4rpck9ePBgceLECSGXy0VcXJyIiIgQv//+u2IfQ2wjw4YNE9HR0SI/P19cvXpVbN68WelSVLpQXO5CL152p3///mLevHmKocJCCPH48WOdnx+FKF3up0+fFnlclz+qvErmp0+fGtT5sZA6bUQf50chSpe78Ef85/fT9TmSqBALZ9KpF69fXDgv8nn5+fmic+fOYv78+YptT58+FZ07dy7ya+/zz9EWbWXW9pf9q+Yu/KX3zz//LPIcbTLlNqLtY/siUz7WxT1Hm7SVW9uM8XgbY2Yh1Mv9YpY///xTdO7cWRw/frzEc4qx5lZ1je9X9aqZC7e9WHga67E2ltzabBNE6jDX91BxMi2Fq1/+9NNP2L9/PypXrgw/Pz94e3sr5i6ZmZlhwYIFGDNmDOzs7NC7d28cP34cFStWhIODQ7Gvq61LrGgzs7bnar1q7qpVqwIAWrZsCeDZCuDm5uZaPdZlkduY24i2j622cr/IGI61LjNrM7e2lXVu8f+XFzKmNqKLzOrmLlSYpWXLlqhRowY2bdoEFxcXVKhQocjrGmtuba2WXRaZnZ2dUalSJXh6egIwrPOjqtzG3ka02SaI1KLvyp1eXwUFBSImJkZMnDhR6dfFhQsXip49e4rz58+LpUuXioMHDxb5NVIIIXbv3i2mTJki/Pz8hK+vr056XIwxM3OzjbyuuY0xM3OzjWgz9/MKe+aSk5PFrVu3mPs1yszcus9NpA4WzqRVaWlpSsNsnj59KgYOHFhkeJNMJhPp6elCiKJDhp6f76SLuSzGmFkI5i7ENlIyY8xtjJmFYO5CbCMlK01uVZi7ZMaYWQjmLsR5zGQoOOaBylzh5VHkcjns7OzwxhtvoEOHDkhLS4OFhQWuXLkCGxsbxb65ubnYtWuX4vIThcN0xP9fwuntt99W7Kut4c3GmJm52UZe19zGmJm52UZ0kVsV5jb+zMyt+9xEmmDhTGVGPBvBUGSuTMOGDZGRkYHg4GBYWlrC29sbGzZsAPBs3ouVlRW2bNmCmJgYpeeVNO/F1DMzt25zG2NmY81tjJmZW7e5jTGzNnLrijHmNsbMAHMTGQMWzlQmxP8vniKRSHD+/HkMGjQIX3/9NTZt2gQA+O6777Bhwwb8888/8PX1xdWrV7FhwwZkZWXh6tWrsLCwQLNmzZiZuQ0qtzFmNtbcxpiZudlGmNuwchtjZubWfW6i0pKIwnFKRK8oOzsb+/btw44dO9C2bVuYm5sjODgY//vf/zBu3DgMGjQIDg4OWLx4Mfbv34+VK1eiSpUqiI+Px6hRozBo0CBmZm6Dy22MmY01tzFmZm5mZm7Dym2MmZmbyEiU2WxpMikvLqySnp4uVq1aJVxdXUVQUJBi+7Zt28TQoUPFyZMnRXx8vGjUqJHiusDp6ekiJiZGadVFbS4AYYyZmZtt5HXNbYyZmZtthLkNK7cxZmZu3ecmKiscqk0aefr0KYD/5oHFx8ejoKAAtra2aNOmDapWrYrMzEzF/h988AEsLCxw+fJl1KxZE15eXpgzZw5kMhlsbW3RsGFDmJubKxaV0MYCEMaYmbnZRl7X3MaYmbnZRpjbsHIbY2bm1n1uorJmru8AZByuX7+OefPmoWnTpvjiiy/w999/46uvvkJubi4qVqwIZ2dnTJ48Gf3798fx48fx6NEjVKlSBRYWFqhYsSKuXr0KAFiyZAnOnz8PS0tLpdfXxoIsxpiZudlGXtfcxpiZudlGmNuwchtjZubWfW4irdF3lzcZtqysLDFr1izh5uYmVq9eLYQQ4sGDB6J///5i7dq14uHDh2LPnj2icePGYuvWreL69etiwoQJ4ssvvxQZGRkiLS1NDB8+XPz222/MzNwGk9sYMxtrbmPMzNxsI8xtWLmNMTNz6z43kbaxcKYSrVq1Sri5uQlnZ2exf/9+xfYjR46Ivn37iqysLMW2X3/9VbRu3Vo8evRI7NmzRzRq1EgMHTpU9OnTR0yePFnk5uYyM3MbRG5jzGysuY0xM3OzjTC3YeU2xszMrfvcRLrAOc5UxMOHD9GrVy9ERkZi8eLF8PHxwV9//YWHDx8CAP755x/k5eWhXLlyivkpH3/8MWQyGS5cuIAOHTqgT58+yMrKwtq1axEYGAhLS0sILS7gboyZmZtt5HXNbYyZmZtthLkNK7cxZmZu3ecm0iUWzlRE5cqVMXPmTISFhcHLywuurq6Ii4vD8ePHATxb9OHvv//GsWPHFPNTHj16hLfffhuVKlVC+fLl4e3tjby8PJw/fx4AIJPJtLr4gzFmZm62kdc1tzFmZm62EeY2rNzGmJm5dZ+bSJdYOFMRUqkULVq0gFwuBwB0794d1apVw5kzZ3D79m1UqVIF/v7+mDt3Ln755Rc8ePAAy5YtQ/ny5fHuu+8CAFxcXNCmTRt8//33kMlkRRaEYGbmZht5fXMbY2bmZhthbsPKbYyZmVv3uYl0iatqU4mkUinkcjmsra3RpUsXhISEIDIyErVr18bnn38OAIiKisL27dvx5ptvYtmyZahcuTIAoEKFCmjXrh3Kly8P8WwuvU5+dTTGzMzNNvK65jbGzMzNNsLchpXbGDMzt+5zE+mCRHDyAakpMDAQ8fHxGDx4MJo1a4b8/HzI5XI8efIEjo6OAICCggJIpVJIJBKD+MI0xswAczPzyxljbmPMDDA3M78cczPzyzA3kfHjUG16qcJhOz179oRMJsO+ffuQl5cHc3NzWFpawtHREUIIyOVymJmZKb4w9fnFaYyZAeZm5pczxtzGmBlgbmZ+OeZm5pdhbqLXB4dq00tJpc9+X6lbty5atGihuLj98yQSiUF9WRpjZoC5dckYMwPGmdsYMwPMrUvGmBlgbl0yxswAcxO9TjhUm9Qil8shlUqNagiOMWYGmFuXjDEzYJy5jTEzwNy6ZIyZAebWJWPMDDA30euChTOVijF+iRpjZoC5dckYMwPGmdsYMwPMrUvGmBlgbl0yxswAcxMZKxbORERERERERCpwcTAiIiIiIiIiFVg4ExEREREREanAwpmIiIiIiIhIBRbORERERERERCqwcCYlaWlpCAoKQlpamr6jaIS5dccYMwPMrUvGmBlgbl0yxsyAceY2xswAc+uSMWYm0gcWzqQkLS0NK1asMLovT+bWHWPMDDC3LhljZoC5dckYMwPGmdsYMwPMrUvGmJlIH1g4ExEREREREanAwpmIiIiIiIhIBRbORERERERERCqwcCYiIiIiIiJSQSKEEPoOQaWTHTZf3xE05vfZKX1H0NiWQeX0HaFUpK3b6DtC6dyL13cCjcmTkvUdoVQqzD+u7wil0rdac31H0NhPQ6z1HaFUAkPM9B2hVH7JvqXvCBo71clO3xFKxdzRVt8RNCZPz9F3hFIZftg4v0fC/tmj7whqM7d8W+1982X3tJiEDBF7nImIiIiIiIhUMNd3ACIiIiIiIn2TSiT6jkAGjIUzERERERGZPAkLZ1KBhTMREREREZk8CVg4U8lYOBMRERERkcljjzOpwsKZiIiIiIhMHuc4kyosnImIiIiIyOSZSXjBISoZC2ciIiIiIjJ5HKpNqrBwJiIiIiIikyfl4mCkAgtnIiIiIiIyeexxJlVYOBMRERERkcnj4mCkCgtnIiIiIiIyeexxJlVYOBMRERERkcnjqtqkCgtnIiIiIiIyeRIuDkYq8GcVIiIiIiIyeVKJRO1baVy+fBmenp4lPn7y5Ek0aNAArq6uitvKlStL+3GojLHHmYiIiIiITJ625jgLIRAWFobvvvtO5X7Xrl1D165dsXTpUq3koFfDHmciIiIiIjJ5UkjUvmli+fLlCA0NxejRo1XuFxMTgwYNGrzKRyAtYo+zBgYNGgRvb28MHTpU31GIiIiIiKgMabI4WFpaGtLS0opst7e3h729vdK2jz76CBMnTsTp06dVvmZMTAweP36M0NBQCCHQrVs3TJ48GZaWlmrnIu1h4UxERERERCZPokHhHBISghUrVhTZPm7cOIwfP15pm6Oj40tfLz8/H2+++SY6deqEPn36ICUlBRMnToREIsHUqVPVzkXaYxRDtWNjYzF06FA0b94cXbt2xd69ewEA9+/fx9ixY9G+fXs4OzujX79+uHHjBgDg8ePHGDVqFJo3b4727dtj+vTpyMnJwYULF9CkSRNkZmYqXn/lypVFGnhJbt26hY8++giurq7w8/NDfHw8gGdzF1asWIFu3brB1dUVbdu2RXBwsOJ59erVw7Zt2+Dl5QVXV1csWbIEhw8fhre3N9zc3F4654GIiIiIiLRHosH/hgwZgsjIyCK3IUOGlOq9zc3NERISggEDBqBcuXKoWbMmPv30Uxw+fLiMPyWVlsEXzpmZmRg2bBjatGmDEydOYNGiRViwYAHOnTuHmTNnolq1ajhy5AjOnDmDGjVqIDAwEMCzYtjOzg4nTpzArl27EBMTg4MHD6Jp06ZwcHBAZGSk4j0iIiLQu3dvtfJER0cjICAAp06dQsWKFRWT9/ft24fdu3cjODgYFy5cwJw5c7Bo0SKkpKQonnvgwAFEREQgJCQEa9euxfbt27F7925s3LgRISEhuHnzZrHvmZaWhrt37xa5ERERERFR2dBkVW17e3u88847RW4vDtNWV3JyMr777jvIZDLFtry8PFhZWZXVx6NXZPBDtaOiomBrawt/f38AgLOzM/r27YvQ0FAEBASgYsWKAICkpCRUqFABcXFxAABbW1v8+eefOHz4MFq2bIldu3ZBKn32O0GvXr2wf/9++Pj44Pr163j06BHatm2rVp5+/frByckJAODt7Y1t27YBALy8vODh4QEHBwc8ePAAFhYWKCgowOPHj+Hg4AAAGDFiBGxsbODs7AwbGxv0798ftra2aNSoERwcHHD37l3Uq1evyHuWNBTkUsAgDY4kERERERGVRJ/Xca5YsSL27t0La2trjB07Fnfv3sXq1avRt29fvWUiZQZfON+7dw8JCQlo1qyZYltBQQEaNWqE+Ph4LF68GPfv34eTkxOsrKwghAAAjB07FlKpFCtWrMCUKVPg5uaGefPmoXbt2ujduzfWrl2L1NRU7N27F127dlV70n2FChUU/21hYYH8/HwAz+YlLFiwACdOnICDgwOcnZ0BQJEHgKLIBwAzMzPY2dkp7kulUqV9nzdkyBD4+voWfeBUsFqZiYiIiIhItdJen7m09uzZgzlz5uDixYuwsrLCunXr8O2338LDwwPlypWDn58fhg0bptNMVDKDL5wdHBzQuHFjbN26VbEtOTkZcrkc3bt3x/z589GjRw8AQHBwMMLDwwH8Nxd54sSJuH//PhYsWIB58+YhODgYNWvWRMOGDXHs2DEcPny4TOYXBwYGIjc3F1FRUbC2tkZqairCwsKU9intteGKW50PALJL9WpERERERPQiTVbVLg0PDw+cO3dOcd/Hxwc+Pj6K+w0aNMCmTZu0moFKz+DnOLdr1w4JCQnYuXMn8vPzkZiYiMGDB2Pjxo3Izc2FtbU1gGfLt2/cuBF5eXkAng1vDggIQGZmJqpUqQJra2ul3uLevXvj559/BgC4ubm9cs709HRYWVnBzMwMqamp+PbbbwFAkYeIiIiIiAyXJouDkekx+MK5YsWKWL9+PcLDw+Hp6YmPP/4YHTt2xJQpUzBv3jzMnTsXbm5umD59Ovr374+kpCRkZmZi+vTpkMvl8PLyQosWLZCWloYZM2YoXrdbt26Ii4tDr169yiRnYc+2u7s7evXqhUqVKqFevXq4detWmbw+ERERERFpjyaLg5HpkYiSJta+5vLz89GqVSts3boVtWrV0necUskOm6/vCBrz++yUviNobMugcvqOUCrS1m30HaF07sXrO4HG5EnJ+o5QKhXmH9d3hFLpW625viNo7Kch1vqOUCqBIWb6jlAqv2Qb34/WpzrZvXwnA2TuaKvvCBqTp+foO0KpDD9snN8jYf/s0XcEtTV5s6Xa+/717wktJiFDZPBznLUhLi4OBw4cQN26dY22aCYiIiIiorIj5RBsUsEkC+cvvvgCqampWL16tWLbtWvXMGDAgBKfM2bMGIwYMUIX8YiIiIiISMe0vTgYGTeTLJx37txZZFvDhg1x8eJFPaQhIiIiIiJ9K+0VcMg0mGThTERERERE9Dyulk2qsHAmIiIiIiKTxznOpAoLZyIiIiIiMnkcqk2qsHAmIiIiIiKTxx5nUoWFMxERERERmTwzFs6kAgtnIiIiIiIyeVIO1SYVWDgTEREREZHJ46rapAoLZyIiIiIiMnmc40yqsHAmIiIiIiKTx1W1SRUWzkREREREZPLMINV3BDJgLJyJiIiIiMjksWwmVVg4ExERERGRyeNQbVKFhTPplI3E+JqckMv1HaF0LCz1naBURH6BviNoTP44Td8RSuXJ8Pf1HaFUBu7P13cEzZmb6TtBqXx5/ht9RygVj0Yz9B1BY0L2r74jlIqQC31H0FiVkBh9RyiVXtXc9B3htcfFwUgV46tiiIiIiIiIyhgvR0WqsHAmIiIiIiKTxznOpAoLZyIiIiIiMnlmEpbOVDIWzkREREREZPJYNpMqLJyJiIiIiMjkcY4zqcIfVoiIiIiIyORJIVH7VhqXL1+Gp6dniY8nJSVh2LBhcHV1RceOHREVFVXaj0JawMKZiIiIiIhMnlSDmyaEENi+fTuGDx+OvLy8Evf77LPPUK9ePZw+fRrffPMNJk+ejMTExNJ8FNICFs5ERERERGTyzCBV+6aJ5cuXIzQ0FKNHjy5xnzt37uDq1auYMGECLC0t4enpiQ4dOiAsLOxVPxaVEc5xJiIiIiIik6fJAOy0tDSkpaUV2W5vbw97e3ulbR999BEmTpyI06dPl/h6cXFxqFatGmxsbBTbateujcuXL2uQirSJhTMREREREZk8qUT90jkkJAQrVqwosn3cuHEYP3680jZHR8eXvl5mZiasra2VtpUrVw45OTlqZyLtYuFMREREREQmT5MB2EOGDIGvr2+R7S/2NqvLxsYGubm5Stuys7OVeqBJv1g4ExERERGRydPkclTFDcl+FU5OTkhKSkJOTo6i5/n27duoU6dOmb0HvRouDkZERERERCZPW6tqq6N27dqoX78+li5dCplMhlOnTiEyMhI9e/bUwrtRabBwJiIiIiIik2cGidq3srBnzx64uroq7gcFBSEuLg6enp6YNWsWAgICULdu3TJ5L3p1HKpNREREREQmT1pGBXFJPDw8cO7cOcV9Hx8f+Pj4KO5Xq1YN69ev12oGKj0WzkREREREZPK0WzaTsWPhTEREREREJk/bPc5k3ExyjvOgQYMQHBys1wzZ2dmYPXs2WrduDXd3d4wePRpJSUl6zUREREREZKr0uTgYGT7+uevJkiVLkJCQgL179yI6OhpVq1bFZ599pu9YREREREQmSdeLg5Fx0WnhHBsbi6FDh6J58+bo2rUr9u7dCwC4f/8+xo4di/bt28PZ2Rn9+vXDjRs3AACPHz/GqFGj0Lx5c7Rv3x7Tp09HTk4OLly4gCZNmiAzM1Px+itXrsT48ePVynLr1i189NFHcHV1hZ+fH+Lj4wEAQgisWLEC3bp1g6urK9q2bavUO12vXj1s27YNXl5ecHV1xZIlS3D48GF4e3vDzc0N3333nVrvn5ubi3HjxqFSpUqwtrbGgAED8NdffyE/P1+t5xMRERERUdmRaHAj06OzwjkzMxPDhg1DmzZtcOLECSxatAgLFizAuXPnMHPmTFSrVg1HjhzBmTNnUKNGDQQGBgJ4Vgzb2dnhxIkT2LVrF2JiYnDw4EE0bdoUDg4OiIyMVLxHREQEevfurVae6OhoBAQE4NSpU6hYsSKWLl0KANi3bx92796N4OBgXLhwAXPmzMGiRYuQkpKieO6BAwcQERGBkJAQrF27Ftu3b8fu3buxceNGhISE4ObNmy99/2+++QbNmjVT3D969Cjee+89mJsXnXaelpaGu3fvFrkREREREVHZkEKi9o1Mj84WB4uKioKtrS38/f0BAM7Ozujbty9CQ0MREBCAihUrAgCSkpJQoUIFxMXFAQBsbW3x559/4vDhw2jZsiV27doFqfRZvd+rVy/s378fPj4+uH79Oh49eoS2bduqladfv35wcnICAHh7e2Pbtm0AAC8vL3h4eMDBwQEPHjyAhYUFCgoK8PjxYzg4OAAARowYARsbGzg7O8PGxgb9+/eHra0tGjVqBAcHB9y9exf16tVT+9hERERgw4YNWLt2bbGPh4SEYMWKFUW2XwoYpPZ7EBERERFRyTiHlVTRWeF87949JCQkKPWyFhQUoFGjRoiPj8fixYtx//59ODk5wcrKCkIIAMDYsWMhlUqxYsUKTJkyBW5ubpg3bx5q166N3r17Y+3atUhNTcXevXvRtWtXWFpaqpWnQoUKiv+2sLBQDJHOz8/HggULcOLECTg4OMDZ2RkAFHkAKIp8ADAzM4OdnZ3ivlQqVdpXFSEEVq5ciZCQEKxcuRLNmzcvdr8hQ4bA19e36AOngtV6HyIiIiIiUo39yKSKzgpnBwcHNG7cGFu3blVsS05OhlwuR/fu3TF//nz06NEDABAcHIzw8HAA/81FnjhxIu7fv48FCxZg3rx5CA4ORs2aNdGwYUMcO3YMhw8fVnt+sSqBgYHIzc1FVFQUrK2tkZqairCwMKV9JJJX/2uVl5eHqVOn4sqVK/j111/x3nvvlbivvb097O3ti2zPfuUUREREREQE8HJUpJrORiS0a9cOCQkJ2LlzJ/Lz85GYmIjBgwdj48aNyM3NhbW1NQAgJiYGGzduRF5eHoBnw5QDAgKQmZmJKlWqwNraWqm3uHfv3vj5558BAG5ubq+cMz09HVZWVjAzM0Nqaiq+/fZbAFDkKSvffvstbty4gW3btqksmomIiIiISPvMNLiR6dFZ4VyxYkWsX78e4eHh8PT0xMcff4yOHTtiypQpmDdvHubOnQs3NzdMnz4d/fv3R1JSEjIzMzF9+nTI5XJ4eXmhRYsWSEtLw4wZMxSv261bN8TFxaFXr15lkrOwZ9vd3R29evVCpUqVUK9ePdy6datMXh94ttjXli1bkJiYCG9vb7i6uipu6enpZfY+RERERESkHi4ORqpIhLoTcg1Ufn4+WrVqha1bt6JWrVr6jqNT2WHz9R1BY0M/P6fvCBrb8D/j/F3RzLuTviOUioi9ru8IGiu4cVvfEUonp2xH0ujKwP06m2VUZn79pMLLdzJAlmO+0XeEUjnWaMbLdzIwzdv+q+8IpWL+tt3LdzIwlVdd0HeEUulV7dVHVurDzn/26DuC2r6uOUD9ff/5RYtJyBAZ378+nhMXF4cDBw6gbt26Jlc0ExERERFR2eGq2qSKURfOX3zxBVJTU7F69WrFtmvXrmHAgJJ/LRozZgxGjBih1VxPnjxBhw4dSnzc19cXs2fP1moGIiIiIiJSHwdgkypGXTjv3LmzyLaGDRvi4sWLekjzn0qVKuk9AxERERERqc+cpTOpYNSFMxERERERUVlg2UyqsHAmIiIiIiKTxznOpAoLZyIiIiIiMnlSo77WEGkbC2ciIiIiIjJ5HKpNqrBwJiIiIiIik8eh2qQKC2ciIiIiIjJ55hyqTSqwcCYiIiIiIpPHodqkitojErKzs7WZg4iIiIiISG+kGtzI9Kj9596yZUtMmTIF0dHRKCgo0GYmIiIiIiIindJW4Xzjxg30798fLi4u6NWrFy5fvlzsfidPnkSDBg3g6uqquK1cubK0H4fKmNpDtdeuXYt9+/Zh6tSpkEql6NatG3r16gUXFxctxiMiIiIiItI+iRbmOMtkMowZMwaDBw/G5s2bcfjwYfj7++PYsWOwtbVV2vfatWvo2rUrli5dWvZB6JWp/YNJ8+bNMXfuXPzxxx8ICAhAamoqRo0ahU6dOmH58uW4c+eONnMSERERERFpjbkGN3WdOXMGeXl5GDp0KCwsLNCjRw/UqVMH+/fvL7JvTEwMGjRo8Kofg7RE48XBzM3N0b59e5QrVw4VK1bE9u3b8csvv2D9+vVwc3PDnDlzUKtWLS1EJSIiIiIi0g6pBj3OaWlpSEtLK7Ld3t4e9vb2ivuxsbFwcnJS2qd27dq4detWkefGxMTg8ePHCA0NhRAC3bp1w+TJk2Fpaal+MNIajQrnCxcuYP/+/Th48CAyMzPh5eWFZcuWoU2bNsjIyMBXX32FMWPGFPsLCpW9nNCj+o6gsZ8/q6/vCBqz/2KvviOUysUtMn1HKJXaIyvrO4LGpPbl9R2hVOSyVH1HKJUtc4zwe2R0qL4jlEqq1Tf6jlAqNSoU/cesoXPY/be+I5SKMV69J3VWe31HIAOlyaraISEhWLFiRZHt48aNw/jx4xX3s7KyYG1trbRPuXLliiy8nJ+fjzfffBOdOnVCnz59kJKSgokTJ0IikWDq1KkafQ7SDrULZy8vL6SkpKBZs2b47LPP0LlzZ6Vx+RUqVED37t0xZ84crQQlIiIiIiLSFk0W/RoyZAh8fX2LbH++txkAbGxskJubq7QtOzsbNjY2StvMzc0REhKiuF+zZk18+umn+P7771k4Gwi1C+e+ffuiX79+cHR0LHGf9u3b4/fffy+TYERERERERLqiyVDtF4dkl8TJyQnBwcFK227fvo0PPvhAaVtycjKCg4OVhmbn5eXByspK/VCkVWr/sBIaGoqUlBSV+1hbW3MMPhERERERGR2JBjd1eXh4QAiB4OBg5OXlISIiAjdv3kSnTp2U9qtYsSL27t2L1atXIz8/H/Hx8Vi9ejX69OlTFh+NyoDahXOFChWQnp6uzSxERERERER6YS6E2jd1WVpaYt26dTh06BDc3d3x448/YuXKlahcuTL27NkDV1dXAICVlRXWrVuHc+fOwcPDAwMHDkTXrl0xbNgwbX1c0pDaQ7U9PDwwatQotGjRAjVq1CgybIBj74mIiIiIyFhpMsdZE3Xr1kVoaNFFIn18fODj46O436BBA2zatElLKehVqV04x8bGwsXFBTk5OUWWT5dINBmwQEREREREZFhY0ZAqahfO/PWDiIiIiIheV1INhmCT6dHoOs43b97E+vXrERsbC7lcDicnJwwcOBBNmzbVVj4iIiIiIiKt09ZQbXo9qN0+oqKi0KdPHzx9+hRdunRB165dkZGRgUGDBuGPP/7QZkYiIiIiIiKtMoNQ+0amR+0e5x9++AGjR4/GuHHjlLavWrUKy5YtQ+vWrcs8HBERERERkS5och1nMj1q9zjHxcWhV69eRbb36NGjyGJhRERERERExkQCofaNTI/ahXO1atVw7dq1IttjYmJQpUqVMg1FRERERESkS1INbmR61B6qPWDAAHz99ddITk5GkyZNAACXLl3Cjz/+CH9/f60FJCIiIiIi0jYWxKSK2oXz4MGDkZmZiTVr1uDJkycAAAcHB4wfPx4DBw7UWkAiIiIiIiJt4xBsUkWjy1GNHj0ao0ePxqNHj2BlZQVbW1tt5SIiIiIiItIZrpZNqqhdOO/atavExywtLeHg4IAmTZrAwsKiLHIRERERERHpjJSFM6mgduG8c+dOnDt3DlZWVqhVqxYAID4+Hjk5OXjrrbeQlpYGe3t7/PTTT6hZs6a28hIREREREZU5qUTfCciQqT0HvkGDBmjdujWOHz+O8PBwhIeHIzo6Gt7e3ujWrRtOnTqFtm3bYv78+drMS0REREREVOZ4OSpSRe3CeefOnZg6dSoqVKig2GZnZ4dJkyZh69atMDMzw5AhQ3DhwgWtBL179y7q1auHtLS0Mn/toKAgjBkzRq8ZiIiIiIhIf3g5KlJF7T93S0tL3Lt3r8j2u3fvQip99jIymQzm5hqtN0ZERERERKR3ZlK52jcyPWpXuX5+fpg+fTrGjh2Lxo0bQy6XIyYmBqtWrUK/fv3w6NEjLFiwAB4eHtrMi82bN2PHjh148uQJevfuja+++grJycmYP38+YmJi8PjxY9SrVw/ffPMN6tevj507d2Lbtm0AgDt37iAkJAS2traYOXMmLl++jNq1a6N27dqvnEEqleL+/fsqcxw4cAAODg44cOAAKlWqhPnz5+PAgQOIiIhAxYoVERAQgBYtWhR5v7S0tGJ7ucuX7hASEREREdELOMWZVFG7x3nixIkYOnQo1qxZg/79++Pjjz/G+vXrMXLkSHz++ee4evUq7OzsMHv2bG3mRVxcHCIiIrB9+3aEh4fj999/x8yZM1GtWjUcOXIEZ86cQY0aNRAYGKh4zsWLFzFmzBhERkaiXr16mDhxImrUqIHTp0/j66+/RlRU1CtnAPDSHNHR0Xj//fdx/vx5tGvXDv7+/mjUqBFOnTqFzp07Y9GiRcW+X0hICLy9vYvciIiIiIiobEglQu0bmR6NxlWPHDkSI0eOxJMnT2BhYaF0Hed27dqhXbt2ZR7wRZMmTYK1tTWcnJxQv3593L17FwEBAahYsSIAICkpCRUqVEBcXJziOZUqVULbtm0BAImJibh69SrWr18PS0tLvP/+++jdu3exw9A1yQDgpTkcHR3x0UcfAQA8PDywd+9e9O/fHwDQtm1b7Nixo9j3GzJkCHx9fYs+MHGg2pmJiIiIiKhkEhbEpIJGhfO9e/fwyy+/ID4+Hl9//TUOHz6M2rVrw8XFRUvxinp+cTILCwvk5eUhPj4eixcvxv379+Hk5AQrKysI8V/Dd3BwUPz3gwcPYGVlhUqVKim2vfPOOxoVzsVlAPDSHIVFNQCYmZnBzs5OcV8qlUIuL36+hL29Pezt7Ytsf6J2YiIiIiIiUoU9yaSK2kO1//rrL/Ts2RM3b95EdHQ0cnNzcf36dQwcOBC//fabNjOqlJ+fjzFjxsDf3x8nT57E5s2b0aZNG6V9JJL/Ziw4OjoiNzcXjx49UmxLTk5+5Rx5eXka5SAiIiIiIsMhkah/I9OjduG8aNEijB49Ghs2bICFhQWAZ3N6x4wZg2XLlmkt4MvIZDLk5ubC2toaABATE4ONGzcqeoFf9Pbbb8Pd3R3fffcdsrOzcePGDezcuVPnOYiIiIiIyHBwVW1SRe3C+dq1a+jatWuR7b169UJ8fHxZZtKIjY0N5s2bh7lz58LNzQ3Tp09H//79kZSUhMzMzGKfs3TpUqSmpqJly5b4/PPP0bFjx1fOUb58eY1zEBERERGRYWCPM6kiEc9PwlWhQ4cO+Prrr9G2bVu4urpiz549qF69Ovbv34/vv/9er8O1TdWTvu31HUFjVm3r6zuCxuy/2KvvCKVy8e2m+o5QKrVHVtZ3BM1l5+o7QanIH6bqO0KpmLs10ncEjdmPDtV3hFJJ/aazviOUSsKGV5+CpWvvx/+l7wilYowzQlNntdd3BJNSfvYv+o6gtqu1e6q9b+Pb+7SYhAyR2ouDDRgwAHPmzMHnn38OALh+/TqOHTuGlStXwt/fX2sBiYiIiIiItI2Lg5EqahfO/v7+KF++PAIDA5GdnY0JEyagatWqGD16NIYOHarFiLqzcOFCbN26tcTHDx48CEdHRx0mIiIiIiIiXZBItVM437hxA3PmzMHNmzdRvXp1BAQEwNnZuch+SUlJmDlzJi5duoQqVargq6++0snlfkk9ahfOSUlJ8PPzw0cffYSsrCzI5XLY2tqioKAAV69eRePGjbWZUye+/PJLfPnll/qOQUREREREOiY1K/vCWSaTYcyYMRg8eDA2b96Mw4cPw9/fH8eOHYOtra3Svp999hlcXFywZs0anD9/HmPHjsXu3btRvXr1Ms9FmlN7cTBvb288ffoUwLMFuQr/oJOSkjBgwACthCMiIiIiItIFiUSofVPXmTNnkJeXh6FDh8LCwgI9evRAnTp1sH//fqX97ty5g6tXr2LChAmwtLSEp6cnOnTogLCwsLL+mFRKKnucw8LCsGXLFgCAEAL+/v4wMzNT2ufhw4d4++23tZeQiIiIiIhIyzSZ45yWloa0tLQi2+3t7WFvb6+4HxsbCycnJ6V9ateujVu3bilti4uLQ7Vq1WBjY6O03+XLl9XORNqlsnDu3r07/v33XwDA1atX0aJFC5QvX15pn/Lly6NLly7aS0hERERERKRlErXH4gIhISFYsWJFke3jxo3D+PHjFfezsrJgbW2ttE+5cuWQnZ2ttC0zM7PY/XJyctQPRVqlsnC2sbHBuHHjAABvv/02evToAUtLS50EIyIiIiIi0hVNhmAPGTIEvr6+RbY/39sMPKuncnOVL1uZnZ2t1LOsyX6kP2ovDubr64tr164hNjYWcrkcwLPh2zKZDDExMZg3b57WQhIREREREWmTJqtqvzgkuyROTk4IDg5W2nb79m188MEHRfZLSkpCTk6Oouf59u3bqFOnjtqZSLvULpxXr16NZcuWwcbGBtnZ2bCzs0N6ejoAcJl0IiIiIiIyatpYVdvDwwNCCAQHB2PAgAE4fPgwbt68iU6dOintV7t2bdSvXx9Lly7F559/jgsXLiAyMlLlpXJJt9Qeyb9161ZMnToVFy5cwBtvvIE9e/YgOjoaTZo0eS0uRUVERERERKZLIhVq39RlaWmJdevW4dChQ3B3d8ePP/6IlStXonLlytizZw9cXV0V+wYFBSEuLg6enp6YNWsWAgICULduXW18VCoFtXucHz58iM6dOwMAGjRogIsXL6Jbt2744osvMH36dMVcaCIiIiIiImMjkWjndevWrYvQ0NAi2318fODj46O4X61aNaxfv147IeiVqd3jXKlSJcV1nGvVqoUbN24AABwdHZGSkqKVcERERERERLqgjR5nen2oXTh7e3tj9uzZuH79Ojw8PLB7926cOXMGwcHBeOutt7SZkYiIiIiISKskUvVvZHrU/mOfNm0aGjdujFu3bsHLywuenp4YOnQowsPDMW3aNG1mJCIiIiIi0iqpmVD7RqbnpXOcZTIZduzYge7duytdcqpOnTqYPn06+vbty+uLERERERGRUeMQbFJFZeGcnp6O4cOH4+bNm2jQoAFcXFwUjz148ABbt27F/v37sW7dOtja2mo7K73gq/OO+o6gsUXtjW9sy8N+9fQdoVRG/2Gm7wil8niV8a2ZEFwzR98RSqV8Awt9RygdIdd3Ao2l9H5P3xFKZeFqmb4jlMqMHSP1HUFjT3dv0XeEUpE/zdB3BI1l/3lX3xFKpXw/N31HeO1xCDaporJ5rFq1Cjk5OTh8+LBS0QwAX375Jfbu3YvHjx9jzZo12sxIRERERESkXRKh/o1MjsrC+fDhw/jyyy/x5ptvFvv4O++8gy+++AKHDh3SSjgiIiIiIiJd4OJgpIrKodoPHz5ErVq1VL5A/fr1eTkqIiIiIiIyaiyISRWVzePNN99EfHy8yhf4559/ULVq1bLMREREREREpFMSc/VvZHpUFs7dunVDUFAQZLLiFweRyWQICgpC+/bttZGNiIiIiIhIJzhUm1RR+XvJqFGjEBkZib59+2LQoEFo3Lgx7OzskJqaiitXrmDz5s0oKCjAmDFjdJWXiIiIiIio7LEgJhVUFs7lypVDaGgovv/+eyxatAiZmZkAACEEKlasiN69e2PMmDGoUKGCTsISERERERFpA3uSSZWXjtC3tbXF119/jRkzZiAxMRFpaWmoVKkSatasCYlEoouMRERERERE2sXCmVRQe2q7paUlnJyctJmFiIiIiIhILyTm7BSkknFNOCIiIiIiMnkSKQtnKhkLZyIiIiIiIg7VJhVYOBMREREREbHHmVRg4UxERERERCaPQ7VJFRbOREREREREHKpNKrBwJiIiIiIikycxZ+VMJWPhTERERERExKHapAILZyIiIiIiIhbOpAILZyIiIiIiMnkSCQtnKhkH8hMREREREUkl6t/KyP79+9GpUye4uLhgxIgRePToUYn7Llu2DI0bN4arq6vidvr06TLLQqqxx7kYcrkcCxcuxL59+yCEQMOGDTFnzhzs3r0bsbGxyMzMxLlz5/Duu+9izpw5cHFxAQAcOnQIa9euRWJiIoQQ6NixI+bNmwcLCwsMGjQILVu2xKFDhxAfHw93d3eMGzcO8+bNQ1xcHNzc3PDDDz/A1tZWvx+eiIiIiMgUmZvp9O1iY2Mxc+ZMrFu3Du+//z4WL16MyZMnY+PGjcXuf+3aNcycORMff/yxTnPSM+xxLsaRI0cQHR2NAwcOIDo6Gg4ODlizZg2AZ8Wxj48Pzp49i169emH06NHIyMjAvXv3MHXqVEyfPh1nzpxBWFgYjh07hqNHjype99dff0VQUBCOHz+OGzduYNKkSQgMDMSxY8eQkJCA8PDwYvOkpaXh7t27RW5ERERERFRGdNzjvGfPHnTo0AHNmjWDlZUVPv/8c1y4cAHx8fHF7h8TE4MGDRqUyXuT5tjjXAw7OzukpKRg9+7d8PLyQkBAAKRSKYKCgtC8eXP4+PgAAIYNG4aQkBCcPHkS7dq1w759+1C9enWkp6fj8ePHqFSpElJSUhSv26dPH1SvXh0A0LBhQ1SvXh01atQAALi4uJRYDIeEhGDFihVFtneCSxl/ciIiIiIi0yTRoCBOS0tDWlpake329vawt7dX3JfJZEhNTS36XhIJYmNj0bhxY8W2cuXKoVq1arh16xZq1aqltH9KSgoePHiAtWvX4tKlS6hYsSL8/f3Rt29ftTPTq2HhXIyWLVti9uzZ2LJlCxYuXIjq1atj2rRpAICaNWsq9pNIJHB0dMSDBw9gYWGBHTt2ICwsDNbW1mjYsCFyc3MhhFDsX6lSJcV/S6VSpb9UUqkUcrm82DxDhgyBr69vke0LW3/+yp+ViIiIiIigUU9ySR1b48aNw/jx4xX3L168iMGDBxfZz8zMDO7u7ihXrpzSdmtra2RnZxfZ//Hjx3B3d8fAgQOxbNkyXLx4EaNHj0aVKlXQvn17tXNT6bFwLkZiYiIaNmyILVu2ID09Hb/++ismTZqEYcOGITk5WbGfEAL3799HtWrVEBERgb1792LHjh1wdHQEAEXP9Kt68ZcrIiIiIiIqYxL1Z7GW1LH14r/ZPTw8cPPmzWJfY/To0cjJyVHalpOTg/LlyxfZt379+ti0aZPivru7O3r37o3Dhw+zcNYRznEuxqlTpzBu3DgkJSXB1tYWFSpUgJ2dHczMzPDnn38iKioKeXl5WLduHaRSKTw9PZGeng4zMzNYWloiLy8PmzZtws2bN5GXl6fvj0NERERERC+jwRxne3t7vPPOO0VumnR21alTB3fu3FHcz87Oxv379+Hk5FRk33PnziE4OFhpW15eHqysrEr9cUkz7HEuRt++fXHnzh34+fkhMzMT7777LpYvX44TJ07A2dkZGzduxOTJk1GvXj2sX78e1tbW8PX1xenTp9GxY0dYWlqiadOm6NmzJ/7++299fxwiIiIiInoJiY5X1e7Zsyf+97//4fTp03B1dcWSJUvQoEEDvPvuu0X2tba2xpIlS1CrVi20bdsWp06dwr59+xASEqLTzKaMhXMxpFIppk6diqlTpyptP3HiBCpXroxVq1YVeY61tTV++OGHEl/z+aEVAIq8xsKFC0sfmIiIiIiIXk0ZXp9ZHfXq1cO3336LOXPmIDk5GU2aNMGyZcsUj8+ePRtJSUlYv349GjdujEWLFmHx4sWYNGkSqlWrhgULFsDZ2VmnmU0ZC2ciIiIiIiIN5jiXlS5duqBLly7FPjZv3jyl+926dUO3bt10EYuKwcKZiIiIiIhIxz3OZFxYOGvg+aXliYiIiIjoNSLluslUMhbOREREREREZrpdHIyMCwtnIiIiIiIiDtUmFVg4ExERERGRyZNwqDapwMKZiIiIiIiIPc6kAgtnIiIiIiIiPVyOiowHC2ciIiIiIiL2OJMKLJyJiIiIiIi4qjapwMKZiIiIiIiIi4ORCiyciYiIiIiIWDiTCiyciYiIiIiIOMeZVGDhTERERERExFW1SQUWzkREREREZPIk5lwcjErGwtmI3SpI1XcEzVm8qe8EGpNaG+eXaBWJlb4jlIqlEa5oWfPsFX1HKJUnXi31HaFU5P+m6DuCxiRGOvyvitw4c8uP7tZ3BJMhT8/RdwSNlfN8S98RSoe9odrHY0wqsHAmIiIiIiLi4mCkAgtnIiIiIiIiFs6kAgtnIiIiIiIiiXFOTyHdYOFMRERERETEHmdSgYUzERERERGRGUsjKhlbBxEREREREXucSQUWzkRERERERJzjTCqwcCYiIiIiImKPM6nAwpmIiIiIiIiFM6nA1kFERERERGRmrv6tjAUHB2PMmDEq9zl16hR69eoFFxcXfPTRR0hISCjzHFQyFs5EREREREQSqfq3MpKZmYnvvvsOCxcuVLnf48ePMXbsWIwbNw5nz55Fx44d4e/vD7lcXmZZSDUWzkRERERERFKp+rcyMmrUKNy7dw/9+/dXud+RI0fw3nvvoUuXLrCwsMAnn3wCmUyGkydPllkWUo1znImIiIiIiDToSU5LS0NaWlqR7fb29rC3t1fcl8lkSE1NLfpWEgmqVq2KJUuWwNHREUFBQXjw4EGJ7xcbGwsnJyelbe+++y5u3bqFVq1aqZ2bSo+FMxERERERkQY9ySEhIVixYkWR7ePGjcP48eMV9y9evIjBgwcX2c/MzAzXrl2Do6OjWu+XlZWFcuXKKW2ztrZGdna22pnp1bBwJiIiIiIi0qBwHjJkMHx9fYtsf763GQA8PDxw8+bNV45Wrlw55OTkKG3LycmBjY3NK782qYeFMxERERERkQarZb84JFvb6tSpg7179yptu337Nj755BOdZTB1XByMiIiIiIhID6tqq6tTp064ceMG9u/fj7y8PGzYsAFSqRTu7u46z2KqWDjrSUhICLy9veHm5oYPP/wQ586d03ckIiIiIiLTpYdVtVWZPXu2oke5SpUq+PHHH7F27Vq4u7vj0KFD+PHHH2FpaamTLMSh2npx+PBhrF+/Hj///DNq166N8PBwjBo1CkeOHEHlypX1HY+IiIiIyPTooSe50PMLihWaN2+e0v3mzZtj165dOkpEL3pte5zlcjm+/fZbtGzZEp6envD390dCQgKCgoIwceJEfPLJJ3BxcYGvry8uXbqkeN6hQ4fQt29fuLu7o3nz5pg+fTry8vIAAIMGDcLq1avxwQcfwMXFBSNHjsTly5fx4YcfwtXVFZ988gkyMjJemu3Bgwf49NNPUadOHUilUvTt2xdmZmZlsnAAERERERGVgoH1OJNheW3/1I8cOYLo6GgcOHAA0dHRcHBwwJo1awA8K459fHxw9uxZ9OrVC6NHj0ZGRgbu3buHqVOnYvr06Thz5gzCwsJw7NgxHD16VPG6v/76K4KCgnD8+HHcuHEDkyZNQmBgII4dO4aEhASEh4e/NNuAAQMwYMAAxf2zZ88iKysL7733XrH7p6Wl4e7du0VuRERERERUNiRmFmrfyPS8tkO17ezskJKSgt27d8PLywsBAQGQSqUICgpC8+bN4ePjAwAYNmwYQkJCcPLkSbRr1w779u1D9erVkZ6ejsePH6NSpUpISUlRvG6fPn1QvXp1AEDDhg1RvXp11KhRAwDg4uKicUF769YtTJ48GRMnTkTVqlWL3aek68TVxLsavRcREREREZWAPcmkwmtbOLds2RKzZ8/Gli1bsHDhQlSvXh3Tpk0DANSsWVOxn0QigaOjIx48eAALCwvs2LEDYWFhsLa2RsOGDZGbmwshhGL/SpUqKf5bKpUqLUMvlUohl8vVznjs2DFMnToVI0aMwIgRI0rcb8iQIcVeJ264J5efJyIiIiIqE3qc40yG77UtnBMTE9GwYUNs2bIF6enp+PXXXzFp0iQMGzYMycnJiv2EELh//z6qVauGiIgI7N27Fzt27ICjoyMAKHqmy1pISAh++OEHLFy4EF26dFG5r66vE0dEREREZHLY40wqvLat49SpUxg3bhySkpJga2uLChUqwM7ODmZmZvjzzz8RFRWFvLw8rFu3DlKpFJ6enkhPT4eZmRksLS2Rl5eHTZs24ebNm4rFwcrK/v37sXTpUgQHB7+0aCYiIiIiIh0w4Os4k/69tj3Offv2xZ07d+Dn54fMzEy8++67WL58OU6cOAFnZ2ds3LgRkydPRr169bB+/XpYW1vD19cXp0+fRseOHWFpaYmmTZuiZ8+e+Pvvv8s027p16yCTyTB06FCl7YGBgfDy8irT9yIiIiIiIjVIzfSdgAzYa1s4S6VSTJ06FVOnTlXafuLECVSuXBmrVq0q8hxra2v88MMPJb7mpk2blO6/+BoLFy5UK5s6K28TEREREZEOmb22pRGVAbYOIiIiIiIiDsEmFVg4a8GHH36IuLi4Yh+rXr069uzZo+NERERERESkikTCodpUMpMrnMePH6/19wgLC9P6exARERERURniqtqkgskVzkREREREREVwqDapwMKZiIiIiIiIi4ORCmwdREREREREvBwVqcDCmYiIiIiIiEO1SQUWzkRERERERFwcjFRg4UxERERERCZPwh5nUoGFMxEREREREXucSQUWzkRERERERGYW+k5ABoyFMxEREREREYdqkwosnImIiIiIiDhUm1Rg4UxERERERCaPi4ORKiyciYiIiIiI2ONMKrBwJiIiIiIi4uJgpIJECCH0HYKIiIiIiIjIUHE8AhEREREREZEKLJyJiIiIiIiIVGDhTERERERERKQCC2ciIiIiIiIiFVg4ExEREREREanAwpmIiIiIiIhIhf8DZ1YLO/6Xzc0AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1008x432 with 2 Axes>"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "source": [
    "print('Spam directory is related to the following topics : \\n', pd.DataFrame([','.join(i) for i in topic_keys]).iloc[0,0],'\\t', pd.DataFrame([','.join(i) for i in topic_keys]).iloc[3,0],'\\t', pd.DataFrame([','.join(i) for i in topic_keys]).iloc[4,0])\n",
    "print('\\n','Spam_2 directory is related to the following topics : \\n', pd.DataFrame([','.join(i) for i in topic_keys]).iloc[0,0],'\\n',pd.DataFrame([','.join(i) for i in topic_keys]).iloc[5,0])"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Spam directory is related to the following topics : \n",
      " NUMd,font,NUM,size,face,color,NUMdNUM,nbsp,width,align,arial,center,table,height,http,div,style,helvetica,verdana,serif \t NUM,money,blockquote,business,people,government,mail,free,report,states,make,new,order,send,one,name,million,company,address,united \t com,received,net,localhost,org,netnoteinc,http,esmtp,content,zzzz,aug,mail,jul,subject,mon,NUM,html,smtp,sep,free\n",
      "\n",
      " Spam_2 directory is related to the following topics : \n",
      " NUMd,font,NUM,size,face,color,NUMdNUM,nbsp,width,align,arial,center,table,height,http,div,style,helvetica,verdana,serif \n",
      " NUM,NUMe,net,NUMc,pro,ssz,cypherpunks,jul,iiu,minder,cpunks,einstein,NUMffont,NUMcfont,sightings,NUMb,locust,waste,envelope,received\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/boyun/opt/anaconda3/lib/python3.8/site-packages/ipykernel/ipkernel.py:287: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "  and should_run_async(code)\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "source": [
    "target_labels = [\"easy_ham\", \"easy_ham_2\", \"hard_ham\", \"spam\", \"spam_2\"]\n",
    "\n",
    "for _topic_index in range(0, len(topic_keys)):\n",
    "    \n",
    "    lmw.plot_categories_by_topic_boxplots(directory,\n",
    "                                          topic_distributions,\n",
    "                                          topic_keys, \n",
    "                                          _topic_index,\n",
    "                                          output_path=output_directory_path + '/boxplot.' + str(_topic_index) + '.pdf',\n",
    "                                          target_labels=target_labels,\n",
    "                                          dim=(4,4))"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUUAAAEPCAYAAADVt9ZXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABXm0lEQVR4nO2deVxN+f/HX7dVicJEdiqFRNEie1HZKpKJ0YIwthh7IZmh7EsqirElTKKM1IiyDGM31ijTgkS0KdqXz++Pfvd8Oy23UPeeW5/n49GDz9nu+3zuua/zWd7v94dHCCGgUCgUCgBAQtQGUCgUCpegokihUCgVoKJIoVAoFaCiSKFQKBWgokihUCgVoKJIoVAoFfhmUXRxcYGmpmaNf97e3t9lmL29PbZs2fJd1wCAO3fuQFNTEwcPHqyyLzAwECYmJkzZxMQEgYGBVY57+/YtNDU18fLlSwCAt7c3NDU1MXfu3Go/c+HChdDU1MSVK1e+236uwa/P3Nzcb9r/NdTXM/AtFBcX4/jx4998vre3N6ytrevRov9Rn3XcmHBxccGiRYu++zpS33rimjVrsGzZMgBAUlIS7O3tERwcjPbt2wMA5OXlv8swb29vSEl9s3nVXs/c3BydOnWql+tJS0vj5s2byMvLY91rfn4+/vnnn3r5DIroOH/+PPbs2YNp06aJ2pQq6Orq4saNG9/9G6NUzze3FFu0aAFlZWUoKytDSUkJANC6dWtmW/Pmzb/LMCUlJSgoKHzXNSrSunVr/Prrr/V2PXV1dcjLy+PGjRus7devX4empma9fQ5FNHA5pkFGRgbKysrg8XiiNqVR0mBjip8/f8aGDRswdOhQ9OvXD05OTkhMTGT2m5iY4PDhw7C1tUXfvn1ha2uLmJgYZn/lrtOJEydgbm6Ofv36wcbGBvfv3wfwv65tSEiIQHvc3d1x/fp1RERE1Mv9SUhIwMTEBFFRUaztkZGRMDc3F3iui4sLNm7cCFdXV+jo6GDYsGGIiIjAxYsXMWrUKOjq6mLp0qUoKipizgkLC4OFhQX69u0Lc3NzhIaGMvu8vb2xcOFCbNmyBQYGBhgyZAg2btyI0tJS5pizZ88y9Tdx4kRcvXoVAPDx40f07t0bt27dYtloamqKM2fO1HgPISEhGDJkCHR1dbFu3ToUFBRUe1xtz8GnT5/g4uICAwMDGBgYYMWKFfj8+XOV62RlZWHs2LFYsGABSkpKAAAHDhzA0KFD0b9/f3h6emLhwoXMsE11XanKwyM11emdO3fg6uqKT58+QVNTE3fu3BFYhzVBCMHGjRuhq6uLIUOGICAggNlXXfe64jP/5csXLFu2DIaGhtDR0YGTkxNevXrF2Fex+8x//q2trdGvXz9MnjwZDx8+ZK6blpaGxYsXM3asWbOGVcdBQUEwNTVFnz59MHr0aJw9e5bZd+nSJYwfPx7a2towMTHB77//XuP92tvbY9euXXByckLfvn1hYWFRpdd08OBBGBsbQ1dXF1OnTsWjR49Y9RUQEABzc3Noa2vDysoK165dY/a7uLhg3bp1WLZsGfr16wdTU1OcP3++RnseP36MKVOmQFtbG2ZmZjhw4ADKyspqPJ5Pg4niokWLcOfOHezcuROnTp2CrKwsnJyckJ+fzxzj5eUFS0tLhIaGonv37pg5cyY+ffpU5VpnzpzBli1bMHv2bJw7dw6Ghob4+eefkZWVhfbt2+PGjRsYO3asQHv09PRgY2MDT09P5OTk1Ms9mpmZ4dq1a8yPtKioCNeuXYOpqWmt5/7xxx/o0qULwsLCmAf14MGD8PLywq5duxAdHY1z584BAM6dOwdXV1dMnToV586dg729Pdzc3Fg/yqtXryInJwd//PEHFi1ahOPHjzOCff36dWzcuBGLFi1CWFgYbG1tsWjRIjx8+BBt27bFwIEDWQ/Xo0eP8OHDB5iZmdVo/6lTp+Dj44MDBw7g1q1b2LhxY7XH1fYcLFy4EC9fvoS/vz+OHj2K+Pj4Ki363NxczJkzBx06dMCuXbsgJSWF4OBg+Pr6YtWqVQgKCsK7d+9YP6DaEFSnurq6WL16NZSUlHDjxg3o6uoKrMOaeP78OdLS0hAcHIyVK1dix44dzHdaG7t378bbt28REBCAkJAQSEhIYPXq1QKPX7x4MYKCgiAtLY1169Yx+5ydnUEIQVBQEPbt24c3b95gyZIlAICYmBhs2LABLi4uiIyMhL29PVxcXPDq1Sukp6djyZIlcHR0xIULF7By5Urs3r27ygu0IgcPHoSuri5CQ0MxfPhw/Pzzz4yY//HHHwgICIC7uzuz39HREcnJyQAAPz8/eHt7Y9GiRTh37hxGjRqFefPmITY2lrl+SEgIFBQUEBISgmnTpmHFihXMS6siGRkZcHJywtChQxEWFoY1a9bgxIkTAkWdgdQDcXFxRENDgyQnJ7PKT548YY7Jzc0lBgYGJCgoiBBCiLGxMXF1dWX2FxYWkkGDBpHjx48TQgixs7MjmzdvJoQQYm1tTTw8PJhjy8rKyNatW8mrV69qte327dtEQ0ODfPnyhXz69IkMGjSIuLm5EUIIOXbsGDE2NmaONTY2JseOHatyjeTkZKKhoUHi4uIIIYTs2bOHTJw4kRQWFhJdXV1y69YtQgghV69eJRMnTiSEEKKhoUEuX75crU2rVq0iY8eOZcqPHz8mGhoa5J9//mG22dnZkU2bNhFCCJk4cSL59ddfWddYt24dsbW1Zezp378/KSwsZPZPmDCB7NixgxBCyLRp08jevXtZ569du5Y4OzsTQggJDQ0l+vr6pKioiBBCyIYNG5h9leHX54MHD5htly5dIlpaWiQ3N5dV37U9By9fviQaGhrkxYsXzP4nT54QX19fpg42bNhApk+fTuzt7Ul+fj5z3MSJE5n6IYSQz58/Ez09PbJnzx6mjivfQ8Xvt7Y6PXPmDDEwMGD21VaHldmzZw/R19cnubm5zDYPDw/Wd8Z/VvhUfObnzp1LZsyYQb58+UIIISQ1NZWp84p1TEj5s+bv789cJyoqimhoaJDCwkJy69YtoqOjw3o2UlNTmef54sWLRFtbm/Ud3Lhxg2RnZ5OYmBiioaFBrl+/zuy7d+8eSUtLq/ae7ezsiIODA2ubpaUl2bp1KyGEkBEjRpA///yTtX/GjBlk8+bNpKysjBgaGpLff/+dtd/JyYksXbqUEFL+nZqZmZHS0lJm/7x588jixYuZ/fzvw8vLi0yfPp11rXPnzrG+05qov5mMCsTHx0NaWhp9+vRhtsnLy6N3797477//mG16enrM/2VkZKCpqcnazychIQHTp09nyjweDytWrPhquxQVFeHq6orly5fDysqqyn4pKalqm9fk/8eXKk/8yMjIYPjw4YiOjsbAgQNx8eLFWrvOfLp06cL8v1mzZgCAzp07s67N7z7Hx8ez7h8ABgwYgPDwcKbcoUMHyMjIMGUFBQUUFxcDAP777z88fvwY+/fvZ/YXFxeje/fuAMq7yuvXr8eNGzcwbNgw/PXXXwLHXyUkJNC3b1+mrK2tjeLiYrx+/Zp1XG3PQYsWLSAtLc0ag9XW1oa2tjZTDgoKQnFxMcaMGcPUE1D+TMyYMYN1vz179qzR5srUpU4rUlsdVoeGhgZrMkRbW7vWYR4+P//8M37++WcYGRlBX18fo0aNqvaZ5dOtWzfm//yx+JKSEsTHxyM/Px+GhoZVzklKSsLw4cPRv39/WFlZoUePHhgxYgSsra3RsmVL9OrVCxYWFnByckLnzp0xYsQIWFlZ4YcffqjRjoq/aQDo27cv/vvvP+Tm5uLdu3dwc3ODu7s7s7+oqAgyMjLIyMhAVlYWdHR0WOcPGDAAFy5cYMq6urqQkPhfB7dv374ICwurYkd8fDzu3r0LXV1dZltZWRkKCgqQlZWFVq1a1XgPDSKKsrKy1W4nhLAGsCUlJVn7y8rKqmwDymd662tQefz48QgNDcW6detgY2PD2teyZctqx7Oys7OZ/ZUxNTXF9u3b4eLigsuXL+PEiRN1skNaWrrKtprusaIY8CGEsAS8uuvxKS0txbJly2BsbMzazhf55s2bY+TIkYiIiGDEeNiwYTVej8fjsb4n/nda2YbanoO6fK+qqqpYsWIFZs+ejYkTJzJ2VVcnFV8K1V2XP8xR0/mV67QitdVhdVT88QLlzze/jmqzT0dHB9HR0bhy5Qr+/vtv7Nq1CydOnMDp06er/azqvn9CCEpKStChQwccPny4yv42bdqgWbNmOHz4MP79919cuXIFV69eRUBAAPz9/WFkZITt27fDyckJ0dHRuHbtGk6cOAFPT09MmDChWjuq+01LSEgw9bp582b07t2bdUyzZs2q/T7491DxO6np+pUpKSmBmZkZfvnllyr7WrRoUe1n8WmQMUU1NTUUFxfj2bNnzLb8/HzExsay3qzPnz9n/l9YWIi4uLhqZ267devGmoQBgHHjxiEyMvKb7Fu/fj1SUlKq+CT27t272jGihw8fQllZudo35PDhw5Geno7AwED88MMPAlsO34qqqiprQJpvk6qqap3OV1NTQ0pKCrp27cr8/fXXX6xWkaWlJf7++29ERUVh9OjRLIGpTGlpKatF/+jRI8jKyrJauvzPFfQcdO/eHUVFRYiPj2f23717F8OHD2dayYMGDcKQIUNgY2ODX3/9lZnQ0dDQYNVJUVER4uLimLK0tDTrBZebm4vMzEymXFudVhatutRhZRISEliTXY8fP4a6ujpj35cvX5h9hBC8ffuWKfv7++PJkyewsLDAtm3bEBQUhJcvX7LusS6oqanh48ePaN68OWO3tLQ0Nm/ejMzMTNy5cwf79u3DgAEDsHz5cpw/fx69e/dGZGQkYmNjsWnTJvTq1QsLFy5EcHAwxowZI/CeK/6mAeDp06fo2bMn463y4cMHVh0GBgbi+vXrUFBQQNu2bWt9zitf/8mTJ9X2ENTU1JCYmMj6rISEBPj6+lYrohVpEFHs1q0bzMzMsHr1aty/fx8vX77EqlWrICkpiXHjxjHHBQUFISwsDAkJCXBzc4OUlBTGjBlT5XrTp0/HH3/8gXPnzuHNmzfYsWMH0tPToa+vj9LSUqSlpdU4+1kdnTt3xoIFC1gPIQD89NNPuH37NrZu3YqEhAQkJiYiODgY3t7emD17drXXat68OQYNGgQvLy+BExPfw5w5c3Dq1CmcOHECr169wsmTJ3H69Gk4ODjU6fxZs2bhjz/+wMmTJ/HmzRucOnUK3t7e6NixI3PMkCFDICUlhTNnzsDCwkLg9Xg8HlxdXfHkyROmvhwcHKq0DGt7DtTU1DBkyBCsXbsWz549Q0xMDDZv3gwjI6Mqorx06VJ8+fIFPj4+rDo5ffo0EhMT8euvvyItLY05XltbG3fu3EFUVBQSExOxdu1a1o+htjqVl5dHXl4e4uPjUVhYWKc6rExmZibWrFmD+Ph4hISEIDg4mHmOtLW18fr1a5w4cQJv3ryBp6cn0yMBgNTUVGzcuBH//vsvkpOTERoaCgUFBVY3uS4MHjwYPXr0wJIlS/Ds2TPExsZixYoVSE5ORseOHSEnJwdfX1+cOnUKKSkpuH79OhITE6GtrQ1FRUWcPHkSvr6+SE5OxoMHD/Do0SPW8EZloqOjERgYiKSkJGzduhXJycn48ccfAZQ/h3v37kVERASSk5Oxd+9eHD9+nGlIzJkzB35+fggPD8erV6+wd+9e3LhxA/b29sz1nz9/jl27diEpKQmHDx/G9evXWfv5TJs2Da9fv8bGjRuRmJiImzdvYt26dZCTk6tVFBuk+wwAnp6e2LRpE+bNm4eSkhLo6+sjMDCQ8WkEABsbGxw+fBgJCQnQ0dHBkSNHqvVNHDduHNLS0rB7925kZGSgV69eOHDgAFq3bo23b99i5MiR2LRp01dFEMyYMQPnz59ntSZ69uyJI0eOwMfHB8HBwSgqKkLXrl2xYsUKTJ48ucZrmZqa4sqVKw0miiYmJnB3d8eBAwfg6emJrl27YuPGjbC0tKzT+aampnBzc8PBgwfh4eGBjh07wt3dnfWCkpSUxJgxY3Dp0iXo6+sLvJ6cnBysra0xZ84clJSUYMKECTVGEtT2HGzbtg0bNmyAg4MDpKWlYWZmBhcXlyrXadWqFZYtW4Zff/0VFhYWGDp0KNatWwdfX19kZWVh0qRJ6NWrF3O8lZUVHj16hJUrV0JWVhYzZsxAVlZWnevUyMgIvXv3xoQJE7Bjxw6Ym5vXWoeVGTRoEKSkpGBjYwNFRUWsXbuW6f4bGRlhzpw58PLyws6dOzFp0iTWtVasWMG4GX3+/Bm9evXC/v37qx3CEYSEhAT27dsHDw8PODg4QEJCAoaGhti2bRskJSXRt29feHh4wN/fHxs2bECbNm0wY8YMTJo0CUC569Du3buxf/9+KCgoYPz48TVGcgHlv9WoqChs3boVmpqaOHToEDp06AAAcHBwQEFBAbZt24b09HR069YNe/bswYABAwAAdnZ2yMvLw7Zt25CRkQENDQ34+fmxximHDBmC169fw8rKCl26dIGPjw9rfJuPiooKfv/9d2zfvh1WVlZQVFTE2LFjsXz58lrrjEeIaLxUTUxMMHPmTNjZ2Yni4ynVsGjRInTt2pWJVBI3rK2tYWxsDGdnZ1Gb0iSxt7dHnz59sGrVqga5vouLC/Ly8rBnz54GuT6fBmspUsSHu3fv4vnz57h27ZpAZ1gKpSlARZGC8PBwhIWFYcWKFVUmSyiUpobIus8UCoXCRWg+RQqFQqlAkxPFkpISvH37luUoS6FQKHyanCimpqZi5MiRSE1NFbUpFAqFg4hEFJ88eQIjI6Ma97979w4zZsyArq4uRo0axcp+QgjBrl27YGRkBD09PXh6etJWH4VCqTeEKoqEEAQHB2PmzJlMsoLqWLp0KZPHbsOGDViyZAmTXigoKAiXLl1CaGgoLl68iKdPn8LPz09Yt0ChUBo5QhXFPXv24OTJk5g3b16NxyQlJeHZs2dYtGgRZGRkYGRkBBMTEyYQ/uzZs3B0dISKigpat24NZ2dnBAUFCesWqiU7Oxve3t71lqeRIh68ffsWLi4uSElJEbUplHpEqKI4ZcoUhISEsFJJVSYhIQHt27dnpVxSVVVlAuHj4+OhpqbG2vfx48dqk9Pm5OTg7du3rL+GGEsMCgpCQkKCyMWZIlwCAwNRUFCAY8eOidoUSj0iVOftdu3a1XpMbm5ulTRCcnJyTMKHvLw8yMnJMfv4x1aXEOLo0aNMAoGGIjs7m8ncERMTg5ycnK+OT6WIHxVfsKmpqUhJSRGYHIIiPnBu9lleXh6FhYWsbfn5+UzLsaJAAv8Tw+pWNnN0dER0dDTr73uWrayOyq1D2lpsGlROO0dbi40Hzomimpoa3r17xxK+xMREJg+duro6kpKSWPuUlZWrbZ21bNkSnTp1Yv2pqKjUq72V87tVzvtIaZxUHobhsosXHfv8OjgniqqqqujZsyd27dqFoqIi3L59G9HR0Rg/fjyA8mSohw4dQkpKCjIzM+Ht7S0wTTuF0hBUfrnW98u2PqFjn18HJ0Tx3LlzrLUUvL29kZCQACMjI6xduxYeHh7Q0NAAAEydOhWjR4/GlClTYG5uDnV1dSxevFhUplcZ/6wprTqlcVE5Hf/EiRNFY0gtVDf2SRFMk0sIwU9KGx0djU6dOn339WJjY1l+kvPmzat2SQVK4yI4OJi1pvHgwYMFJiIWFZs3b2Z17VVUVKpN4kv5H5xoKYozPXv2ZFqHzZo1o4LYRLh//77AMlcQp7FPrkBFsR6YPn06eDwea8lNSuOm8hKZgpbMFCXKysoCy5Sq0CSz9QB/YojSdKi4MmB1Za7QoUMH1oJe1JeydmhLkUL5Bvr16yewzBViY2NZ5RcvXojIEvGBiiKF8g2Iy/xk5ZXuqlv5jsKGiiKF8g08ffqUVX7y5ImILBGMuIg3l6CiSOEc4pB1qHJYaXVhplxAXMSbS1BRpHCO8+fPIyEhAWFhYaI2pUaysrIElrmCnp4eJCUlAQCSkpKsheUp1UNFkcIpsrOzGZ+/+/fvc7q1KA6YmZmBx+MBACQkJGBubi5ii7gPFUUKpzh//jwzDkYI4Wxrkd/6qqnMFRQVFWFoaAgejwcDAwOa1q4OUFGkcIoHDx4ILHMFCQkJgWUuYWRkBFlZWQwaNEjUpogF3P0mKU0SflevpjJXqOyXqKOjIxpD6sCtW7dQWFiImzdvitoUsYCKIoVTVMyWBAD9+/cXkSWC4apYVyY7Oxt37twBIQR3796lY7R1gIoihVNYWFgwXVEJCQlYWFiI2KLqqezawlVXl4sXL6KsrAwAUFpaisjISBFbxH2oKFI4haKiItM6HDBgAGcnBnr27Mkq9+rVS0SWCOb+/fuMKJaVlXE2mw+XoKJI4RwWFhZQU1PjbCsRAF69esUqV1wig0vQML+vh2bJoXAORUVFODs7i9oMgWRnZwssc4Xc3FxWOS8vT0SWiA+0pUihNGIqZ8WpvNAapSpUFCmUb0BGRkZgmStUTghBE0TUDhVFCuUbUFNTY5V79OghIksEIy5+n1yCiiKF8g0kJiayyvHx8SKyRDC0pfj1UFGkUL4BOqvbeKGiSKF8A4WFhaxycXGxiCyh1DdUFCmUb+DZs2esMlcjWihfj1D9FGNjY+Hu7o64uDh07twZHh4eVbod9+/fx+zZs1nbioqK0KlTJyZEacSIEfj06RMzaNy2bVsavkQRKvwokZrKFPFFaKJYVFSE+fPnw8HBAYGBgbh48SKcnJxw5coVKCgoMMfp6enh4cOHTPn9+/ewsbGBm5sbgPKlJD98+IAHDx5wNgU8hcIVeDwea3KFzj7XjtC6z3fv3kVxcTGmT58OaWlpjBs3Durq6oiIiBB43urVq2FpaYkhQ4YAAGJiYtC1a1cqiBRKHai8/ABdjqB2hCaK8fHxVXy7VFVV8fLlyxrPiYqKQnx8PBYvXsxse/78OQghsLGxwcCBA+Hk5ISEhIQGs5tCEWfGjx/PKnM5npwrCE0U8/Ly0KxZM9Y2OTk55Ofn13jOvn37MGfOHNZ5EhIS0NbWho+PD65cuYJevXph9uzZ1V4nJycHb9++Zf2lpqbW301Rmizi4hT9+fNngWVKVYQ2pigvL1/FjSE/P7/GbnBsbCzi4+MxceJE1vbKkzDLli3DiRMnEBMTU6VrcPToUfj4+NSD9RQKG3Fxij569CirfOTIEaxZs0ZE1ogHQhNFNTU1HDlyhLUtMTEREyZMqPb46OhoDBs2jDUJA5R/qX369GEEsLS0FKWlpdXGnjo6OlYR1dTUVEybNu3bb4RCESPS0tIElilVEVr32dDQEIQQHDlyBMXFxQgPD0dcXBxMTU2rPf7x48fVpqJPSUmBp6cnPn78iIKCAmzevBldu3aFlpZWlWNbtmyJTp06sf5UVFTq/d4o9Ut2dja8vb05nTq/VatWAssU8UVooigjI4MDBw4gMjISBgYG8PPzg6+vL1q3bo1z585VWZsjJSUFbdu2rXKd5cuXo1+/fpg4cSKMjIyQnJwMPz8/zi4xSfl6Ll68iMTERE77nn769ElgmSK+1Ln7fPfuXRgYGHzXh2loaODkyZNVtltaWsLS0pK1LTw8vNpryMrKwt3dHe7u7t9lC4WbVF5oydzcnJNLEojLmCLl66lzS3Hu3LkYMWIEtm3bhri4uIa0idKEuXjxIiMwZWVlnG4tUhondRbFW7duYcWKFUhMTISNjQ0sLCywf/9+vHv3riHtozQx7t+/j9LSUgDlk2h0oSWKsKmzKMrKymLcuHHYt28fbty4AQcHBzx8+BDjxo3DtGnTEBwcXGU9CArla9HT02PGhyUlJWkEBkXofNNES1paGt6/f4/379+jpKQELVu2RFBQEExMTBAVFVXfNlKaEGZmZsz/eTwezM3NRWgNpSlS54mW5ORkhIeHIzw8HPHx8dDR0YGtrS3Gjh0LRUVFAMDu3bvh5uaGUaNGNZjBXCQ7OxsBAQFwdHTk5KSAOKGoqAglJSVkZGRAUVGR1idF6NRZFE1NTdG9e3eMHz8elpaW6Ny5c5Vj+vfv3yRXC6voQjJ58mRRmyPWZGdnIyMjAwCQkZGBnJwcKowUoVLn7vOpU6cQHh6OBQsWsASxtLSUSbg5bNgw7N+/v/6t5DCVXUi47HAsDoSEhAgsU76Oyi8Ufq+OUjN1FkVbW9tqHVTfvXvXpMPmqAtJ/VI5g/Xjx49FZEnj4KeffhJYplRFYPf59OnT+OOPPwCUO6c6OTlViRxJT09Hx44dG85CjlOdCwlXu9DiMPZJnaLrl8ouTffv34empqaIrBEPBIri2LFjmVRbz549w8CBA9G8eXPWMc2bN2/SM4R6enr4559/WGWuIg5jnzIyMigqKmKVKd9OxSz2APDvv/826Z5dXRAoivLy8li4cCEAoGPHjhg3bhx9SCuhra3NEkWuLnUpLuFzFQWxujLl66At769HoCgeP34cNjY2kJWVRV5eHoKDg2s8tqm+fc6cOcMqnz59mpP56qob++Rqa5FSfwwYMAD37t1jlSmCESiKBw8exNixYyErK4uDBw/WeByPx2uyoigu+erEZexTQkKCtTKehARdhfd7GD58OEsUR4wYITpjxASBonj58uVq/08RP/T09HD79m2UlpZyOnyOLh1av1y9erVKuak2YOqKwNdwfn5+nf+aKpXHEHV0dERjSC2YmZkx64hISEg06cmxpsS///7LKj948EBElogPAluKurq6tS7IQwgBj8fDixcv6tUwccHMzIzlW1dTJnFRo6ioCENDQ9y8eRMGBgacnGSh1D+05f31CBTFo0ePcnaVMq5w69YtVvnmzZucHKsDygU8NTWVthIpFAEIFEVDQ0Nh2SG2VOccy1VRVFRUhLOzs6jNEAiPx2O5jdCXMkXYCBRFIyMjhIeHo3Xr1hg4cKDAB7Ryi6mpoKenh5s3bzLDCFydwBAXqF8dRdQIFMWVK1cyS4yuWrVKKAaJG0ZGRozzNiEEgwYNErFFFArlexAoihXXTK68fjKlHHEaU6RQKLXzVZ6x58+fx08//QQDAwMMHjwYDg4OrBC3pkh1Y4oUCkV8qbMoHjt2DGvXroW2tjbWrVuHVatWQV1dHfPnzxcY/tfYoWuKUCiNizpn3t6/fz82btyI8ePHM9ssLS3Rp08feHt7N9kuo5mZGW7fvg2ArilCoTQG6txSzM3NRa9evaps79evX7XJZ5sKioqKaN26NQCgVatW1Cn6O6kc60xjnynCps5P3KRJk+Dr64vCwkJmGyEEv//+OywsLOp0jdjYWNja2kJHRwcWFhZVsizzuXXrFnr16gVdXV3mz9fXl/nMXbt2wcjICHp6evD09ERJSUldb6Peyc7ORnp6OoDyhLt0OYLvo3///qwyzepCETYCu8+TJk1ifBPLysrw/Plz/PPPP9DQ0ICkpCTi4+ORmZmJwYMH1/pBRUVFmD9/PhwcHBAYGIiLFy/CyckJV65cYdx++Dx//hyjR4/Grl27qlwnKCgIly5dQmhoKGRkZLBgwQL4+fkxeR+FTVhYGONLRwhBWFgYZwPuxSHztoWFBR48eMD4fdb1hUuh1BcCRdHY2JhVNjExYZW/ZlLh7t27KC4uxvTp0wEA48aNQ2BgICIiIvDjjz+yjo2Jiam2qw4AZ8+ehaOjI1RUVAAAzs7OWLVqlchEsbqAe66Kojhk3lZUVESvXr3w/Plz9O7dm7PiTWm8CBTF+hSa+Ph4qKmpsbapqqri5cuXVY6NiYlBZmYmTp48CUIIxowZgyVLlkBGRqbKdVRVVfHx40d8+vQJSkpKrOvk5ORU6c7yl1eoL8Ql4F5cMm8D/8tJ+fHjRxFbQmmK1Hn2mRCCqKgoxMfHM8lKgfJucUxMjMAktACQl5eHZs2asbbJyclVSTtWUlICFRUVmJqawtraGh8/fsTixYvB4/GwcuVK5OXlQU5Ojjmef82CgoIqn3n06FH4+PjU9RYbNeKSefvt27eMKKalpSElJaVJL4xGET51FkVPT0+cOHECqqqqSEhIgKamJt69e4ecnBzY2NjUer68vDxrkgYoz9coLy/PNkhKCkePHmXKXbt2xdy5c7F9+3asXLkScnJyLAHk/7/ydQDA0dGxSiROamoqZ7u3DYm4ZN4ODAxklY8dOwYXFxcRWUNpitR59jkiIgKenp4ICwtDhw4dsGvXLly/fh0jRoyo0gKsDjU1NSQlJbG2JSYmQl1dnbXtw4cP2LJlC2vBouLiYsjKygIA1NXVWddJTEyEsrJytV3Bli1bolOnTqw//lhkU0NPT49xb5GQkOCsk3nl4Y36Hu6gUGqjzqKYnZ0NfX19AICmpiaePHkCGRkZLFy4ENHR0bWeb2hoCEIIjhw5guLiYoSHhyMuLq5KUlYlJSWEhYVh3759KCkpwatXr7Bv3z5YW1sDKHcYP3ToEFJSUpCZmQlvb29YWVl9zT3XK9LS0gLLXMHMzIwZ7ySEcNbJvPILti4vXAqlPqmzKLZt2xbv3r0DAHTr1o3JtN2iRQtkZmbWer6MjAwOHDiAyMhIGBgYwM/PD76+vmjdujXOnTsHXV1dAICsrCwOHDiA+/fvw9DQEHZ2dhg9ejRmzJgBAJg6dSpGjx6NKVOmwNzcHOrq6li8ePFX33h9Ubmlq6GhISJL6g6X03FVHmKpXKZQGpo6jylaWFhg5cqV2LRpE4YNG4ZFixahR48euHr1KlRVVet0DQ0NDZw8ebLKdktLS1haWjLlXr164dixY9VeQ0JCAosWLcKiRYvqanqDkpCQwCr/999/IrJEMGFhYVXKTXFslUKpjTqL4qJFi9C8eXPk5eXB2NgYdnZ22LZtG5SUlLBly5aGtJHT8JNB1FTmCg8fPmSV//33X06KIs28TRE1dRZFSUlJzJkzhyk7OztzPrW9MKjsUsTVlQ3FJaN1nz59WOGflVdLpFAamjqLIlDeujh58iTi4+MhLS0NdXV1ODk5VXHKbkpISkqy/Da52lIcMGAAa1F0rsYU870M+HB14orSeKnzRMtff/0FOzs75OfnY/To0TA2NkZmZiasrKxw+fLlhrSR01QUxOrKXGH48OGs8ogRI0RjSC1UThJSU9IQCqWhqHNLcceOHXB1dYW9vT1r+8GDB7F169YqcdEUbnH16tUqZS6OKerp6eHWrVsoKyvjtD9ljx49WJNq4uB1QKkbdW4ppqWlYciQIVW2GxsbM646FO5S3UQLFzEzM2OczCUlJTnrT1nZN1aUvrKU+qXOomhqalqtO83Zs2erdM2aEjIyMgLLXEFcJloUFRVhaGgIHo8HAwMDziatuHbtGqtcuSVOEV8Edp8rOkXn5+fj/PnzuHHjBvr16wcJCQnExcXhxYsXLB/DpkbFcMTqylyhTZs2TKIFfpmrmJmZITU1lbOtRKA8RVzlMheHIyhfj0BRrJhkQV5enpVcoaysDD169ECPHj0azjpKvVF5yQguLyGhqKjIeXcvcUkZR/l6BIripk2bhGUHpYFp06YNK7kCl1uKFIoo+So/xbi4OPz++++Ij49HWVkZVFVVYW9vX2VdDQr3yMrKElimfB008qbxUueJlmvXrsHa2hqfPn2Cubk5Ro8ejdzcXNjb2+PGjRsNaSOlHqjs2sJVVxdxhYpi46HOLcXdu3dj3rx5VZYo2Lt3L7y8vKp112kKKCkpscbnWrVqJTpjBGBmZoabN28yC0JxeRJDHKg8e0/HFBsPdW4pJiQkVLuy2rhx46pdZ6WpkJ2dzSpzeQKD35qhrRoKpWbqLIrt27fH8+fPq2yPiYlp0oP2lQWGq4Jz8eJFlihGRkaK2CIKhZvUufs8bdo0rF+/Hh8+fEC/fv0AAI8ePYKfnx+cnJwazECuo6SkxEqyy9Xus7is0SIu9O7dm9VI0NLSEqE1lPqkzqLo4OCA3Nxc+Pv7MzOXbdu2hbOzM+zs7BrMQK5TeRa3LlnIRYGenh5u376N0tJSSEpK0omW78TW1hbu7u6sMqVxUGdRPHv2LKZMmYJ58+YhIyMDsrKyUFBQaEjbxAJxCZ8zMzPDnTt3AJRnL+fyREt2djYCAgLg6OjI2TA/RUVFprWopaXFWTspX0+dxxQ9PDyYVlGbNm2oIIoZ4hJTDADnz59HQkJClSUUuIatrS3U1NRoK7GRUWdR7Nu3L6KiohrSFkoDY2ZmBlVVVc63EvnJcO/du4ecnBwRW1Qz79+/R2JiIt6/fy9qUyj1SJ27zxISEti5cyf27duHTp06VcmQfPr06Xo3jlK/iENM8fnz51llLi+wdfToUWbZXhoS23iosyj269ePmXWmiCfiMFZ3//79KmUuimJsbCyzHk9+fj7i4uKgqakpYqso9YFAUczPz4enpyciIyMhIyODkSNHYsWKFXQ8UUw5c+YMEhISEBISgunTp4vanGoRl4mro0ePssq0tdh4ECiKe/bswZUrVzBr1ixISkoiMDAQ2dnZ2L17t5DMo9QX2dnZzHonjx49Qk5ODidbixISEqyQOX4Wbq4hLqs4Ur4egaIYGRmJ7du3Y+DAgQAAfX19TJs2DcXFxXSVNTHjzJkzrDJXW4va2tp4/PgxU+bqEqfisooj5esR+Br+8OEDVFVVmbK2tjYIIcjIyPimD4uNjYWtrS10dHRgYWFR40ptz549w7Rp06Cnp4cRI0bAx8eH1Y0aMWIEdHR0oKurC11dXU7PpnKFynX96NEj0RjSSBCXVRwpX4/AliI/+oEPj8eDtLQ0iouLv/qDioqKMH/+fDg4OCAwMBAXL16Ek5MTrly5whqjzM/Px88//4x58+YhICAAycnJcHJygrKyMmxtbZGZmYkPHz7gwYMHrMzglMbB06dPWWW6xClF2AhtwObu3bsoLi7G9OnTIS0tjXHjxkFdXR0RERGs496/fw8dHR3Y2dlBUlIS3bp1w6hRo5g1MWJiYtC1a1cqiF+JsrKywDJXoGn+KaKmVpeckJAQlgCVlpbizz//rJL4oDa3ifj4eKipqbG2qaqqVkk7pqqqCl9fX6ZcVFSEv//+m4kaeP78OQghsLGxwdu3b6GlpYXVq1dXuTYA5OTkVHH+rZiSvykxadIk+Pn5MWUbGxsRWkOpT+7evcuEcNYFb2/varcbGhrCwMCgvswSWwSKYocOHaosa/rDDz8gJCSEtY3H49Uqinl5eWjWrBlrm5ycnMBZu6KiIixbtgxycnKYMmUKgPLZSG1tbSxfvhyKiorw9fXF7NmzER4eDjk5Odb5R48ehY+Pj0C7mgrVdUupX92307p1a1byj6acPq+xIVAUL1++XG8fJC8vj8LCQta2/Pz8GrvBaWlpcHZ2hoSEBA4fPswI6uzZs1nHLVu2DCdOnEBMTEyVzC+Ojo6sFQiB8pYiF52BG5rqnKJp6rBvZ9y4cTh27BhTHj9+vMhsMTAwqLGFt3LlStayuzIyMpyPahI1QhtTVFNTQ1JSEmtbYmIi1NXVqxwbHx8PGxsbdO3aFUeOHIGioiKz78iRI6wfeGlpKUpLS6tdhL5ly5bo1KkT609FRaUe70p8qNwq7NWrl4gsaRyEh4ezypXDE7nCzJkzWeWmnPu0rghNFA0NDZk40eLiYoSHhyMuLg6mpqas47KzszFz5kyMHTsWW7ZsqSJ2KSkp8PT0xMePH1FQUIDNmzeja9euNMlnLSQnJ7PKr1+/FpEljYPKeTO/1U2toenZsyfzfxkZGTpkUgeEJooyMjI4cOAAIiMjYWBgAD8/P/j6+qJ169Y4d+4cdHV1AQB//vknPnz4gJMnTzJ+iLq6uli6dCkAYPny5ejXrx8mTpwIIyMjJCcnw8/PjzrP1gJd4rTpwu8d0VZi3fiqdZ+/Fw0NjSoTNwBgaWkJS0tLAOUZvh0cHGq8hqysLNzd3VlZjymNB3FZHVGcaN68OdTU1GgrsY5wM7CUUu+IywJbubm5rPKXL19EZAmlqUJFsYkgLtlnlJSUBJYplIaGiiKFU6SnpwssUygNDRVFCqcQlxYtpfFCRZFCoVAqQEWRwikqu1ZRVyuKsKGiSOEUP/30E6tsZ2cnIksoTRUqihROMWDAAFaZ79RPoQgLKooUTpGdnc38n8fjcXrdZ0rjhIoihVNcvHiR+T8hBJGRkSK0htIUoaJI4RTVpTijUIQJFUUKp5CVlWWVKycmplAaGiqKFE5ReQyx4hgjhSIMqChSKBRKBYSaOoxCoXw9ISEhSElJ+ebz+efWtGBVXenYsSOsra2/6xriABVFCqUGuLJKXkpKCpLeJEO21bctS1sqXb6g27vPBd9sQ2FW2jefK25QUaRQxADZVsroNEp0y9K+jTotss8WNlQUKZQaELRK3i+//FJlG10lr3FARZEidOqjWyrqhdsnTZqEM2fOMOUff/xRZLZQ6hc6+0yhfANDhw5llQcNGiQiSyj1DW0pUoROY1m8vXXr1sjMzKStxEYGbSlSOIU4Ld7eqlUrqKmp0VZiI4O2FCmcQliLt3+v7x9A/f8aK1QUKZxDRUUFqampDdpK/F7fP4D6/zVWhCqKsbGxcHd3R1xcHDp37gwPDw/07du3ynHv3r3DmjVr8OjRI7Rp0wZubm4YPnw4gPJ0Urt378apU6dQXFwMa2trrFy5ElJSVN+5QH20wD59+oRmzZrh4sWLrFRiX0ttLTBR+/4BTcv/T1wQmpIUFRVh/vz5cHBwQGBgIC5evAgnJydcuXIFCgoKrGOXLl0KHR0d+Pv748GDB1iwYAH+/PNPdO7cGUFBQbh06RJCQ0MhIyODBQsWwM/PDwsXLhTWrVAEQFtgFHFHaKJ49+5dFBcXY/r06QCAcePGITAwEBEREazZu6SkJDx79gyHDh2CjIwMjIyMYGJigtOnT2PJkiU4e/YsHB0doaKiAqDcYXbVqlVNXhS/1vcPaDj/P9oCo4gzQhPF+Ph4qKmpsbapqqri5cuXrG0JCQlo37495OXlWcc9efKk2uuoqqri48eP+PTpE5SUlFjXysnJqZKKKjU1lVW+e/cuQkJCarS7uLgYpaWltd9gBaqLdpCUlIS0tLTA86ytrWsUpIawMyEhocq2V69eCfyc2uz8/Pkz8tPSkBC8r8bzSVkZSNnX2VoZnoQkeBI1O0+UlRTjM2purQrLTkCwrbXZCdRuK1fsbCwITRTz8vKqJAyVk5NDfn4+a1tubm61xxUUFDDXkZOTY/bxj+Xvr8jRo0fh4+NTL/ZzHQkJCZSVldW4v7pF5Xk8XrXX+R5atWpV67oqpSAoqdnUOiElwYOkpABbJWXRqlWrGncLy06gFltrsROo3Vau2NlYEJooysvLo7CwkLUtPz+f1SKsy3EVBRL4nxhWvg4AODo6YuLEiaxtqampmDZtGlMW5EhcF6prFe7evfubr1cT32vnxYsXERERwZQtLCwwcuTI+jCNxbx58+r9mg2BuNgJiJetjQGhOW+rqakhKSmJtS0xMRHq6upVjnv37h1L+Coep66uzrpOYmIilJWV0bJlyyqf2bJlS3Tq1In1xx+LbGqYmZmxyg0hiBRKY0BoomhoaAhCCI4cOYLi4mKEh4cjLi4OpqamrONUVVXRs2dP7Nq1C0VFRbh9+zaio6Mxfvx4AIClpSUOHTqElJQUZGZmwtvbG1ZWVsK6jSpUbhU2RCuxvhg7diyA8lYihUKpHqF1n2VkZHDgwAG4u7vDy8sLnTp1gq+vL1q3bo1z587B3d0dDx8+BFA+K+rm5gYjIyO0atUKHh4e0NDQAABMnToVGRkZmDJlCgoKCjB69GgsXrxYWLch1piZmVVpMVIoFDY8Ut0IfCPm7du3GDlyJKKjo9GpUydRm0OhUDhGkwsD4butVHbNoVAoX4+KikqjiyZrXHdTB9LSyiMdKs5AUyiUb6Mx9riaXPe5oKAAz549g7KyMiQlJevlmnw3n+PHj3N6dpvaWb9QO2lLsVHQrFkz6OnpNci1VVRUxOKtSe2sX6idjQuaZJZCoVAqQEWRQqFQKkBFkUKhUCpARbEeaNmyJRYuXFhtqCGXoHbWL9TOxkmTm32mUCgUQdCWIoVCoVSAiiKFQqFUgIoihUKhVICKYiNDUPZtCoVSO00uouVbIYRUm75f1OTn5yMyMhJFRUXQ0tKClpYWysrKvntZAQq3KSoqQlZWFoqKitC5c2dRm9OooLPPNVBaWgpJSUnk5+dDRkam3uKk65MvX75g8uTJUFFRASEE9+/fx5EjRxosjLE+4NdrRbj4wklKSsLZs2fRq1cv9O7dG126dBG1SQxfvnzBzJkzoaCggLt372Lx4sWYPXs2J+tRHKGiWA38hys2NhabNm1CYWEhZs2aBSMjIzRv3lzU5gEoF5clS5agXbt2WLNmDUpKSrBp0ybk5eVh06ZNojavWioKYmxsLKSkpJhlJrj0g05MTIStrS169+6N9PR06OnpYerUqejZs6eoTUNRUREmT54MPT09LF26FFeuXMHKlSsRGBiI/v37i9q8RgHtY1WC/+P89OkT5s+fD319ffTs2ROenp6IiIjAly9fRG0igPJV9woKCpjFrKSkpNCuXTu8fPnyq5c6FRZ8QZw7dy4WLlyIBQsWYPv27QDKVxbkwvu5rKwMSUlJWLBgAY4ePYolS5YgKysLJ06cQGxsrKjNQ2xsLJSUlODm5obmzZtj/PjxGDx4MGMbHVP+fuiYYgX4LZmcnBzEx8dj/PjxWLhwIQDgwIED8PPzg4SEBMzNzaGgoCAyOwkhyM3NhbS0ND59+sTYra6ujgcPHnCue1qxhRgWFobi4mIEBATgwYMH2LZtG0pLS7Fq1SpGGEVl63///QdfX1+8ffsWAwYMAACMGjUKABAaGoqTJ0/CxsYG2traIrEPKF/WIyUlBQ8fPoSuri6A8hdKUVERgO9fopZCRZGFpKQknj9/jlWrVqGgoAD5+fmwtbVFx44dmTGb/fv3Iz8/H5MmTWKtPy1MeDweFBQUYG9vDzk5OUZw4uPjUVxczBy3evVq/Pjjj9DR0RGJnUB5y4Vv361bt/Du3TsYGxujQ4cOUFJSgry8PNzd3SEpKYnly5eLTBCTk5Nhb28PMzMzlJSU4PXr17h27RqGDx+OUaNGgcfj4dChQ7h9+7ZIRbFt27b48ccf8cMPPzDbsrOzWcdcvnwZurq6TWad5nqHUEhpaSkhhJDc3Fxib29P/P39ydWrV4mBgQHZvHkzycjIYI7duXMncXZ2JmVlZaIyt0Y2bdpEVqxYQQghZMWKFcTY2JgUFRWJzB5+vRJCyJw5c4iRkREZNGgQMTU1ZbYXFBSQK1euEB0dHeLp6SkKM0laWhoJDAwkXl5ehBBC4uPjyS+//EKWLFlCrly5whx379491j2JipycHFbZ1taW3LlzhxBCiLOzM5kwYQIn7BRXaFsb5V2Ot2/fYt68eUyrcPjw4fD390doaCj8/f2RmZkJAFiyZAm8vLw4MwYG/G/dmYKCAnTp0gWenp54/vw5IiMjIS0tjZKSEpHYxe/KnT17Fi1btsSNGzfg5eWFDh06YPr06SgsLISsrCyMjIywfft2jBgxQqj28b+/AwcOwMvLCw8ePMCXL1+gpqaGhQsXorS0FGFhYYiMjAQA6OnpQUJCQuTjdi1atABQbn9OTg4KCgrQsWNHrF27Fq9fv8apU6cgISHBmedT3GjSoljxoenUqRPS09MRGhqKly9fghACHR0d7Nu3D+fPn8e2bdvw+fNnABD52Fdl+OLTsmVL+Pj44PHjxwgNDWUEUZTp4qOjo+Hj4wMejwcJCQn06dMHS5YsAY/Hw/z58xlhNDExgZGRkVB/yPyXiaurK+zs7FBQUIDLly8jNzcXampqWLRoETIyMvDy5UvWeVwZt+PxeJCSkkJBQQFsbW3x+PFjnD59mvneufJ8ihvc+HZFQGlpKXg8Hr58+YKUlBQAQHh4OHr37o01a9Ywq/3p6upi586dyMjIYLnjcOmB49uiqqoKDQ0NHD9+XGSCWHnmu0uXLtDX18e1a9fw5MkTNGvWDFpaWliyZAlyc3Mxa9YslhAKq14TEhKwfv167Nq1CwkJCVi0aBH69euH06dPIyoqCnl5eVBTU4OHhwcWLFggFJu+FkIIpKSk0KJFC3Tv3p0zL0Jxp0n6KfIjPmJjY7F27Vrk5ubC0NAQ69evBwBMnDgRkpKS8PX1Rbt27ao9V9jUtWXKP66srAw8Hk+o4l1xlvnvv/+GnJwctLS0UFxcjO3bt+PZs2dYs2YN9PT0UFJSgpiYGPB4PPTt21doNgLAy5cvYWdnh9GjR+PWrVtQU1PDvHnz0K9fP2zevBmxsbEYPXo0JkyYgGbNmgEQ3fdeF+7duwddXV1ISUmhqKgIMjIyojZJrGmSogiUtxTs7e1hZ2eHjh07Ys2aNZg8eTLc3d0BAJMmTUJmZibOnDmD1q1bC92+r42oqShISUlJ6N69uzDMZKgoGnPnzkV8fDwUFBSgoqKC7du3o7CwEHv27EFMTAxWrFgBQ0NDodrH58uXL5g1axbGjBkDR0dHREREYNeuXejRowfmz5+PPn36wM3NDTIyMnBzcxO6fdVF/NQEIQSEEM6KtbjSJGuzpKQEx44dw4wZMzB//nwYGBhg5MiRCA4OhqurKwDgzJkzGDx4MBQVFYVuHyEEkpKSiI2Nxdy5czFt2jRERUUhNzdX4PEAcPz4cWzYsAE5OTnCNJn5Ya5fvx4FBQUICwvDlClT8PbtWyxfvhwyMjJYvHgxevTogY0bNyI7O1skExY5OTn4/PkzHB0dkZubiwsXLmDKlClIT0/Hpk2bcPbsWWzYsAFr164FAKGOcfLdl5KSkrBr1y5cuHABb968EXgOv95PnDiBffv2CcPMRk+TFEUpKSmkpKRARkYGpaWlWLx4MYYNG4ZDhw4hNDQUCxcuRHJyMjZu3AhJSUmhRoiQr4yoqditPn78OHbs2IFly5aJJPX858+fkZCQgNWrV0NOTg4fP36EhoYGioqKsHbtWvB4PCxbtgw7d+6EoqKiSFo4rVq1Qp8+fZCWlobZs2dDXl4eTk5OGDVqFN6/f4+4uDimToU9mSYhIYHExET8+OOPePToEby9vXHw4MFqI2kqf++7d+/G8OHDhWZro0ZYvj+ihO+rV1JSwmy7efMmSU1NJS4uLuSXX34hhBDy4sUL8tNPPxE3NzeR+Hnx7cvOzib37t0jO3bsYPbt37+fmJiYkNOnT5PPnz8TQgjLVzIwMJDo6+uTmJgYodlbuY7evHlDbGxsSHp6OvH39yfjx48nqampZOfOnURTU5MYGxuTvLw8odnHJyUlhdy6dYtcunSJEEJIcXExefPmDbG1tWWOWbFiBfHy8mLqVJh+qPzvvbS0lERFRZHDhw8TQgi5dOkScXZ2Jm5ubuTFixfM8RXrPTAwkBgYGJBnz54Jzd7GTqOeoiorKwMhBNLS0oiLi8O+ffvQunVr6OvrY8yYMQDK37j8kK7Dhw9DS0sLrq6uzGSFMFszXxNRY21tDXl5eQDlLQUvLy8cPnwYvXv3FoqtxcXFkJaWBgC8evUKHTp0QOfOnbFz506UlZXh0qVL8PDwQLt27dCmTRssWrQIw4cPF3oU0H///YeffvoJffv2xYMHDzBw4EBMnz4dXbt2RW5uLlxdXVFQUID4+Hh4enoK/Xvnd5m/JsSQb1vF711LS0so9jYFGm33uaioCFu3bsU///yD169fY/bs2ZCTk0NWVhb8/f0REBAAoDyWNCQkBJMnT8bz58+xYsUKkfwwACAvLw+bN2+GhYUF1q5di+LiYgQGBjKO43PmzMHo0aNx9+5dRlx+//13eHl54ciRI0L7YaSlpTGCuGjRIvz888+YNGkSLly4wExKZWRkIDc3F9HR0fD19YWOjo7Qf7gZGRnw9vaGs7MzDh48iMuXL0NSUhJHjhxhxhVLSkogKSmJkJAQSElJobS0VKjfu4SEBBNi2LJlS1aIIVAujNbW1oiPj8ft27eZc6kgNiAibqk2GJmZmcTOzo4sXryYbNiwgfj7+xNCCHn//j0JCAgg48ePJ+fOnSO5ubnkwoUL5MSJE6S4uJgQQph/hUlycjJxcHAgLi4uTNft4cOHxNDQkHh6erJCDfn7S0pKiKurK3n06JHQ7Lx27RrR1NQksbGxJCgoiPz000/k7du3ZOXKlWT69OkkNDSUEELIqlWriLGxMTExMSHnz58Xmn18srOzyW+//UYGDx5MHj9+zGzPzc0l5ubmxMfHp8o5ovjevyXE8NGjR2Tw4MG0y9xANEpR5D88aWlpxNnZmQwaNIiJCSaEkA8fPpDDhw+T8ePHkyNHjrDOrTju2NBUHrcaO3YsIzj8ff/++y8ZNGgQcXFxYcW8ijK2dfXq1aR///5k8eLF5N69e8x2Dw8PMm3aNHL+/HlSVlZGUlJSSEpKCiGk/F6FNU6Xm5tLPn78SLy9vcm4cePIxo0bWfv9/f3Jr7/+ytom7Fh2/vfn6elJ9PX1iYODAzNWHB8fTxYtWkSWLl1KLly4UO35Hz58EJqtTY1G2X3mx6f+8MMPWL9+PQYOHIiMjAxcv34dhBC0bdsW48aNw5gxYxATE8NyuxBWhu3vjagRpW+ah4cHJk+ejAsXLiApKYnZvnr1amhrazOz+O3bt0eHDh0AQGiO5C9fvsTMmTOxevVqyMvLY+rUqfjw4QMOHDjAxID/+++/kJWVZZ0nrFlmvicDf8jE1dUV9vb2KCgoQHR0NBN7vXjxYmRkZCAuLo51Pv+8tm3bCsXepkijdt7mj9mkp6dj/fr14PF4sLGxYVwXPn36BEVFRaG7X4hjRE11bNy4EWfPnsWxY8fQq1cv1vbx48cLPWXZq1evYGtri7lz50JbWxt6enpISUlBaGgo4uLi8PLlS+jo6ODdu3f4/fffqwhjQ8N3zE5ISMChQ4fQpk0bWFpaQl1dHZs2bcKLFy9gbW0NU1NTNG/eHO/evYOKigpnvu+mQqMWRYAtjL/99hsAwMLCAqampswxwhREPlyPqKkr69atQ3h4OI4fPy7ydP2HDh3CmzdvsH79erx58warV6+GlJQU7t+/j5kzZ+Lff/+FoqIivLy8ICUlxSSjEAb8Z6yxhRg2Rhp9TVfsSq9btw5ZWVl49uwZ6xhhCyLXI2q+ht9++w2WlpaYMGECnj9/LlJb2rdvj4iICMybNw+zZ89G8+bNsWDBAhgYGEBGRgbm5uYoKCiAn58fSkpKhNpS5A+VrFu3DgsWLMBvv/2GJUuWICEhAf7+/nj27BlcXFzQuXNn/Pfff4wgAtzJytNUEHs/xcpv0epafRWFcc+ePYzQiKKFCPwvoqZ79+5MRI2trS2mTZsGBwcHfP78GatWrcLGjRsBfF08bH1BKiSWqO1H6e7ujuLiYsZ1SFQMHToUixcvRlZWFsaOHQsLCwsA5fkSu3TpgpEjR0JCQgLnzp2DhIQE5s+fL1T7agoxjIyMxKZNmzB58mRs2LCBGeMW1fPZ1BF7UZSQkMD79+8RHR0NOzu7Gh8iCQkJlJSUMCnahZlNhO/oXFHcZs6cCVVVVaxduxYdO3bEpEmTEBsbiwEDBqB169bo2LEjc76wBTEiIgL37t3D2rVrISkpWSdh5Au4KH/ICgoKmDZtGsrKynD37l1ERkYiOjoa79+/x+jRoyEjIwMrKytIS0tjyJAhDW5P5XqrGGK4ePFidOnSBU5OTigtLcUff/wh0hBDyv8Q+3Y5IQRXrlzBjRs3BGaYJv+few4o/9H/9ddfDW5bWVkZSktLmYiaZcuW4bfffsNff/0FIyMjtGvXrtqIml9//VVkGZ6LioqQlpaGV69eYefOnYwzsyBb+PVeXFzMiR9yUVERbt26hSNHjqC0tBQhISGQkZFBUVERFBQUMHnyZGZWvKHg19u7d+9w+/ZtREVFQU5ODh4eHigoKEBZWRk2b94MoHxtnQkTJmDlypVUEDmAWE60VM4VmJiYiOnTp8PFxQVjx46tcjypFDy/efNmnD17Fmpqag1mY1FREXbu3IlBgwaha9eusLe3x+DBg1FQUICkpCRYW1vDwcEB69atw7NnzyApKYmCggKEhIRAWlpaJIPrFZOTenl54fr16xgxYgTmzZtXY4uRf05mZiamTp2K48ePsxZVEhX8zNP8VrYwE6/y60lQiOGcOXPQp08fJsQwNDQUUlJSdFKFA4ilKALloWbKysqM4AUEBOD69evw8PBg+XBVFsQ9e/bg0KFDDR4alZWVhUWLFqFNmzb44Ycf0LZtW8yZMwepqam4dOkSTp06hTlz5mDkyJG4fv06MjMzMXnyZEhJSYk8c/KyZcuQk5ODN2/eoFmzZhg0aBCWL19eRRj5dmZlZWHKlClwcXGBsbGxyOyuCWG2vPiflZGRgV9//RV6enpwcHBAZmYm3NzcUFpaiqVLl+LJkye4c+cOCCHYtGlTleEViggRgoN4vZOXl0fGjx9PJk2aRIKDg0l6ejp5//49mTlzJhPSVVJSUm0WGWGERolLRE118LPbEEJIVlYWCQ0NJQsXLiRbtmxhZXPhh8RlZmYSExMTcu3aNZHZzDXEJcSQUj1i007nj2mVlZVBTk4Ov//+O/T09BAdHY3JkycjOTkZUlJS8PHxAVA+OcFvHQQEBMDHx0dowfPiEFHDp3KuSEII9PT0AABKSkoYO3YshgwZgqioKOzcuRMlJSWQkJBgWog2NjZwd3fHsGHDhGo31+DXY15eHgoLC9GqVSsoKSkhLCyMOUZeXh7W1tbIyMhgnUsqjHdTRI9YiCJ/0DoxMRHu7u6Mg7OLiws8PT0xceJE7N27F4QQPHz4EPfv32fOffLkCXbs2IH9+/cLNZsIXxhbt24NV1dXyMnJ4dSpU/j7778BAMrKyvjpp5+wZcsWkS2XWnGhejc3N7x48QLFxcVMhhagPIuQsbExeDwe4uLi8OrVKwDlC7BbWFhQQcT/6pHrIYaUuiE2Y4ovX76Eo6MjLC0t0aJFCyxcuJC1PzY2FqmpqVi/fj2GDRvGRK/k5uYiLy8PysrKojCbsxE1FfHx8cHff/+NU6dOgRDCjG0GBARARkYG9+/fx6ZNm7B161aoqamhrKwM+/fvR79+/WBkZCQyu7kE10MMKXVHLESxuLgYS5YsgZaWFubNm4fPnz9j586dKCgogLy8PGuBoffv32PGjBnYunWr0FeJq4mKwrhkyRL0798fS5YsEbVZAIDIyEisX78eTk5OmDVrFkpKSpCcnAxXV1ekpKRAXV0dMTExWL9+PWtmn64ax4bLIYaUr4OzAxkVW0/S0tLo0KEDXr58ib179+L06dNo164dtLS0cO/ePTx9+hTa2tooLS2FnJwclJWVBfos1ifiFlFTeYazU6dO0NPTw4EDBzBs2DBoaGige/fu+OOPP3Dp0iUoKChAQUEB2traTBefx+NRQaxE+/bt4efnhw8fPiAxMRHdunXDrFmzsG/fPibE8OrVq/Dz88PcuXOpIHIYTo4pVkyrxV+VTkdHB58/f0ZsbCxsbW1x8uRJrFq1Ci1btmSERVJSEvfv30dcXJzQusv8iJrAwEAANY8PVYyokZCQQFFRkUgF8fbt23jy5Ak0NTXh5uaGwYMHY82aNaxFkkxNTWFkZARtbW1mGx3/qh5+iKGWlhYWLlwIf39/6OvrQ0pKCl26dIG1tTVMTExw/fp17N+/X9TmUgTAue5zxbRarq6uaN26NTp06AB3d3dISUnh06dPyMvLQ/PmzbF161bExcUhKCiI1fpJTU2FioqKUOwlhODkyZP4+++/4ePjU+MsYsVWYUREBIqLi2FlZSUUGwF2i3bevHmIj4+HvLw81NXVsXnzZqSlpcHLywuvX7/G+vXrRZ7xRlzhhxhmZ2cjOjoaL168wJkzZyAjI4MvX74gIiICQ4YMafCIGsq3w7mWIn+W2cnJCSNHjsSgQYPw6NEjrFy5EgUFBfjy5QtcXFzg6OiIV69e4eTJk8wypHy3iIYWRP6CWEB5y2ngwIF4/vw5Ll68WO3xpJID+apVq9CnT58GtbEyfEFct24d8vPzERwcjIkTJ+L58+dwcXHBDz/8gF9++QXt2rXD6tWrkZubK5IZcXGHCyGGlO+DUy1FUr48AtasWQNlZWUsXboUz58/h4+PD9LS0tClSxfs2LEDKSkpKCoqQteuXZluqbD9vLgeUVPd56enp2Px4sXw8PBAt27d4OXlhVevXjF1u27dOqSnp+PLly+0pfgdiDLEkPL9cKKlWLHVJSEhgezsbPTs2ROFhYXYsGED+vbtC3t7e1y4cAEWFhaQk5ND9+7dmQkMYT9w+fn5mDlzJmxsbHDmzBlkZGTAzMwMZWVlzDICpaWlVQTRy8tLqILI/3EC5XXMT0TQqlUr+Pn54cqVK1i5ciV69erFrGjYtm1bKojfiZSUFCOI1DFb/BC5KPInVdLS0vDkyRMkJSVBR0cHo0aNwvr169GmTRvMnTsXWlpaGD58OPT19VmJV4W9DKk4RNTw4f8Yt27diqVLl+LRo0eYM2cOgHJXnA0bNqB9+/Zo3749fvnlF3h4eNBZ5XqGTkyJHyJ9hRFCICkpibi4OMydOxddu3ZFTk4O2rdvj6ioKPB4PPz4448AAH9/f7Ro0QLr1q0DINzEq/zPSkxMxOHDh5kEpS4uLsjKykJgYCD27t0LaWlpJqKGHyrHj6gJDAwUmiDu378fRUVFWLhwIVasWIH4+Hhoa2vD19cXo0ePBo/HY+4pOjoa+/fvx+7duznj10mhiBKRiiKPx0N2djZWrVoFOzs7ODk54fHjx5g1axb09fWRnp6OVatWoVevXvj48SNCQ0MB/E9MhQU/hKtiRA1/QalWrVrB2dmZiah5+fIlzp07x4iimpoaoqKihBpR0759e6xYsQLZ2dlQVFRk6i08PByBgYGQk5NjJqyys7Oxdu1aDBw4UGj2UShcRuQTLZmZmfjll18QEBCA0tJS2NraokePHvjll18QFBQEeXl5KCoqYuLEiSJLqyWOETVRUVFYvnw5FBQUcO7cObRq1Qo8Hg9nz55FWFgY+vTpAy0tLfTp0wcdOnQQeaghhcIVRD6mWFxcjP/++w9//fUX7Ozs0KVLF2zatAnXr1/HgwcPMGvWLCYWt7S0VGiCWPFdUTmixsrKCrGxsWjevDnu37+Pp0+fAoBIImpqYtSoUdi9ezeys7Px559/MoI3YcIEjBkzBrdv34ampiZrXWYKhcIBUWzXrh2mTZsGNzc3yMrKYufOnQCAW7duVfHnEsVC9VyPqBHEiBEjsH37dmzfvh0HDhxgttvY2GDPnj3o2rWrCK2jULiJyLvPAJCRkYF9+/bhwYMH0NbWxsePH5Gamorg4GBIS0uLbKF6cYioqQuRkZFYuXIl5syZgwULFojaHAqF03BCFIHyFF8PHjzA48eP0bZtW0yaNElkY4iJiYmwt7fH1KlTIScnh7Nnz6JHjx7w9PREeno6Vq9ejZycHDRv3hxHjhxhUskDwk8SW1fCw8OxbNkyXLp0CZ07dxa1ORQKZ+GMKFaHsNesEKeImm8hPT2dE4tKUShcRuRjioIQliCKW0TNt8IXRA6/BykUkSMev+YGhN8aTUtLw/v379GiRQsmosbd3Z2JqElISMDw4cOhoqIikoia+oTONFMoNdOkRVFcImooFIrw4PSYojDIzs6Go6MjLCwsWBE1CxYswM2bN/H06VNWRI2wZ8MpFIpwadItRaC8xdeyZUs4OTmhtLQUGzZswKhRozBmzBjk5OTAwMBA5BE1FApFeDT5X3fFiJqAgAAmoub06dN48OABjh49yhwrzIgaCoUiGsRvlqCe4WJEDYVCER1NfkwR4FZEDYVCES1UFP8fLkXUUCgU0UFFUQDU7YZCaXpQUaRQKJQKNPmJFgqFQqkIFUUKhUKpABVFCoVCqQAVRQqFQqkAFUUKhUKpABVFyleRl5eH3bt3w9zcHH379oWJiQk2b97MrGVTG7m5uQgODm5gKymUb4e65FDqzJcvXzB16lTIy8tjwYIF6N69O169eoUtW7agWbNmCAwMRLNmzQRew8fHB5cvX0ZISIiQrKZQvg4aqkGpM9u3b0dZWRmOHj3KiF/nzp2hrq4OU1NTnDlzBtOmTRN4DfoOpnAd2n2m1ImioiKEhYXBzs6uSmuwffv2CAgIwNixY5GXl4f169djyJAh0NLSwvDhw7F3714AQEhICHx8fBATEwNNTU28ffsWAHDw4EEYGxtDV1cXU6dOxaNHj5hrE0Lg5eWFQYMGYcCAAfD09IS9vT3T0iSEICAgAObm5tDW1oaVlRWuXbvGnG9vb4/169djzJgxMDIygo+PD0aMGMES53v37kFHRwe5ubkNVX0UcYJQKHUgPj6eaGhokKdPnwo8zs3NjVhZWZHHjx+TN2/ekAMHDjDn5efnk82bNxNLS0vy8eNHUlJSQk6ePEmGDRtGrly5QpKSksi+fftI3759yZs3bwghhPj7+xNDQ0MSHR1N4uLiyKxZs4impiY5c+YMIYSQvXv3Ej09PXL+/HmSmJhI9uzZQ3r16kVevHhBCCHEzs6OaGlpkb///ps8fvyYpKamkp49e5J79+6xbF66dGkD1RxF3KAtRUqdyM7OBgC0aNFC4HH9+/eHh4cH+vbti86dO2PWrFmQl5dHfHw8mjVrBnl5eUhKSkJZWRmSkpLw9/fHsmXLMGLECHTr1g1z587FgAEDcOLECQDAsWPHMH/+fJiYmEBDQwPbtm1jWqqEEBw9ehRz587FuHHj0L17dzg7O2PQoEE4cOAAY5OhoSGGDh2Kvn37ol27dtDX10dERASA8nyakZGRsLCwaIhqo4ghVBQpdaJVq1YA/ieONWFpaYnU1FR4enpizpw5GDp0KPLy8lBWVlbl2NzcXLx79w5ubm7Q1dVl/u7cuYOkpCRkZmbi48eP0NbWZs5RUlJCt27dAJSnfMvKyoKOjg7rugMGDEB8fDxT7tKlC2u/lZUVLly4gNLSUty4cQM8Hg+DBw/+muqgNGLoRAulTnTt2hVKSkp4+vQp+vbtW2X/pk2boKysjPj4eNy8eRMTJkzAhAkTsH79elhZWVV7Tb5Qbt68Gb1792bta9asGaSlpVnHVaammW5CCOucyseZm5vjt99+w927d3H+/HmMGTOG+SwKhbYUKXVCQkICVlZWCAwMRGFhIWvf27dvERQUBGlpafz555/YsmULli5dirFjx0JaWhqfP39mra3Np0WLFlBWVsaHDx/QtWtX5i8wMBDXr19HixYtoKKigpiYGOacz58/4/Xr1wAABQUFtG3bljUxAwAPHz6EqqpqjfeioKAAY2NjREVF4fr167TrTGFBRZFSZ+bPn4+ysjI4ODjgxo0bSE5ORlRUFJycnKCpqYlJkyZBTk4Oly5dQnJyMh48eABnZ2cQQlBUVAQAkJeXR3p6OpKTk1FSUoJZs2Zh7969iIiIQHJyMvbu3Yvjx4+je/fuAIDp06dj3759uHLlCuLj4+Hq6oq8vDxGXOfMmQM/Pz+Eh4fj1atX2Lt3L27cuAF7e3uB92JlZYXg4GC0aNEC/fv3b9iKo4gVtPtMqTNKSko4ceIE9u7di3Xr1iE9PR1t27aFmZkZ5s2bBwUFBezYsQNbtmzB6dOn0bZtW1hZWaFly5ZMa8/c3BzBwcEYO3Ysjh8/DgcHBxQUFGDbtm1IT09Ht27dsGfPHgwYMAAA4OjoiLS0NLi4uKCkpAS2trbo2LEj0921s7NDXl4etm3bhoyMDGhoaMDPzw96enoC72Xo0KGQk5PD+PHjG7bSKGIHjWihcJpr166hd+/eUFZWBgCUlJRg4MCB2LdvH/T19b/5utnZ2RgyZAjOnj0LNTW1+jKX0gigLUUKpzl9+jQ+f/4MV1dXyMrK4ujRo2jZsiX69ev3TdfLz8/HtWvXEB4ejv79+1NBpFSBthQpnObjx4/YsGEDbt++jZKSEvTv3x9r165lxhy/lqKiIgwZMgStWrXCvn37BE7IUJomVBQpFAqlAnT2mUKhUCpARZFCoVAqQEWRQqFQKkBFkUKhUCpARZFCoVAqQEWRQqFQKvB/B0BB0CbOf0UAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARkAAAEPCAYAAAB2n7CRAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABOkElEQVR4nO2deVxN+f/HX7dVWSJCZKtGSClSwqBIqGTJZClR9i27LNEwJcYghewxYQjZakSNbzK2aqylmhZLES1UWtzb7fP7o989021zUfeeW5/n49GDzzmfc+77nnPP67w/n8/78/5wCCEEFAqFUk/ISNoACoXSsKEiQ6FQ6hUqMhQKpV6hIkOhUOoVKjIUCqVeoSJDoVDqlUYhMm5ubtDR0anxz9fX97vO7+joiG3bttWRtd8OIQRnz57F58+fq91///596Ojo4MiRI1X2BQYGwtzcnCmbm5sjMDCwSr309HTo6OggKSkJAODr6wsdHR3Mmzev2s9ctGgRdHR0cPPmzWr3FxUVYe7cudDT08PUqVO/+B2/Fjc3NyxZsqTOz9uQ8PX1xYQJEwBUvb+1kZubiytXrnyxntx3WygFrF+/HitWrAAApKWlwdHREUFBQVBXVwcAKCsrf9f5fX19IScn+UsZHR0Nd3d3WFlZQVFRscZ6vr6+sLS0hIaGRp18rry8PO7cuYOioiKha1lcXIy///671mPDw8Nx584dnD59Gm3btq0Teyjfjrq6Om7fvo1WrVp9se6vv/6KwsJC2NjY1FqvUXgyzZs3h5qaGtTU1NCyZUsAgKqqKrOtadOm33X+li1bolmzZnVg6fchalylqqoqfv755zr7XG1tbSgrK+P27dtC26OioqCjo1PrsQUFBWjTpg169+5NRYYFyMrKQk1NTaSXpqi/t0YhMl+ioKAAW7ZswY8//og+ffrAxcUFqampzH5zc3McO3YM9vb20NfXh729PeLi4pj9lZtLp06dgqWlJfr06QM7OzvExMQA+M8VvXDhQrV2XLhwARMmTMChQ4cwePBgmJiYYOXKlSgqKmLqREZGwtbWFvr6+rCyssL58+eZc0+fPh0A0Ldv3xo/AwA2bdqEqKgohIaGfsPVqoqMjAzMzc0RHh4utD0sLAyWlpY1Hufr64vNmzfjzZs3zHVxc3PDypUrMWnSJBgbG+POnTvg8XjYsWMHBg0ahH79+lW5PzExMbCzs4O+vj4GDx6M7du3g8/nM/uLi4uxbt069O3bFwMGDMCuXbuYfdV9nqOjI3bt2gUXFxfo6+vDxsZGyCNzc3PDL7/8grVr18LAwABDhgxBaGgorl+/jhEjRsDQ0BDLly8Hl8sFALx79w7z5s1Dv379YGRkhCVLliAnJ6fGazJr1iy4uLigX79+CA4OhpubGzZu3IgVK1agT58+sLCwwNWrV4WOWbx4MXbs2IF+/fphwIABOHHiBGJiYmBjYwMDAwPMmjULHz9+/OK9rNxcunHjBqytraGnpwdzc3McPnyY+czg4GCEhYV98UVCRQbAkiVLcP/+fezcuRNnz56FoqIiXFxcUFxczNTx8fHB2LFjERwcjG7dusHZ2bnam3b+/Hls27YNs2fPxuXLl2FiYoK5c+fiw4cPjCs6ZsyYGm1JSkpCbGwsjh07hl9++QU3btzAH3/8AQD4999/sWTJEkyePBlXr17FwoULsW3bNoSEhEBdXZ3pWwoPD6/1M4yMjGBnZwcvLy/k5+d/41UTZuTIkYiMjERpaSkAgMvlIjIyEhYWFjUe4+zsjOXLl6N9+/ZC1+XKlSuYMmUKjh8/DkNDQ+zZsweRkZHYtWsXzp49i27dusHR0REFBQXg8/lYuHAhhg0bhtDQUGzfvh1BQUFCInvr1i2oqanh0qVLWLFiBfz9/REVFcXsr/x5AHDkyBEYGhoiODgYQ4cOxdy5c/HixQvmmD/++AOdO3fGlStXMHjwYKxfvx5HjhyBj48Pdu3ahYiICFy+fBkA4OHhARkZGZw7dw6BgYHIyMiAt7d3jdclKioKRkZGOHv2LIYOHQqg/AXUrFkzXLhwAdOmTcOqVatw//595pibN2+iqKgIwcHBmDx5Mry9vbFlyxZs2rQJR44cQVxcHI4fPy7q7QQAZGdnY9myZXBycsK1a9ewevVq7N69G3fv3oWzszNGjx4NMzOzKh5sZSTfkSBhkpKScOfOHZw7dw56enoAgB07dsDMzAxXrlzBTz/9BAAYM2YMpk2bBgDYvHkzzMzMEBoaWqWz8tSpU7C3t4ednR0AYOXKlQCA/Px8tGrVCmpqarXaw+PxsGXLFqipqeGHH37Ajz/+yHhNhw8fho2NDaZMmQIA6Ny5M169eoUjR47AysoKKioqAMqbQ02aNKn1c1atWoWbN29ix44d2Lx5s8jXqyYGDhwIHo+HmJgYDBgwAHfv3kXnzp1r7fdp2rQpmjZtyrjoArp168Z0RJaUlCAgIAABAQHo168fAGDDhg2IiorCpUuXYG1tjY8fP6Jt27bQ0NCAhoYGjhw5gtatWzPn6969O5YtWwYA6NSpEw4ePIiEhAT8+OOPVT5PQL9+/bBo0SIA5fcwKioKQUFBWLVqFQCgS5cumD9/PgBg8uTJOH/+PFxdXaGrqwsA0NfXR3JyMgAgIyMDvXr1QseOHaGgoICdO3eisLCwxuvSpEkTzJ07FzIy//kAHTt2xKZNmyAjIwMtLS08ePAAp0+fhomJCQBAUVER69atg5ycHBwdHbF//344ODjAyMgIADB06FDGHlF5//49eDwe1NXV0bFjR3Ts2BFt2rRB165d0bRpUzRp0gRlZWVf/E03epFJTk6GvLw8evfuzWxTVlZGr1698O+//zLbBDcLABQUFKCjoyO0X0BKSgpmzJjBlDkcDvPDFIWmTZsK3bRmzZoxzaV///0XSUlJCAkJYfaXlpZ+U6eziooK1q5di5UrV8LW1rbKfjk5OZSVlVXZLmiHV/5MBQUFDB06FBERERgwYACuX79ea1OpNjp37sz8/9WrV+ByuXBxcQGHw2G2f/78GWlpaWjZsiXmzp0Ld3d3+Pr6YujQobCysoK+vn615wPK++gqjsBV3g8I32+gXDQq3u+KxwgEvVOnTsw2BQUFprnk6uqKZcuW4fr16zA1NYWFhQWsra1r/P4aGhpCAgMAhoaGQtv09fWFRnY6duzI3BOBPRUFXkFBAR8+fKjxM6ujZ8+esLGxgYuLCzp16oRhw4bB1tYWbdq0+arzNHqRqWkUhhAi1LElKysrtL+srKzKNqB8pKXiw/C1yMvL17iPz+fD0dERkydP/ubzV8Ta2hrBwcHYuHEj43kJaNGiBQoKCqock5eXx+yvjIWFBXbs2AE3Nzf89ddfOHXq1DfZVdELE/StHD16VMg7AcB0ti9fvhzjx49HREQEIiMj4ezsDFdXV2ZYvbr7VPHeVuf1VXe/Kz7k1d2nmu778OHD8b///Q8RERGIioqCh4cHLl++jKNHj1Zbv7rf5Jfsqe5FU1movhYOh4MdO3bAxcWFubanTp2Cl5cXxo0bJ/J5Gn2fjJaWFng8Hp49e8ZsKy4uRkJCArp168Zsi4+PZ/7/+fNnJCYmVtvh1bVrV6FOYQCwsrJCWFhYndj68uVLdOnShfl78OAB8zB/i7h5eHggIyOjSkxMr1698PDhwyr1Hz58CDU1tWrfZkOHDkV2djYCAwPRpk0boev3rXTu3BlycnLIzc1lvnPnzp3h6+uLJ0+e4M2bN9i0aRM6duyIWbNm4ffff2f6w76HivcbAJ4+fYoePXp89XkIIdi2bRvev3+PSZMmYc+ePfDz88Pff/9dY+evKPY8efLkm+z5GhISErB161b07NkTixYtQlBQEEaPHs140qL+3hq9yHTt2hUjR47EunXrEBMTg6SkJKxZswaysrKwsrJi6p05cwZXrlxBSkoK3N3dIScnh9GjR1c534wZM/DHH3/g8uXLePXqFX777TdkZ2ejf//+4PP5yMrKQklJyTfZ6uzsjP/973/w9/fHy5cvERYWBi8vL+YNL4hRiYuLq7XNX5FOnTph4cKFSE9PF9o+depU3Lt3D9u3b0dKSgpSU1MRFBQEX19fzJ49u9pzNW3aFAMHDoSPjw9Gjhz5Td+xunNOmTIFnp6eiIyMxMuXL7F582ZERERAS0sLqqqqCAsLg6enJ16+fIm4uDj8/fffTP/atxIREYHAwECkpaVh+/bteP36NdM/9zVwOBykpKRg8+bNiI+Px8uXLxESEoKOHTuKFIsiID4+Hrt27UJaWhqOHTuGqKgoODo6frU9X4OKigpOnz6NvXv34vXr14iNjcWjR4+Ya6usrIyMjAxkZGTUep5GLzIA4OXlBT09PcyfPx/29vYoKSlBYGAgE1MDAHZ2djh27BgmTJiAd+/eISAgoNrYGCsrK7i6umL37t2wsbFBdHQ0Dh06BFVVVbx9+xaDBw/+5qHj3r17Y8+ePQgNDYWVlRW8vb0xZ84c5qHv3r07zMzM4OzsjLNnz4p83pkzZ1Z5K/bo0QMBAQGIj4/H5MmTMX78ePz+++9YtWoVnJycajyXhYUFCgsL60xkAGD16tWwtLTEunXrMHbsWMTHx+PQoUPo3LkzmjRpggMHDiApKQnjxo3DzJkz0b17d2zYsOG7PtPKygrh4eGwtbVFdHQ0jh49ig4dOnzTuTw9PdGmTRvMnDkTY8eOxdu3b3Hw4MGvas4MHjwYL1++hK2tLc6fPw8/Pz+hfqf6QDBiGR4eDmtrayxZsgQjRoxgmqHjx49HdnY2xowZg6ysrBrPw6GZ8b6Mubk5nJ2d4eDgIGlTKGLA0dERvXv3xpo1ayRtCoDyuJyioiLs2bNH0qZ8E9SToVAo9QoVGQqFUq/Q5hKFQqlXqCdDoVDqlUYnMqWlpUhPT2fm2FAolPql0YlMZmYmhg8fjszMTEmbQqE0ChqdyFAoFPEiEZF58uQJTE1Na9z/5s0bzJw5E4aGhhgxYgQiIyOZfYQQ7Nq1C6ampjAyMoKXlxdt+lAoLEasEyQJITh37twX8+EuX74cBgYGOHDgAGJjY7Fw4UJcunQJnTp1wpkzZ3Djxg0EBwdDQUEBCxcuhL+/PzMtXxIsXbqU+f/u3bslZseX2LBhAz59+oTmzZtjy5YtkjaH0kgQqyezZ88enD59msnDUR1paWl49uwZlixZAgUFBZiamsLc3Bznzp0DAFy8eBFOTk5o3749VFVVsXjxYpw5c0ZcX0Gq+fTpEwBUO7uaQqkvxCoykydPxoULF4Ryt1QmJSUF6urqQgmpNTU1kZiYCKA8/4uWlpbQvvfv34uUWrA+qOjFVFdmC5Xn8ri7u0vIEkpjQ6zNpXbt2n2xTmFhYZX8HkpKSszM5aKiIigpKTH7BHWrm9mcn59fJb1kYx1VEngxAqg3QxEXrEtapaysXGXdoOLiYsazqSg4wH/iUt2yJsePH4efn189WkuhUL4E60RGS0sLb968QUlJCeOlpKamQltbG0D58htpaWlMvtfU1FSoqalVm6nNyckJ48ePF9qWmZnJ5OqlUCj1D+viZDQ1NdGjRw/s2rULXC4X9+7dQ0REBJMTdezYsTh69CgyMjKQm5sLX1/fanPUAuUpIgXJpQV/7du3F+fXoVAaPawQmcuXLzNLUQDla7qkpKTA1NQUGzZsgKenJ7p37w4AmDJlCkaNGoXJkyfD0tIS2tracHV1lZTpFArlC0ikuWRiYsIseAaUeydjx45lyurq6swiUpWRkZHBkiVL6PrGFIqUwApPhkKhNFyoyFAolHqFigyFQqlXqMhQKJR6hYoMhUKpV6jIUFhHXl4efH19q0wJoUgnVGQorOP69etITU2tk6V9KZKHigyFVeTl5eHevXsghODevXvUm2kAUJGhsIrr16+Dz+cDAPh8PvVmGgBUZCisIjo6utYyRfqgIkNhFbKysrWWKdIHFRkKqyguLq61TJE+qMhQWEXlVBw0NYf0Q0WGwiocHByEyo6OjhKyhFJXUJGhsIrmzZvXWqZIH1RkGgkKCgq1ltnC9evXISNT/rOUkZGhQ9gNACoyjQQul1trmS3ExMSgrKwMAFBWViaU3IwinVCRobAKIyMjZthaVlYWRkZGEraI8r1QkaGwipEjR4LD4QAoby5ZWlpK2CLK90JFppFQecG8ymW2oKKiAhMTE3A4HBgbG1e71A1FumDdukuU+kEwH6imMpsYOXIkMjMzqRfTQKCeTCOhdevWtZbZhIqKChYvXky9mAYCFZlGQm5ubq1lCqW+oCLTSFBRUam1TKHUF1RkvhNB4FhNZbaQnZ1da5lCqS/E2vGbkJCATZs2ITExEZ06dYKnpyf09fWF6sTExGD27NlC27hcLjQ0NJjoz2HDhuHjx4/MUGfbtm0lFhkqCByrqUyhNHbEJjJcLhcLFizA9OnTERgYiOvXr8PFxQU3b95Es2bNmHpGRkZ4+PAhU3779i3s7Ozg7u4OoLwv4d27d4iNjYWysrK4zJd65OXlhaJ85eXlJWgNpTEhNt/+wYMH4PF4mDFjBuTl5WFlZQVtbW2EhobWety6deswduxYDB48GAAQFxeHLl26sEZgFBUVay2zBWmZVkBpeIjNk0lOToaWlpbQNk1NTSQlJdV4THh4OJKTk7F//35mW3x8PAghsLOzQ3p6OnR1dbFu3boq5waA/Pz8KomoMzMzv/ObCPP58+dayxRKY0dsIlNUVFQlylRJSanWzGf79+/HnDlzhI6TkZGBnp4eVq5cCRUVFezduxezZ89GSEgIlJSUhI4/fvw4/Pz86vaLUCiUr0JsIqOsrFzlLV9cXFxjsychIQHJyckYP3680PbKncIrVqzAqVOnEBcXV2UynZOTU5XjMzMzMW3atG/9GhQK5SsRW5+MlpYW0tLShLalpqZCW1u72voREREYMmSIUKcwAAQEBAhN/+fz+eDz+dXmR2nRogU0NDSE/hprOscffvhBqNy9e3cJWUJpbIgsMg8ePPiuDzIxMQEhBAEBAeDxeAgJCUFiYiIsLCyqrf/48WP07du3yvaMjAx4eXnh/fv3KCkpgbe3N7p06QJdXd3vsu9bEQyj11RmCx8/fhQqf/jwQTKGUBodIovMvHnzMGzYMPz6669ITEz86g9SUFDAoUOHEBYWBmNjY/j7+2Pv3r1QVVXF5cuXYWhoKFQ/IyMDbdu2rXKelStXok+fPhg/fjxMTU3x+vVr+Pv7S2zpjMpxPn369JGIHV8iKyur1jKFUl9wCCFElIqfP39GeHg4rl69itu3b6Nr166wsbGBtbU1OnToUN921hnp6ekYPnw4IiIioKGh8d3nCwwMFGq+9e/fn5V9Pt7e3kIja+3bt4ebm5sELaI0FkT2ZBQVFWFlZYX9+/fj9u3bmD59Oh4+fAgrKytMmzYNQUFBKCwsrE9bWcmTJ0+Eyo8fP5aQJbVDVwGgSIpv6vjNysrC27dv8fbtW5SWlqJFixY4c+YMzM3NER4eXtc2shppWfFQQ0ODGeJXUlJCx44dJWwRpbEg8hD269evERISgpCQECQnJ8PAwAD29vYYM2YMM6N39+7dcHd3x4gRI+rNYLYhLSse5uXlMVG+XC4X+fn5NF8LRSyILDIWFhbo1q0brK2tMXbsWHTq1KlKnb59+yI+Pr5ODWQ7TZo0QUlJiVCZjVy/fl2oHBYWhkmTJknIGkpjQmSROXv2LHr37l0llQGfz8fz58/Ru3dvDBkyBEOGDKlzI9mMtEwriImJYVJu8vl8xMTEUJGhiAWR+2Ts7e2rxFoAwJs3b1g5miIuKg/OiThYJ3boUiMUSVGrJ3Pu3Dn88ccfAMofHhcXlyodm9nZ2bQTUQoYOXIk7t+/D4AuNUIRL7WKzJgxY5jYimfPnmHAgAFo2rSpUJ2mTZs26h+sqqqqUL5ctiboFiw1cufOHbrUCEWs1CoyysrKWLRoEQCgY8eOsLKyYu0aypJCQ0NDSGSq6xBnC3SpEYokqFVkTp48CTs7OygqKqKoqAhBQUE11m2s/TIJCQlCZTaPrgmWGqFQxEmtInPkyBGMGTMGioqKOHLkSI31OBxOoxUZVVVVoXB9VVVVCVpTO3l5eThx4gScnJxoc4kiNmoVmb/++qva/1P+Q5rWM7p+/TpSU1NpjAxFrNQ6hF1cXCzyX2OlsufCVk8mLy8P9+7dAyEE9+/fr5KWlEKpL2r1ZAwNDb+YH4UQAg6Hg+fPn9epYdJC5bwsbM3Tcv36dSYYr7S0lHozFLFRq8gcP36ctUmY2IKRkRHu3LnDiC1bg9yio6OrlKnIUMRBrSJjYmIiLjukFkGQW2lpKWRlZVk7PCxNHdSUhkWtImNqaoqQkBCoqqpiwIABtXo1d+/erXPjpIGKQW4mJiasHbWRlmYdpeFRq8isXr2aSeS9Zs0asRgkjZiamiI2NhYDBw6UtCk1YmRkhL///luoTKGIg1pFpuJyIpWXFqH8x927d/H582fcuXOHtf0cpqamQiLDZkGkNCy+KjPe1atXMXXqVBgbG2PQoEGYPn260A+3MZKXl4f79++DEIIHDx6wdmj47t27THOXw+Hgzp07EraI0lgQWWR+//13bNiwAXp6eti4cSPWrFkDbW1tLFiwoNbpBg2d69evM+kdysrKEBYWJmGLqicmJoaxkxAilPycQqlPRE5adfDgQfzyyy+wtrZmto0dOxa9e/eGr68va5sJ9Y20JIMyMjLCvXv3wOfzWZ9Phk5/aFiI7MkUFhaiZ8+eVbb36dOn2mRWjQVpSQY1cuRIprnE9nwyFac/UKQfkUVm4sSJ2Lt3r1B6SUIIDh8+DBsbm3oxThqo+PByOBzWPryCoXYOh8PqfDLS0sdFEZ1am0sTJ05kHqCysjLEx8fj77//Rvfu3SErK4vk5GTk5uZi0KBBIn1YQkICNm3ahMTERHTq1Amenp5VVmAEyjspnZ2dhZJyz5o1CwsXLgQhBLt378bZs2fB4/EwYcIErF69GnJyIrf86hQVFRW0adMGmZmZaNOmDWsfXkA68slU18fFxuYnRXRqfTLNzMyEyubm5kLlr2kacLlcLFiwANOnT0dgYCCuX78OFxcX3Lx5k4nFERAfH49Ro0Zh165dVc5z5swZ3LhxA8HBwVBQUMDChQvh7+/PJNcSN3l5ecySr9nZ2axeakQa8slISx8XRXRqFZm6fHAfPHgAHo+HGTNmAACsrKwQGBiI0NBQ/PTTT0J14+Liqu3/AYCLFy/CyckJ7du3BwAsXrwYa9askZjIVFxqhBBC37zfCQ0abHiI3MYghCA8PBzJycnMmwYo91Di4uJqTWoFAMnJydDS0hLapqmpiaSkpCp14+LikJubi9OnT4MQgtGjR2PZsmVQUFCoch5NTU28f/8eHz9+RMuWLYXOk5+fX6VNX3H+Tl1A37x1Cw0abHiILDJeXl44deoUNDU1kZKSAh0dHbx58wb5+fmws7P74vFFRUVVFj5TUlKqkoumtLQU7du3h4WFBSZMmID379/D1dUVHA4Hq1evRlFREbPcKvDfYmoVF1gTcPz4cfj5+Yn6Fb8JaRoalgYEQYOCWe1sjqKmiIbIo0uhoaHw8vLClStX0KFDB+zatQtRUVEYNmyYSKsmKisrV1n4rLi4GMrKykLb5OTkcPz4cUybNg1KSkro0qUL5s2bxzRLlJSUhARF8P/K5wEAJycnRERECP2dPHlS1K8sEtI0NCwN0KDBhofIIpOXl4f+/fsDAHR0dPDkyRMoKChg0aJFiIiI+OLxWlpaSEtLE9qWmpoKbW1toW3v3r3Dtm3bmHWbAYDH40FRUREAoK2tLXSe1NRUqKmpVdvZ2qJFC2hoaAj9Cfpy6gppGRqWFqQl7ogiOiKLTNu2bfHmzRsAQNeuXZlMeM2bNxcpr62JiQkIIQgICACPx0NISAgSExNhYWEhVK9ly5a4cuUK9u/fj9LSUrx48QL79+/HhAkTAJRHGR89ehQZGRnIzc2Fr68vbG1tRf7C9cHIkSOhqanJei8mLy8Pvr6+rI49oZ5hw0NkkbGxscHq1atx//59DBkyBBcuXMCFCxewY8cOaGpqfvF4BQUFHDp0CGFhYTA2Noa/vz/27t0LVVVVXL58GYaGhgAARUVFHDp0CDExMTAxMYGDgwNGjRqFmTNnAgCmTJmCUaNGYfLkybC0tIS2tjZcXV2/8evXDYKhYbZ7MdIQSUs9w4YHh4i4eDOfz8eRI0fwww8/wMzMDL6+vjh16hRatmyJbdu2VRtUx0bS09MxfPhwREREQENDQ9LmiI28vDxs2bIFpaWlkJeXh7u7O2sfYDp3qWEhssg0FBqryAQFBQmNgg0YMICO2lDEwlfF4v/zzz84ffo0kpOTIS8vD21tbbi4uFSJf6GwDxrPQ5EUIvfJ/Pnnn3BwcEBxcTFGjRoFMzMz5ObmwtbWli78JgXQURuKpBDZk/ntt9+wdu1aODo6Cm0/cuQItm/fXmVeE4VdCFZVAOioDUW8iOzJZGVlYfDgwVW2m5mZMUPbFPZCR20okkJkkbGwsMDp06erbL948SKGDh1ap0ZR6gdpieehNCxqbS5VjD8pLi7G1atXcfv2bfTp0wcyMjJITEzE8+fPMXbs2Ho3lPL9SEOqB0rDo1aRqTgfSFlZWWhZlLKyMvzwww/44Ycf6s86CoUi9dQqMlu3bhWXHVINDR6jUGrmq+JkEhMTcfjwYSQnJ6OsrAyamppwdHRE375968s+qaBiuD6NPaFQhBG54zcyMhITJkzAx48fYWlpiVGjRqGwsBCOjo64fft2fdrIaiomvr5//z6rJx9SKJJAZE9m9+7dmD9/fpU0l/v27YOPj0+1w9uNgevXrwtF0lJvhkIRRmRPJiUlpdqlT6ysrKpNodlYkKYkS9KQ6oHS8BBZZNTV1REfH19le1xcHFq3bl2nRkkTrVq1qrXMJqQh1QOl4SFyc2natGnw8PDAu3fv0KdPHwDAo0eP4O/vDxcXl3ozkO1UTtglSgIvSVB50TRLS0s6EkYRCyKLzPTp01FYWIgDBw7gw4cPAMqz5S1evBgODg71ZiDb6d+/v1B2fUGKUrZBF02jSAqRRebixYuYPHky5s+fj5ycHCgqKlZZlK0xMnLkSCZPi5ycHGtD9mmqB4qkELlPxtPTk/FgWrduTQXm/1FRUcGAAQPA4XBgYmLC2iYITfVAkRQii4y+vj7Cw8Pr0xapRRomHtIE3RRJIXJzSUZGBjt37sT+/fuhoaHBLFEi4Ny5c3VunLQgDRMPVVRUYGhoiOjoaBgYGLDW46I0PEQWmT59+jCjShTppJGlc6awhFpFpri4GF5eXggLC4OCggKGDx+OVatW0f4YKSQvLw8PHz4EADx8+BA2NjbUm6GIhVr7ZPbs2YObN29i1qxZmDlzJm7duoUNGzaIyzapQRoiaStOfygtLaUBeRSxUasnExYWhh07dmDAgAEAymNApk2bBh6PB3l5ebEYKA1Iwyzs6OjoKmW22kppWNTqybx7905odUg9PT0QQpCTk/NNH5aQkAB7e3sYGBjAxsYGT548qbbes2fPMG3aNBgZGWHYsGHw8/MT6k8YNmwYDAwMYGhoCENDQ4mOlFSOpGWrN6OqqlprmUKpL2oVGcFCYAI4HA7k5eXB4/G++oO4XC4WLFiA0aNHIzo6GvPmzYOLiws+ffokVK+4uBhz587F6NGjcf/+fQQEBCA4OBhnz54FUB62/+7dO9y5cwcPHz7Ew4cPJer6VxdJy0YEMU41lSmU+kLkOJnv5cGDB+DxeJgxYwbk5eVhZWUFbW1thIaGCtV7+/YtDAwM4ODgAFlZWXTt2hUjRoxAbGwsgPIJmV26dBFKDSpJqoukZSOVg+/YHIwnDX1cFNH54hD2hQsXhB5oPp+PS5cuVZltPG3atFrPk5ycXGWlSU1NzSppIjQ1NbF3716mzOVycevWLdjb2wMA4uPjQQiBnZ0d0tPToauri3Xr1lW7imV+fn6VH2pmZmatdn4tRkZGQsu/svXhNTU1FZpjNXDgQAlaUzvS0MdFEZ1aRaZDhw5VlkFp06YNLly4ILSNw+F8UWSKiorQpEkToW1KSkooLi6u8Rgul4sVK1ZASUkJkydPBlAeFKinp4eVK1dCRUUFe/fuxezZsxESEgIlJSWh448fPw4/P79a7fpeRo4cyTy8fD6ftZG0d+/eFSrfuXOHlQ8wnS3e8KhVZOpy+VllZWV8/vxZaFtxcXGNzZ6srCwsXrwYMjIyOHbsGCNQs2fPFqq3YsUKnDp1CnFxcVW8CCcnJ6EVFoByT+ZLgvg1FBQUVCmz8aGo3Ixj6wRJOlu84SG2PhktLS2kpaUJbUtNTYW2tnaVusnJybCzs0OXLl0QEBAAFRUVZl9AQIDQA8Pn88Hn86GgoFDlPC1atICGhobQX/v27evwW5V7SxUJCAio0/PXFdLSJyMtfVwU0RGbyJiYmIAQgoCAAPB4PISEhCAxMREWFhZC9fLy8uDs7IwxY8Zg27ZtVcQjIyMDXl5eeP/+PUpKSuDt7Y0uXbpAV1dXXF9FiKysrFrLbMHU1FSozNY+GTpbvOEhNpFRUFDAoUOHEBYWBmNjY/j7+2Pv3r1QVVXF5cuXYWhoCAC4dOkS3r17h9OnTzNxMIaGhli+fDkAYOXKlejTpw/Gjx8PU1NTvH79Gv7+/kJD7ZSq3L17l5mFzeFwcOfOHQlbVD10tnjDg0Ma2ay59PR0DB8+HBEREdDQ0Pju8y1btkwoUFAwW51trFmzRqhPTFFREdu2bZOgRTUTFBSEO3fuYODAgbQ/pgEgNk+moVJZo8vKyiRkSe1IUzPE1NQUioqKrG3SUb4OKjKNBGlqhty9exefP39mbZOO8nVQkflOKo58VVdmCyoqKjAxMQGHw4GxsTErh9kBuiJnQ4SKzHdSOfKZzWtQSUOa0OpW5KRIN1RkvpMXL14IlVNTUyVjSANBmlbkpIgGFZlGhDSsIClNK3JSRIOKTCNBWvLeSMuKnBTRoSLTSJCWvDeVV+Bk64qcFNGhItNIkJY5QSNHjmTiedi8IidFdKjINBKkZYKktKzICdDkWqJCRaaRoKenJ1TW19eXkCVfRhqG2gHg6tWrSElJwZUrVyRtCquhItNIuHjxolA5ODhYMoaIgGBFTrZ7MYImZ0xMDPVmaoGKTCOhctrRuk5DWpekp6fDzc0NGRkZkjalRq5evSoUz0O9mZqhIvOdVE75WbnMFion66rr5F11SWBgIEpKSvD7779L2pQaESS2r6nMJiTdd0RF5jupnKO4tpzFkmTcuHFC5cppSdlCeno642VlZmay1psRTDatqcwmJB2ESUWmkVBdjl82EhgYKFRmqzfTu3dvoXLljnW2kJeXh3v37kl0wikVmUbCP//8I1Rmq3svLX1HldPCsnXZZjasgU5FppFQOZkWW5NrSUvfUeUllmtaclnSVLcGurihIkNhFQ4ODkJlR0dHCVlSO0ZGRpCRKX98ZGRkWBvcyIY10KnIUFiFhoYG0xRRUFBAx44dJWxR9YwcOZIRGVlZWdYGDrJhDXQqMhRWkZeXBy6XC6B8BVG2BrmpqKgwy/Do6uqyNnCQDZHeVGQorOLMmTO1ltlEeno6AOD169cStqRmBIItgMfjid0GKjIUVhEfHy9UjouLk5AltZOeno6cnBwAQE5ODmvjeZ4+fSpUfvz4sdhtoCJDoXwD0rI8ceUleySxzJpYRSYhIQH29vYwMDCAjY1NjcN+b968wcyZM2FoaIgRI0YgMjKS2UcIwa5du2BqagojIyN4eXmhtLRUXF+BUs8oKirWWmYL0rI8MRsQm8hwuVwsWLAAo0ePRnR0NObNmwcXFxd8+vSpSt3ly5dDR0cH9+/fx5YtW7Bs2TKm3XvmzBncuHEDwcHBuH79Op4+fQp/f39xfQ1KPdOmTRuhspqamoQsodQVYhOZBw8egMfjYcaMGZCXl4eVlRW0tbURGhoqVC8tLQ3Pnj3DkiVLoKCgAFNTU5ibm+PcuXMAylMWODk5oX379lBVVcXixYtZ3TlI+Toq920IOlcp0oucuD4oOTkZWlpaQts0NTWRlJQktC0lJQXq6upQVlYWqidoWlU+j6amJt6/f4+PHz+iZcuWQufKz8+vMgRaOUz9wYMHuHDhQo1283g8JixbVJYuXVplm6ys7BdDzydMmABjY+Nq9zUUO4Gvt1Wa7QS+bGttdgJ1f+/ry86aEJvIFBUVoUmTJkLblJSUqsxaLiwsrLZeSUkJc56K6RQEdQX7K3L8+HH4+fnVif0UCuXbEJvIKCsr4/Pnz0LbiouLhTwWUepVFBzgP3GpfB4AcHJyqpLSIDMzE9OmTWPKxsbG36TOAqp7K+zevfubz1cT1M66RVrsBL7PVnHaWRNi65PR0tJCWlqa0LbU1FRoa2tXqffmzRshIalYT1tbW+g8qampUFNTqzbiskWLFtDQ0BD6Y+uEu/qmshvM1lnD0kKvXr2EymxN9cAGxCYyJiYmIIQgICAAPB4PISEhSExMhIWFhVA9TU1N9OjRA7t27QKXy8W9e/cQEREBa2trAMDYsWNx9OhRZGRkIDc3F76+vrC1tRXX16hC5beCuN8SovLrr7/WWmYL0nI958yZI1R2cXGRkCW1w4brKTaRUVBQwKFDhxAWFgZjY2P4+/tj7969UFVVxeXLl2FoaMjU9fX1RUpKCkxNTbFhwwZ4enqie/fuAIApU6Zg1KhRmDx5MiwtLaGtrQ1XV1dxfQ2pRuC9UC+mbhB4M9SLqR0OkUQIoARJT0/H8OHDERERAQ0NDUmbQ6E0eMTW8csWBEN9bM24RqFIE+3bt4ecXO0y0uhERhD+XXGEiUKhfBuitAgaXXOppKQEz549g5qaGrPm8vciGBY/efIkq0evqJ11C7WTejLV0qRJk3pLldi+fXup6OehdtYt1M7aoakeKBRKvUJFhkKh1CtUZCgUSr1CRaYOaNGiBRYtWsTaZNICqJ11C7VTNBrd6BKFQhEv1JOhUCj1ChUZCoVSr1CRoVAo9QoVmQZGWVmZpE2gUIRodBG/3wohBBwOR9JmVKG4uBhhYWHgcrnQ1dWFrq4uysrKmHWaKQ0TLpeLDx8+gMvlolOnTpI2p1bo6FIN8Pl8yMrKori4GAoKCnU2z6ku+fTpEyZNmoT27duDEIKYmBgEBATU27SJukBwXSvCRgFPS0vDxYsX0bNnT/Tq1QudO3eWtEkMnz59grOzM5o1a4YHDx7A1dUVs2fPZuV1BKjIVIvgZiUkJGDr1q34/PkzZs2aBVNTUzRt2lTS5gEof1iXLVuGdu3aYf369SgtLcXWrVtRVFSErVu3Stq8aqkoMAkJCZCTk2PSqrLpAUlNTYW9vT169eqF7OxsGBkZYcqUKejRo4ekTQOXy8WkSZNgZGSE5cuX4+bNm1i9ejUCAwPRt29fSZtXLdSnroTgx/7x40csWLAA/fv3R48ePeDl5YXQ0NBqF6OTBDIyMigpKWESTMvJyaFdu3ZISkr66qVRxIVAYObNm4dFixZh4cKF2LFjBwCAw+FIZAnVypSVlSEtLQ0LFy7E8ePHsWzZMnz48AGnTp1CQkKCpM1DQkICWrZsCXd3dzRt2hTW1tYYNGgQYxsb++Ron0wFBG/a/Px8JCcnw9raGosWLQIAHDp0CP7+/pCRkYGlpSWaNWsmMTsJISgsLIS8vDw+fvzI2K2trY3Y2FjWNUcqejBXrlwBj8fDiRMnEBsbi19//RV8Ph9r1qxhhEZStv7777/Yu3cv0tPT0a9fPwDAiBEjAADBwcE4ffo07OzsJJpuU0FBARkZGXj48CGTspbD4YDL5QIAK/viqMhUQFZWFvHx8VizZg1KSkpQXFwMe3t7dOzYkWnzHjx4EMXFxZg4caLQ+k/ihMPhoFmzZnB0dISSkhLzACcnJ4PH4zH11q1bh59++gkGBgYSsRMof7MK7Lt79y7evHkDMzMzdOjQAS1btoSysjI2bdoEWVlZrFy5UmIC8/r1azg6OmLkyJEoLS3Fy5cvERkZiaFDh2LEiBHgcDg4evQo7t27J1GRadu2LX766Seh5Xzz8vKE6vz1118wNDREq1atxG1e9RAK4fP5hBBCCgsLiaOjIzlw4AD53//+R4yNjYm3tzfJyclh6u7cuZMsXryYlJWVScrcGtm6dStZtWoVIYSQVatWETMzM8LlciVmj+C6EkLInDlziKmpKRk4cCCxsLBgtpeUlJCbN28SAwMD4uXlJQkzSVZWFgkMDCQ+Pj6EEEKSk5PJ0qVLybJly8jNmzeZetHR0ULfSVLk5+cLle3t7cn9+/cJIYQsXryYjBs3jhV2CmCfbyUBZGRkkJ6ejvnz5zNey9ChQ3HgwAEEBwfjwIEDyM3NBQAsW7YMPj4+rOlDAP7LW1xSUoLOnTvDy8sL8fHxCAsLg7y8PEpLSyVil8B1v3jxIlq0aIHbt2/Dx8cHHTp0wIwZM/D582coKirC1NQUO3bswLBhw8Rqn+D+HTp0CD4+PoiNjcWnT5+gpaWFRYsWgc/n48qVKwgLCwMAGBkZQUZGRuL9Hs2bNwdQbn9+fj5KSkrQsWNHbNiwAS9fvsTZs2chIyPDmt9noxaZijdBQ0MD2dnZCA4ORlJSEgghMDAwwP79+3H16lX8+uuvKCgoAACJ9x1URvAwt2jRAn5+fnj8+DGCg4MZgflSesT6JCIiAn5+fuBwOJCRkUHv3r2xbNkycDgcLFiwgBEac3NzmJqaivXBEIjz2rVr4eDggJKSEvz1118oLCyElpYWlixZgpycnCrrtbOl34PD4UBOTg4lJSWwt7fH48ePce7cOea+s+b3KWkDJAWfzweHw8GnT5+QkZEBAAgJCUGvXr2wfv16ZjUDQ0ND7Ny5Ezk5OULD12y5gcB/tmhqaqJ79+44efKkxASm8shW586d0b9/f0RGRuLJkydo0qQJdHV1sWzZMhQWFmLWrFlCwiKu65qSkgIPDw/s2rULKSkpWLJkCfr06YNz584hPDwcRUVF0NLSgqenJxYuXCgWm74WQgjk5OTQvHlzdOvWjTUvlso0yjgZQURsQkICNmzYgMLCQpiYmMDDwwMAMH78eMjKymLv3r1o165dtceKG1E9J0G9srIycDgcsYphxVGkW7duQUlJCbq6uuDxeNixYweePXuG9evXw8jICKWlpYiLiwOHw4G+vr7YbASApKQkODg4YNSoUbh79y60tLQwf/589OnTB97e3khISMCoUaMwbtw4NGnSBIDk7rsoREdHw9DQEHJycuByuVBQUJC0ScJIoiOIDSQnJxNTU1Oyd+9ecvHiRaKrq0s8PDyY/RMmTCDDhg0T6vQVJ6WlpYQQQoqKipj/i1KfEEJSU1Prza6aqNjROHfuXDJ8+HBia2tL5s6dSwoKCkh2djbZuHEjmThxIrl3757Y7RNQUFBA7O3tSUBAACGEkJCQEDJixAgyf/588vTpU0IIIRs2bCCbN2+WiH2i3GsBZWVlrOrgrYlGKTI8Ho9s2rSJHDx4kBBCyJs3b8iSJUuIrq4ucXNzY+qtX7/+q256XSEYuXr+/DmZPn06sbe3Jzdu3CCfPn2qtT4hhAQGBpKZM2eSvLw8sdhamU2bNhEnJydSVFRETp8+TaysrMjcuXNJfn4+ycnJIW5ubsTa2pp8/PhRIg9IRkYGGTNmDCGEkE+fPpHFixeTw4cPk0mTJpGpU6eS4OBgQsh/11Sco4iC65Gamkp27txJ/vzzT/Ly5csa61e07eTJk2Tfvn31buO3wE7/r56Rk5NDRkYGFBQUwOfz4erqiiFDhuDo0aMIDg7GokWL8Pr1a/zyyy+QlZUVawQt+cqIY1KhGXXy5En89ttvWLFihURSLRYUFCAlJQXr1q2DkpIS3r9/j+7du4PL5WLDhg3gcDhYsWIFdu7cCRUVFYk0P1q1aoXevXsjKysLs2fPhrKyMlxcXDBixAi8ffsWiYmJzDUlYu7cl5GRQWpqKn766Sc8evQIvr6+OHLkSLWRxpXv++7duzF06FCx2fpVSFTixIQgVqSiV3Lnzh2SmZlJ3NzcyNKlSwkh5Z7D1KlTibu7u0TesgL78vLySHR0NPntt9+YfQcPHiTm5ubk3LlzpKCggBBS1YPp378/iYuLE5u9la/Rq1eviJ2dHcnOziYHDhwg1tbWJDMzk+zcuZPo6OgQMzMzUlRUJDb7BGRkZJC7d++SGzduEELKPdlXr14Re3t7ps6qVauIj4+PRDwYwX3n8/kkPDycHDt2jBBCyI0bN8jixYuJu7s7ef78OVO/4nUPDAwkxsbG5NmzZ2Kz92thTxd0PVBWVgZCCOTl5ZGYmIj9+/dDVVUV/fv3x+jRowGUvxEEIeTHjh2Drq4u1q5dy3SeivNt+zURxxMmTICysjKA8jeZj48Pjh07hl69eonFVh6PB3l5eQDAixcv0KFDB3Tq1Ak7d+5EWVkZbty4AU9PT7Rr1w6tW7fGkiVLMHToULFHSf/777+YOnUq9PX1ERsbiwEDBmDGjBno0qULCgsLsXbtWpSUlCA5ORleXl5iv++CiOivmdIgsK3ifdfV1RWLvd9Cg20ucblcbN++HX///TdevnyJ2bNnQ0lJCR8+fMCBAwdw4sQJAOVzQS5cuIBJkyYhPj4eq1atksgPDQCKiorg7e0NGxsbbNiwATweD4GBgUwg4Jw5czBq1Cg8ePCAeVgPHz4MHx8fBAQEiO2HlpWVxQjMkiVLMHfuXEycOBHXrl2DqqoqACAnJweFhYWIiIjA3r17YWBgIPYHIScnB76+vli8eDGOHDmCv/76C7KysggICEBBQQGcnJxQWloKWVlZXLhwAXJycuDz+WK97zIyMsyUhhYtWghNaQDKhWbChAlITk7GvXv3mGOlRWAANNzmUm5uLnFwcCCurq5ky5Yt5MCBA4QQQt6+fUtOnDhBrK2tyeXLl0lhYSG5du0aOXXqFOHxeIQQwvwrTl6/fk2mT59O3NzcGFf94cOHxMTEhHh5eQmNcgn2l5aWkrVr15JHjx6Jzc7IyEiio6NDEhISyJkzZ8jUqVNJeno6Wb16NZkxYwbTcbpmzRpiZmZGzM3NydWrV8Vmn4C8vDyyefNmMmjQIPL48WNme2FhIbG0tCR+fn5VjpHEff+WKQ2PHj0igwYNYnUTqSINUmQENyMrK4ssXryYDBw4kJnTQwgh7969I8eOHSPW1tbMUKYAcY4mVW73jxkzhnmABfv++ecfMnDgQOLm5iY0Z0WSQ5fr1q0jffv2Ja6uriQ6OprZ7unpSaZNm0auXr1KysrKSEZGBsnIyCCElH9XcfVzFBYWkvfv3xNfX19iZWVFfvnlF6H9Bw4cID///LPQNnHPRRPcPy8vL9K/f38yffp0pq8tOTmZLFmyhCxfvpxcu3at2uPfvXsnNlu/lwbZXBLML2nTpg08PDwwYMAA5OTkICoqCoQQtG3bFlZWVhg9ejTi4uKEIk7FlQHveyOOJRkY5unpiUmTJuHatWtIS0tjtq9btw56enrMKJ26ujo6dOgAAGILDExKSoKzszPWrVsHZWVlTJkyBe/evcOhQ4eYOVz//PMPFBUVhY4T1yiSYKRS0EReu3YtHB0dUVJSgoiICGbulKurK3JycpCYmCh0vOC4tm3bisXeuqBBR/wK2rzZ2dnw8PAAh8OBnZ0dM9T38eNHqKioiH24Uhojjqvjl19+wcWLF/H777+jZ8+eQtutra3FnmLixYsXsLe3x7x586CnpwcjIyNkZGQgODgYiYmJSEpKgoGBAd68eYPDhw9XEZr6RhARnZKSgqNHj6J169YYO3YstLW1sXXrVjx//hwTJkyAhYUFmjZtijdv3qB9+/asud/fSoMWGUBYaDZv3gwAsLGxgYWFBVNHnAIjICUlBY6OjnBwcEDHjh2xfv16TJo0CZs2bQIATJw4Ebm5uTh//jzTmcpGNm7ciJCQEJw8eVLi6SmPHj2KV69ewcPDA69evcK6desgJyeHmJgYODs7459//oGKigp8fHwgJyfHTM4UB4LfWEOb0iAK0mu5iFRsOm3cuBEfPnzAs2fPhOqIW2BKS0vx+++/Y+bMmViwYAGMjY0xfPhwBAUFYe3atQCA8+fPY9CgQVBRURGrbV/L5s2bMXbsWIwbNw7x8fEStUVdXR2hoaGYP38+Zs+ejaZNm2LhwoUwNjaGgoICLC0tUVJSAn9/f5SWlorVkxE0jTdu3IiFCxdi8+bNWLZsGVJSUnDgwAE8e/YMbm5u6NSpE/79919GYAD2zPr+VqQ+TqayylfnlVQUmj179jAPriQ8GOC/iONu3boxEcf29vaYNm0apk+fjoKCAqxZswa//PILgOoz/Nc3pMJEyy/9yDdt2gQej8cMtUuKH3/8Ea6urvjw4QPGjBkDGxsbAOX5Yjp37ozhw4dDRkYGly9fhoyMDBYsWCBW+/Lz85mh88LCQly7dg2TJ09GWFgYtm7dikmTJmHLli1MH6Gkfp91jdSLjIyMDN6+fYuIiAg4ODjUeFNkZGRQWlrKpCQU52xVQeBaRbFwdnaGpqYmNmzYgI4dO2LixIlISEhAv379oKqqio4dOzLHi1tgQkNDER0djQ0bNkBWVlYkoREIoiQfjGbNmmHatGkoKyvDgwcPEBYWhoiICLx9+xajRo2CgoICbG1tIS8vj8GDB9e7PZWvW8UpDa6urujcuTNcXFzA5/Pxxx9/SHRKQ30i3X4Yyn/UN2/exO3bt2vNAEf+P/cGUP4Q/fnnn/VuW1lZGfh8PhNxvGLFCmzevBl//vknTE1N0a5du2ojjn/++WeJZWDjcrnIysrCixcvsHPnTiY4rTZbBNedx+Ox4sHgcrm4e/cuAgICwOfzceHCBSgoKIDL5aJZs2aYNGkSM+pVXwiu25s3b3Dv3j2Eh4dDSUkJnp6eKCkpQVlZGby9vQGU52YeN24cVq9e3eAEBpDSjt/KuVJSU1MxY8YMuLm5YcyYMVXqk0qTyby9vXHx4kVoaWnVm41cLhc7d+7EwIED0aVLFzg6OmLQoEEoKSlBWloaJkyYgOnTp2Pjxo149uwZZGVlUVJSggsXLkBeXl4inX0Vkx35+PggKioKw4YNw/z582v0aATH5ObmYsqUKTh58qRQkmtJIcgMJ/ACxZnISXCdapvSMGfOHPTu3ZuZ0hAcHAw5OTmp7+StDqkUGaA8tF1NTY0RkBMnTiAqKgqenp5CMQSVBWbPnj04evRovYdif/jwAUuWLEHr1q3Rpk0btG3bFnPmzEFmZiZu3LiBs2fPYs6cORg+fDiioqKQm5uLSZMmQU5OTuKZzVasWIH8/Hy8evUKTZo0wcCBA7Fy5coqQiOw88OHD5g8eTLc3NxgZmYmMbtrQpyegeCzcnJy8PPPP8PIyAjTp09Hbm4u3N3dwefzsXz5cjx58gT3798HIQRbt26t0pxuUIgh4K/OKSoqItbW1mTixIkkKCiIZGdnk7dv3xJnZ2cmhLy0tLTaWcriCMWWlojj6hDMniaEkA8fPpDg4GCyaNEism3bNqHZwoIQ/NzcXGJubk4iIyMlZjPbkJYpDeJCavwyQZ9AWVkZlJSUcPjwYRgZGSEiIgKTJk3C69evIScnBz8/PwDlnaWCt9eJEyfg5+cntslk0hBxLKByrhxCCLOWdsuWLTFmzBgMHjwY4eHh2LlzJ0pLSyEjI8N4MHZ2dti0aROGDBkiVrvZhuA6FhUV4fPnz2jVqhVatmyJK1euMHWUlZUxYcIE5OTkCB1LKvQXNkSkQmQEnWipqanYtGkTE7Dm5uYGLy8vjB8/Hvv27QMhBA8fPkRMTAxz7JMnT/Dbb7/h4MGDYp2tKhAaVVVVrF27FkpKSjh79ixu3boFAFBTU8PUqVOxbds2iS2vUnHhNXd3dzx//hw8Ho+ZAQyUz1I3MzMDh8NBYmIiXrx4AaB8QTEbGxsqMPjvOrJ9SoOkkJo+maSkJDg5OWHs2LFo3rw5s3ysgISEBGRmZsLDwwNDhgxhonsLCwtRVFQENTU1SZjN2ojjivj5+eHWrVs4e/YsCCFM39CJEyegoKCAmJgYbN26Fdu3b4eWlhbKyspw8OBB9OnTB6amphKzm02wfUqDJJEKkeHxeFi2bBl0dXUxf/58FBQUYOfOnSgpKYGysjLc3d2Zum/fvsXMmTOxfft2sWfBr4mKQrNs2TL07dsXy5Ytk7RZAICwsDB4eHjAxcUFs2bNQmlpKV6/fo21a9ciIyMD2traiIuLg4eHh9DIHSuz4ksQNk9pkDSsbQhWfLvLy8ujQ4cOSEpKwr59+3Du3Dm0a9cOurq6iI6OxtOnT6Gnpwc+nw8lJSWoqamJbdVEaYs4rjyCoaGhASMjIxw6dAhDhgxB9+7d0a1bN/zxxx+4ceMGmjVrhmbNmkFPT49p0nE4HCowlVBXV4e/vz/evXuH1NRUdO3aFbNmzcL+/fuZKQ3/+9//4O/vj3nz5jUagQFY2idTMQ1Cfn4+AMDAwAAFBQVISEiAvb09Tp8+jTVr1qBFixbMgyorK4uYmBgkJiaKrXkkiDgODAwEUHP7umLEsYyMDLhcrkQF5t69e3jy5Al0dHTg7u6OQYMGYf369UJJqy0sLGBqaiq0wHxD7z/4VgRTGnR1dbFo0SIcOHAA/fv3h5ycHDp37owJEybA3NwcUVFROHjwoKTNFSusay5VTIOwdu1aqKqqokOHDti0aRPk5OTw8eNHFBUVoWnTpti+fTsSExNx5swZobdzZmYm2rdvLxZ7CSE4ffo0bt26BT8/vxpHCSp6LaGhoeDxeLC1tRWLjYCwxzV//nwkJydDWVkZ2tra8Pb2RlZWFnx8fPDy5Ut4eHhIfEa1tCKY0pCXl4eIiAg8f/4c58+fh4KCAj59+oTQ0FAMHjy43iOO2QTrPBnBKJKLiwuGDx+OgQMH4tGjR1i9ejVKSkrw6dMnuLm5wcnJCS9evMDp06eZZUsEw4j1LTCCBOVA+Zt9wIABiI+Px/Xr16utTyoFBK5Zswa9e/euVxsrIxCYjRs3ori4GEFBQRg/fjzi4+Ph5uaGNm3aYOnSpWjXrh3WrVuHwsJC1izYLk2wYUoD22CVJ0PK04Fi/fr1UFNTw/LlyxEfHw8/Pz9kZWWhc+fO+O2335CRkQEul4suXbowzRBxxxmwPeK4us/Pzs6Gq6srPD090bVrV/j4+ODFixfMtd24cSOys7Px6dMn6sl8B5Kc0sBGWOHJVPQKZGRkkJeXhx49euDz58/YsmUL9PX14ejoiGvXrsHGxgZKSkro1q0b06Eq7htYXFwMZ2dn2NnZ4fz588jJycHIkSNRVlbGpM3k8/lVBMbHx0esAiP4sQPl11gwMa9Vq1bw9/fHzZs3sXr1avTs2ZNZsaFt27ZUYL4TOTk5RmAaeqCdKEhcZASdvFlZWXjy5AnS0tJgYGCAESNGwMPDA61bt8a8efOgq6uLoUOHon///kKJnMS9bIk0RBwLEPy4t2/fjuXLl+PRo0eYM2cOgPKh6y1btkBdXR3q6upYunQpPD096ahRHUM7yiU8hE0IgaysLBITEzFv3jx06dIF+fn5UFdXR3h4ODgcDn766ScAwIEDB9C8eXNs3LgRgHgTOQk+KzU1FceOHWMSHrm5ueHDhw8IDAzEvn37IC8vz0QcC0LzBRHHgYGBYhOYgwcPgsvlYtGiRVi1ahWSk5Ohp6eHvXv3YtSoUeBwOMx3ioiIwMGDB7F7927WxBVRGhYSFRkOh4O8vDysWbMGDg4OcHFxwePHjzFr1iz0798f2dnZWLNmDXr27In3798jODgYwH/iJC4EIeMVI44FCb5btWqFxYsXMxHHSUlJuHz5MiMyWlpaCA8PF2vEsbq6OlatWoW8vDyoqKgw1y0kJASBgYFQUlJiOtDz8vKwYcMGDBgwQGz2URoXEu/4zc3NxdKlS3HixAnw+XzY29vjhx9+wNKlS3HmzBkoKytDRUUF48ePl1gaBGmMOA4PD8fKlSvRrFkzXL58Ga1atQKHw8HFixdx5coV9O7dG7q6uujduzc6dOgg8akNlIaLxPtkeDwe/v33X/z5559wcHBA586dsXXrVkRFRSE2NhazZs1i5tLw+XyxCUxF7a0ccWxra4uEhAQ0bdoUMTExePr0KQBIJOK4JkaMGIHdu3cjLy8Ply5dYgRk3LhxGD16NO7duwcdHR2hdZEolPpA4iLTrl07TJs2De7u7lBUVMTOnTsBAHfv3q0STyCJhdfYHnFcG8OGDcOOHTuwY8cOHDp0iNluZ2eHPXv2oEuXLhK0jtJYkHhzCShfGH3//v2IjY2Fnp4e3r9/j8zMTAQFBUFeXl5iC69JQ8SxKISFhWH16tWYM2cOFi5cKGlzKI0MVogMUJ6SITY2Fo8fP0bbtm0xceJEifXBpKamwtHREVOmTIGSkhIuXryIH374AV5eXsjOzsa6deuQn5+Ppk2bIiAggEmdCIg/6ZSohISEYMWKFbhx4wY6deokaXMojQjWiEx1iDvnqTRFHH8L2dnZrEjyTWlcSLxPpjbEJTDSFnH8rQgEhsXvFUoDRDqejnpE4C1lZWXh7du3aN68ORNxvGnTJibiOCUlBUOHDkX79u0lEnFcl9CRJIo4adQiIy0RxxSKNMPqPhlxkJeXBycnJ9jY2AhFHC9cuBB37tzB06dPhSKOxT3aRaFIO43akwHKPZIWLVowaxJv2bIFI0aMwOjRo5Gfnw9jY2OJRxxTKNJMo39aKkYcnzhxgok4PnfuHGJjY3H8+HGmrjgjjimUhoL09VrWMWyMOKZQGhKNvk8GYFfEMYXS0KAi8/+wKeKYQmlIUJGpBTpMTaF8P1RkKBRKvdLoO34pFEr9QkWGQqHUK1RkKBRKvUJFhkKh1CtUZCgUSr1CRYbyVRQVFWH37t2wtLSEvr4+zM3N4e3tzeRC/hKFhYUICgqqZyspbIIOYVNE5tOnT5gyZQqUlZWxcOFCdOvWDS9evMC2bdvQpEkTBAYGokmTJrWew8/PD3/99RcuXLggJqspkoaGslJEZseOHSgrK8Px48cZMenUqRO0tbVhYWGB8+fPY9q0abWeg77TGh+0uUQRCS6XiytXrsDBwaGKt6Kuro4TJ05gzJgxKCoqgoeHBwYPHsysX75v3z4AwIULF+Dn54e4uDjo6OggPT0dAHDkyBGYmZnB0NAQU6ZMwaNHj5hzE0Lg4+ODgQMHol+/fvDy8oKjoyPjCRFCcOLECVhaWkJPTw+2traIjIxkjnd0dISHhwdGjx4NU1NT+Pn5YdiwYUJiFx0dDQMDAxQWFtbX5WvcEApFBJKTk0n37t3J06dPa63n7u5ObG1tyePHj8mrV6/IoUOHmOOKi4uJt7c3GTt2LHn//j0pLS0lp0+fJkOGDCE3b94kaWlpZP/+/URfX5+8evWKEELIgQMHiImJCYmIiCCJiYlk1qxZREdHh5w/f54QQsi+ffuIkZERuXr1KklNTSV79uwhPXv2JM+fPyeEEOLg4EB0dXXJrVu3yOPHj0lmZibp0aMHiY6OFrJ5+fLl9XTlKFRkKCIRGxtLunfvTl68eFFrveDgYPLs2TOhbQYGBiQ4OJgQQsiePXvI+PHjmX3Dhg0jly5dEqo/c+ZM4u3tTQghZPDgweT48ePMvg8fPpA+ffqQ8+fPk7KyMmJiYkIOHz4sdLyLiwsjGg4ODsTZ2Vlov6OjI/n5558JIYRwuVxibGxMbt68+YUrQPlWaHOJIhKtWrUCUJ6utDbGjh2LzMxMeHl5Yc6cOfjxxx9RVFSEsrKyKnULCwvx5s0buLu7w9DQkPm7f/8+0tLSkJubi/fv30NPT485pmXLlujatSuA8hQdHz58gIGBgdB5+/Xrh+TkZKbcuXNnof22tra4du0a+Hw+bt++DQ6Hg0GDBn3N5aB8BbTjlyISXbp0QcuWLfH06VPo6+tX2b9161aoqakhOTkZd+7cwbhx4zBu3Dh4eHjA1ta22nMKhMfb2xu9evUS2tekSRPIy8sL1atMTSNZhBChYyrXs7S0xObNm/HgwQNcvXoVo0ePZj6LUvdQT4YiEjIyMrC1tUVgYCA+f/4stC89PR1nzpyBvLw8Ll26hG3btmH58uUYM2YM5OXlUVBQILS2lYDmzZtDTU0N7969Q5cuXZi/wMBAREVFoXnz5mjfvj3i4uKYYwoKCvDy5UsAQLNmzdC2bVuhjmIAePjwITQ1NWv8Ls2aNYOZmRnCw8MRFRUFGxub7708lFqgIkMRmQULFqCsrAzTp0/H7du38fr1a4SHh8PFxQU6OjqYOHEilJSUcOPGDbx+/RqxsbFYvHgxCCHgcrkAAGVlZWRnZ+P169coLS3FrFmzsG/fPoSGhuL169fYt28fTp48iW7dugEAZsyYgf379+PmzZtITk7G2rVrUVRUxIjVnDlz4O/vj5CQELx48QL79u3D7du34ejoWOt3sbW1RVBQEJo3b46+ffvW74Vr5NDmEkVkWrZsiVOnTmHfvn3YuHEjsrOz0bZtW4wcORLz589Hs2bN8Ntvv2Hbtm04d+4c2rZtC1tbW7Ro0YLxRiwtLREUFIQxY8bg5MmTmD59OkpKSvDrr78iOzsbXbt2xZ49e9CvXz8AgJOTE7KysuDm5obS0lLY29ujY8eOTPPGwcEBRUVF+PXXX5GTk4Pu3bvD398fRkZGtX6XH3/8EUpKSrC2tq7fi0ahEb8UdhMZGYlevXpBTU0NAFBaWooBAwZg//796N+//zefNy8vD4MHD8bFixehpaVVV+ZSqoF6MhRWc+7cORQUFGDt2rVQVFTE8ePH0aJFC/Tp0+ebzldcXIzIyEiEhISgb9++VGDEAPVkKKzm/fv32LJlC+7du4fS0lL07dsXGzZsYPpsvhYul4vBgwejVatW2L9/f60dxJS6gYoMhUKpV+joEoVCqVeoyFAolHqFigyFQqlXqMhQKJR6hYoMhUKpV6jIUCiUeuX/ALB1+RDRsRyBAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 288x288 with 1 Axes>"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### PCA plot using Gensim package\n",
    "\n",
    "From the Mallet's LDA model, it is observed that __topic 0,3,4 and 5__ make up the the spam emails. \n",
    "From the Gensim's interactive PCA plot below, __topic 1,5,7 and 9__ are clustered in the same region, which indicate there is a clear pattern on what topics/words dictate spam directory.  It also appears that words belonging to clusters of __topic 1,5,7 and 9__ are very similar to the __topic 0,3,4 and 5__ of Mallet's model. Thus we can conclude that those clusters make up the spam directories. \n",
    "Below are some of the observed characteristics of topics from the spam directory. \n",
    "\n",
    "- Topic 1 from the PCA plot below mainly talk about font sizes, styles of the email using words such as \"font\",\"width\",\"color\",\"verdana\". \n",
    "- Topic 5 talks about sending money to a certain address using words such as \"send\", \"free\", \"money\", \"address\", \"investment\", \"services\"\n",
    "- As seen in the PCA plot below, the overall term frequencies of the words in Topic 1 and 5 are very low, meaning they are very specific to each individual topic. This tells that these words are very unique in spam directories and can be used as features for modeling. \n",
    "- Looking at the shape of the PCA plot, the number of clusters could be reduced down to 3 for further research. "
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "source": [
    "import gensim\n",
    "from gensim.corpora import Dictionary\n",
    "import pickle\n",
    "import pandas as pd\n",
    "import pyLDAvis\n",
    "import gensim\n",
    "import gensim.corpora as corpora\n",
    "from gensim.corpora import Dictionary\n",
    "from gensim.models.coherencemodel import CoherenceModel\n",
    "from gensim.models.ldamodel import LdaModel\n",
    "import pyLDAvis\n",
    "import pyLDAvis.gensim_models\n",
    "from pprint import pprint\n",
    "import spacy\n",
    "import re \n",
    "import pyLDAvis\n",
    "import matplotlib.pyplot as plt \n",
    "import pandas as pd\n",
    "import gensim.corpora as corpora\n",
    "\n",
    "# Create Dictionary\n",
    "dictionary = corpora.Dictionary(combined_df['text'])\n",
    "\n",
    "# Create Corpus\n",
    "texts = combined_df['text']\n",
    "\n",
    "# Term Document Frequency\n",
    "doc_term_matrix = [dictionary.doc2bow(text) for text in texts]\n",
    "\n",
    "# Creating the object for LDA model using gensim library\n",
    "LDA = gensim.models.ldamodel.LdaModel\n",
    "\n",
    "# Build LDA model\n",
    "lda_model = LDA(corpus=doc_term_matrix, id2word=dictionary, num_topics=15, random_state=100,\n",
    "                chunksize=1000, passes=50)\n",
    "\n",
    "lda_model.print_topics()"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/boyun/opt/anaconda3/lib/python3.8/site-packages/h5py/__init__.py:46: DeprecationWarning: `np.typeDict` is a deprecated alias for `np.sctypeDict`.\n",
      "  from ._conv import register_converters as _register_converters\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "[(0,\n",
       "  '0.096*\"http\" + 0.013*\"web\" + 0.010*\"software\" + 0.010*\"site\" + 0.007*\"windows\" + 0.006*\"url\" + 0.006*\"internet\" + 0.005*\"news\" + 0.005*\"file\" + 0.004*\"...\"'),\n",
       " (1,\n",
       "  '0.011*\"business\" + 0.009*\"blockquote\" + 0.009*\"/blockquote\" + 0.007*\"company\" + 0.007*\"government\" + 0.007*\"money\" + 0.005*\"security\" + 0.005*\"information\" + 0.005*\"grants\" + 0.005*\"new\"'),\n",
       " (2,\n",
       "  '0.036*\"com\" + 0.026*\"exmh\" + 0.024*\"received\" + 0.020*\"org\" + 0.020*\"redhat\" + 0.020*\"localhost\" + 0.018*\"listman\" + 0.017*\"mx1\" + 0.014*\"taint\" + 0.014*\"spamassassin\"'),\n",
       " (3,\n",
       "  '0.057*\"font\" + 0.053*\"/font\" + 0.049*\"http\" + 0.044*\"/td\" + 0.031*\"nbsp\" + 0.026*\"img\" + 0.024*\"/tr\" + 0.024*\"width=\" + 0.020*\"arial\" + 0.020*\"href=\"'),\n",
       " (4,\n",
       "  '0.028*\"net\" + 0.027*\"linux\" + 0.027*\"localhost\" + 0.024*\"received\" + 0.023*\"ilug\" + 0.022*\"org\" + 0.017*\"esmtp\" + 0.016*\"lugh\" + 0.015*\"aug\" + 0.014*\"rpm\"'),\n",
       " (5,\n",
       "  '0.030*\"zoo\" + 0.030*\"petting\" + 0.028*\"net\" + 0.017*\"0xdeadbeef\" + 0.013*\"cache\" + 0.009*\"bulletin\" + 0.008*\"gentle\" + 0.008*\"request\" + 0.006*\"resent\" + 0.006*\"tobacco\"'),\n",
       " (6,\n",
       "  '0.072*\"com\" + 0.028*\"received\" + 0.023*\"net\" + 0.017*\"netnoteinc\" + 0.015*\"jul\" + 0.014*\"mail\" + 0.014*\"localhost\" + 0.013*\"email\" + 0.013*\"aug\" + 0.011*\"content\"'),\n",
       " (7,\n",
       "  '0.054*\"perl\" + 0.033*\"use\" + 0.016*\"pudge\" + 0.011*\"alb\" + 0.011*\"osdn\" + 0.010*\"cpu59\" + 0.009*\"messages\" + 0.009*\"change\" + 0.009*\"preferences\" + 0.008*\"ftp\"'),\n",
       " (8,\n",
       "  '0.016*\"devel\" + 0.013*\"eff\" + 0.012*\"hughes\" + 0.012*\"spamassassin-devel\" + 0.011*\"tony\" + 0.010*\"mindspring\" + 0.010*\"family\" + 0.008*\"spamassassin-devel-request\" + 0.007*\"instructed\" + 0.006*\"belphegore\"'),\n",
       " (9,\n",
       "  '0.033*\"social\" + 0.033*\"dvd\" + 0.023*\"iiu\" + 0.014*\"copy\" + 0.011*\"gray\" + 0.010*\"kingdom\" + 0.009*\"burn\" + 0.008*\"javascript\" + 0.007*\"events\" + 0.007*\"earnings\"'),\n",
       " (10,\n",
       "  '0.058*\"net\" + 0.030*\"sourceforge\" + 0.026*\"received\" + 0.019*\"localhost\" + 0.017*\"esmtp\" + 0.017*\"usw\" + 0.015*\"pro\" + 0.013*\"aug\" + 0.013*\"example.sourceforge.net\" + 0.012*\"mailto\"'),\n",
       " (11,\n",
       "  '0.058*\"////////////////////////////////////////////////////////////////////////////\" + 0.044*\"normal\" + 0.026*\"microsoft\" + 0.023*\"priority\" + 0.021*\"iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii\" + 0.018*\"mimeole\" + 0.014*\"importance\" + 0.012*\"msmail\" + 0.012*\"sslv3\" + 0.011*\"outlook\"'),\n",
       " (12,\n",
       "  '0.082*\"org\" + 0.053*\"localhost\" + 0.032*\"com\" + 0.031*\"sep\" + 0.030*\"received\" + 0.025*\"dogma\" + 0.025*\"slashnull\" + 0.023*\"jmason\" + 0.019*\"taint\" + 0.019*\"esmtp\"'),\n",
       " (13,\n",
       "  '0.062*\"com\" + 0.043*\"fork\" + 0.041*\"xent\" + 0.028*\"localhost\" + 0.025*\"received\" + 0.023*\"org\" + 0.018*\"mailto\" + 0.017*\"esmtp\" + 0.016*\"admin\" + 0.015*\"sep\"'),\n",
       " (14,\n",
       "  '0.014*\"one\" + 0.012*\"get\" + 0.011*\"n\\'t\" + 0.010*\"people\" + 0.009*\"/span\" + 0.008*\"time\" + 0.008*\"make\" + 0.007*\"would\" + 0.007*\"sans\" + 0.007*\"report\"')]"
      ]
     },
     "metadata": {},
     "execution_count": 15
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "source": [
    "# Visualize the topics\n",
    "pyLDAvis.enable_notebook()\n",
    "vis = pyLDAvis.gensim_models.prepare(lda_model, doc_term_matrix, dictionary)\n",
    "vis"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/Users/boyun/opt/anaconda3/lib/python3.8/site-packages/ipykernel/ipkernel.py:287: DeprecationWarning: `should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "  and should_run_async(code)\n"
     ]
    },
    {
     "output_type": "execute_result",
     "data": {
      "text/html": [
       "\n",
       "<link rel=\"stylesheet\" type=\"text/css\" href=\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v1.0.0.css\">\n",
       "\n",
       "\n",
       "<div id=\"ldavis_el406101403484648204008810904140\"></div>\n",
       "<script type=\"text/javascript\">\n",
       "\n",
       "var ldavis_el406101403484648204008810904140_data = {\"mdsDat\": {\"x\": [0.054326539567039644, 0.2362369052221656, 0.2160715081528144, 0.2213812352602446, 0.06851223411823464, 0.19804741189857797, -0.01970657986951019, 0.1917266516963101, -0.004467118495716355, 0.14832487875773084, -0.22831049264111278, -0.27831854496629965, -0.23974608969120037, -0.28855791881326975, -0.27552062019600837], \"y\": [-0.2660013438314006, 0.10873553760303195, -0.006305973285223762, 0.09935814593680958, -0.2377344386090619, 0.12253214726302045, -0.26727358157701564, 0.13803904506274214, -0.2009825333429885, 0.15330564103658098, 0.03437498024023909, 0.05829504721616995, 0.07239401453230399, 0.083843699257733, 0.10741961249705972], \"topics\": [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], \"cluster\": [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], \"Freq\": [22.999359125099357, 12.349813708533311, 11.366934999747205, 9.967681266926169, 8.867432234503779, 7.409811399693028, 6.970846805013293, 6.95566645431438, 5.612169408720856, 4.340695419676248, 1.1622979283521824, 0.7517348050970853, 0.5169389474429479, 0.4451928678095793, 0.28342462907058]}, \"tinfo\": {\"Term\": [\"http\", \"font\", \"/font\", \"com\", \"net\", \"org\", \"/td\", \"localhost\", \"fork\", \"xent\", \"nbsp\", \"received\", \"img\", \"/tr\", \"width=\", \"arial\", \"href=\", \"sep\", \"src=\", \"face=\", \"linux\", \"esmtp\", \"sourceforge\", \"aug\", \"ilug\", \"width=3d\", \"taint\", \"spamassassin\", \"size=\", \"helvetica\", \"font\", \"/font\", \"/td\", \"nbsp\", \"/tr\", \"width=\", \"arial\", \"href=\", \"src=\", \"face=\", \"width=3d\", \"size=\", \"helvetica\", \"height=\", \"/table\", \"color=\", \"border=\", \"verdana\", \"bgcolor=\", \"color=3d\", \"ffffff\", \"size=3d\", \"sans-serif\", \"face=3d\", \"//clickthru.online.com/click\", \"height=3d\", \"align=\", \"src=3d\", \"align=3d\", \"href=3d\", \"img\", \"table\", \"center\", \"=20\", \"http\", \"fork\", \"xent\", \"xent.com\", \"fork-request\", \"//xent.com/mailman/listinfo/fork\", \"khare\", \"rohit\", \"lair\", \"//xent.com/pipermail/fork/\", \"fork.xent.com\", \"crackmice\", \"slack\", \"listinfo\", \"leitl\", \"crank\", \"barrera\", \"murphy\", \"maya\", \"smtp1\", \"beberg\", \"garym\", \"adam\", \"eugen\", \"argote\", \"bitbitch\", \"fees\", \"news.com\", \"magnesium\", \"hydrogen\", \"endeavors\", \"pdt\", \"mailman\", \"friends\", \"beenthere\", \"request\", \"admin\", \"gary\", \"errors\", \"com\", \"mailto\", \"postfix\", \"subscribe\", \"localhost\", \"sep\", \"help\", \"received\", \"taint\", \"org\", \"esmtp\", \"delivered\", \"spamassassin\", \"aug\", \"list\", \"subject\", \"http\", \"version\", \"mon\", \"net\", \"ist\", \"jmason\", \"wed\", \"thu\", \"jul\", \"jun\", \"einstein\", \"zzzzteana\", \"grp\", \"scd\", \"pop3\", \"securityfocus\", \"yahoogroups.com\", \"temple\", \"yahoogroups\", \"qmqp\", \"secprog\", \"nov\", \"base64\", \"nnfmp\", \"sentto\", \"password\", \"outgoing\", \"egroups\", \"compliance\", \"forteana\", \"rom\", \"btinternet\", \"banned\", \"moderator\", \"5,000\", \"egroups.com\", \"webmaster\", \"forteana-unsubscribe\", \"//docs.yahoo.com/info/terms/\", \"mandark\", \"yahoo\", \"returns\", \"aol.com\", \"removal\", \"adult\", \"emails\", \"nextpart\", \"removed\", \"transaction\", \"msn\", \"wish\", \"fax\", \"webnote\", \"groups\", \"email\", \"zzzz\", \"remove\", \"netnoteinc\", \"may\", \"mail\", \"please\", \"rates\", \"labs\", \"com\", \"smtp\", \"receive\", \"hotmail\", \"jul\", \"aol\", \"click\", \"credit\", \"received\", \"net\", \"mime\", \"free\", \"content\", \"transfer\", \"aug\", \"message\", \"charset\", \"mon\", \"subject\", \"reply\", \"localhost\", \"list\", \"esmtp\", \"return\", \"date\", \"version\", \"type\", \"ilug\", \"lugh\", \"rpm\", \"tuatha\", \"linux.ie\", \"freshrpms.net\", \"egwn\", \"freshrpms\", \"zzzlist\", \"rpm-zzzlist\", \"matthias\", \"rpm-list\", \"rpm-list-request\", \"//lists.freshrpms.net/mailman/listinfo/rpm-zzzlist\", \"ilug-admin\", \"//www.linux.ie/mailman/listinfo/ilug\", \"maintainer\", \"listmaster\", \"auth02\", \"1.1\", \"ilug.linux.ie\", \"//lists.freshrpms.net/mailman/listinfo/rpm-list\", \"rpm-zzzlist-admin\", \"//lists.freshrpms.net/pipermail/rpm-zzzlist/\", \"rpm-zzzlist.freshrpms.net\", \"rpm-zzzlist-request\", \"jinny\", \"ns2\", \"saou\", \"gnu\", \"irish\", \"eircom\", \"linux\", \"feb\", \"root\", \"group\", \"clean\", \"net\", \"localhost\", \"aug\", \"list\", \"admin\", \"received\", \"esmtp\", \"org\", \"users\", \"fri\", \"mon\", \"mailto\", \"jul\", \"http\", \"tue\", \"com\", \"wed\", \"ist\", \"/span\", \"sans\", \"/center\", \"lockergnome\", \"quot\", \"height\", \"cc0000\", \"mortgage\", \"weight\", \"banner\", \"/script\", \"spambayes\", \"/select\", \"0.2\", \"skip\", \"extreme\", \"0.99\", \"noshade\", \"commissions\", \"ham\", \"consolidation\", \"chicago\", \"guido\", \"refund\", \"pricing\", \"e-mails\", \"multipart/alternative\", \"currency\", \"financially\", \"telecommunications\", \"orders\", \"comcast\", \"margin\", \"weeks\", \"/title\", \"ordered\", \"total\", \"decide\", \"report\", \"program\", \"order\", \"amazing\", \"people\", \"one\", \"make\", \"remember\", \"ever\", \"tim\", \"away\", \"method\", \"much\", \"get\", \"every\", \"time\", \"day\", \"would\", \"send\", \"n't\", \"say\", \"like\", \"money\", \"see\", \"want\", \"work\", \"address\", \"could\", \"good\", \"even\", \"know\", \"way\", \"...\", \"first\", \"name\", \"rssfeeds\", \"utf\", \"newsisfree\", \"awl\", \"txt\", \"unlimited\", \"boingboing\", \"var\", \"yellow\", \"featured\", \"guidelines\", \"oracle\", \"sleep\", \"dekalb\", \"boing\", \"quicktopic\", \"ref\", \"diveintomark\", \"millionaire\", \"domains\", \"xxx\", \"baseline\", \"fark\", \"hormone\", \"userland\", \"pda\", \"instruction\", \"lookup\", \"sync\", \"directions\", \"nonsense\", \"tests\", \"paragraph\", \"hits\", \"align\", \"att\", \"supplied\", \"cgi\", \"shopping\", \"status\", \"www\", \"archives\", \"dogma\", \"slashnull\", \"org\", \"oct\", \"jmason\", \"fastmail\", \"jalapeno\", \"required\", \"sep\", \"level\", \"yyyy\", \"taint\", \"ist\", \"localhost\", \"date\", \"spam\", \"spamassassin\", \"encoding\", \"example\", \"plain\", \"imap\", \"tue\", \"thu\", \"received\", \"fetchmail\", \"drop\", \"single\", \"esmtp\", \"com\", \"path\", \"type\", \"return\", \"message\", \"content\", \"http\", \"postfix\", \"subject\", \"blockquote\", \"/blockquote\", \"grants\", \"guide\", \"grant\", \"federal\", \"assistance\", \"trading\", \"fund\", \"funds\", \"venture\", \"promotion\", \"planning\", \"nigeria\", \"marshall\", \"assist\", \"datapower\", \"areas\", \"republic\", \"pacific\", \"associates\", \"sponsorship\", \"nations\", \"generation\", \"strategies\", \"unsubscribed\", \"equity\", \"securities\", \"responded\", \"develop\", \"united\", \"estate\", \"government\", \"states\", \"president\", \"u.s.\", \"foreign\", \"capital\", \"investment\", \"billion\", \"resources\", \"freedom\", \"bank\", \"financial\", \"security\", \"america\", \"development\", \"technology\", \"company\", \"law\", \"business\", \"american\", \"market\", \"companies\", \"industry\", \"dollars\", \"million\", \"money\", \"said\", \"must\", \"world\", \"sales\", \"based\", \"information\", \"state\", \"also\", \"year\", \"new\", \"free\", \"help\", \"services\", \"sourceforge\", \"usw\", \"example.sourceforge.net\", \"cypherpunks\", \"cpunks\", \"minder\", \"lists.sourceforge.net\", \"razor\", \"list1\", \"razor-users\", \"list2\", \"mm2\", \"sf.net\", \"spamassassin-talk\", \"tlsv1\", \"cipher\", \"sha\", \"cbc3\", \"razor-users-request\", \"spamassassin-talk-request\", \"fw2\", \"//thinkgeek.com/sf\", \"thinkgeek\", \"2.0.9-sf.net\", \"edh\", \"//www.geocrawler.com/redir-sf.php3\", \"//example.sourceforge.net/lists/listinfo/razor-users\", \"razor-users-admin\", \"habeas\", \"//example.sourceforge.net/lists/listinfo/spamassassin-talk\", \"sponsored\", \"des\", \"pro\", \"talk\", \"debian\", \"lists\", \"exim\", \"https\", \"net\", \"cdt\", \"received\", \"esmtp\", \"aug\", \"jul\", \"localhost\", \"helo\", \"thu\", \"mailto\", \"spamassassin\", \"wed\", \"sep\", \"users\", \"com\", \"org\", \"example\", \"http\", \"admin\", \"broadcast\", \"islands\", \"ordering\", \"documents\", \"array\", \"lifetime\", \"usd\", \"sony\", \"rss\", \"listing\", \"editors\", \"unison\", \"paypal\", \"wing\", \"les\", \"broker\", \"databases\", \"convert\", \"rating\", \"recommendation\", \"underwriting\", \"winning\", \"incredibly\", \"resource\", \"presented\", \"3rd\", \"plc\", \"disc\", \"overnight\", \"streaming\", \"divx\", \"film\", \"tips\", \"titles\", \"weblog\", \"2.0\", \"links\", \"palm\", \"audio\", \"site\", \"interface\", \"web\", \"ultimate\", \"desktop\", \"software\", \"download\", \"print\", \"sites\", \"design\", \"writing\", \"press\", \"http\", \"play\", \"book\", \"file\", \"latest\", \"windows\", \"url\", \"search\", \"news\", \"issue\", \"page\", \"end\", \"code\", \"community\", \"blog\", \"internet\", \"digital\", \"july\", \"system\", \"newsletter\", \"user\", \"...\", \"new\", \"use\", \"files\", \"list\", \"also\", \"using\", \"users\", \"information\", \"exmh\", \"listman\", \"mx1\", \"redhat.com\", \"exmh-users\", \"exmh-workers\", \"deepeddy.com\", \"exmh-workers-request\", \"deepeddy.vircio.com\", \"exmh-users-request\", \"garrigues\", \"exmh-workers-admin\", \"cwg-exmh\", \"exmh-users-admin\", \"munnari.oz.au\", \"welch\", \"sequences\", \"vircio\", \"//listman.spamassassin.taint.org/mailman/listinfo/exmh-workers\", \"nmh-1.0.4\", \"07/13/2001\", \"//listman.spamassassin.taint.org/mailman/listinfo/exmh-users\", \"//listman.redhat.com/mailman/listinfo/exmh-users\", \"//listman.redhat.com/mailman/listinfo/exmh-workers\", \"envelope-sender\", \"qmail-ldap-1.03\", \"panasas\", \"brent\", \"mar\", \"kre\", \"int\", \"unseen\", \"workers\", \"corp\", \"redhat\", \"2.0.1\", \"spamassassin.taint.org\", \"example.com\", \"austin\", \"folder\", \"chris\", \"https\", \"taint\", \"users\", \"spamassassin\", \"com\", \"received\", \"example\", \"org\", \"localhost\", \"aug\", \"sep\", \"esmtp\", \"mailto\", \"wed\", \"edt\", \"postfix\", \"smtp\", \"mon\", \"list\", \"////////////////////////////////////////////////////////////////////////////\", \"iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii\", \"sslv3\", \"evergo\", \"cats\", \"//////////////////////////////////////////////////////////////////+piiiiiiii\", \"johnhall\", \"outblaze\", \"antiabuse\", \"employment\", \"d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3\", \"smtpd32\", \"jmhall\", \"x-antiabuse\", \"co2\", \"perfectpresence\", \"inexpensive\", \"0200a8c0\", \"homeowner\", \"relay1\", \"submitting\", \"hostname\", \"sol\", \"dhcp\", \"ranging\", \"coral\", \"nospam\", \"louie\", \"luis\", \"servants\", \"irs\", \"importance\", \"alter\", \"normal\", \"socialism\", \"mimeole\", \"missouri\", \"msmail\", \"priority\", \"qualify\", \"produced\", \"hotmail.com\", \"microsoft\", \"originalarrivaltime\", \"filetime\", \"asmtp\", \"outlook\", \"domain\", \"behalf\", \"build\", \"ucsc\", \"mailer\", \"fork-admin\", \"original\", \"transfer\", \"iso\", \"encoding\", \"7bit\", \"express\", \"iiu\", \"kingdom\", \"javascript\", \"earnings\", \"w3c\", \"webdev\", \"comparisons\", \"award\", \"shopper\", \"iiu.taint.org\", \"airport\", \"burner\", \"plants\", \"vip\", \"alarms\", \"panels\", \"mini\", \"dakota\", \"ipod\", \"mid\", \"2fwww\", \"iiu-request\", \"trends\", \"extensions\", \"//iiu.taint.org/mailman/listinfo/iiu\", \"graphic\", \"sean\", \"rod\", \"medication\", \"tradesignals\", \"dvd\", \"gray\", \"social\", \"visual\", \"burn\", \"lesson\", \"island\", \"wake\", \"draw\", \"scroll\", \"sincerely\", \"events\", \"strip\", \"copy\", \"catalog\", \"pcs\", \"drives\", \"backup\", \"ibm\", \"drive\", \"data\", \"pudge\", \"alb\", \"cpu59\", \"visa\", \"bulkmail\", \"announcements\", \"//use.perl.org/my/messages/\", \"//use.perl.org/comments.pl\", \"======================================================================\", \"bugtraq\", \"1997-2002\", \"//use.perl.org/article.pl\", \"relax\", \"damian\", \"rafael\", \"lynch\", \"rcvd\", \"porters\", \"perl5\", \"maillist\", \"ziggy\", \"thai\", \"gnat\", \"summaries\", \"yapc\", \"gaim\", \"trna\", \"_______________________________\", \"toc\", \"bangkok\", \"perl\", \"sid\", \"preferences\", \"headlines\", \"osdn\", \"ftp\", \"slashdot\", \"posted\", \"subscribed\", \"use\", \"headline\", \"reserved\", \"fixes\", \"messages\", \"change\", \"daily\", \"log\", \"article\", \"september\", \"copyright\", \"forged\", \"stop\", \"rights\", \"updates\", \"receiving\", \"aug\", \"exim\", \"http\", \"please\", \"list\", \"may\", \"spamassassin-devel\", \"spamassassin-devel-request\", \"instructed\", \"belphegore\", \"//example.sourceforge.net/lists/listinfo/spamassassin-devel\", \"perkel\", \"spamassassin-devel-admin\", \"luth\", \"eff.org\", \"svanstrom\", \"//lists.sourceforge.net/lists/listinfo/spamassassin-devel\", \"spamassassin-devel.example.sourceforge.net\", \"emc\", \"assignee\", \"hormones\", \"8.12.3/8.12.3/debian\", \"serenity\", \"dinosaurs\", \"mgm\", \"uucp\", \"maynard\", \"list=spamassassin-devel\", \"fsck\", \"ulises\", \"//www.hughes-family.org/bugzilla/show_bug.cgi\", \"outbound\", \"psi\", \"ctyme\", \"assignedto\", \"virtues\", \"eff\", \"campus\", \"marc\", \"moon\", \"hughes\", \"eggs\", \"devel\", \"tony\", \"penguin\", \"bugzilla\", \"mindspring\", \"family\", \"sadev\", \"craig\", \"bug\", \"0.01\", \"developers\", \"pdt\", \"zoo\", \"petting\", \"0xdeadbeef\", \"bulletin\", \"gentle\", \"tobacco\", \"veuve\", \"gkm\", \"cigarettes\", \"glen\", \"10.\", \"inventory\", \"forwarded-by\", \"patterns_run_cb\", \"utmost\", \"pregnancy\", \"naked\", \"securing\", \"gartner\", \"data/nice/002\", \"breast\", \"sarun\", \"deadbeef\", \"mccready\", \"12.\", \"anyhow\", \"+ok\", \"-sarun\", \"chemicals\", \"complement\", \"bert\", \"cache\", \"resent\", \"jar\", \"net\", \"request\", \"google\", \"skin\", \"chinese\", \"list\", \"bottle\", \"subscribe\", \"unsubscribe\"], \"Freq\": [89586.0, 54670.0, 50294.0, 92429.0, 45666.0, 53206.0, 41686.0, 56909.0, 21757.0, 21028.0, 29573.0, 57206.0, 24731.0, 23173.0, 23039.0, 19399.0, 18680.0, 23526.0, 17829.0, 17643.0, 11964.0, 32980.0, 8581.0, 23415.0, 9691.0, 14185.0, 13856.0, 16254.0, 13299.0, 13090.0, 54669.73544048654, 50293.354561745036, 41685.77309812116, 29572.695706348102, 23172.257199938907, 23039.01332203311, 19398.434825641885, 18679.395199741826, 17828.902512946206, 17642.423496510528, 14184.640430080966, 13299.029112480252, 13089.141812708858, 12607.710801972731, 11145.884607382493, 10889.630218278377, 10794.184104348426, 10718.176333867154, 10510.014246918487, 8740.70211204233, 8367.143889509625, 8235.23071275846, 8119.524429949889, 7445.681656524386, 6767.414589886231, 6685.643846150003, 6667.653670813312, 6095.475667712941, 5451.657005315653, 5000.4832664363, 24631.15071003912, 11824.483684402001, 11132.606851757626, 8866.020294863243, 46264.3673597823, 21756.587635284042, 21027.6810133761, 5156.000477516142, 4075.300383512926, 4005.6784104076164, 2184.808322408155, 2147.2360590524663, 2026.5379627394427, 1814.2384774790114, 1678.6098259293078, 1493.8056455406713, 981.6012946585344, 870.0960888727055, 607.0140787957803, 598.1747083420022, 581.9630668133533, 559.7600306319353, 500.7297950300663, 488.1121453200032, 480.0868052658784, 475.27135131178954, 437.70125989393586, 428.6504379032304, 419.7382926556822, 415.4885541262646, 403.89281604401174, 388.7395627977774, 367.8140425995578, 346.9928051700072, 341.7108011173452, 4628.141750799271, 2639.299822119558, 2257.5288474850104, 1816.7406688756062, 2020.0992679028875, 8034.659036056767, 602.262949904705, 2029.4703763428718, 31919.455364852845, 8965.28619932813, 6858.391773134265, 1255.0518384356037, 14134.8495658013, 7601.86382670197, 2623.3990949796653, 12758.610736771712, 4839.842495278905, 11697.117951870956, 8484.016193735737, 4129.068281512943, 4677.517692970516, 4749.166649375079, 4184.149101058132, 3759.006253501244, 7362.400837264806, 3404.6454913794632, 3575.0284805826645, 4486.973170140816, 2991.326792408415, 2869.000873316669, 2885.748953738072, 2719.781484401528, 2716.270918797877, 2902.498908568675, 1070.0601154539513, 759.4394263352262, 679.98543728309, 633.0782920046573, 614.6685099115316, 586.3065913005934, 581.03213968497, 494.3436686965737, 384.46016373184307, 370.23883942382247, 356.2821605599385, 337.235575366859, 330.76945248188036, 299.71693534266046, 260.53478619702935, 240.24293793677, 234.49210261396536, 208.05669548843824, 158.34282699154775, 155.91640836802463, 152.02443070511998, 139.1372601008342, 135.8148287900771, 135.74985866215758, 376.56979158112534, 130.64782092446112, 1790.1027516055556, 123.77195492504804, 123.55780681110468, 2325.7354911345346, 3821.332993351012, 355.4922771585853, 266.7059761936231, 473.5802290462608, 271.91766214720565, 583.8241710283471, 901.432825290274, 1543.9064687164912, 456.318494444292, 926.6542700256338, 1164.891363090024, 851.1776303429519, 2931.4324547666274, 784.0300925978787, 6221.97563258542, 4753.657423463686, 1493.363132278771, 7807.140501304698, 4643.050963229377, 6713.900687289221, 3358.2623979202185, 852.6450652321806, 4020.755603839823, 33760.17083954297, 4670.936443363796, 1958.727686687673, 2088.4856701678277, 6960.252296022568, 1413.1862974957335, 2845.2318690465195, 1288.800528900611, 13281.46646492156, 10917.14159319311, 3175.5895656407715, 3497.35110500238, 5059.9012756571765, 2146.6914561536632, 6014.542677709678, 4374.844197619315, 2605.6850220187425, 4268.0869265021265, 4073.02351721152, 2424.48723756659, 6408.151073607298, 3715.4835787905304, 4591.946350609639, 3150.7137702013447, 3128.550590564318, 2936.7389464453345, 2724.7947506874225, 9690.239797435626, 6432.296084671092, 5922.288317039485, 5337.951947656016, 5018.925354366568, 4977.483164938809, 4816.171341591094, 4470.966022194337, 2080.1384849560663, 1619.2261546473014, 1491.9655019445863, 1383.61733151767, 1240.898128561489, 1227.27095569102, 1178.6641482196053, 1088.8332913622228, 1076.2069115764825, 1067.2236914142763, 1051.5187610607693, 789.8068548370005, 729.2549349714802, 686.7698805354767, 669.7959725984821, 617.4969696200748, 617.4969696200748, 578.1531894186323, 462.95099356150394, 461.86106178950854, 417.83293791213543, 395.3247710086878, 2465.988548326179, 921.6464792976169, 11301.89108475768, 726.5429288332482, 2643.4254449348405, 2376.7593776078447, 1251.452802305245, 11585.549598563313, 11088.028931518593, 6305.488233446581, 5152.765556423134, 4685.392208161474, 9792.582536971828, 6985.008199435932, 9032.871179450452, 2286.8403426093946, 2631.2532113919833, 3062.4700638496934, 3170.72265992981, 2998.7908456968075, 5131.778774630613, 2646.9174145866536, 4233.978142886814, 2489.4908085046177, 2314.1367602738487, 3229.6941686604196, 2518.643851984428, 2145.3205044857523, 1754.641010780111, 1429.2514236965826, 1018.9245605638116, 861.1365036246282, 795.8748815990124, 795.757385163413, 334.4562956699976, 282.6123606930469, 276.6741335638337, 273.212197222347, 273.02520698583777, 240.54267647475618, 222.2120361939175, 221.3483733032976, 199.1558601771804, 195.4032162910326, 178.30716446650675, 172.58347514720376, 166.7392878924466, 163.3234528796914, 159.03836855487822, 155.2541822165191, 722.6984433155666, 148.79665288643932, 142.21840130952833, 139.90723573979548, 138.8134186948248, 1002.4336153826222, 381.5862522877151, 479.9575580716652, 782.3871453117417, 675.7197793809245, 233.15845275710453, 777.4111926270987, 277.4753524396797, 2434.036335022442, 1895.472494845319, 2372.8300063380893, 317.69938128458796, 3625.2174208791607, 5256.639107852707, 2953.300167962043, 695.0004744770309, 987.5055864247868, 486.1841428168274, 639.6583288044758, 544.3720094336215, 1746.1614425638259, 4406.614007149836, 1458.274495972819, 3096.1631867617516, 1418.6287204949658, 2623.0384523169646, 1956.1454324380272, 3948.600767006325, 1122.546247979988, 2332.197431705108, 2108.671187197518, 1740.5713466917994, 1779.2086182184212, 1695.987302840648, 1874.5810304320873, 1603.6856920158527, 1429.2184392426616, 1530.833971154427, 1566.0130021365364, 1557.1552235755512, 1949.3943863303261, 1522.8718999664977, 1560.3904345223077, 2936.5008184964663, 1012.3429153486571, 556.7073333458345, 545.1755697980467, 407.0305092956578, 375.1981448542444, 347.27997644444343, 269.1945450703359, 263.6729412898226, 255.1028609725071, 244.98144402418086, 241.75775640107136, 202.62588210265108, 188.21274858748788, 186.19318866735262, 184.16238239119357, 180.40393946483147, 171.78182243670747, 166.26593109099989, 165.84163999580335, 160.24614944134947, 154.09639204850967, 154.0370005778887, 146.7001278269805, 142.70128065209047, 140.88677036444324, 128.91633645197132, 127.11554493353724, 124.47776568222986, 122.73217056510627, 627.6354944895982, 806.5409551967307, 418.91194525899215, 739.8027221721092, 1459.070812137145, 401.9480489408359, 396.81535486244104, 322.3632640924498, 333.90181115656947, 1142.294951933627, 3830.824188109564, 289.5771015520651, 7778.643610988583, 7771.927501387471, 25149.894512007515, 5274.67607424812, 7117.389295180228, 465.9882276012701, 4433.185300422894, 1090.8047069196693, 9463.422290628921, 896.4447536836454, 4503.588774429751, 5892.813503386209, 5390.259766377076, 16177.81530508146, 5393.704935808419, 2121.4337687304246, 5560.068174656062, 2393.9033419316866, 3916.910295010366, 2497.326260035307, 2811.7192494650444, 3971.0573220470396, 3959.9629167890666, 9057.245897444875, 2838.2026684679477, 2867.019617088997, 2922.617316944589, 5728.001761831805, 9700.551743093585, 3104.3958467372704, 2836.655169879072, 3023.214721982419, 3185.1985226862407, 3208.772027399099, 4598.785072358751, 2975.7292507944626, 2863.7523900018527, 2602.6814395953365, 2555.6866702514394, 1391.6180167196667, 986.1720521996106, 773.0903018621599, 676.6873559520476, 493.5481922057854, 426.33117023893135, 366.9424561562995, 356.37739271383833, 327.0620555433751, 305.66635108577947, 287.8952582071698, 259.8887814946198, 257.73636713033363, 243.47162370431775, 242.74691970826538, 238.85142370742403, 232.43807967710288, 232.2944588726573, 217.8455207573104, 212.5192914994606, 198.43093363344144, 195.42441883972432, 187.6444939697219, 186.42690795897437, 180.81627524891624, 176.37951421915992, 175.6663811695836, 170.66460864020533, 1119.6364877650458, 319.7567745924612, 2045.9499682034957, 1205.95957746867, 616.4151186559124, 815.0466545472002, 490.0160810233016, 434.62067018719523, 939.759827110228, 526.4876371404389, 424.76077952053737, 408.11087608841916, 628.6325602778647, 1080.1042231759939, 1452.0827784430896, 615.2352515795409, 591.7925476067593, 1087.7874659866893, 2128.467039414006, 571.5299898001412, 3118.6815671842087, 668.9022753773019, 1114.976268079542, 1054.0716424517645, 638.8451548892099, 801.5653095409431, 1217.3103232912972, 1875.7407745510072, 1111.0168211872917, 822.8532983952045, 1288.8364396186612, 827.888154762838, 766.7768307181855, 1399.6387482682173, 901.858713303939, 1145.0780601461915, 870.3489268213964, 1307.638001692872, 1072.4495298703916, 940.2765395647627, 806.1407539901464, 8580.952060784064, 4797.270110005494, 3643.7178517752577, 3290.016174834063, 3137.136656326263, 2980.6638461658276, 2943.048233145513, 2556.598322430531, 2312.5762293426537, 1638.2669789722913, 1532.8327450688198, 1513.3322606759148, 1263.5948712506686, 1145.3032649731974, 904.6672098110881, 899.0512504799348, 895.5682885926365, 893.9963240381367, 869.0908206369936, 768.4724876934275, 757.5230293173505, 714.4997161280364, 714.4997161280364, 711.3667797781593, 696.7028412830236, 606.8177311001152, 578.324298367859, 533.574351162613, 517.095099259357, 507.6612005182585, 1289.3987645939897, 1148.0385851821359, 4252.767381093292, 1935.133276032002, 1738.662239340496, 2479.985494249473, 1862.4908447646928, 3100.4825075550425, 16785.423034680683, 1382.5909525182385, 7512.101960868574, 4875.387873722013, 3881.0067458426415, 3296.326554802459, 5568.911394573985, 1493.9807147633144, 2926.837935181151, 3339.1296776642303, 2973.5585020316303, 2189.988867293019, 2314.0221793413607, 1768.4216491553395, 2694.200855035065, 2360.812724487717, 1802.1590971921023, 1900.6814327103227, 1773.703267739527, 880.8981838916231, 476.56882630009716, 408.8556106789003, 314.8980015290034, 302.13601970381677, 283.3598734623374, 264.9139265018522, 256.7287917247916, 253.24316257734841, 243.33104581022215, 239.07698877251514, 228.40166305388783, 227.26117620799397, 225.4573027548616, 216.22608172890583, 209.3252250250495, 199.44133920128232, 188.37660763989382, 174.48307606343036, 173.30397447891886, 171.5683934084047, 162.3102529340639, 154.63883209039838, 152.50547594630922, 149.74306191394166, 148.0960946378602, 135.64497768322545, 133.25784050137275, 117.56681717927954, 112.99853541018506, 316.70210483177044, 311.2275191930518, 837.3228034183213, 166.02321689072159, 498.4107610505204, 278.91160700393596, 766.9285134942819, 308.68768073040525, 325.71613147356885, 2292.0086375405835, 418.0144966727817, 3013.453748182811, 281.8270468106992, 329.3936174630933, 2433.548420076896, 847.0431174256333, 426.79321137098736, 902.5183813474634, 462.09533001319863, 382.6830654418719, 540.3669725329642, 22234.171799860374, 548.560976801808, 581.7977207095132, 1218.526150994721, 775.9257475921858, 1566.8018889014243, 1382.9156913264694, 966.9523187108467, 1258.1776466701765, 670.4607300051724, 845.5954433033417, 910.4018452393573, 908.334172008505, 484.0259596814551, 442.7505628167815, 1329.3943656280508, 564.317737119405, 573.9091190022571, 853.8031844679963, 664.8059422529049, 720.4490775314143, 974.141798952963, 942.7041016560629, 842.2577021149319, 606.1770331794359, 903.153902066645, 676.742168984838, 643.8828652183265, 660.3926670212339, 644.5499699814193, 4653.816042849514, 3154.1401930445513, 3125.225919181614, 1724.448899441189, 1611.5632641611053, 1505.7988626630972, 625.0301084583061, 620.6791636198602, 586.8818727531847, 566.1115238963187, 453.5505279302942, 415.10049185783134, 397.8160104455112, 383.1848206618584, 365.11553280573236, 315.9233487310469, 290.5770119343772, 285.2320291292644, 257.2270256038299, 252.64754489470212, 225.10592937611563, 216.91894082490506, 213.1463525739916, 212.01655672898698, 190.13303928228171, 190.13303928228171, 185.1558429955126, 179.75472490377857, 175.01611404254288, 171.91516232637898, 1961.7897278672237, 405.02270204470096, 1452.7136211679983, 1690.787471026783, 3613.4285671271614, 402.3118110943664, 1866.7029104259093, 1236.739673135719, 285.2063666863865, 489.59456880814696, 751.1729575461501, 1748.4488677799768, 2604.463633767713, 1600.3233233284536, 2599.0219409904184, 6489.88297419838, 4375.716646060898, 1894.792305114353, 3645.8404625445855, 3531.1236958881004, 2339.5447353504464, 2198.163958526311, 2112.868942109282, 1709.9793809561884, 1426.9561415008368, 1164.3707476286163, 1430.3998865656304, 1048.4842587999428, 1148.0823120024547, 1071.4097090741059, 2787.142149407582, 996.7893742327144, 579.6334998750912, 307.1313434737523, 225.31832037009292, 198.96371650756183, 172.39559546394943, 172.04780539996395, 171.299716823957, 166.93536650344743, 158.9007115071696, 146.3994558443668, 144.1876382763204, 95.273185177139, 93.88114567350698, 92.4847155317286, 91.3737272721968, 91.30562757849832, 88.29698759114198, 86.43254566460298, 85.96394584794776, 83.64642254646722, 74.15920556384357, 71.39576367592812, 70.39835517691434, 66.58148495149283, 65.7481877538064, 62.830244685812495, 60.47543456191053, 52.84396783439331, 100.59693473505612, 695.0905718176841, 133.20446815383318, 2136.285957643903, 85.03254252902694, 864.4432646358852, 90.3894738867108, 586.3405132606514, 1097.1449358757966, 178.11211672563712, 482.3261163638976, 350.1221424764489, 1241.605712167007, 279.9430335913604, 273.8727523418811, 161.6663655150695, 513.0493379784019, 307.4231192995493, 183.39845625445014, 309.44332807481607, 191.25718330847104, 284.94437366192375, 197.76946293175462, 195.81325442783537, 227.333197656182, 207.08034875025075, 200.51027399218358, 182.30137914945345, 177.11339215524018, 729.3193368572936, 324.46445092341787, 238.04767586998184, 203.8772046898154, 183.65819809705872, 165.62233874695028, 161.48052667425236, 157.43871315659226, 132.44715889348268, 120.0632680037823, 95.97293224531126, 92.92287489824608, 77.12960122486912, 72.2570332669614, 70.28458277694094, 67.2252311567583, 65.08525193710847, 62.89403468646892, 62.768159947818056, 61.93952080589493, 60.49835429350367, 58.68259162919965, 58.59814269009845, 58.27741058165015, 58.1141559137038, 54.696844383396595, 54.087482716242825, 49.96702646361529, 49.86890052712846, 48.54289824457732, 1018.5664467667563, 351.74649657497037, 1021.3046280060638, 129.30747664617644, 270.5912218484238, 137.02682323100188, 199.99666093687935, 124.65880432290743, 108.93041670121646, 103.18172405167013, 180.44592924286474, 216.94224337251325, 126.85845379015551, 437.83606685196867, 116.33459832404004, 86.89589736810163, 94.91703617278176, 125.3896833590471, 117.56799110734437, 119.81643731950221, 101.04173034367561, 334.71489817043187, 238.58217144969544, 224.6301812110219, 142.26868467347347, 93.6554880217324, 90.2449981979208, 63.833257087296644, 58.116280711917405, 52.15954295555052, 50.17150116358471, 48.474525603277485, 48.21099232721248, 39.727703636013665, 32.04876630619534, 31.38580840330775, 31.26179054624608, 28.25485006198213, 28.19765205291294, 27.640591665858206, 26.821739049957564, 26.284779165131905, 25.143287335384862, 24.8545698984725, 22.749531725998157, 21.213492194256848, 19.88773828843766, 19.04192061217371, 17.86654934855804, 17.270098702209424, 16.784109905068487, 1144.923730217342, 77.15292089475709, 182.5487859778092, 123.02178875393433, 229.48187953542643, 174.27443291597635, 103.9349396585874, 147.9720365448635, 143.27022270895174, 706.5823392029497, 57.14625089864459, 151.6466474661188, 60.67137087283864, 192.14350541307937, 186.47174645002562, 130.00960007929027, 95.58854360710478, 95.80602755414735, 94.85218275493705, 103.91519284549295, 91.8670066062961, 89.26645493773113, 94.61329812448032, 72.85078574406037, 83.83968796962797, 124.89009431998244, 93.44824611498301, 119.5479736304112, 96.97893046612622, 96.83993561775061, 94.20132500568376, 218.43063987374651, 155.8039632466905, 124.90287387186793, 111.70908023008128, 103.93245695526967, 67.04261645385978, 63.174850337074425, 62.653494344616774, 56.03389834861324, 54.12647692546528, 52.886864290227116, 51.98549553403627, 43.3044746582733, 42.67352253514745, 40.097125403728704, 37.693539739445555, 35.13398787566436, 34.16304149294971, 33.701573826886126, 32.87386262833414, 31.426821318633916, 31.11193515247406, 29.23941837659442, 26.274709189389796, 25.909014795894798, 25.69292645691564, 24.95473778473387, 24.75331102786379, 23.461450820506986, 22.307134915478844, 237.82291712626582, 84.88624881245372, 67.34946931716689, 109.35214992949565, 219.42495898486797, 38.69045010019215, 288.872473888417, 209.90548008884616, 98.60767922393508, 53.054107144757715, 175.58182187554797, 175.23685962412824, 53.278327460002494, 73.42755420359545, 70.61208855336805, 56.54514246132514, 57.309063443526014, 55.32921573532597, 353.2018028591285, 349.28121141901784, 194.38479046575574, 104.94671338846734, 99.12872148923778, 68.10255044929559, 57.20428862234912, 46.51105092504741, 35.85331676678137, 34.446083100229416, 33.08811301807523, 32.12122265312249, 31.6970183169946, 31.15660468946814, 27.58249162857499, 25.942476122768333, 22.90988148065789, 22.32466167128049, 21.402301631771078, 19.840091289170005, 19.037170246029522, 17.010959866437467, 16.6914321119937, 15.528263084893107, 15.393468992250725, 14.527577578755821, 14.181829809330708, 14.181829809330708, 13.875302812315192, 13.309653785994964, 30.815140888115756, 150.52195012881802, 72.52033892349009, 48.228254331206955, 328.07378595171826, 91.02005621868479, 36.38970726075913, 26.79386732172824, 24.211231352335957, 53.29445285088388, 25.579236052723633, 26.821543093791362, 26.644350417548676], \"Total\": [89586.0, 54670.0, 50294.0, 92429.0, 45666.0, 53206.0, 41686.0, 56909.0, 21757.0, 21028.0, 29573.0, 57206.0, 24731.0, 23173.0, 23039.0, 19399.0, 18680.0, 23526.0, 17829.0, 17643.0, 11964.0, 32980.0, 8581.0, 23415.0, 9691.0, 14185.0, 13856.0, 16254.0, 13299.0, 13090.0, 54670.70451922055, 50294.32364029233, 41686.74217622408, 29573.66478498177, 23173.226278125057, 23039.982400089906, 19399.40390399974, 18680.364278818834, 17829.87159098168, 17643.392574852107, 14185.609508119038, 13299.998190778537, 13090.11089089783, 12608.679880006446, 11146.853685582659, 10890.599296714132, 10795.153182522938, 10719.145412357015, 10510.983325012237, 8741.67119104609, 8368.112967824285, 8236.199791111205, 8120.493508228507, 7446.6507349117, 6768.383667935831, 6686.612924133701, 6668.622749382451, 6096.4447457097685, 5452.626083608791, 5001.452344926548, 24731.716129622764, 11952.337444748046, 11622.023665018927, 9123.515802609885, 89586.24845765886, 21757.52521819915, 21028.618596225853, 5156.938060374217, 4076.2379663435836, 4006.615993248886, 2185.7459053221287, 2148.17364195824, 2027.475545577671, 1815.1760602972722, 1679.5474087626963, 1494.7432285224832, 982.5388778056694, 871.0336725170691, 607.9516614476144, 599.1122909826537, 582.9006495485256, 560.697613489059, 501.6673777250258, 489.04972942256586, 481.02438796820445, 476.2089340255837, 438.63884372750186, 429.58802056600285, 420.67587528435047, 416.426136902554, 404.83042253843956, 389.677239523675, 368.75162562097063, 347.9303878488972, 342.64838400314494, 5333.172216871256, 3104.9304894671723, 2652.315433818503, 2262.287628168126, 2772.057983965777, 15306.776078592708, 655.4881642799933, 2941.174138731595, 92429.06508020684, 19601.85854435169, 15881.21548417655, 1758.1561585909099, 56909.41232860292, 23526.37557883707, 5346.8938364284995, 57206.23824811243, 13856.144400969775, 53206.37253866756, 32980.544340466804, 11449.854190569058, 16254.285030910749, 23415.20079211568, 17434.919378518833, 14465.462446586922, 89586.24845765886, 12784.308120906535, 15642.664625184794, 45666.3697918345, 12927.13220408721, 13328.258281703682, 13954.041174807984, 13824.313522189683, 17145.53360386046, 2903.469261085047, 1071.030469097633, 760.4097776350806, 680.9557886116784, 634.048643331293, 615.6388616741098, 587.2769427745267, 582.00249106368, 495.3140227732533, 385.43051520355044, 371.2091912373661, 357.25251193543374, 338.2059283024182, 331.7398067847565, 300.6872869241792, 261.5051375226118, 241.2132924542462, 235.4624551034667, 209.02704701732918, 159.31317940336487, 156.88675968268095, 152.99478771723778, 140.10761161912416, 136.78518176215755, 136.7202129141916, 379.36112197736736, 131.61817233490572, 1803.417494919191, 124.74230622490239, 124.52815813283536, 2356.950365226424, 3927.345652977395, 362.4024131686795, 272.0623131174587, 489.9823600981104, 279.0237628411453, 614.4154121979138, 973.6655555988584, 1726.0508487609216, 482.3663791113784, 1028.970320009534, 1319.9424612898454, 963.570661070139, 3796.550115476854, 885.850975032884, 8914.139303414528, 6715.156120608236, 1836.0891030864063, 12759.437411380599, 7120.219844557407, 11004.941285782334, 4994.238890548197, 1009.6327604974225, 6702.878770708111, 92429.06508020684, 8135.012476164119, 2840.216982253047, 3128.2442930900243, 17145.53360386046, 1985.2165336216337, 5492.305274901685, 1826.96937120815, 57206.23824811243, 45666.3697918345, 7093.468228309651, 8472.301450949448, 15385.898924056419, 4097.182936449015, 23415.20079211568, 14857.005833123238, 5935.4678011697115, 15642.664625184794, 14465.462446586922, 5433.823608653399, 56909.41232860292, 17434.919378518833, 32980.544340466804, 11560.740058396796, 14441.447719410471, 12784.308120906535, 9646.946258261323, 9691.178694207045, 6433.234981467271, 5923.227213789563, 5338.8908444661265, 5019.864251147471, 4978.422061577863, 4817.110238230148, 4471.904918833391, 2081.077381595122, 1620.1650512863562, 1492.9043985765668, 1384.5562281567247, 1241.8370252005438, 1228.2098523300747, 1179.6030449402447, 1089.7721881388559, 1077.1458086805924, 1068.16258820338, 1052.457657699824, 790.7457548966569, 730.1938317209374, 687.7087771745314, 670.7348692375368, 618.4358662591295, 618.4358662591295, 579.0920860576871, 463.8898902379175, 462.7999601064371, 418.7718345441159, 396.2636751796216, 2480.594413815093, 926.6896872760366, 11964.852799728971, 739.9119581974685, 3180.58563964903, 3459.1666163834175, 1625.4943617971403, 45666.3697918345, 56909.41232860292, 23415.20079211568, 17434.919378518833, 15306.776078592708, 57206.23824811243, 32980.544340466804, 53206.37253866756, 6447.883118217678, 10228.435115397015, 15642.664625184794, 19601.85854435169, 17145.53360386046, 89586.24845765886, 13881.777274920723, 92429.06508020684, 13954.041174807984, 12927.13220408721, 3230.6574746446586, 2519.6071598766594, 2146.2838132795523, 1755.6043169036143, 1430.2147301403713, 1019.8878687649336, 862.0998126093053, 796.838189200783, 796.7206927057267, 335.41960282736375, 283.57567557529273, 277.6374375621307, 274.1755055584818, 273.9885113002428, 241.5059811244357, 223.1753413615603, 222.311677540837, 200.11916997329686, 196.36652475417526, 179.27046838436732, 173.54678173777717, 167.70259365632654, 164.2867568252251, 160.0016736891934, 156.21748986042527, 727.2301139809889, 149.75998838621894, 143.18170754718395, 140.87054176509142, 139.77672550986077, 1021.1916136014768, 387.4505872619886, 489.2759867275878, 823.4691755445612, 716.1615962379996, 237.75078716172717, 837.3555506423377, 286.3476918960979, 2878.3112640225513, 2233.4614670030774, 2880.0402243090766, 332.81365981697485, 4882.678762564142, 7619.1086115870285, 4139.062397558371, 808.7862906859103, 1219.8078600534686, 542.1425609199459, 754.4700861647544, 628.4227313080463, 2577.6489172602523, 7995.680127936346, 2084.586487444678, 5552.973442755683, 2114.6849889274395, 4744.253169580861, 3324.911972762232, 8625.89287145514, 1591.2590202087395, 4787.11144224769, 4215.687510007149, 3165.363811787038, 3333.7696116437273, 3127.782968089956, 3759.690199705776, 2877.116387565136, 2362.925099525918, 2695.422691956986, 2826.6809419109477, 2906.71020973202, 6258.568387090535, 3048.479320699452, 3536.5703925655102, 2937.4453921757577, 1013.2874891002874, 557.6519069924673, 546.1201437603147, 407.9750837218582, 376.1427249670901, 348.2245501374109, 270.1391271174821, 264.6175214857197, 256.04744865776564, 245.9260315040911, 242.7023399127569, 203.5704607138522, 189.15732466578487, 187.13776256600275, 185.10695608424487, 181.34851751910773, 172.72639627070782, 167.2105117633815, 166.78621976764657, 161.19073059126032, 155.04096916839853, 154.9815742174472, 147.6447067510187, 143.645855049737, 141.8313617137449, 129.86091858375545, 128.06013884482195, 125.42236864850639, 123.6767602500113, 635.0366485786727, 818.4522330987472, 424.6253219814778, 765.024529742944, 1533.5031013773557, 418.40804482434095, 420.91913696684327, 339.9631642734209, 353.5435502412632, 1393.6625844919508, 5293.517740449377, 313.57564846480216, 13582.13844470717, 13573.335217387545, 53206.37253866756, 9089.254836008766, 13328.258281703682, 547.0854280881324, 7825.289237001945, 1550.65921003648, 23526.37557883707, 1247.1924401701688, 9802.907019864097, 13856.144400969775, 12927.13220408721, 56909.41232860292, 14441.447719410471, 4149.014164052809, 16254.285030910749, 5216.304654683258, 11321.552602794509, 5994.806113117107, 7871.390096732942, 13881.777274920723, 13824.313522189683, 57206.23824811243, 8371.958742663493, 8556.028724456675, 8978.061353123827, 32980.544340466804, 92429.06508020684, 10656.856831637499, 9646.946258261323, 11560.740058396796, 14857.005833123238, 15385.898924056419, 89586.24845765886, 15881.21548417655, 14465.462446586922, 2603.641942553734, 2556.64717314624, 1392.578519077939, 987.1325560017775, 774.0508049419349, 677.6478594645582, 494.5086953169698, 427.2916733813026, 367.90295931205003, 357.3378957171743, 328.0225582729711, 306.6268613715519, 288.8557623462781, 260.8492842233558, 258.6968698395651, 244.43212779796795, 243.70742178993913, 239.81192923715867, 233.39858286651003, 233.25496246128773, 218.8060241162076, 213.4797957649507, 199.39143613151109, 196.38492289300038, 188.60499805960765, 187.3874279601202, 181.77678469131985, 177.34001716199077, 176.6268859887399, 171.62511218497775, 1143.0102814273705, 322.92085755720603, 2110.1500190455286, 1245.3827013517696, 628.5778355505851, 836.4287750243581, 498.3135018006653, 441.25805414379784, 989.0813447335976, 547.2568759254995, 438.14345768757767, 420.5792348466079, 665.845084386955, 1202.9674213759713, 1655.7242494742588, 664.4887159450278, 644.1918492605629, 1299.4527567948655, 2806.158656309635, 643.7770642172763, 4905.946304449013, 779.497551227817, 1470.770932402823, 1400.0866951234045, 747.8399566515883, 1013.8285135273994, 1865.580609702534, 4215.687510007149, 1992.5243982076054, 1202.8609046793003, 2679.8182726756454, 1249.6929186950072, 1121.5896052719522, 5221.662810139358, 1813.8633866596901, 3953.580045777035, 1850.9541551375437, 7760.51652605563, 8472.301450949448, 5346.8938364284995, 1690.3771179913672, 8581.897248912313, 4798.215298072615, 3644.663039743455, 3290.9613638823885, 3138.0818450630363, 2981.6090347527643, 2943.99342119729, 2557.543510501663, 2313.521417314951, 1639.212166887209, 1533.777933041117, 1514.2774486441117, 1264.5400593629217, 1146.2484528807431, 905.6123983131406, 899.9964393110615, 896.5134770975737, 894.9415125430739, 870.0360085519112, 769.4176756093523, 758.4682172896479, 715.4449042255404, 715.4449042255404, 712.3119677505463, 697.6480301975793, 607.7629190401705, 579.2694862827766, 534.5195390775306, 518.0402872189238, 508.6063884341835, 1335.1957640773394, 1188.9950021587192, 4674.5200134790475, 2297.179257327315, 2050.551156669382, 3159.074245942528, 2496.1812053309695, 4922.857091926741, 45666.3697918345, 1800.9023816535462, 57206.23824811243, 32980.544340466804, 23415.20079211568, 17145.53360386046, 56909.41232860292, 2866.830803595249, 13824.313522189683, 19601.85854435169, 16254.285030910749, 13954.041174807984, 23526.37557883707, 6447.883118217678, 92429.06508020684, 53206.37253866756, 11321.552602794509, 89586.24845765886, 15306.776078592708, 881.8513775232747, 477.52202171502597, 409.808808507695, 315.8511965391305, 303.0892149900507, 284.3130690246149, 265.86712216302755, 257.6819874050196, 254.19635540912486, 244.28424177547197, 240.03018363740094, 229.35485629539306, 228.21437044455186, 226.41049702889305, 217.17927796919267, 210.2784231217039, 200.39453379384847, 189.32980173125748, 175.4362753761178, 174.25716899657442, 172.52158775332222, 163.26344751202316, 155.59202659536902, 153.45866977320716, 150.696257736933, 149.04929206422725, 136.5981715605107, 134.21103484276153, 118.52001411690802, 113.95172885858473, 319.71266137628106, 315.76676841220683, 887.9808005872511, 168.68279356963197, 538.6732748564646, 292.58269438763176, 874.1972717820271, 328.55746897833285, 350.42517820964275, 3043.6999743321294, 466.15205968714884, 4441.805343716088, 301.57508114910314, 361.9287096630204, 3723.5469261651424, 1129.5779827912795, 496.3182997598914, 1305.6983073491267, 563.526126945631, 444.5258204072029, 699.2159120013492, 89586.24845765886, 714.8620285824178, 776.2320690034147, 2163.098864480988, 1173.675963581069, 3147.338038289423, 2822.5690556434192, 1717.2812010629264, 2569.509763323623, 1036.5367207384813, 1615.6462620775133, 1866.2164174445777, 1977.577783494745, 671.1747803313715, 575.0717407526861, 4495.7406467229, 901.4974651945518, 939.0399912055743, 2917.4999157553057, 1610.1079078623327, 2068.921314870103, 6258.568387090535, 7760.51652605563, 5017.221146956101, 1392.9756073200253, 17434.919378518833, 3953.580045777035, 2744.459088692982, 6447.883118217678, 5221.662810139358, 4654.7509944806525, 3155.0751446756913, 3126.160871472697, 1725.3838510899168, 1612.4982157964346, 1506.7338142858578, 625.9650600810664, 621.6141152426205, 587.816824375945, 567.0464755316477, 454.48547955305463, 416.03544348059177, 398.75096206827163, 384.11977229718764, 366.05048468706036, 316.85830133633243, 291.5119636356102, 286.1669807520248, 258.1619772265903, 253.58249675677698, 226.04088106149786, 217.85389246023422, 214.08130420932076, 212.95150835174735, 191.06799090504208, 191.06799090504208, 186.09079461827295, 180.68967889647507, 175.9510731837548, 172.85011394913934, 1986.763479843264, 408.06455251847854, 1527.7329184588295, 1790.6123328519664, 4065.969940322315, 469.0830204167794, 3367.910698731935, 2068.6474258014164, 313.2384421991136, 647.4997814186946, 1166.9746733278037, 4922.857091926741, 13856.144400969775, 6447.883118217678, 16254.285030910749, 92429.06508020684, 57206.23824811243, 11321.552602794509, 53206.37253866756, 56909.41232860292, 23415.20079211568, 23526.37557883707, 32980.544340466804, 19601.85854435169, 13954.041174807984, 6991.490370201833, 15881.21548417655, 8135.012476164119, 15642.664625184794, 17434.919378518833, 2788.1015972980185, 997.7488221092009, 580.5929565984377, 308.0907915255108, 226.2777688450211, 199.9231643840484, 173.3550435817369, 173.00725540742496, 172.2591649656371, 167.8948240589396, 159.86015946388468, 147.3589123558382, 145.14708633799367, 96.23263329472559, 94.84059367779919, 93.44416348960968, 92.33319075080843, 92.26507569270923, 89.25644427417186, 87.39199477940375, 86.9234032581231, 84.60587424781879, 75.11866516794707, 72.35521375377803, 71.35781660838084, 67.54093556352412, 66.70763834239374, 63.78969305782923, 61.434884715129954, 53.80341819868925, 103.82520299366675, 816.0604821056311, 145.76247542145313, 3917.224185591401, 90.20688621478597, 1598.4523785372394, 97.8345966321595, 1133.9186869178286, 2760.3323218361497, 248.3654795742521, 1031.0860671752437, 683.5692686520109, 4905.271507533255, 539.1673885385902, 531.9624152026054, 233.9256693528443, 1593.2025793407504, 776.6482693920063, 313.89249137001815, 1461.0423694208782, 473.556114962715, 3594.6534537344614, 723.1638460178194, 940.4221619651211, 4097.182936449015, 2746.0630944330273, 5216.304654683258, 3032.518114104487, 1113.5736258260024, 730.2935074154507, 325.4386229062959, 239.02185075573223, 204.8513796074873, 184.63237393602478, 166.59650956849572, 162.45470385132342, 158.41288792269418, 133.42133324360785, 121.03743844133493, 96.94710464153565, 93.89704593813988, 78.10377262189559, 73.23120664390352, 71.25875641220745, 68.19940452404671, 66.05942302105997, 63.86820703650931, 63.742334679737155, 62.91369470788676, 61.47252644389754, 59.656762066752265, 59.57231680036197, 59.25158465885424, 59.08832633569227, 55.67101716650474, 55.06165609859963, 50.94119928730492, 50.84307782130413, 49.517068795187384, 1086.1772322302877, 368.91349559519387, 1122.579950611013, 140.16411827603844, 315.78190010539754, 151.17291655354262, 230.84957100424907, 142.36924394888322, 126.32701034899759, 120.30479397365096, 251.0270399541974, 363.27806672553083, 184.35643796649015, 1398.5945530107144, 176.8472680084299, 108.4867363879067, 135.11628803548277, 272.60067881332947, 288.5926592587843, 708.4312829848602, 1381.8471231498577, 335.688765765584, 239.55603912304798, 225.604048806174, 143.24256149262607, 94.62935577525073, 91.21887923225334, 64.80712468244876, 59.09014829835303, 53.13341077823298, 51.14537163502477, 49.44839435684326, 49.184859926851786, 40.701613730581904, 33.02263475647488, 32.359679079458886, 32.23566328036516, 29.22871970988204, 29.17151964806505, 28.614459377164554, 27.795610492668498, 27.258647178939444, 26.11715500059501, 25.828437623328917, 23.723400028563148, 22.18736036156862, 20.861607373928074, 20.01578822150766, 18.840423602726254, 18.243972973413324, 17.757981082028138, 1225.252355410482, 94.04687839721429, 301.43133829577096, 207.0572619349052, 481.2875176086554, 337.58029317330016, 181.6247156749006, 334.94802429928455, 351.3896227857586, 5017.221146956101, 93.06499129162512, 485.68916033295534, 105.84145002346243, 1096.6530939600361, 1656.455991502676, 1015.7658199292283, 535.2750600293334, 673.959256094259, 660.8670610037481, 974.116615290346, 634.4282800338276, 673.1197032904369, 1013.0040701371713, 305.0418609820342, 746.5408189025055, 23415.20079211568, 2496.1812053309695, 89586.24845765886, 4994.238890548197, 17434.919378518833, 7120.219844557407, 219.40679636978132, 156.7801197427253, 125.8790419884909, 112.68523672581665, 104.90861345130448, 68.01877311231154, 64.15100688606645, 63.62965080903043, 57.01005564206336, 55.10263340624344, 53.86302078626192, 52.96165203007107, 44.28067706240445, 43.649679027365316, 41.07329125502006, 38.669696687748896, 36.11015398127834, 35.13919810318729, 34.67773066796414, 33.85002145388638, 32.40297836166107, 32.08809168181151, 30.21557838047812, 27.25086602397333, 26.88517126738284, 26.669087453938268, 25.930895628923295, 25.72946753767591, 24.437607268556135, 23.28330333563882, 251.61127779920474, 92.69395401208054, 73.96314676998962, 125.70214086485478, 268.31571477958653, 41.697205505333045, 385.2122609792857, 400.3394276135672, 150.28625916250104, 67.62957985986323, 565.9142144983715, 1233.299332209214, 103.92007787102447, 255.24643755794034, 432.61180736302884, 204.46703319673375, 471.2673976395594, 5333.172216871256, 354.19025177350113, 350.2696602936007, 195.373239335123, 105.93517030042567, 100.11717158428733, 69.09100181840232, 58.19273847012553, 47.49949979441467, 36.84176822727732, 35.434535531957486, 34.07658135319481, 33.10970252467205, 32.68546718636186, 32.14505359156441, 28.57095344573866, 26.930926549824605, 23.898333534403516, 23.31312494515332, 22.39076838595146, 20.828540191266274, 20.02562161631466, 17.999408768533737, 17.679880981360956, 16.516711954260366, 16.3819193048548, 15.5160406757947, 15.17027871142698, 15.17027871142698, 14.863759672757126, 14.298113190675869, 34.123469967846425, 216.3909666650477, 99.80481339854506, 77.31971688232463, 45666.3697918345, 2772.057983965777, 262.1652445126995, 182.11867445742465, 117.9984367084958, 17434.919378518833, 217.54561422673814, 1758.1561585909099, 2205.7355675558183], \"Category\": [\"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic9\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic10\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic11\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic12\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic13\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic14\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\", \"Topic15\"], \"logprob\": [30.0, 29.0, 28.0, 27.0, 26.0, 25.0, 24.0, 23.0, 22.0, 21.0, 20.0, 19.0, 18.0, 17.0, 16.0, 15.0, 14.0, 13.0, 12.0, 11.0, 10.0, 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, -2.8572, -2.9406, -3.1283, -3.4716, -3.7155, -3.7213, -3.8933, -3.9311, -3.9777, -3.9882, -4.2063, -4.2708, -4.2867, -4.3242, -4.4474, -4.4707, -4.4795, -4.4865, -4.5062, -4.6905, -4.7342, -4.7501, -4.7642, -4.8508, -4.9464, -4.9585, -4.9612, -5.0509, -5.1626, -5.2489, -3.6545, -4.3883, -4.4486, -4.6763, -3.0241, -3.1567, -3.1908, -4.5965, -4.8317, -4.8489, -5.4551, -5.4725, -5.5303, -5.641, -5.7187, -5.8353, -6.2552, -6.3758, -6.7359, -6.7505, -6.778, -6.8169, -6.9283, -6.9539, -6.9704, -6.9805, -7.0629, -7.0838, -7.1048, -7.115, -7.1433, -7.1815, -7.2368, -7.2951, -7.3104, -4.7045, -5.2661, -5.4224, -5.6396, -5.5335, -4.1529, -6.7437, -5.5289, -2.7734, -4.0433, -4.3112, -6.0095, -3.588, -4.2083, -5.2722, -3.6904, -4.6598, -3.7773, -4.0985, -4.8186, -4.6939, -4.6787, -4.8054, -4.9125, -4.2403, -5.0115, -4.9627, -4.7355, -5.1409, -5.1827, -5.1769, -5.2361, -5.2374, -5.0881, -6.086, -6.4289, -6.5394, -6.6109, -6.6404, -6.6876, -6.6967, -6.8582, -7.1096, -7.1473, -7.1858, -7.2407, -7.2601, -7.3586, -7.4987, -7.5798, -7.6041, -7.7237, -7.9967, -8.0122, -8.0374, -8.126, -8.1502, -8.1507, -7.1304, -8.189, -5.5714, -8.243, -8.2448, -5.3097, -4.8131, -7.188, -7.4753, -6.9012, -7.456, -6.6919, -6.2575, -5.7194, -6.9383, -6.2299, -6.0011, -6.3149, -5.0782, -6.397, -4.3256, -4.5948, -5.7527, -4.0987, -4.6183, -4.2495, -4.9423, -6.3131, -4.7623, -2.6344, -4.6124, -5.4814, -5.4173, -4.2135, -5.8079, -5.1081, -5.9, -3.5674, -3.7634, -4.9982, -4.9017, -4.5324, -5.3898, -4.3595, -4.6779, -5.196, -4.7026, -4.7493, -5.2681, -4.2962, -4.8412, -4.6294, -5.0061, -5.0132, -5.0764, -5.1513, -3.7512, -4.161, -4.2436, -4.3475, -4.4091, -4.4174, -4.4504, -4.5248, -5.2899, -5.5404, -5.6223, -5.6977, -5.8065, -5.8176, -5.858, -5.9373, -5.9489, -5.9573, -5.9721, -6.2583, -6.3381, -6.3981, -6.4231, -6.5044, -6.5044, -6.5703, -6.7925, -6.7949, -6.895, -6.9504, -5.1198, -6.104, -3.5974, -6.3418, -5.0503, -5.1566, -5.7981, -3.5726, -3.6165, -4.1809, -4.3828, -4.4779, -3.7407, -4.0786, -3.8215, -5.1952, -5.0549, -4.9031, -4.8684, -4.9242, -4.3869, -5.049, -4.5792, -5.1103, -5.1833, -4.733, -4.9817, -5.1421, -5.3431, -5.5482, -5.8867, -6.0549, -6.1337, -6.1339, -7.0006, -7.1691, -7.1903, -7.2029, -7.2036, -7.3303, -7.4095, -7.4134, -7.5191, -7.5381, -7.6296, -7.6623, -7.6967, -7.7174, -7.744, -7.7681, -6.2302, -7.8106, -7.8558, -7.8722, -7.88, -5.903, -6.8688, -6.6395, -6.1508, -6.2974, -7.3614, -6.1572, -7.1874, -5.0158, -5.2659, -5.0413, -7.052, -4.6175, -4.2459, -4.8225, -6.2692, -5.918, -6.6266, -6.3522, -6.5135, -5.348, -4.4223, -5.5281, -4.7752, -5.5557, -4.9411, -5.2344, -4.532, -5.7898, -5.0586, -5.1593, -5.3512, -5.3292, -5.3771, -5.277, -5.4331, -5.5483, -5.4796, -5.4569, -5.4625, -5.2379, -5.4848, -5.4605, -4.6486, -5.7136, -6.3115, -6.3325, -6.6247, -6.7061, -6.7834, -7.0381, -7.0589, -7.0919, -7.1324, -7.1456, -7.3222, -7.396, -7.4068, -7.4178, -7.4384, -7.4873, -7.52, -7.5225, -7.5569, -7.596, -7.5964, -7.6452, -7.6728, -7.6856, -7.7744, -7.7885, -7.8094, -7.8236, -6.1916, -5.9408, -6.5959, -6.0272, -5.348, -6.6373, -6.6501, -6.8579, -6.8227, -5.5928, -4.3827, -6.9652, -3.6744, -3.6753, -2.501, -4.0629, -3.7633, -6.4894, -4.2367, -5.6389, -3.4784, -5.8351, -4.2209, -3.9521, -4.0412, -2.9422, -4.0406, -4.9737, -4.0102, -4.8529, -4.3605, -4.8106, -4.692, -4.3468, -4.3496, -3.5223, -4.6826, -4.6725, -4.6533, -3.9805, -3.4536, -4.593, -4.6832, -4.6195, -4.5673, -4.5599, -4.2, -4.6353, -4.6737, -4.7082, -4.7264, -5.3343, -5.6787, -5.9221, -6.0553, -6.3709, -6.5173, -6.6673, -6.6965, -6.7824, -6.85, -6.9099, -7.0123, -7.0206, -7.0775, -7.0805, -7.0967, -7.1239, -7.1245, -7.1887, -7.2135, -7.2821, -7.2973, -7.338, -7.3445, -7.375, -7.3999, -7.4039, -7.4328, -5.5517, -6.8049, -4.9489, -5.4775, -6.1486, -5.8693, -6.3781, -6.498, -5.7269, -6.3063, -6.521, -6.561, -6.129, -5.5877, -5.2918, -6.1505, -6.1893, -5.5806, -4.9093, -6.2242, -4.5273, -6.0669, -5.5559, -5.6121, -6.1128, -5.8859, -5.4681, -5.0357, -5.5595, -5.8597, -5.411, -5.8536, -5.9303, -5.3285, -5.768, -5.5293, -5.8036, -5.3965, -5.5948, -5.7263, -5.8802, -3.513, -4.0945, -4.3696, -4.4717, -4.5193, -4.5704, -4.5831, -4.7239, -4.8242, -5.1689, -5.2355, -5.2483, -5.4286, -5.5269, -5.7628, -5.769, -5.7729, -5.7746, -5.8029, -5.9259, -5.9403, -5.9987, -5.9987, -6.0031, -6.024, -6.1621, -6.2102, -6.2907, -6.3221, -6.3405, -5.4084, -5.5245, -4.215, -5.0024, -5.1095, -4.7543, -5.0407, -4.531, -2.8421, -5.3386, -3.6461, -4.0784, -4.3065, -4.4698, -3.9454, -5.2611, -4.5886, -4.4569, -4.5728, -4.8787, -4.8236, -5.0925, -4.6715, -4.8036, -5.0736, -5.0204, -5.0895, -5.5748, -6.1891, -6.3423, -6.6035, -6.6448, -6.709, -6.7763, -6.8077, -6.8214, -6.8613, -6.8789, -6.9246, -6.9296, -6.9376, -6.9794, -7.0118, -7.0602, -7.1173, -7.1939, -7.2007, -7.2107, -7.2662, -7.3146, -7.3285, -7.3468, -7.3578, -7.4457, -7.4634, -7.5887, -7.6283, -6.5977, -6.6152, -5.6255, -7.2436, -6.1443, -6.7248, -5.7133, -6.6234, -6.5697, -4.6185, -6.3202, -4.3449, -6.7144, -6.5585, -4.5586, -5.614, -6.2994, -5.5505, -6.2199, -6.4085, -6.0635, -2.3463, -6.0484, -5.9896, -5.2503, -5.7016, -4.9989, -5.1238, -5.4816, -5.2183, -5.8477, -5.6157, -5.5418, -5.5441, -6.1736, -6.2627, -5.1632, -6.0201, -6.0032, -5.606, -5.8562, -5.7758, -5.4741, -5.507, -5.6196, -5.9485, -5.5498, -5.8384, -5.8882, -5.8629, -5.8872, -3.6534, -4.0423, -4.0515, -4.6461, -4.7138, -4.7817, -5.661, -5.668, -5.724, -5.76, -5.9817, -6.0703, -6.1128, -6.1503, -6.1986, -6.3433, -6.4269, -6.4455, -6.5488, -6.5668, -6.6822, -6.7193, -6.7368, -6.7421, -6.8511, -6.8511, -6.8776, -6.9072, -6.9339, -6.9518, -4.5172, -6.0949, -4.8176, -4.6659, -3.9064, -6.1016, -4.5669, -4.9786, -6.4456, -5.9052, -5.4772, -4.6323, -4.2338, -4.7208, -4.2359, -3.3208, -3.715, -4.5519, -3.8975, -3.9294, -4.3411, -4.4034, -4.443, -4.6546, -4.8355, -5.0389, -4.8331, -5.1437, -5.053, -5.1221, -2.8484, -3.8766, -4.4188, -5.0539, -5.3637, -5.488, -5.6314, -5.6334, -5.6378, -5.6636, -5.7129, -5.7948, -5.8101, -6.2244, -6.2391, -6.2541, -6.2662, -6.267, -6.3005, -6.3218, -6.3272, -6.3546, -6.475, -6.5129, -6.527, -6.5827, -6.5953, -6.6407, -6.6789, -6.8138, -6.17, -4.2371, -5.8893, -3.1143, -6.3381, -4.0191, -6.277, -4.4073, -3.7807, -5.5988, -4.6025, -4.9229, -3.657, -5.1466, -5.1685, -5.6956, -4.5408, -5.0529, -5.5695, -5.0464, -5.5275, -5.1289, -5.4941, -5.504, -5.3548, -5.4481, -5.4803, -5.5755, -5.6044, -3.7533, -4.5632, -4.8729, -5.0279, -5.1323, -5.2357, -5.261, -5.2864, -5.4592, -5.5574, -5.7813, -5.8136, -5.9999, -6.0652, -6.0928, -6.1373, -6.1697, -6.2039, -6.2059, -6.2192, -6.2428, -6.2733, -6.2747, -6.2802, -6.283, -6.3436, -6.3548, -6.434, -6.436, -6.4629, -3.4192, -4.4825, -3.4166, -5.4832, -4.7448, -5.4252, -5.0471, -5.5198, -5.6547, -5.7089, -5.15, -4.9658, -5.5023, -4.2636, -5.5889, -5.8807, -5.7924, -5.514, -5.5784, -5.5594, -5.7299, -4.1577, -4.4962, -4.5565, -5.0132, -5.4313, -5.4684, -5.8147, -5.9085, -6.0166, -6.0555, -6.0899, -6.0954, -6.2889, -6.5037, -6.5246, -6.5285, -6.6297, -6.6317, -6.6517, -6.6817, -6.7019, -6.7463, -6.7579, -6.8464, -6.9163, -6.9808, -7.0243, -7.088, -7.122, -7.1505, -2.9278, -5.6251, -4.7639, -5.1586, -4.5351, -4.8103, -5.3272, -4.9739, -5.0062, -3.4105, -5.9253, -4.9494, -5.8655, -4.7127, -4.7427, -5.1033, -5.4109, -5.4086, -5.4186, -5.3274, -5.4506, -5.4793, -5.4211, -5.6825, -5.542, -5.1435, -5.4335, -5.1872, -5.3964, -5.3979, -5.4255, -4.4351, -4.7729, -4.994, -5.1056, -5.1778, -5.6162, -5.6756, -5.6839, -5.7956, -5.8302, -5.8534, -5.8706, -6.0533, -6.0679, -6.1302, -6.192, -6.2624, -6.2904, -6.304, -6.3288, -6.3739, -6.3839, -6.446, -6.5529, -6.5669, -6.5753, -6.6045, -6.6126, -6.6662, -6.7166, -4.35, -5.3802, -5.6116, -5.1269, -4.4305, -6.1659, -4.1555, -4.4749, -5.2304, -5.8502, -4.6534, -4.6554, -5.846, -5.5252, -5.5643, -5.7865, -5.7731, -5.8082, -3.5029, -3.5141, -4.1001, -4.7165, -4.7735, -5.1489, -5.3233, -5.5303, -5.7905, -5.8306, -5.8708, -5.9004, -5.9137, -5.9309, -6.0528, -6.1141, -6.2384, -6.2643, -6.3065, -6.3823, -6.4236, -6.5361, -6.5551, -6.6273, -6.636, -6.6939, -6.718, -6.718, -6.7398, -6.7815, -5.942, -4.3559, -5.0861, -5.494, -3.5767, -4.8589, -5.7757, -6.0818, -6.1831, -5.3941, -6.1282, -6.0808, -6.0874], \"loglift\": [30.0, 29.0, 28.0, 27.0, 26.0, 25.0, 24.0, 23.0, 22.0, 21.0, 20.0, 19.0, 18.0, 17.0, 16.0, 15.0, 14.0, 13.0, 12.0, 11.0, 10.0, 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, 1.4697, 1.4697, 1.4697, 1.4697, 1.4697, 1.4697, 1.4697, 1.4697, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4696, 1.4695, 1.4695, 1.4695, 1.4656, 1.4589, 1.4267, 1.4411, 0.8089, 2.0915, 2.0915, 2.0913, 2.0913, 2.0913, 2.0911, 2.0911, 2.0911, 2.091, 2.091, 2.0909, 2.0906, 2.0905, 2.09, 2.09, 2.0899, 2.0899, 2.0897, 2.0896, 2.0896, 2.0896, 2.0894, 2.0893, 2.0893, 2.0893, 2.0892, 2.0891, 2.089, 2.0888, 2.0888, 1.9497, 1.9291, 1.9304, 1.8722, 1.7751, 1.447, 2.0068, 1.7205, 1.0283, 1.3093, 1.2519, 1.7544, 0.6987, 0.9618, 1.3795, 0.5911, 1.0397, 0.5767, 0.7338, 1.0716, 0.8459, 0.4961, 0.6644, 0.7439, -0.4073, 0.7685, 0.6155, -0.2287, 0.6279, 0.5556, 0.5155, 0.4657, 0.2491, 2.1741, 2.1736, 2.1732, 2.173, 2.1729, 2.1729, 2.1728, 2.1728, 2.1725, 2.1719, 2.1718, 2.1717, 2.1716, 2.1715, 2.1712, 2.1707, 2.1704, 2.1703, 2.1698, 2.1684, 2.1683, 2.1681, 2.1675, 2.1673, 2.1673, 2.1671, 2.1671, 2.1671, 2.1667, 2.1666, 2.1611, 2.1471, 2.1552, 2.1546, 2.1404, 2.1487, 2.1234, 2.0974, 2.0629, 2.1189, 2.0697, 2.0495, 2.0504, 1.9159, 2.0524, 1.8149, 1.829, 1.9679, 1.6832, 1.7469, 1.6803, 1.7776, 2.0055, 1.6634, 1.1673, 1.6196, 1.8029, 1.7704, 1.2729, 1.8346, 1.5168, 1.8255, 0.7142, 0.7434, 1.3708, 1.2897, 1.0624, 1.5281, 0.8153, 0.9519, 1.3512, 0.8756, 0.9071, 1.3674, -0.0094, 0.6285, 0.2028, 0.8745, 0.6449, 0.7035, 0.9102, 2.3057, 2.3057, 2.3057, 2.3056, 2.3056, 2.3056, 2.3056, 2.3056, 2.3054, 2.3052, 2.3052, 2.3051, 2.3051, 2.3051, 2.305, 2.305, 2.305, 2.3049, 2.3049, 2.3046, 2.3045, 2.3045, 2.3044, 2.3043, 2.3043, 2.3042, 2.3038, 2.3038, 2.3036, 2.3035, 2.2999, 2.3004, 2.2488, 2.2876, 2.1208, 1.9305, 2.0443, 0.9342, 0.6702, 0.9939, 1.0869, 1.122, 0.5408, 0.7537, 0.5325, 1.2692, 0.9481, 0.675, 0.4842, 0.5623, -0.5539, 0.6486, -0.7775, 0.5821, 0.5855, 2.4225, 2.4224, 2.4223, 2.4222, 2.4221, 2.4218, 2.4217, 2.4216, 2.4216, 2.4199, 2.4194, 2.4193, 2.4193, 2.4193, 2.4188, 2.4185, 2.4184, 2.418, 2.4179, 2.4174, 2.4172, 2.417, 2.4169, 2.4167, 2.4166, 2.4165, 2.4163, 2.416, 2.4159, 2.4159, 2.4042, 2.4075, 2.4036, 2.3716, 2.3647, 2.4033, 2.3485, 2.3913, 2.2551, 2.2587, 2.2291, 2.3763, 2.125, 2.0516, 2.0852, 2.2712, 2.2115, 2.3138, 2.2577, 2.2792, 2.0333, 1.827, 2.0655, 1.8386, 2.0236, 1.8302, 1.8923, 1.6414, 2.0739, 1.7037, 1.73, 1.8247, 1.7949, 1.8107, 1.7268, 1.8383, 1.92, 1.857, 1.8322, 1.7986, 1.2564, 1.7287, 1.6046, 2.602, 2.6014, 2.6007, 2.6006, 2.6, 2.5999, 2.5996, 2.5989, 2.5988, 2.5987, 2.5985, 2.5985, 2.5977, 2.5974, 2.5973, 2.5972, 2.5971, 2.5969, 2.5967, 2.5967, 2.5965, 2.5963, 2.5963, 2.5959, 2.5958, 2.5957, 2.5951, 2.595, 2.5948, 2.5947, 2.5906, 2.5877, 2.5888, 2.5688, 2.5526, 2.5622, 2.5434, 2.5492, 2.5452, 2.4035, 2.279, 2.5227, 2.045, 2.0448, 1.853, 2.0582, 1.975, 2.4419, 2.0341, 2.2506, 1.6917, 2.2722, 1.8246, 1.7474, 1.7276, 1.3445, 1.6175, 1.9316, 1.5296, 1.8235, 1.541, 1.7267, 1.5729, 1.3508, 1.3522, 0.7593, 1.5206, 1.509, 1.4801, 0.8518, 0.3481, 1.369, 1.3783, 1.2611, 1.0624, 1.0348, -0.367, 0.9277, 0.9827, 2.6631, 2.6631, 2.6627, 2.6625, 2.6622, 2.662, 2.6615, 2.6612, 2.6608, 2.6607, 2.6605, 2.6603, 2.6601, 2.6597, 2.6597, 2.6595, 2.6595, 2.6594, 2.6593, 2.6593, 2.659, 2.6589, 2.6586, 2.6585, 2.6583, 2.6583, 2.6581, 2.658, 2.658, 2.6578, 2.6428, 2.6536, 2.6325, 2.6313, 2.6439, 2.6375, 2.6466, 2.6483, 2.6123, 2.6247, 2.6324, 2.6333, 2.6059, 2.5557, 2.5322, 2.5864, 2.5786, 2.4856, 2.387, 2.5444, 2.2104, 2.5104, 2.3865, 2.3796, 2.5059, 2.4285, 2.2365, 1.8536, 2.0793, 2.2838, 1.9314, 2.2517, 2.2831, 1.3468, 1.9647, 1.4243, 1.9089, 0.8826, 0.5966, 0.9253, 1.923, 2.6655, 2.6654, 2.6654, 2.6653, 2.6653, 2.6653, 2.6653, 2.6652, 2.6652, 2.665, 2.665, 2.665, 2.6649, 2.6648, 2.6646, 2.6646, 2.6646, 2.6646, 2.6645, 2.6644, 2.6644, 2.6643, 2.6643, 2.6643, 2.6643, 2.6641, 2.664, 2.6638, 2.6638, 2.6638, 2.6307, 2.6306, 2.5711, 2.4941, 2.5006, 2.4236, 2.3728, 2.2033, 1.6648, 2.4013, 0.6355, 0.7539, 0.8683, 1.0167, 0.3414, 2.0139, 1.1131, 0.8957, 0.967, 0.8137, 0.3465, 1.3719, -0.8697, -0.4496, 0.8279, -1.1874, 0.5104, 2.8792, 2.8782, 2.8779, 2.8772, 2.8771, 2.8769, 2.8766, 2.8765, 2.8765, 2.8763, 2.8763, 2.8761, 2.876, 2.876, 2.8758, 2.8757, 2.8755, 2.8752, 2.8748, 2.8747, 2.8747, 2.8744, 2.8741, 2.874, 2.8739, 2.8738, 2.8732, 2.8731, 2.8722, 2.8718, 2.8708, 2.8658, 2.8215, 2.8643, 2.8025, 2.8324, 2.7493, 2.8179, 2.8071, 2.5966, 2.7712, 2.4923, 2.8125, 2.786, 2.4549, 2.5924, 2.7293, 2.5109, 2.6818, 2.7304, 2.6225, 1.4867, 2.6154, 2.5919, 2.3063, 2.4664, 2.1827, 2.1668, 2.3059, 2.1662, 2.4446, 2.2328, 2.1625, 2.1022, 2.5533, 2.6187, 1.6618, 2.4118, 2.3878, 1.6515, 1.9957, 1.8253, 1.0201, 0.7722, 1.0957, 2.0482, -0.0801, 1.1151, 1.4304, 0.6016, 0.7882, 3.1369, 3.1368, 3.1368, 3.1366, 3.1366, 3.1365, 3.1356, 3.1356, 3.1355, 3.1355, 3.1351, 3.1349, 3.1348, 3.1347, 3.1346, 3.1342, 3.1339, 3.1339, 3.1335, 3.1334, 3.133, 3.1328, 3.1328, 3.1327, 3.1322, 3.1322, 3.1321, 3.1319, 3.1318, 3.1317, 3.1245, 3.1297, 3.0868, 3.0798, 3.0191, 2.9836, 2.547, 2.6227, 3.0434, 2.8576, 2.6966, 2.102, 1.4656, 1.7436, 1.3039, 0.4809, 0.5665, 1.3495, 0.4565, 0.3573, 0.8337, 0.7666, 0.3893, 0.698, 0.8569, 1.3446, 0.73, 1.0883, 0.5252, 0.3476, 4.4544, 4.4538, 4.4531, 4.4517, 4.4505, 4.45, 4.4492, 4.4492, 4.4492, 4.449, 4.4488, 4.4482, 4.4481, 4.4448, 4.4446, 4.4445, 4.4443, 4.4443, 4.444, 4.4437, 4.4437, 4.4434, 4.4419, 4.4414, 4.4412, 4.4405, 4.4403, 4.4396, 4.439, 4.4368, 4.4232, 4.2943, 4.3647, 3.8485, 4.3957, 3.8401, 4.3756, 3.7952, 3.5321, 4.1223, 3.695, 3.7857, 3.0809, 3.7993, 3.7909, 4.0853, 3.3216, 3.528, 3.9174, 2.9026, 3.5481, 1.9199, 3.1582, 2.8856, 1.5631, 1.87, 1.1961, 1.6433, 2.6162, 4.8892, 4.8875, 4.8865, 4.8858, 4.8853, 4.8847, 4.8845, 4.8844, 4.8832, 4.8825, 4.8804, 4.8801, 4.878, 4.8771, 4.8768, 4.8762, 4.8757, 4.8752, 4.8751, 4.8749, 4.8746, 4.8741, 4.8741, 4.874, 4.8739, 4.8729, 4.8727, 4.8712, 4.8712, 4.8707, 4.8263, 4.8429, 4.796, 4.8099, 4.7361, 4.7923, 4.7471, 4.7577, 4.7424, 4.737, 4.5604, 4.375, 4.5167, 3.7292, 4.4717, 4.6686, 4.5374, 4.114, 3.9925, 3.1134, 2.2749, 5.2621, 5.2609, 5.2607, 5.2582, 5.2547, 5.2543, 5.2499, 5.2484, 5.2465, 5.2458, 5.2451, 5.245, 5.2408, 5.2351, 5.2344, 5.2343, 5.2311, 5.231, 5.2304, 5.2293, 5.2286, 5.227, 5.2266, 5.2231, 5.2201, 5.2172, 5.2151, 5.2119, 5.2101, 5.2086, 5.1972, 5.067, 4.7635, 4.7444, 4.5244, 4.6038, 4.7068, 4.448, 4.3678, 3.3048, 4.7773, 4.101, 4.7085, 3.5232, 3.0808, 3.2092, 3.5423, 3.3142, 3.3238, 3.027, 3.3326, 3.2447, 2.8941, 3.833, 3.0785, 0.0313, 1.9799, -1.3542, 1.3235, 0.0718, 0.9397, 5.41, 5.4082, 5.4066, 5.4057, 5.4051, 5.4, 5.3991, 5.399, 5.3971, 5.3965, 5.3961, 5.3958, 5.3921, 5.3918, 5.3904, 5.3889, 5.387, 5.3862, 5.3859, 5.3852, 5.3838, 5.3835, 5.3816, 5.3779, 5.3774, 5.3771, 5.376, 5.3757, 5.3737, 5.3716, 5.3581, 5.3264, 5.3207, 5.2751, 5.2133, 5.3396, 5.1266, 4.7688, 4.993, 5.1717, 4.2441, 3.4631, 4.7463, 4.1685, 3.6018, 4.1291, 3.3075, 0.846, 5.8632, 5.8632, 5.8609, 5.8566, 5.8561, 5.8516, 5.8488, 5.845, 5.8388, 5.8377, 5.8365, 5.8357, 5.8353, 5.8347, 5.8308, 5.8286, 5.8237, 5.8227, 5.8208, 5.8174, 5.8154, 5.8095, 5.8084, 5.8043, 5.8037, 5.8002, 5.7986, 5.7986, 5.7972, 5.7943, 5.764, 5.503, 5.5466, 5.394, 0.9301, 2.4497, 3.8913, 3.9495, 4.2821, 0.0756, 3.7254, 1.6832, 1.4497]}, \"token.table\": {\"Topic\": [15, 15, 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 11, 1, 3, 8, 14, 8, 12, 10, 10, 10, 10, 4, 4, 4, 14, 8, 13, 13, 13, 8, 14, 4, 2, 2, 7, 5, 1, 5, 5, 5, 1, 1, 5, 8, 1, 4, 5, 14, 5, 5, 11, 10, 15, 4, 15, 15, 13, 4, 9, 4, 10, 8, 12, 9, 3, 12, 1, 2, 3, 4, 8, 10, 11, 14, 1, 4, 8, 13, 13, 2, 1, 2, 3, 5, 7, 8, 11, 2, 4, 8, 10, 2, 3, 12, 12, 13, 6, 9, 1, 1, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1, 8, 11, 2, 5, 2, 3, 7, 2, 6, 7, 13, 11, 15, 1, 2, 3, 5, 2, 3, 4, 6, 10, 7, 2, 1, 9, 2, 6, 7, 8, 9, 13, 4, 8, 11, 14, 14, 7, 7, 7, 6, 8, 4, 9, 2, 3, 4, 8, 10, 13, 1, 6, 8, 10, 4, 12, 2, 5, 6, 7, 10, 6, 1, 2, 3, 4, 12, 13, 3, 6, 7, 3, 5, 2, 3, 1, 2, 3, 4, 6, 7, 8, 9, 14, 6, 2, 2, 4, 8, 2, 3, 8, 9, 11, 14, 14, 15, 1, 5, 7, 2, 7, 2, 6, 9, 6, 6, 1, 2, 3, 6, 7, 9, 10, 1, 3, 15, 15, 10, 9, 9, 3, 2, 4, 5, 8, 9, 10, 14, 13, 4, 8, 14, 1, 3, 4, 6, 7, 8, 9, 10, 11, 15, 13, 15, 2, 3, 12, 12, 1, 2, 3, 5, 6, 7, 4, 10, 15, 2, 14, 5, 7, 2, 6, 12, 15, 11, 8, 5, 1, 2, 3, 8, 10, 1, 2, 4, 7, 9, 6, 11, 13, 1, 2, 3, 4, 5, 7, 8, 10, 11, 13, 1, 2, 3, 4, 8, 11, 15, 5, 2, 7, 15, 1, 2, 4, 10, 15, 8, 4, 5, 7, 8, 10, 11, 1, 3, 6, 10, 11, 2, 3, 4, 5, 7, 8, 9, 10, 13, 1, 1, 1, 2, 3, 4, 6, 8, 10, 5, 10, 5, 2, 7, 9, 1, 2, 5, 7, 15, 1, 3, 7, 11, 12, 12, 15, 3, 5, 1, 2, 3, 4, 6, 8, 11, 9, 3, 4, 5, 8, 9, 10, 12, 1, 2, 8, 9, 13, 11, 1, 8, 10, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 8, 2, 5, 8, 14, 2, 1, 3, 5, 7, 14, 5, 10, 8, 11, 1, 3, 5, 6, 8, 12, 13, 12, 13, 2, 3, 4, 5, 7, 12, 15, 9, 7, 1, 2, 3, 4, 5, 6, 8, 10, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 2, 4, 8, 13, 14, 5, 9, 10, 10, 6, 2, 3, 4, 6, 8, 10, 1, 3, 8, 1, 6, 7, 8, 9, 1, 4, 5, 9, 10, 4, 8, 14, 7, 4, 8, 9, 10, 14, 2, 6, 7, 11, 1, 2, 4, 5, 7, 9, 14, 6, 9, 6, 2, 9, 9, 1, 2, 3, 4, 6, 8, 10, 3, 5, 7, 1, 3, 8, 9, 11, 6, 1, 3, 4, 8, 9, 6, 12, 1, 2, 3, 4, 8, 9, 12, 1, 2, 4, 6, 12, 2, 3, 4, 6, 8, 10, 4, 12, 5, 10, 12, 8, 9, 2, 3, 4, 8, 10, 13, 6, 14, 14, 3, 14, 3, 3, 4, 3, 4, 8, 3, 5, 8, 10, 3, 4, 8, 14, 11, 1, 2, 3, 4, 6, 11, 1, 2, 4, 5, 7, 8, 9, 10, 14, 2, 10, 7, 1, 2, 4, 7, 8, 14, 1, 2, 3, 4, 6, 8, 10, 2, 7, 2, 2, 3, 4, 5, 6, 7, 8, 9, 10, 2, 7, 12, 13, 2, 3, 4, 5, 6, 7, 9, 10, 11, 2, 3, 4, 5, 7, 8, 10, 1, 2, 3, 4, 5, 6, 8, 10, 1, 2, 8, 10, 11, 8, 2, 3, 4, 8, 13, 14, 10, 10, 10, 10, 10, 10, 10, 3, 8, 11, 12, 5, 1, 1, 1, 2, 3, 5, 6, 7, 14, 6, 1, 2, 6, 1, 3, 4, 5, 8, 9, 6, 2, 4, 7, 2, 2, 3, 4, 6, 8, 10, 1, 2, 4, 7, 8, 9, 10, 1, 4, 5, 8, 9, 10, 1, 11, 3, 9, 1, 7, 5, 2, 4, 5, 6, 7, 8, 9, 10, 1, 2, 4, 8, 10, 13, 5, 8, 10, 14, 1, 2, 7, 8, 1, 2, 3, 4, 6, 7, 8, 13, 2, 2, 11, 2, 2, 3, 3, 15, 1, 2, 3, 4, 5, 6, 7, 8, 9, 7, 14, 4, 4, 1, 2, 3, 4, 5, 6, 8, 10, 13, 2, 5, 6, 7, 14, 1, 3, 4, 10, 13, 7, 7, 8, 13, 10, 15, 1, 2, 4, 2, 7, 15, 1, 2, 3, 4, 5, 6, 8, 9, 10, 15, 15, 13, 4, 2, 3, 4, 5, 6, 7, 8, 9, 10, 2, 4, 6, 9, 15, 6, 7, 11, 7, 7, 12, 10, 12, 2, 3, 4, 6, 7, 9, 2, 3, 7, 13, 3, 7, 6, 5, 8, 5, 2, 5, 6, 13, 3, 6, 13, 5, 1, 1, 2, 3, 4, 8, 10, 11, 13, 1, 2, 3, 4, 5, 6, 7, 8, 10, 15, 1, 6, 8, 10, 11, 6, 14, 11, 2, 3, 11, 1, 2, 8, 11, 1, 1, 1, 2, 3, 4, 6, 8, 9, 10, 13, 1, 8, 10, 8, 14, 2, 1, 3, 4, 12, 11, 12, 12, 12, 4, 4, 4, 2, 3, 4, 6, 8, 10, 1, 6, 5, 7, 11, 9, 1, 2, 6, 7, 11, 1, 3, 4, 5, 7, 9, 11, 14, 6, 8, 10, 4, 9, 10, 1, 2, 3, 4, 5, 6, 7, 9, 15, 5, 7, 12, 2, 4, 2, 11, 2, 3, 9, 12, 9, 1, 2, 3, 4, 8, 11, 1, 2, 4, 5, 7, 8, 9, 10, 13, 2, 3, 4, 6, 8, 10, 2, 4, 6, 8, 10, 2, 6, 15, 12, 4, 2, 4, 6, 8, 10, 11, 11, 1, 2, 3, 4, 8, 1, 7, 9, 13, 3, 2, 12, 2, 3, 4, 5, 6, 7, 8, 9, 10, 10, 1, 2, 3, 4, 8, 10, 13, 2, 1, 2, 4, 6, 8, 9, 10, 15, 2, 7, 2, 9, 4, 6, 12, 2, 5, 6, 7, 9, 9, 2, 3, 4, 5, 6, 7, 8, 9, 10, 9, 11, 13, 1, 4, 8, 9, 10, 4, 2, 3, 4, 5, 8, 9, 10, 13, 15, 8, 8, 14, 2, 9, 10, 4, 3, 4, 5, 8, 9, 8, 2, 3, 4, 6, 8, 10, 5, 4, 8, 9, 10, 13, 6, 11, 4, 11, 14, 13, 2, 1, 2, 3, 4, 5, 8, 10, 1, 2, 3, 4, 7, 8, 11, 13, 13, 2, 4, 5, 1, 2, 3, 4, 8, 9, 10, 13, 4, 2, 3, 4, 5, 7, 8, 9, 10, 3, 8, 10, 7, 14, 5, 6, 1, 5, 6, 7, 11, 7, 4, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 2, 14, 15, 12, 1, 2, 3, 4, 5, 6, 7, 8, 10, 11, 13, 4, 5, 6, 7, 8, 10, 13, 5, 6, 7, 8, 14, 1, 2, 3, 4, 6, 7, 8, 9, 11, 12, 1, 3, 5, 6, 7, 6, 1, 2, 3, 4, 8, 10, 11, 3, 11, 8, 2, 3, 9, 14, 12, 9, 11, 8, 3, 1, 2, 3, 4, 6, 8, 10, 11, 3, 5, 7, 2, 14, 5, 3, 11, 1, 2, 3, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 5, 10, 2, 2, 3, 4, 7, 8, 10, 10, 1, 2, 3, 4, 5, 7, 8, 9, 10, 15, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 7, 1, 1, 2, 3, 4, 5, 6, 8, 10, 15, 2, 3, 4, 8, 10, 13, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1, 2, 4, 5, 6, 7, 9, 13, 2, 6, 1, 9, 13, 1, 3, 7, 10, 3, 6, 9, 1, 3, 4, 5, 7, 8, 9, 10, 11, 5, 11, 3, 4, 2, 4, 6, 10, 14, 2, 3, 4, 5, 6, 7, 8, 9, 10, 6, 2, 3, 5, 7, 10, 5, 6, 9, 2, 5, 6, 1, 2, 3, 4, 6, 8, 10, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 1, 11, 8, 13, 11, 14, 3, 1, 3, 4, 8, 10, 11, 9, 7, 1, 2, 3, 4, 5, 6, 8, 9, 13, 6, 9, 10, 12, 2, 6, 3, 1, 2, 3, 4, 5, 6, 7, 8, 10, 15, 9, 4, 6, 12, 6, 1, 2, 4, 8, 10, 14, 15, 10, 14, 2, 3, 4, 5, 7, 8, 11, 11, 14, 8, 13, 13, 15, 1, 2, 3, 4, 6, 8, 11, 7, 12, 2, 4, 6, 7, 9, 10, 15, 9, 1, 3, 4, 5, 7, 8, 9, 10, 11, 13, 3, 13, 2, 3, 4, 8, 9, 13, 2, 3, 4, 6, 8, 10, 1, 7, 9, 10, 13, 15, 9, 2, 7, 1, 3, 5, 7, 9, 10, 5, 2, 7, 8, 9, 10, 2, 3, 7, 11, 14, 1, 4, 7, 8, 3, 7, 8, 9, 11, 2, 3, 5, 7, 9, 7, 14, 13, 10, 3, 9, 11, 6, 5, 13, 11, 3, 5, 6, 7, 9, 8, 8, 8, 8, 13, 1, 3, 5, 6, 10, 1, 2, 3, 4, 6, 8, 10, 13, 1, 3, 7, 13, 14, 9, 4, 10, 10, 6, 5, 13, 11, 2, 4, 5, 8, 10, 11, 3, 6, 1, 2, 3, 4, 5, 7, 8, 10, 14, 1, 3, 4, 7, 8, 10, 2, 3, 4, 5, 8, 11, 1, 5, 6, 7, 8, 10, 11, 7, 1, 2, 3, 4, 7, 10, 15, 1, 4, 6, 7, 9, 2, 11, 15, 1, 2, 8, 13, 9, 4, 7, 7, 1, 2, 3, 4, 5, 6, 7, 8, 10, 3, 6, 1, 2, 6, 7, 8, 9, 11, 13, 12, 2, 3, 1, 2, 3, 4, 8, 9, 10, 4, 4, 4, 4, 4, 4, 4, 9, 6, 8, 14, 2, 3, 4, 5, 7, 8, 10, 13, 15, 1, 6, 7, 5, 1, 4, 15, 2, 4, 5, 6, 7, 8, 9, 10, 3, 6, 12, 12, 1, 3, 4, 5, 6, 8, 9, 10, 3, 15, 7, 1, 2, 7, 13, 3, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1, 2, 3, 4, 5, 8, 3, 2, 4, 5, 6, 8, 10, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 10, 14, 11, 1, 3, 5, 6, 7, 13, 8, 8, 12, 2, 6, 6, 13, 1, 12, 2, 3, 4, 6, 7, 8, 10, 1, 2, 3, 4, 6, 7, 9, 1, 3, 9, 12, 1, 1, 1, 2, 12, 15, 5, 2, 4, 6, 13, 1, 2, 3, 4, 6, 8, 10, 6, 2, 3, 4, 8, 10, 11, 13, 2, 11, 2, 7, 12, 2, 11, 1, 2, 3, 4, 7, 8, 9, 10, 11, 9, 8, 2, 5, 6, 8, 1, 2, 3, 4, 6, 8, 10, 14, 14, 14, 14, 8, 8, 1, 2, 8, 10, 11, 5, 7, 8, 7, 1, 1, 11, 1, 2, 3, 7, 9, 10, 2, 7, 11, 2, 6, 7, 8, 9, 2, 3, 4, 5, 6, 7, 8, 13, 7, 9, 5, 6, 12, 1, 2, 3, 4, 5, 6, 8, 10, 11, 1, 2, 3, 4, 7, 13, 15, 1, 13, 13, 6, 7, 14, 6, 1, 2, 3, 4, 5, 6, 7, 8, 9, 1, 2, 3, 5, 1, 2, 3, 4, 6, 8, 10, 1, 2, 5, 6, 7, 8, 9, 10, 13, 1, 2, 6, 7, 5, 3, 6, 8, 13, 8, 1, 2, 3, 4, 6, 8, 10, 11, 3, 5, 6, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1, 4, 9, 9, 14, 8, 15, 13, 1, 2, 6, 10, 14, 5, 7, 12, 7, 2, 3, 1, 2, 3, 4, 7, 8, 11, 12, 13, 4, 2, 3, 4, 6, 8, 10, 6, 1, 2, 3, 4, 5, 6, 8, 10, 11, 1, 7, 2, 11, 14, 1, 2, 9, 9, 9, 2, 7, 6, 7, 10, 1, 2, 3, 15, 7, 1, 3, 4, 10, 12, 13, 14, 2, 6, 9, 9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 2, 3, 4, 8, 9, 10, 13, 14, 6, 1, 2, 4, 5, 6, 8, 9, 10, 2, 4, 5, 6, 7, 8, 9, 10, 8, 6, 15, 14, 6, 7, 1, 1, 2, 3, 4, 6, 8, 9, 10, 15, 12, 10, 14, 13, 4, 12, 14, 12, 4, 5, 12, 1, 2, 3, 4, 5, 7, 8, 9, 10, 2, 3, 4, 5, 6, 7, 8, 9, 10, 1, 3, 4, 7, 9, 12, 6, 8, 9, 3, 9, 2, 3, 4, 8, 13, 1, 2, 3, 4, 6, 8, 10, 3, 5, 6, 7, 10, 5, 10, 1, 1, 1, 2, 3, 4, 5, 7, 8, 9, 10, 9, 9, 1, 2, 3, 4, 6, 8, 9, 10, 2, 3, 4, 5, 6, 7, 8, 9, 10, 7, 10, 11, 2, 3, 4, 5, 6, 7, 8, 9, 10, 2, 3, 4, 5, 7, 8, 9, 10, 2, 3, 6, 9, 10, 1, 2, 3, 4, 6, 11, 2, 2, 6, 3, 4, 6, 3, 3, 13, 1, 2, 3, 5, 6, 7, 6, 2, 3, 4, 6, 8, 10, 13, 15, 4, 2, 3, 4, 6, 10, 3], \"Freq\": [0.922857138376405, 0.922857138376405, 0.028281228078468477, 0.1578635782007167, 0.09203382696721944, 0.14140614039234237, 0.311413070762345, 0.020611742497866854, 0.04441910398765105, 0.15562664490637454, 0.04809406582835599, 0.9953824041006323, 0.9996048934159766, 0.9997955689269824, 0.9957587252493371, 0.9978084703012358, 0.9913390004746728, 0.9988077451483645, 0.9815813646589129, 0.9949490955629479, 0.9955318074095268, 0.9960804351458165, 0.9954990380881286, 0.9989693643616947, 0.9990149465681459, 0.9976782293242225, 0.9839774900541405, 0.9979804116054128, 0.9759100680857092, 0.9815510989607144, 0.9875457415152481, 0.9987447094643824, 0.9670758553635574, 0.9992914224208872, 0.9998462559801279, 0.9993520957426688, 0.9997468664612631, 0.9994018436557136, 0.9999736821136755, 0.997969940213931, 0.9957125799546267, 0.9997964889036307, 0.9999234146597114, 0.999982196348639, 0.9439210417746936, 0.055853316081342815, 0.9999470821149227, 0.01956305589934044, 0.6993792484014207, 0.27877354656560127, 0.996392143248811, 0.9940998261749159, 0.9862886830882511, 0.9953951645533771, 0.992971200458178, 0.9990568967433099, 0.9684070023915742, 0.9156436264189589, 0.9707089709244965, 0.04443188284668946, 0.9535765626327969, 0.14070004056288185, 0.856991156155735, 0.9981581556818575, 0.9760457796501754, 0.9929601003151689, 0.9937760570586138, 0.005272021522857368, 0.09233250700060039, 0.20675886389063014, 0.3413005169486478, 0.19653633632984938, 0.061994683271831684, 0.040560351289549455, 0.06001612955039025, 0.982681615189367, 0.9717744991973138, 0.0013152824261637452, 0.0268536828675098, 0.9786685860811085, 0.9553925314819014, 0.9985435769388934, 0.10825360026521622, 0.005319587236619961, 0.32103708973001466, 0.49871130343312137, 0.031119585334226773, 0.024204121926620824, 0.011171133196901918, 0.5249309167877192, 0.3060735961606054, 0.11589638411716414, 0.053048401298273554, 0.021503544855482216, 0.9748273667818604, 0.9902307073012898, 0.9823354142622701, 0.9976788766207544, 0.9514163999339559, 0.048255526795827786, 0.9999066149929521, 0.9998851776008126, 0.046287161985115, 0.02959343143310631, 0.01467024806085612, 0.07613352872961539, 0.32375719858441093, 0.010117412455762842, 0.2896109315462113, 0.0174525364861909, 0.1712372058137861, 0.020993630845707895, 0.04116285746830098, 0.04116285746830098, 0.9124433405473383, 0.042065581105352044, 0.955489627964425, 0.016554080958855612, 0.05718682513059211, 0.9255236172451091, 0.11802474536974134, 0.0218089203400609, 0.8582451592647494, 0.9866378622220303, 0.9926902875333903, 0.9667414718369659, 0.1385239319452559, 0.12945691094520279, 0.7117611485041694, 0.019645212166781743, 0.01470251411952475, 0.981392817478277, 0.04783534714330249, 0.9248167114371815, 0.028701208285981494, 0.9966143083884884, 0.9983933585830335, 0.9999276315908113, 0.9964062891842375, 0.10386384542837987, 0.05638323037540621, 0.20179261397513804, 0.00445130766121628, 0.4896438427337908, 0.14244184515892097, 0.06839779509561317, 0.23511742064117025, 0.6925276753430832, 0.9411723393064793, 0.9851160640389128, 0.9941410001587366, 0.9989713116841278, 0.9963162617690109, 0.9607845856997576, 0.03824018251541324, 0.06848822942066662, 0.9302984496307216, 0.2028169667286848, 0.25688440826974923, 0.2692695252104354, 0.16574703050621725, 0.09993508152139871, 0.005338412474433692, 0.019154737068274755, 0.054271755026778466, 0.01276982471218317, 0.9098500107430508, 0.9995651533375468, 0.9910809786929478, 0.07687514861567948, 0.8482775019661184, 0.04771560948559416, 0.01060346877457648, 0.01590520316186472, 0.99794890598138, 0.12472456102459818, 0.047688802744699305, 0.1650766248854976, 0.19809194986259712, 0.45854618023749333, 0.9573160328008656, 0.012014806728453386, 0.04205182354958685, 0.9446641790246475, 0.9942597454487224, 0.9957676807932588, 0.9984548832648871, 0.9977699185638083, 0.04992913605544843, 0.021398201166620754, 0.06865256207624158, 0.04368799404851737, 0.017831834305517297, 0.6838508456165883, 0.023181384597172484, 0.08559280466648302, 0.0053495502916551885, 0.993285844548173, 0.9978704032605679, 0.8031693129451029, 0.18609481604286643, 0.010608730605769107, 0.082830907762464, 0.19751985697202953, 0.028672237302391382, 0.10513153677543507, 0.5830021584819581, 0.9939190195119885, 0.05861068648307288, 0.9084656404876297, 0.9999064478572716, 0.03654590902339378, 0.9611574073152563, 0.9965752944491865, 0.9997534443798733, 0.0573841447274174, 0.1721524341822522, 0.770338670128664, 0.9939201871904316, 0.996483446853682, 0.036071686700535975, 0.05153098100076568, 0.036071686700535975, 0.010306196200153137, 0.10692678557658879, 0.7497757735611407, 0.009017921675133994, 0.9998931759000137, 0.8779767897362857, 0.11951516509499177, 0.9487845303399174, 0.9961830753107364, 0.999034556678172, 0.9939203314219074, 0.992094564982414, 0.023115411622615375, 0.05778852905653843, 0.027738493947138447, 0.16874250484509223, 0.3906504564221998, 0.1664309636828307, 0.16411942252056916, 0.9776055662827482, 0.05914571712981938, 0.1330778635420936, 0.7836807519701068, 0.03627547093039329, 0.13894189809188373, 0.1889062259771424, 0.019164399736811548, 0.18753734028165586, 0.1594751835241818, 0.04859544218977214, 0.0075288713251759655, 0.2114928399526703, 0.0013688856954865393, 0.9933492543609249, 0.9911722396086816, 0.06016810967841548, 0.0823353079809896, 0.858187248571084, 0.9904464945709702, 0.04667804859428012, 0.04830872278097986, 0.13493828894940366, 0.1251542438292052, 0.009376376573523517, 0.635759098539562, 0.06931897496081041, 0.23106324986936802, 0.6978110146054914, 0.07551733092632676, 0.9169961612482534, 0.013597485515912444, 0.9858176999036522, 0.022618387295693647, 0.3053482284918642, 0.6559332315751157, 0.005654596823923412, 0.9943530959689804, 0.9989479619283739, 0.998724263022426, 0.008884434916072007, 0.03997995712232403, 0.10439211026384608, 0.7679483430579741, 0.07884935988013905, 0.9579226751627742, 0.0018069142350146644, 0.010583354805085892, 0.02779206180522555, 0.0018069142350146644, 0.9471614393523714, 0.038239436992487046, 0.008824485459804703, 0.07666970970040096, 0.010866573028403286, 0.0380330055994115, 0.055540262145172346, 0.5058993443223307, 0.11591011230296838, 0.024751638564696372, 0.057351357649906225, 0.0012073970031559206, 0.11228792129350061, 0.11877749549261553, 0.2560876498564193, 0.43905553653015045, 0.1487667071205383, 0.03571748800629006, 0.0015163084530972195, 0.9418882105353023, 0.9958104782936967, 0.1694937709167126, 0.618652263846001, 0.20339252510005515, 0.18680782452488043, 0.127680577312877, 0.04113199806052413, 0.6435443863219504, 0.9771517962415799, 0.9988928408296546, 0.7696120204421375, 0.09289481621643707, 0.03199026783612402, 0.012303949167740007, 0.06398053567224804, 0.028914280544189017, 0.26018218734674753, 0.5179974268729858, 0.21229701220875272, 0.009285718372767057, 0.9911367733456528, 0.0106190513340462, 0.014664404223206657, 0.04095919800274963, 0.2937937535752782, 0.029328808446413314, 0.05461226400366617, 0.4591475529197119, 0.0955714620064158, 0.0010113382222901143, 0.9999449711904915, 0.9999232193671642, 0.03927336056953522, 0.3453350953220371, 0.36525307240427246, 0.045808101556862844, 0.10495616277825376, 0.029146675860696386, 0.07021600829098722, 0.9859321744728626, 0.012904871393623856, 0.9930409485227385, 0.10876451580012965, 0.1683615107591048, 0.7211236390035993, 0.11499287905582829, 0.00928513930264452, 0.12142105241919758, 0.752810524999025, 0.0007142414848188092, 0.12151843205061234, 0.1083331476345635, 0.7583320334419444, 0.009265334995061351, 0.0017817951913579524, 0.9910454802671965, 0.9092108746542586, 0.9917572456448187, 0.9968493697647282, 0.14480791866612616, 0.1652162160005802, 0.32887256214120214, 0.12069493041427122, 0.2085675991919205, 0.02651778761929061, 0.005199566199860904, 0.9929762682942802, 0.07150034996542269, 0.012870062993776084, 0.3188915608457852, 0.006435031496888042, 0.2702713228692978, 0.006435031496888042, 0.31317153284855137, 0.25869592618014087, 0.0041062845425419185, 0.12010882286935111, 0.5102058544108333, 0.10676339810608987, 0.9919909968819525, 0.036858899488802055, 0.018429449744401027, 0.9443696823570344, 0.06325778157136844, 0.00451841296938346, 0.09488667235705266, 0.5575026463762361, 0.034061882384583006, 0.08376442504780107, 0.02989103964361366, 0.1157408860618994, 0.01668337096387739, 0.9973225267482102, 0.9996552527574326, 0.9995027717749102, 0.07835564794301995, 0.6307629659413106, 0.2859981149920228, 0.9981434348795793, 0.04324100953469098, 0.7055400163318566, 0.2145629840202388, 0.03612540037075449, 0.9716485567916342, 0.9917467980552297, 0.9981167140904778, 0.9997078774935071, 0.9946193006014171, 0.44006206079187027, 0.07383591624024669, 0.26777825623129464, 0.05020842304336775, 0.006891352182423024, 0.03150332426250525, 0.12798225481642758, 0.9864062719654395, 0.9690323087780157, 0.06947222915742833, 0.04197280511594628, 0.026052085934035626, 0.48123992072593585, 0.3068356787786418, 0.07309057442604439, 0.9602209188134224, 0.9930410587182829, 0.9970972497072785, 0.031437291386637596, 0.13205047285092048, 0.21666802808103314, 0.10005922038258003, 0.04071613950516059, 0.37350825933815684, 0.06176666060986947, 0.043832170589739206, 0.018915346829169035, 0.04728836707292259, 0.18726193360877347, 0.009930557085313743, 0.6710219287647715, 0.024589950877919747, 0.026008601890107425, 0.005674604048750711, 0.006147487719479937, 0.0028373020243753553, 0.9615449344892242, 0.023896014415747856, 0.07461408582876372, 0.8480646748772556, 0.04047692237769535, 0.013167191616840657, 0.9673554487755754, 0.027938063502543697, 0.9984582844273425, 0.9986104100085734, 0.9938816819923325, 0.36061594595684443, 0.13580958972218288, 0.13388816787402336, 0.23904234538238875, 0.0591273905092719, 0.07152929516557413, 0.021026160711029418, 0.01261569642661765, 0.9655212998504709, 0.07275616522382765, 0.035490812304306174, 0.015970865536937776, 0.05501075907167456, 0.8198377642294725, 0.04697057609999529, 0.008288925194116815, 0.013814875323528026, 0.9090187962881441, 0.019340825452939237, 0.04932345598683242, 0.19988979531505768, 0.7502357252733983, 0.9963576881202326, 0.012731625463701171, 0.025463250927402342, 0.37770488875646807, 0.45833851669324216, 0.12095044190516113, 0.02018032363328111, 0.06054097089984333, 0.9189808916078782, 0.9812699917052312, 0.2052141100142475, 0.04658914930053187, 0.04880768021960481, 0.04437061838145892, 0.0288409019479483, 0.6256257191785708, 0.9675804183168324, 0.9945279917694866, 0.9909766373221073, 0.9957945265669217, 0.006255617126298698, 0.9915153145183436, 0.9973050710320008, 0.015535108912265923, 0.1616829344613079, 0.050507510492011486, 0.11765452152512297, 0.5727374987133489, 0.05190640655520131, 0.02996582619569778, 0.07496336805084941, 0.13414497440678316, 0.7910608049576477, 0.016738593919969666, 0.3605235613531928, 0.021888930510729564, 0.20343829533501592, 0.3952883333408221, 0.995286062788989, 0.1965335757088854, 0.008852863770670514, 0.023902732180810387, 0.020361586672542183, 0.7498375613757926, 0.13457137909806394, 0.8628400189228805, 0.1397453816309178, 0.011292556091387296, 0.167976771859386, 0.287960180330376, 0.005646278045693648, 0.21738170475920543, 0.16938834137080944, 0.07401032211137941, 0.09621341874479324, 0.06660928990024147, 0.05180722547796559, 0.7030980600581044, 0.23597396233942758, 0.12143484243222648, 0.1771850058972621, 0.33508536405504646, 0.08239804034140487, 0.04791942771627801, 0.06168422427933463, 0.9381526050842087, 0.9941832524538451, 0.005500322281902324, 0.99584391567624, 0.9990711215834785, 0.9957081079479688, 0.2630340460508867, 0.2846317300931292, 0.14589164055024711, 0.13616553118023064, 0.16648810745145848, 0.0037188065238298288, 0.051667000436977584, 0.9459035464615896, 0.9822828511446301, 0.04796484502406765, 0.9353144779693192, 0.9950865352977788, 0.9953032903896222, 0.9997695219384152, 0.9990378713515955, 0.9949393121123191, 0.004316439531940647, 0.6979922332621262, 0.11958529743770918, 0.1810606661017473, 0.0012339946264679183, 0.9504969901566914, 0.01627563339309403, 0.03255126678618806, 0.9710781960131368, 0.9946703296902978, 0.07457386516923457, 0.116174196124823, 0.2116440800689845, 0.1000708422065307, 0.4589455866713305, 0.03853302544734229, 0.21916000533316837, 0.03590151676606915, 0.018218680149945536, 0.15003618947013972, 0.06965965939685058, 0.00643012240586313, 0.48761761577795404, 0.010181027142616624, 0.0026792176691096377, 0.9981077278241622, 0.9944104143243289, 0.9957267112374172, 0.0061200048521311435, 0.689860546943005, 0.24650019543305995, 0.01292001024338797, 0.03842003046060107, 0.005780004582568303, 0.006155144011705137, 0.25724257042022847, 0.13923360247167482, 0.2117915316342876, 0.17367815221205432, 0.1478144190003081, 0.06406807535336431, 0.006193468006772205, 0.9909548810835528, 0.9986311988746146, 0.15915871053549996, 0.06603787989584847, 0.05268190418657574, 0.5679999669693483, 0.005564989878863635, 0.08124885223140907, 0.01112997975772727, 0.044148919705651504, 0.012242977733499997, 0.03303254751426107, 0.3495944611925963, 0.5973385675495543, 0.016516273757130533, 0.026233639778806904, 0.03853065842512264, 0.04508906836982437, 0.8099636281706631, 0.009017813673964873, 0.03853065842512264, 0.018035627347929746, 0.014756422375578884, 0.9964595127296414, 0.03885662719578078, 0.04173489587694973, 0.012472497618398769, 0.6994192895240541, 0.19188457874459644, 0.0028782686811689466, 0.011992786171537278, 0.014397318611562742, 0.2333602194585814, 0.028264674574847104, 0.03948221729673955, 0.012012486694310019, 0.3459772822177378, 0.15916544869960775, 0.16737986974792268, 0.13632095855616871, 0.22333433635797856, 0.01015156074354448, 0.5979752685602153, 0.031904905193996934, 0.9998180792747574, 0.08132422420554367, 0.044868537492713745, 0.08132422420554367, 0.745939435816366, 0.03725691059662838, 0.008813462721783058, 0.9998386606541267, 0.9996910286215799, 0.9970848355696689, 0.9981545154113398, 0.9995129768251696, 0.9975111652220562, 0.9990120635494565, 0.7462461221489499, 0.09339301649036195, 0.15894772998840448, 0.9788767732363558, 0.9947335518593151, 0.9999210710272303, 0.9999126137461168, 0.04378499086938577, 0.02513582809168442, 0.40055158313845496, 0.03486582606265903, 0.030000827077171728, 0.3227115993706581, 0.1418958037433798, 0.9936665102132077, 0.13343437103618142, 0.012795076674702326, 0.8517865329158978, 0.005189033043458394, 0.8831734239966187, 0.025945165217291972, 0.04566349078243387, 0.008302452869533431, 0.030096391652058685, 0.9959091619023877, 0.01756965792480211, 0.9825493316408564, 0.9990439585169352, 0.9979487150861032, 0.23626490058056712, 0.11514629128395688, 0.18000566490137243, 0.338988770398205, 0.08098463225157963, 0.04849522226274532, 0.999866998948441, 0.011095193286858174, 0.23531055762545044, 0.016180490210001505, 0.11233883202943902, 0.5635433590283381, 0.06194816251829147, 0.03733015835075811, 0.31012746937552893, 0.09260750821630379, 0.07035299073796722, 0.43503992231845034, 0.053123686883771164, 0.483116838061046, 0.515073982991154, 0.0126675774658413, 0.9849041479691611, 0.10141588029080177, 0.8977799238857862, 0.993820271050401, 0.09906578599677306, 0.03444340241609659, 0.499593351235382, 0.024930462701174678, 0.21059680334413344, 0.013449328562475812, 0.09447333234129351, 0.022962268277397727, 0.13227332011132262, 0.018896188587331803, 0.1417214144049885, 0.028344282880997704, 0.09448094293665901, 0.5763337519136199, 0.21621628920592856, 0.021621628920592856, 0.75675701222075, 0.0061776082630265305, 0.9999871134051272, 0.006020306472049108, 0.9833167237680209, 0.008027075296065478, 0.0803873370797416, 0.02206711213953691, 0.5816260271063657, 0.14343622890698993, 0.00788111147840604, 0.006304889182724832, 0.012609778365449664, 0.14501245120267112, 0.9999758603888135, 0.7259765582737159, 0.27379687340608716, 0.9996962968418907, 0.9996740736463643, 0.9943477723392687, 0.9940492824979197, 0.9790283803363264, 0.19546046721630997, 0.007435996035403097, 0.4127567958064227, 0.008380249500216189, 0.16193946921544522, 0.0031868554437441843, 0.12652996428495428, 0.03824226532493021, 0.04591432472653658, 0.9700906896861046, 0.02853207910841484, 0.9997976435434528, 0.9997143549582028, 0.01662033322615465, 0.17940183217643405, 0.21821519859280697, 0.2572240983412523, 0.0057682332961360265, 0.12699889918102877, 0.12807433250742703, 0.06442823291785833, 0.003030766647122319, 0.8513316218762063, 0.05014486523894395, 0.020736598407082085, 0.07804501582301802, 0.9597698126055568, 0.1303393618934153, 0.06813193917155799, 0.21328259218922502, 0.06813193917155799, 0.5154329311239605, 0.99754565901362, 0.9962559366548874, 0.9993826804090473, 0.9586988980051042, 0.9989318040401818, 0.9378865270731814, 0.03356277229530975, 0.9183994964443848, 0.04729299732520918, 0.9974613369485447, 0.992947916405197, 0.9888413589136724, 0.0675364686130049, 0.04402377213292171, 0.11918936034850679, 0.08216937014582262, 0.551172624402801, 0.004252296171929937, 0.06878714395769017, 0.041772556512488214, 0.02113641332518116, 0.9894841041152732, 0.9595158928874991, 0.9679253683319718, 0.996811024429507, 0.10537786409309281, 0.022853030767176753, 0.07194472648926015, 0.6047589067832515, 0.04401324444048856, 0.04316683589355609, 0.0338563418772989, 0.0702519093953952, 0.0033856341877298896, 0.11824603241219618, 0.07628776284657818, 0.47679851779111365, 0.18309063083178764, 0.13731797312384073, 0.0052128995098535994, 0.9695993088327696, 0.02464279768294429, 0.9986424599842465, 0.9995845698680444, 0.98794674139871, 0.04337060094314545, 0.9541532207492, 0.0037581306255758183, 0.10031317900575454, 0.6871597305379785, 0.004625391539170238, 0.06215369880760008, 0.1419417028582867, 0.04515432180736172, 0.8850247074242898, 0.059829476394754284, 0.009030864361472345, 0.9985964013704517, 0.9988526809343977, 0.9962345120667891, 0.9921676168542666, 0.9979918796962519, 0.9929131195125604, 0.08596143285428876, 0.19341322392214969, 0.09670661196107484, 0.6124752090868073, 0.18352411137333818, 0.21733118452105837, 0.5940385710242263, 0.9991294447241452, 0.9999460784147971, 0.9999083356341012, 0.12906237770850956, 0.1796408770807633, 0.08162323346970604, 0.5211329521527386, 0.03488172370500258, 0.023021937645301704, 0.030347099623352247, 0.05741651309932555, 0.4905651917248564, 0.1520508962532628, 0.051618754447602125, 0.04918743630333101, 0.004301562870633511, 0.17580300427806522, 0.009725272577084458, 0.006732881014904625, 0.0024313181442711146, 0.9999151351041188, 0.9672892452855695, 0.024835804946521378, 0.006535738143821416, 0.9859232094176595, 0.9956333906903557, 0.9738688762885813, 0.9928388631025297, 0.3196681289274303, 0.6674670532004745, 0.012467057028169783, 0.37304193107284717, 0.02779528113876116, 0.0863116624835215, 0.512018336766653, 0.9999269672262023, 0.9997096153624214, 0.516418544101283, 0.082177790975135, 0.01570553543901317, 0.057285577734908014, 0.05133600389766991, 0.021219774605233853, 0.2481854121897788, 0.006329096370945605, 0.0013394912954382233, 0.014828787152833795, 0.6297156188189694, 0.355078355385664, 0.1788937335982374, 0.8162026595419581, 0.9973259367925595, 0.4920430074857413, 0.031185824418110367, 0.065836740438233, 0.408880809037447, 0.9992494883555784, 0.9982287841774351, 0.9889909870398701, 0.991428780593058, 0.9998783745255105, 0.9994887729878018, 0.9983650481980603, 0.2573878279569578, 0.04827609803733661, 0.19488298523493253, 0.3572431254762909, 0.08892965427930428, 0.05323075020432642, 0.9959276530146596, 0.004043391064165724, 0.0747493615211037, 0.07229856278270687, 0.8516525615929029, 0.9961950068501348, 0.10162602215089678, 0.03209242804765162, 0.012034660517869356, 0.8544608967687243, 0.9855610887052904, 0.04826048888309479, 0.32441773082524833, 0.18365797158288852, 0.05113313703089805, 0.2681138271283044, 0.12352387035554024, 0.0007660395060808697, 0.9930167724936192, 0.9933704566920941, 0.012079948239180118, 0.9875357685529746, 0.06006623679576101, 0.8967031064510036, 0.040759232111409256, 0.042484657147477234, 0.008674877637442995, 0.38302921260709843, 0.02602463291232899, 0.15992915439285935, 0.003336491399016537, 0.08052065909626575, 0.29561313795286515, 0.9664840684133255, 0.04852988104120847, 0.9503768370569993, 0.9883541341328821, 0.005643808565410879, 0.9941165658788019, 0.019263145578651057, 0.9727888517218783, 0.0216591262364008, 0.030322776730961117, 0.07364102920376271, 0.8663650494560319, 0.9989068112227555, 0.18899784242108122, 0.030589246172198117, 0.5323985464732577, 0.14566307701046724, 0.026947669246936437, 0.07538064235291679, 0.08393335060818374, 0.04727280666437935, 0.016400769659070388, 0.09840461795442232, 0.023154027753981725, 0.005788506938495431, 0.6463832747986564, 0.03569579278738849, 0.041484299725883925, 0.2313738231171123, 0.05461381448367813, 0.17900335228786288, 0.4169524930127835, 0.07240585036362993, 0.045563083188224385, 0.2328348441594539, 0.11309486118612333, 0.5664966323594178, 0.0402541031340439, 0.04728259733205156, 0.20693298740799732, 0.155199740555998, 0.620798962223992, 0.9957248646828674, 0.9980816778795049, 0.21525693300364754, 0.1355766043700213, 0.5339782475381525, 0.07225250138811871, 0.04291633519626573, 0.9920970763731176, 0.9921834199124526, 0.06841431868506498, 0.15840860149073868, 0.4059366223768561, 0.17491435783163672, 0.19223665335547668, 0.315215542226252, 0.05750553810884327, 0.6112625717495562, 0.01490884321340381, 0.9994939636163057, 0.9996587410639487, 0.9955794340160107, 0.10011741891487791, 0.052004454348010785, 0.13337196795374195, 0.5540066361155435, 0.01202824114171678, 0.06084874930515547, 0.02087253609886147, 0.04209884399600873, 0.024764025880005133, 0.9950817854283308, 0.004326499253832747, 0.13188363242717752, 0.5998914999883267, 0.13964149315818797, 0.09264675988379778, 0.028495219223519126, 0.0031329822182926785, 0.9997654494138248, 0.16358859340884682, 0.005964167468030873, 0.04260119620022052, 0.0570856029082955, 0.011928334936061746, 0.6611705650274226, 0.04260119620022052, 0.0161884545560838, 0.11028662552047261, 0.8885063351790187, 0.9984346429034369, 0.9945700253715737, 0.03968964902436682, 0.04630459052842796, 0.9062469860563758, 0.02245042472850432, 0.16677458369746065, 0.7184135913121382, 0.051315256522295585, 0.039288243274882555, 0.9953816086291087, 0.11551851396639937, 0.07958870492079233, 0.10925168680728185, 0.48714136450206746, 0.022978366250097523, 0.015458173659156514, 0.049090146079753795, 0.09462909010267434, 0.025902885591019026, 0.8773763368495667, 0.035461103575406216, 0.08693689908809266, 0.007522031529049709, 0.9446000037924422, 0.004095328276927063, 0.04229053281887947, 0.0015044063058099416, 0.9998278337611872, 0.23997816732981347, 0.21307812897472683, 0.2955562849547153, 0.042443557319326475, 0.08700929250461928, 0.05179261116128622, 0.06142844579594413, 0.005563547378344146, 0.0030398764026004096, 0.9997746217903803, 0.9994927994305052, 0.9660904832670908, 0.9988132806461063, 0.9947428382357453, 0.9996592332586736, 0.998911599960325, 0.08768391574075064, 0.019309454368901767, 0.01867635750434761, 0.7850401120471538, 0.08895010946985896, 0.9996625599805566, 0.24837719142806342, 0.11260000301882069, 0.19483596027975714, 0.2842763496938953, 0.09785727478336648, 0.062045975446232186, 0.9996557784132816, 0.016813785419979764, 0.300779939179638, 0.37363967599955034, 0.12890568821984488, 0.17934704447978417, 0.9917215547758652, 0.9876203659247376, 0.9998080310340244, 0.9766438120331237, 0.9901044434312521, 0.9616678189737201, 0.9979617022170277, 0.0019991019877972967, 0.08823309227959886, 0.6100895793668659, 0.09786512912989492, 0.04906886697320637, 0.1012272552002813, 0.05161317859403929, 0.1610725505120614, 0.13241888435879315, 0.40393323548102444, 0.179989533991889, 0.0019473365346881343, 0.030044620820902646, 0.07928441605515976, 0.011127637341075053, 0.9713763979791574, 0.8499385119738609, 0.1487956015657157, 0.0009662052049721799, 0.04055737868943456, 0.45735459113305765, 0.05642322117045865, 0.16177037462163144, 0.17034099049562515, 0.025150676344517283, 0.08723662586029321, 0.0011223425549277488, 0.9989362548028703, 0.02971687502767718, 0.07755379580393801, 0.04976972565610975, 0.7134466012742334, 0.0949490397225783, 0.013529634158942457, 0.006281615859508998, 0.014254435988885802, 0.9868684696618757, 0.012728312162448956, 0.9945946724475982, 0.08112148092696465, 0.9058565370177719, 0.9810414020323619, 0.016350690033872697, 0.07139113079183564, 0.1618198964614941, 0.004759408719455709, 0.7581058174561595, 0.003399577656754078, 0.997306230106312, 0.9993942019479418, 0.05210513272052787, 0.022892551572630843, 0.6520866070657976, 0.02036453974252437, 0.061374509430918274, 0.010673827727116222, 0.09845201627247989, 0.005196468761885529, 0.054633144550634345, 0.00912893160871782, 0.013201839557222695, 0.9986696808390207, 0.9567021788552295, 0.9687158100418961, 0.983418041207748, 0.041327304229167486, 0.16450151715974812, 0.2944738697110876, 0.11105871657675302, 0.05088508468607593, 0.21437697714967174, 6.730831307682002e-05, 0.08339499990218, 0.038029196888403304, 0.0005384665046145601, 0.0013461662615364002, 0.01276611544444353, 0.4477259059444124, 0.002735596166666471, 0.005471192333332942, 0.17599002005554296, 0.17872561622220942, 0.17507815466665413, 0.8656593291392841, 0.030234425098614705, 0.03341699616162678, 0.07001656338626563, 0.9804563143288312, 0.21833653822325944, 0.010396978010631401, 0.428110859261293, 0.00428110859261293, 0.007746767929490064, 0.014270361975376433, 0.03037548477615841, 0.033025694857299746, 0.25319699390596473, 0.9854770139930723, 0.040201961582329436, 0.03430567388358779, 0.26854910336996063, 0.003752183081017414, 0.652343829942599, 0.9927605522486859, 0.07302492706356106, 0.2578428409251992, 0.4477358462429922, 0.16451754803701885, 0.05244261171359984, 0.003101444778761281, 0.0012687728640387059, 0.4585685565877014, 0.5405228279560356, 0.9997957362130093, 0.4718029573381008, 0.038875150042839764, 0.178472279742128, 0.3110012003427181, 0.9839625753206742, 0.07154933163693353, 0.9199199781891453, 0.9991563972340369, 0.9947322133367097, 0.006200989557995724, 0.22854162546221352, 0.27284354055181187, 0.1957467013049784, 0.12798331025883958, 0.09474089200979033, 0.0733890310575164, 0.0004474940918141244, 0.05455812354545462, 0.5002742719885382, 0.4450045207446646, 0.11932970987444708, 0.8671292250876488, 0.9989481061373029, 0.48239790587350934, 0.5167919064750941, 0.009718453297960377, 0.08940977034123547, 0.900900620720927, 0.08108163939647112, 0.02909628208007337, 0.0333637367851508, 0.677361446824108, 0.010474661548826413, 0.09659965650584358, 0.012802364115232282, 0.0011638512832029347, 0.01706981882030971, 0.04112274533983703, 0.9949252908309596, 0.9971302191063661, 0.9987558115599637, 0.03741080936702123, 0.20201837058191463, 0.03491675540921981, 0.684202135756855, 0.024109188258747014, 0.017458377704609906, 0.999628659074045, 0.03767725902039568, 0.2031094086268715, 0.012288582942036747, 0.14294172422199347, 0.45780767960474633, 0.0023186005551012728, 0.08462892026119645, 0.0027823206661215273, 0.05645792351671599, 0.9624102018197087, 0.11055908877763339, 0.00848279453536829, 0.3916223477161694, 0.011027632895978778, 0.4411053158391511, 0.0022620785427648776, 0.0011310392713824388, 0.001696558907073658, 0.007068995446140242, 0.010462113260287557, 0.0022620785427648776, 0.00904831417105951, 0.002827598178456097, 0.9930215852871768, 0.9999775210483177, 0.0002189795257556925, 0.09825611320657922, 0.2390599482674895, 0.2537096785405453, 0.019007422835594107, 0.012919792019585857, 0.3675571339809298, 0.0020803054946790787, 0.007182528444786713, 0.13182399394034128, 0.6118608327540098, 0.1420125309274069, 0.0811164290893301, 0.02978187734680719, 0.003291681180436584, 0.2501122178508568, 0.018813180734788815, 0.10514764027114845, 0.04458466119340363, 0.1655817619466002, 0.048708098066782, 0.1685454821993409, 0.05734154402041797, 0.12151253036236886, 0.01958632514854726, 0.21327025404689262, 0.024907474925184542, 0.015567171828240339, 0.053317563511723155, 0.09924072040503215, 0.09612728603938409, 0.4895875539981586, 0.007394406618414161, 0.9982620500891897, 0.9988309786368648, 0.552136906886126, 0.41301579648962183, 0.03415920121342737, 0.07292031600761624, 0.9253690806036935, 0.9967441573555226, 0.9977029299568113, 0.9977142800708015, 0.9889193031702627, 0.0110229858633628, 0.13963969741941562, 0.20065228916208533, 0.008169049940441134, 0.0839880447001604, 0.0025528281063878543, 0.014295837395771983, 0.0020422624851102835, 0.0033186765383042107, 0.5452840835244457, 0.9944074824343605, 0.9893919443113605, 0.9964343371848293, 0.9982714775812576, 0.116181141254447, 0.24446448472289892, 0.5803556061715984, 0.05478997002340399, 0.004180760764838055, 0.06509947886104295, 0.07704313324885526, 0.07048068578302431, 0.6899757265574652, 0.008137434857630368, 0.04934960494304868, 0.0111561606919126, 0.011943654387812313, 0.016799865512527212, 0.9971061675259935, 0.0038193910998721927, 0.1572894698401912, 0.8239468254542466, 0.011458173299616577, 0.0034721737271565384, 0.9800177857729004, 0.01682433966992104, 0.9980263759809355, 0.009792481515523425, 0.9812066478554471, 0.007833985212418739, 0.001240452916650815, 0.21984208736461489, 0.02354981067520411, 0.16977289691070926, 0.47268774020860604, 0.0443743838820087, 0.06852562627437685, 0.08825823481930911, 0.22755737652207408, 0.06273778137758118, 0.02871051012194393, 0.02126704453477328, 0.040407384616069235, 0.012760226720863967, 0.3051820890739966, 0.004253408906954656, 0.20841703644077814, 0.47851558807981204, 0.5193192428773153, 0.5215177847269523, 0.4758070625596497, 0.9941779585772116, 0.9749114979995515, 0.9937890093653187, 0.09289465251885341, 0.4468985986042137, 0.036404661122253365, 0.08159665423953341, 0.020085330274346686, 0.3219929509606203, 0.9956124362557444, 0.9946197823701349, 0.08541473665315188, 0.08727157875430736, 0.10583999976586211, 0.03589894728900586, 0.08789052612135918, 0.008046315771673728, 0.012378947341036504, 0.5236294725258441, 0.05384842093350879, 0.05782854384373463, 0.9404747393533683, 0.9941383741172662, 0.9824132698457241, 0.011775086743927392, 0.9867522691411155, 0.9949700431435539, 0.025523473224534757, 0.17847663997450408, 0.2230488818188203, 0.13953457604736463, 0.02496045543281708, 0.291267870915279, 0.007694576486808272, 0.06868817058955677, 0.040818789899531686, 0.9643785446397607, 0.9946788169290728, 0.09217716684041319, 0.09217716684041319, 0.8019413515115947, 0.9941383788204556, 0.025125759032512938, 0.867776214943805, 0.006937709583604319, 0.07425224311100838, 0.011812856858569516, 0.010312811543195608, 0.00337510195959129, 0.3393523818092703, 0.6587428588062306, 0.09851149817347446, 0.00634897389475615, 0.005734557066231362, 0.742420334467453, 0.14090625934168488, 0.002457667314099155, 0.0032768897521322066, 0.9845451718365453, 0.9850221774710732, 0.06447651347181745, 0.9345013661421642, 0.9785262629265358, 0.9963751919234557, 0.0038366545249352087, 0.2732365266019075, 0.09491549672557104, 0.1726494536220844, 0.41652723255492247, 0.03669843458633678, 0.0021685438619199007, 0.99703740600732, 0.9858678705926408, 0.044763882708187035, 0.081134537408589, 0.06294921005838802, 0.026578555357986054, 0.7679803627123338, 0.012589842011677604, 0.004196614003892535, 0.9956209402096885, 0.07048120999301306, 0.6723747248765279, 0.009010381959334056, 0.10512112285889731, 0.029834375820906096, 0.023627223804475968, 0.05946852093160477, 0.0012013842612445408, 0.009410843379748903, 0.019422378890120075, 0.9989622785144321, 0.9598402941568127, 0.09255167294941473, 0.02089876485954526, 0.07762398376402525, 0.03284091620785684, 0.3313946999156463, 0.4418595998875284, 0.4318309267217648, 0.09684397277603883, 0.12738319696094053, 0.18739119829745873, 0.06649365100877568, 0.09004348574105041, 0.09289014260529804, 0.016587525465231793, 0.1426527190009934, 0.1360177088149007, 0.6071034320274836, 0.9654328064761415, 0.9953797277557586, 0.017499821625693912, 0.9799900110388591, 0.05005606909004734, 0.012871560623155032, 0.06149745631062959, 0.0986819647775219, 0.7722936373893019, 0.004290520207718343, 0.992206443327741, 0.02619286858108825, 0.09268245805615843, 0.010074180223495481, 0.8603349910865141, 0.008059344178796385, 0.02717064152265211, 0.5593529401463314, 0.014491008812081126, 0.39741591667132486, 0.0010868256609060844, 0.06760052349520579, 0.0032088856089496414, 0.019467239360961158, 0.9098260329908551, 0.41315658659520704, 0.05237196168108258, 0.06594987767247436, 0.00096985114224227, 0.4674682505607741, 0.009850181131407756, 0.07432409399153125, 0.8484587838189863, 0.011193387649326996, 0.05596693824663498, 0.9979556214718178, 0.964100907186369, 0.997948201322695, 0.9944104143243289, 0.9967425611598262, 0.27781638623161214, 0.7166857499887965, 0.9940199109333251, 0.9991506659001812, 0.9579823064338738, 0.9809717186859475, 0.844861650071405, 0.11687417902511815, 0.019809182885613247, 0.01782826459705192, 0.9918131220408175, 0.9997874872902723, 0.9992605186127244, 0.9990280260317009, 0.9988092348572613, 0.9579619045213733, 0.12041333536732222, 0.689736035042644, 0.18449294658618962, 0.003168771983350585, 0.0021125146555670566, 0.007359337248746491, 0.22303511628683248, 0.23215999525083644, 0.17118762393580614, 0.1583218942087814, 0.13131435014865472, 0.07649515392046234, 0.0001223642772950723, 0.022771695223567024, 0.7876327524386711, 0.040185344512177104, 0.11251896463409589, 0.03616681006095939, 0.992785553651459, 0.11116658672694697, 0.888594862487742, 0.9991979459590731, 0.9925639451727768, 0.9937396049297635, 0.9827620168766249, 0.984071827369115, 0.05440250472431322, 0.022255570114491773, 0.8593122905317656, 0.012364205619162094, 0.018546308428743143, 0.032146934609821445, 0.9673817643253316, 0.032654236770475326, 0.03649060379878905, 0.0021785435103754656, 0.8131413652476426, 0.03812451143157065, 0.08605246865983089, 0.004357087020750931, 0.016339076327815994, 0.002723179387969332, 0.0010892717551877328, 0.05040407706554798, 0.8945275286115643, 0.014483930191249422, 0.020856859475399166, 0.015063287398899398, 0.0046348576611998146, 0.22948849462359147, 0.4460947160926911, 0.19120974010738698, 0.08373477550419738, 0.048768605513433635, 0.0005520974209067958, 0.015634167354475332, 0.8456347409065101, 0.005211389118158444, 0.022235260237476028, 0.07886568865479779, 0.01320218576600139, 0.018761000825370397, 0.9940077491074145, 0.01190451288929727, 0.7287004859509238, 0.11688067200400956, 0.038238738371682145, 0.06962336326164767, 0.0014429712593087602, 0.03282759614927429, 0.06771313730341164, 0.012897740438745073, 0.7035717409335438, 0.11801432501451742, 0.09737794031252531, 0.11021512515708341, 0.16031290931939407, 0.7314276487697354, 0.623855800677709, 0.016471440281919707, 0.0452964607752792, 0.31295736535647445, 0.9970111185383985, 0.029670647300341007, 0.9700019309726867, 0.9964507895543159, 0.026382394073333777, 0.16443583978166396, 0.2725604056559827, 0.1236080037075212, 0.03684885204996783, 0.26148845011045246, 0.0036329854133771104, 0.07162171243514874, 0.039357341978252025, 0.9795740511108736, 0.016556181145535892, 0.23593192470355712, 0.044422328919079795, 0.007897302918947519, 0.2773927650280316, 0.013820280108158159, 0.3129306281632954, 0.014807442973026597, 0.09378047216250178, 0.9815238097949635, 0.99945365591714, 0.9934978979867188, 0.0940078444273533, 0.00408729758379797, 0.01760682036097587, 0.8309790395367719, 0.01760682036097587, 0.029554305605923783, 0.005659335116027958, 0.9997928133186067, 0.9995982625007109, 0.9993259782213301, 0.9992809058031272, 0.9989043819380194, 0.9981141409389278, 0.9976782293242225, 0.9952935776470936, 0.9998483743129509, 0.48113897741738754, 0.5100073160624308, 0.11191832842830021, 0.01606002919150496, 0.041153824803231466, 0.1701359342475057, 0.5575841384925629, 0.042157576627700526, 0.046172583925576764, 0.005520635034579831, 0.009033766420221542, 0.2776682133738543, 0.05921454694428017, 0.6625627685116754, 0.9997590259758236, 0.9999392268181722, 0.9981569091317802, 0.9444754668675069, 0.17658972953576016, 0.05341682210156446, 0.7057304849418458, 0.02953636045615917, 0.006912765213143636, 0.00628433201194876, 0.018224562834651403, 0.0025137328047795037, 0.9983461153299164, 0.1329955313626513, 0.8561587331470678, 0.9807187764803422, 0.26844759012947905, 0.08094189810845465, 0.0064054739510287844, 0.05415537067687973, 0.01222863208832768, 0.011646316274597791, 0.5630993918768031, 0.002329263254919558, 0.9964940430267426, 0.9436744345409468, 0.9924437970434685, 0.0966344486715132, 0.010267410171348279, 0.8769576216939824, 0.016307063213317853, 0.997825654846087, 0.04549240105158824, 0.07108187664310663, 0.06128837363894527, 0.06128837363894527, 0.5500157654917717, 0.01579597258735703, 0.019587006008322715, 0.04328096488935826, 0.09856686894510786, 0.03411930078869118, 0.02676762594892448, 0.00390987794759571, 0.3582049719681916, 0.006015196842454939, 0.588286251192093, 0.016541791316751084, 0.9980683457028904, 0.3231266955900469, 0.0724293461306814, 0.010413843780526375, 0.4022294028372289, 0.09835769186995115, 0.09342705563100805, 0.13013207205300895, 0.019671127170803677, 0.012105309028186878, 0.052960726998317595, 0.022697454427850396, 0.36013294358855963, 0.030263272570467195, 0.0862503268258315, 0.004539490885570079, 0.13769789019562576, 0.14375054470971918, 0.998243764580962, 0.9692564595029446, 0.9850675249716974, 0.32892068526145274, 0.16091083883294088, 0.010648511393356383, 0.017747518988927305, 0.47681667683584694, 0.004141087764083038, 0.9995729203207735, 0.9994272511114545, 0.9893470316248999, 0.053741611145314706, 0.9447209538176375, 0.17012792208182348, 0.8187406250187755, 0.27885442147097883, 0.7170542266396598, 0.23434903340997265, 0.13510711859614868, 0.1709723223784734, 0.32557139955197245, 0.007574018190056151, 0.0799727214773576, 0.04633517010387292, 0.06702368883935847, 0.0042711174260375495, 0.146532182462519, 0.015770279726907876, 0.0062424023919010335, 0.007228044874832776, 0.7530308569598511, 0.02680557966798486, 0.2634605544510512, 0.6915839554340094, 0.01838096891518962, 0.9999249480515547, 0.9998543270996777, 0.5545834346801793, 0.054909250958433596, 0.2416007042171078, 0.1482549775877707, 0.9979048919530693, 0.9994515455644128, 0.24776363631339585, 0.17618747471174814, 0.5726092928131815, 0.015545184482713387, 0.16164044907617617, 0.05061394189395307, 0.11765715459191128, 0.5725932407566277, 0.051940071375890705, 0.02998526106381208, 0.9971977235211248, 0.12993219163425357, 0.5741847371084186, 0.08334344931695735, 0.06330660235727587, 0.1288258626610196, 0.010694513407928156, 0.009588184434694209, 0.9978535323518013, 0.990778213993893, 0.04186784199594711, 0.04721267288904674, 0.9095120569757872, 0.04434251272653232, 0.9422783954388119, 0.1197782675614975, 0.022022013318481602, 0.0719743849921106, 0.012085251211361855, 0.10178467131346985, 0.008056834140907904, 0.6536778099656613, 0.010205323245150012, 0.9851080265411265, 0.9973533757175365, 0.9998954486535682, 0.020968836586235537, 0.178596642648282, 0.5112057747058112, 0.28874329000356524, 0.009105291295098407, 0.2878010316112862, 0.0001230444769607893, 0.01808753811323603, 0.34206364595099426, 0.1829671372406937, 0.15989629781054568, 0.9935881823486892, 0.9820578515920932, 0.995024115659527, 0.9818424842653123, 0.9989108357113979, 0.9981574694027798, 0.0795745564456046, 0.34888098441636345, 0.009204519588857249, 0.5543496152385963, 0.008016839641907926, 0.9977040648129882, 0.0337029229800593, 0.9654015049176987, 0.9977525003561509, 0.9999511162502079, 0.9997630183213282, 0.9989787051466974, 0.10254355502622899, 0.038040351058117206, 0.32637518588993314, 0.4972811109336481, 0.02205237742499548, 0.013782735890622176, 0.02328601478768147, 0.9683770287566845, 0.007226694244452869, 0.022961081366524135, 0.81942359126783, 0.016503277232189223, 0.00861040551244655, 0.13202621785751378, 0.05348231499393023, 0.19164496206158332, 0.026741157496965116, 0.49322579383291215, 0.03714049652356266, 0.032683636940735145, 0.031198017079792636, 0.13222016762388308, 0.9967922480006789, 0.9916479647292948, 0.24951664562081236, 0.059667023952802954, 0.6888829129096341, 0.021568615670051763, 0.25986034071706593, 0.281567216743977, 0.10908742156199257, 0.026960769587564704, 0.19798883102252643, 0.07459146252559568, 0.028343373156157765, 0.9893768165590455, 0.10636142818500993, 0.7138160019903074, 0.13309397965396966, 0.007394109980776092, 0.020475996869841487, 0.003412666144973581, 0.015356997652381113, 0.5890896787416156, 0.4069556717876861, 0.9695068991926887, 0.9431740330477599, 0.05464232433274176, 0.9799894607919318, 0.9886593702237234, 0.07643530640591911, 0.020565553293072407, 0.12030848676447359, 0.22142245712207959, 0.1521850943687358, 0.01610968341290672, 0.05929734532835878, 0.04078834736459361, 0.2927163752047306, 0.9892625651391362, 0.0014223159343170937, 0.0016733128639024632, 0.007529907887561084, 0.010248161096680083, 0.34930351906994084, 0.00043302089140901756, 0.023022277393246103, 0.42529868551222344, 0.0036806775769766494, 0.18793106687151362, 0.01001227915785713, 0.014365444009099362, 0.029166204503322947, 0.014800760494223585, 0.022201140741335376, 0.8423373987153717, 0.061814940887639676, 0.0021765824256211154, 0.0026118989107453386, 0.09696389448646238, 0.052329720833963825, 0.013082430208490956, 0.8372755333434212, 0.9944430983983383, 0.9973470915160121, 0.986007450849773, 0.013440002427939905, 0.9572252413951842, 0.9979804116054128, 0.02343696845994856, 0.1967547969477163, 0.10004113388922488, 0.11638914275326308, 0.28645183673270463, 0.21172841568601677, 0.0646686722320803, 0.0005786905792579892, 0.06271413176325133, 0.8964431775570632, 0.038735199030243465, 0.0009004184967826411, 0.04376033894363636, 0.11147180990169098, 0.07743599072330715, 0.5575391332078115, 0.008644017569113356, 0.06428988067028057, 0.007743599072330714, 0.10444854562678638, 0.023590964615705198, 0.04842447040697578, 0.007883053322065825, 0.9425879472241564, 0.984095629952177, 0.011856573854845506, 0.999323774371595, 0.9842092054002938, 0.93181457924619, 0.037468205640936635, 0.0474597271451864, 0.03247244488881175, 0.3571968937769292, 0.5245548789731128, 0.9279212389575268, 0.07045991389767578, 0.989557766488035, 0.9969770686821928, 0.051827824414411555, 0.9453395173188668, 0.11178412267745695, 0.15840151881587242, 0.5240185838176857, 0.1422928897837498, 0.0009762805474013708, 0.0073221041055102805, 0.05540392106502779, 0.9903929067878977, 0.9492506510227682, 0.9998331405357257, 0.1909697834433192, 0.1726003776424718, 0.1906816359013451, 0.2860584722947644, 0.10358904133968048, 0.05611673379945139, 0.9976099429578817, 0.10251948891630379, 0.17217883831140604, 0.28247280818698467, 0.11081226860619692, 0.005804945782925189, 0.294082699752835, 0.027780811961141978, 0.003317111875957251, 0.0012439169534839692, 0.023911181199400475, 0.9743806338755694, 0.5933826871227801, 0.4033312926706441, 0.9540981184644587, 0.019895543017473358, 0.04310700987119227, 0.9350905218212477, 0.9969766812367389, 0.9940927507824465, 0.020122303686785754, 0.9798687012695672, 0.9969619910442503, 0.004901185333684267, 0.992490030071064, 0.25252347026221883, 0.14870322844884698, 0.5861990072693877, 0.01224081453694777, 0.9925959389313168, 0.19341607676421457, 0.21964198547800637, 0.18685959958576662, 0.036060624481463736, 0.11473835062283914, 0.23931141701335024, 0.009834715767671927, 0.02409152749125532, 0.4857277086839859, 0.4899791547118545, 0.9967385129986256, 0.0055807785186003465, 0.023319681667008593, 0.14230985222430884, 0.13174623574267247, 0.1606466959282814, 0.012756065185372222, 0.11799360296469305, 0.053814650000789055, 0.1678219825950533, 0.04325103351915269, 0.14091465759465877, 0.15225325281149094, 0.04350092937471169, 0.22765486372765786, 0.1619201260058713, 0.34800743499769354, 0.01643368443044664, 0.04350092937471169, 0.006283467576347245, 0.9955038378969349, 0.006203586396748579, 0.005583227757073721, 0.3546900522341, 0.0037221518380491474, 0.004962869117398863, 0.2741985187362872, 0.10235917554635156, 0.24814345586994316, 0.018582896793804343, 0.24339951094630002, 0.28384463926222714, 0.016032303116223355, 0.06449358299026213, 0.07943277453037935, 0.2346546183374509, 0.059392395635100154, 0.999746718728294, 0.998729394062261, 0.9800162970821746, 0.9748885992570387, 0.9957831835408771, 0.9968826586855647, 0.9998931433138603, 0.07955064837156667, 0.26634214130303296, 0.22973476329133855, 0.16387277122756358, 0.11310741154895321, 0.06304603990902924, 0.037780691409505116, 0.04654143144649181, 0.9795036545541185, 0.9831874046553619, 0.9959220286388105, 0.9448831071287673, 0.9913254728226147, 0.06421044209242946, 0.9203496699914889, 0.007134493565825495, 0.996574956371173, 0.021071966927612054, 0.09833584566218959, 0.8779986219838356, 0.03989477843204164, 0.04589399323385241, 0.2384687883719782, 0.07589006724290627, 0.5336301566210683, 0.03449548511041194, 0.013498233304074238, 0.00029996074009053865, 0.017997644405432316, 0.11249831472885194, 0.022018018784851757, 0.10217736842345268, 0.5356571132502216, 0.0017201577175665434, 0.03371509126430425, 0.032682996633764325, 0.12522748183884436, 0.03405912280781756, 0.17042619867863937, 0.10626309877981213, 0.0056283420963883545, 0.03917326099086294, 0.6783277894567245, 0.996419435376883, 0.06497445043904608, 0.007425651478748123, 0.9244936091041412, 0.9925599618740572, 0.0066540332639601595, 0.02581290830338243, 0.7720166758899377, 0.1253769831878575, 0.07296097551058095, 0.003160764282046828, 0.013042816609182353, 0.20682180623132015, 0.17729630929185242, 0.1783712667046971, 0.1651851224404688, 0.1569437822753261, 0.1022642818752924, 0.015786869000170022, 0.9496408890871507, 0.009714996307796937, 0.012143745384746172, 0.012143745384746172, 0.999095426148304, 0.9972912139820462, 0.9999573610746377, 0.9999570333499812, 0.16267715566970742, 0.05401389934345754, 0.18968410534143618, 0.049883424687781376, 0.012073695147361098, 0.002224101737671781, 0.02097010209804822, 0.4978810604188115, 0.01048505104902411, 0.9937701782938401, 0.9922612958915368, 0.04318370055638615, 0.03409239517609433, 0.8826142306699977, 0.009848914161982806, 0.006818479035218866, 0.00303043512676394, 0.006818479035218866, 0.012879349288746747, 0.023339215266773745, 0.08120768051726755, 0.11829465272200393, 0.5422371108554558, 0.02397864582202782, 0.10582575689454946, 0.01758434026948707, 0.04795729164405564, 0.03964469442575266, 0.03861931577642442, 0.9510824715787235, 0.010473034786826962, 0.09067779053442246, 0.038062282446547704, 0.05298866771970366, 0.033211207232772016, 0.05186918882421697, 0.48100276542745085, 0.0735124474702931, 0.12426215739902338, 0.05448130624701926, 0.11761598297025481, 0.061548148794470264, 0.09569472449551199, 0.5528794324927928, 0.07398424735225706, 0.04531798626990105, 0.021921258474742834, 0.030984855728723042, 0.056239702740099616, 0.04274217408247571, 0.026995057315247817, 0.8615922459783262, 0.011247940548019924, 0.1490502230626359, 0.03778205907798122, 0.04930558709676549, 0.04004898262266009, 0.7237153416387302, 0.9871911091640766, 0.9999705831258947, 0.9998180974129931, 0.9926129090246528, 0.9729217485869185, 0.01858761780864827, 0.00814799684762664, 0.9962885263436005, 0.9982775141359826, 0.9464848300014417, 0.08374059377417806, 0.04160016593943039, 0.12858233108551212, 0.22258790087071845, 0.052945665741093224, 0.47002784892603167, 0.9976663620675889, 0.18729138165644393, 0.12363679442680287, 0.11843425604745719, 0.45945554628574264, 0.0733455900931281, 0.037947927002286025, 0.9538257650617415, 0.9966395128958482, 0.9994822962352816, 0.03276171038300047, 0.7079507780035647, 0.14385369195444753, 0.10558205755248788, 0.009679596249522866, 0.9981460290536176], \"Term\": [\"+ok\", \"-sarun\", \"...\", \"...\", \"...\", \"...\", \"...\", \"...\", \"...\", \"...\", \"...\", \"//////////////////////////////////////////////////////////////////+piiiiiiii\", \"////////////////////////////////////////////////////////////////////////////\", \"//clickthru.online.com/click\", \"//docs.yahoo.com/info/terms/\", \"//example.sourceforge.net/lists/listinfo/razor-users\", \"//example.sourceforge.net/lists/listinfo/spamassassin-devel\", \"//example.sourceforge.net/lists/listinfo/spamassassin-talk\", \"//iiu.taint.org/mailman/listinfo/iiu\", \"//listman.redhat.com/mailman/listinfo/exmh-users\", \"//listman.redhat.com/mailman/listinfo/exmh-workers\", \"//listman.spamassassin.taint.org/mailman/listinfo/exmh-users\", \"//listman.spamassassin.taint.org/mailman/listinfo/exmh-workers\", \"//lists.freshrpms.net/mailman/listinfo/rpm-list\", \"//lists.freshrpms.net/mailman/listinfo/rpm-zzzlist\", \"//lists.freshrpms.net/pipermail/rpm-zzzlist/\", \"//lists.sourceforge.net/lists/listinfo/spamassassin-devel\", \"//thinkgeek.com/sf\", \"//use.perl.org/article.pl\", \"//use.perl.org/comments.pl\", \"//use.perl.org/my/messages/\", \"//www.geocrawler.com/redir-sf.php3\", \"//www.hughes-family.org/bugzilla/show_bug.cgi\", \"//www.linux.ie/mailman/listinfo/ilug\", \"//xent.com/mailman/listinfo/fork\", \"//xent.com/pipermail/fork/\", \"/blockquote\", \"/center\", \"/font\", \"/script\", \"/select\", \"/span\", \"/table\", \"/td\", \"/title\", \"/title\", \"/tr\", \"0.01\", \"0.01\", \"0.01\", \"0.2\", \"0.99\", \"0200a8c0\", \"07/13/2001\", \"0xdeadbeef\", \"1.1\", \"10.\", \"12.\", \"1997-2002\", \"2.0\", \"2.0\", \"2.0.1\", \"2.0.1\", \"2.0.9-sf.net\", \"2fwww\", \"3rd\", \"5,000\", \"5,000\", \"7bit\", \"7bit\", \"7bit\", \"7bit\", \"7bit\", \"7bit\", \"7bit\", \"8.12.3/8.12.3/debian\", \"=20\", \"=20\", \"=20\", \"======================================================================\", \"_______________________________\", \"adam\", \"address\", \"address\", \"address\", \"address\", \"address\", \"address\", \"address\", \"admin\", \"admin\", \"admin\", \"admin\", \"adult\", \"adult\", \"airport\", \"alarms\", \"alb\", \"align\", \"align\", \"align=\", \"align=3d\", \"also\", \"also\", \"also\", \"also\", \"also\", \"also\", \"also\", \"also\", \"also\", \"also\", \"alter\", \"alter\", \"alter\", \"amazing\", \"amazing\", \"america\", \"america\", \"america\", \"american\", \"american\", \"american\", \"announcements\", \"antiabuse\", \"anyhow\", \"aol\", \"aol\", \"aol\", \"aol\", \"aol.com\", \"aol.com\", \"archives\", \"archives\", \"archives\", \"areas\", \"argote\", \"arial\", \"array\", \"article\", \"article\", \"article\", \"article\", \"article\", \"article\", \"asmtp\", \"asmtp\", \"asmtp\", \"assignedto\", \"assignee\", \"assist\", \"assistance\", \"associates\", \"att\", \"att\", \"audio\", \"audio\", \"aug\", \"aug\", \"aug\", \"aug\", \"aug\", \"aug\", \"austin\", \"austin\", \"austin\", \"austin\", \"auth02\", \"award\", \"away\", \"away\", \"away\", \"away\", \"away\", \"awl\", \"backup\", \"backup\", \"backup\", \"backup\", \"backup\", \"bangkok\", \"bank\", \"bank\", \"bank\", \"banned\", \"banner\", \"barrera\", \"base64\", \"based\", \"based\", \"based\", \"based\", \"based\", \"based\", \"based\", \"based\", \"based\", \"baseline\", \"beberg\", \"beenthere\", \"beenthere\", \"beenthere\", \"behalf\", \"behalf\", \"behalf\", \"behalf\", \"behalf\", \"belphegore\", \"bert\", \"bert\", \"bgcolor=\", \"billion\", \"billion\", \"bitbitch\", \"blockquote\", \"blog\", \"blog\", \"blog\", \"boing\", \"boingboing\", \"book\", \"book\", \"book\", \"book\", \"book\", \"book\", \"book\", \"border=\", \"bottle\", \"bottle\", \"breast\", \"brent\", \"broadcast\", \"broker\", \"btinternet\", \"bug\", \"bug\", \"bug\", \"bug\", \"bug\", \"bug\", \"bug\", \"bugtraq\", \"bugzilla\", \"bugzilla\", \"bugzilla\", \"build\", \"build\", \"build\", \"build\", \"build\", \"build\", \"build\", \"build\", \"build\", \"build\", \"bulkmail\", \"bulletin\", \"burn\", \"burn\", \"burn\", \"burner\", \"business\", \"business\", \"business\", \"business\", \"business\", \"business\", \"cache\", \"cache\", \"cache\", \"campus\", \"campus\", \"capital\", \"capital\", \"catalog\", \"catalog\", \"catalog\", \"catalog\", \"cats\", \"cbc3\", \"cc0000\", \"cdt\", \"cdt\", \"cdt\", \"cdt\", \"cdt\", \"center\", \"center\", \"center\", \"center\", \"center\", \"cgi\", \"cgi\", \"cgi\", \"change\", \"change\", \"change\", \"change\", \"change\", \"change\", \"change\", \"change\", \"change\", \"change\", \"charset\", \"charset\", \"charset\", \"charset\", \"charset\", \"charset\", \"chemicals\", \"chicago\", \"chinese\", \"chinese\", \"chinese\", \"chris\", \"chris\", \"chris\", \"chris\", \"cigarettes\", \"cipher\", \"clean\", \"clean\", \"clean\", \"clean\", \"clean\", \"clean\", \"click\", \"click\", \"click\", \"click\", \"co2\", \"code\", \"code\", \"code\", \"code\", \"code\", \"code\", \"code\", \"code\", \"code\", \"color=\", \"color=3d\", \"com\", \"com\", \"com\", \"com\", \"com\", \"com\", \"com\", \"comcast\", \"comcast\", \"commissions\", \"community\", \"community\", \"community\", \"companies\", \"companies\", \"companies\", \"companies\", \"companies\", \"company\", \"company\", \"company\", \"company\", \"company\", \"comparisons\", \"complement\", \"compliance\", \"consolidation\", \"content\", \"content\", \"content\", \"content\", \"content\", \"content\", \"content\", \"convert\", \"copy\", \"copy\", \"copy\", \"copy\", \"copy\", \"copy\", \"copy\", \"copyright\", \"copyright\", \"copyright\", \"copyright\", \"copyright\", \"coral\", \"corp\", \"corp\", \"corp\", \"could\", \"could\", \"could\", \"could\", \"could\", \"could\", \"could\", \"could\", \"could\", \"cpu59\", \"cpunks\", \"crackmice\", \"craig\", \"craig\", \"craig\", \"crank\", \"credit\", \"credit\", \"credit\", \"credit\", \"ctyme\", \"currency\", \"cwg-exmh\", \"cypherpunks\", \"d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3d3\", \"daily\", \"daily\", \"daily\", \"daily\", \"daily\", \"daily\", \"daily\", \"dakota\", \"damian\", \"data\", \"data\", \"data\", \"data\", \"data\", \"data\", \"data/nice/002\", \"databases\", \"datapower\", \"date\", \"date\", \"date\", \"date\", \"date\", \"date\", \"date\", \"date\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"deadbeef\", \"debian\", \"debian\", \"debian\", \"debian\", \"debian\", \"decide\", \"decide\", \"deepeddy.com\", \"deepeddy.vircio.com\", \"dekalb\", \"delivered\", \"delivered\", \"delivered\", \"delivered\", \"delivered\", \"delivered\", \"des\", \"des\", \"des\", \"design\", \"design\", \"design\", \"design\", \"design\", \"desktop\", \"desktop\", \"desktop\", \"desktop\", \"desktop\", \"devel\", \"devel\", \"devel\", \"develop\", \"developers\", \"developers\", \"developers\", \"developers\", \"developers\", \"development\", \"development\", \"development\", \"dhcp\", \"digital\", \"digital\", \"digital\", \"digital\", \"digital\", \"digital\", \"dinosaurs\", \"directions\", \"disc\", \"diveintomark\", \"divx\", \"divx\", \"documents\", \"dogma\", \"dogma\", \"dogma\", \"dogma\", \"dogma\", \"dogma\", \"dogma\", \"dollars\", \"dollars\", \"dollars\", \"domain\", \"domain\", \"domain\", \"domain\", \"domain\", \"domains\", \"download\", \"download\", \"download\", \"download\", \"download\", \"draw\", \"draw\", \"drive\", \"drive\", \"drive\", \"drive\", \"drive\", \"drive\", \"drive\", \"drives\", \"drives\", \"drives\", \"drives\", \"drives\", \"drop\", \"drop\", \"drop\", \"drop\", \"drop\", \"drop\", \"dvd\", \"dvd\", \"e-mails\", \"e-mails\", \"earnings\", \"edh\", \"editors\", \"edt\", \"edt\", \"edt\", \"edt\", \"edt\", \"edt\", \"eff\", \"eff\", \"eff.org\", \"eggs\", \"eggs\", \"egroups\", \"egroups.com\", \"egwn\", \"einstein\", \"eircom\", \"eircom\", \"email\", \"email\", \"email\", \"email\", \"emails\", \"emails\", \"emails\", \"emc\", \"employment\", \"encoding\", \"encoding\", \"encoding\", \"encoding\", \"encoding\", \"encoding\", \"end\", \"end\", \"end\", \"end\", \"end\", \"end\", \"end\", \"end\", \"end\", \"endeavors\", \"envelope-sender\", \"equity\", \"errors\", \"errors\", \"errors\", \"errors\", \"errors\", \"errors\", \"esmtp\", \"esmtp\", \"esmtp\", \"esmtp\", \"esmtp\", \"esmtp\", \"esmtp\", \"estate\", \"estate\", \"eugen\", \"even\", \"even\", \"even\", \"even\", \"even\", \"even\", \"even\", \"even\", \"even\", \"events\", \"events\", \"events\", \"events\", \"ever\", \"ever\", \"ever\", \"ever\", \"ever\", \"ever\", \"ever\", \"ever\", \"evergo\", \"every\", \"every\", \"every\", \"every\", \"every\", \"every\", \"every\", \"example\", \"example\", \"example\", \"example\", \"example\", \"example\", \"example\", \"example\", \"example.com\", \"example.com\", \"example.com\", \"example.com\", \"example.com\", \"example.sourceforge.net\", \"exim\", \"exim\", \"exim\", \"exim\", \"exim\", \"exim\", \"exmh\", \"exmh-users\", \"exmh-users-admin\", \"exmh-users-request\", \"exmh-workers\", \"exmh-workers-admin\", \"exmh-workers-request\", \"express\", \"express\", \"express\", \"extensions\", \"extreme\", \"face=\", \"face=3d\", \"family\", \"family\", \"family\", \"family\", \"family\", \"family\", \"family\", \"fark\", \"fastmail\", \"fastmail\", \"fastmail\", \"fax\", \"fax\", \"fax\", \"fax\", \"fax\", \"fax\", \"featured\", \"feb\", \"feb\", \"federal\", \"fees\", \"fetchmail\", \"fetchmail\", \"fetchmail\", \"fetchmail\", \"fetchmail\", \"fetchmail\", \"ffffff\", \"file\", \"file\", \"file\", \"file\", \"file\", \"file\", \"files\", \"files\", \"files\", \"files\", \"files\", \"files\", \"filetime\", \"filetime\", \"film\", \"film\", \"financial\", \"financial\", \"financially\", \"first\", \"first\", \"first\", \"first\", \"first\", \"first\", \"first\", \"first\", \"fixes\", \"fixes\", \"fixes\", \"fixes\", \"fixes\", \"fixes\", \"folder\", \"folder\", \"folder\", \"folder\", \"font\", \"foreign\", \"foreign\", \"foreign\", \"forged\", \"forged\", \"forged\", \"forged\", \"forged\", \"forged\", \"forged\", \"forged\", \"fork\", \"fork-admin\", \"fork-admin\", \"fork-request\", \"fork.xent.com\", \"forteana\", \"forteana-unsubscribe\", \"forwarded-by\", \"free\", \"free\", \"free\", \"free\", \"free\", \"free\", \"free\", \"free\", \"free\", \"freedom\", \"freedom\", \"freshrpms\", \"freshrpms.net\", \"fri\", \"fri\", \"fri\", \"fri\", \"fri\", \"fri\", \"fri\", \"fri\", \"fri\", \"friends\", \"friends\", \"friends\", \"friends\", \"fsck\", \"ftp\", \"ftp\", \"ftp\", \"ftp\", \"ftp\", \"fund\", \"funds\", \"fw2\", \"gaim\", \"garrigues\", \"gartner\", \"gary\", \"gary\", \"gary\", \"garym\", \"generation\", \"gentle\", \"get\", \"get\", \"get\", \"get\", \"get\", \"get\", \"get\", \"get\", \"get\", \"gkm\", \"glen\", \"gnat\", \"gnu\", \"good\", \"good\", \"good\", \"good\", \"good\", \"good\", \"good\", \"good\", \"good\", \"google\", \"google\", \"google\", \"google\", \"google\", \"government\", \"government\", \"government\", \"grant\", \"grants\", \"graphic\", \"gray\", \"gray\", \"group\", \"group\", \"group\", \"group\", \"group\", \"group\", \"groups\", \"groups\", \"groups\", \"groups\", \"grp\", \"guide\", \"guidelines\", \"guido\", \"habeas\", \"ham\", \"headline\", \"headline\", \"headline\", \"headline\", \"headlines\", \"headlines\", \"headlines\", \"height\", \"height=\", \"height=3d\", \"helo\", \"helo\", \"helo\", \"helo\", \"helo\", \"helo\", \"helo\", \"help\", \"help\", \"help\", \"help\", \"help\", \"help\", \"help\", \"help\", \"help\", \"help\", \"helvetica\", \"hits\", \"hits\", \"hits\", \"homeowner\", \"hormone\", \"hormones\", \"hostname\", \"hotmail\", \"hotmail\", \"hotmail\", \"hotmail.com\", \"hotmail.com\", \"hotmail.com\", \"hotmail.com\", \"href=\", \"href=3d\", \"http\", \"http\", \"http\", \"http\", \"http\", \"http\", \"http\", \"http\", \"http\", \"https\", \"https\", \"https\", \"hughes\", \"hughes\", \"hydrogen\", \"ibm\", \"ibm\", \"ibm\", \"ibm\", \"iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii\", \"iiu\", \"iiu-request\", \"iiu.taint.org\", \"ilug\", \"ilug-admin\", \"ilug.linux.ie\", \"imap\", \"imap\", \"imap\", \"imap\", \"imap\", \"imap\", \"img\", \"img\", \"importance\", \"importance\", \"importance\", \"incredibly\", \"industry\", \"industry\", \"industry\", \"industry\", \"inexpensive\", \"information\", \"information\", \"information\", \"information\", \"information\", \"information\", \"information\", \"instructed\", \"instruction\", \"int\", \"int\", \"interface\", \"interface\", \"interface\", \"internet\", \"internet\", \"internet\", \"internet\", \"internet\", \"internet\", \"internet\", \"internet\", \"inventory\", \"investment\", \"investment\", \"ipod\", \"irish\", \"irish\", \"irs\", \"irs\", \"island\", \"island\", \"island\", \"island\", \"islands\", \"iso\", \"iso\", \"iso\", \"iso\", \"iso\", \"iso\", \"issue\", \"issue\", \"issue\", \"issue\", \"issue\", \"issue\", \"issue\", \"issue\", \"issue\", \"ist\", \"ist\", \"ist\", \"ist\", \"ist\", \"ist\", \"jalapeno\", \"jalapeno\", \"jalapeno\", \"jalapeno\", \"jalapeno\", \"jar\", \"jar\", \"jar\", \"javascript\", \"jinny\", \"jmason\", \"jmason\", \"jmason\", \"jmason\", \"jmason\", \"jmhall\", \"johnhall\", \"jul\", \"jul\", \"jul\", \"jul\", \"jul\", \"july\", \"july\", \"july\", \"july\", \"jun\", \"khare\", \"kingdom\", \"know\", \"know\", \"know\", \"know\", \"know\", \"know\", \"know\", \"know\", \"know\", \"kre\", \"labs\", \"labs\", \"labs\", \"labs\", \"labs\", \"labs\", \"labs\", \"lair\", \"latest\", \"latest\", \"latest\", \"latest\", \"latest\", \"latest\", \"latest\", \"latest\", \"law\", \"law\", \"leitl\", \"les\", \"lesson\", \"lesson\", \"lesson\", \"level\", \"level\", \"level\", \"level\", \"level\", \"lifetime\", \"like\", \"like\", \"like\", \"like\", \"like\", \"like\", \"like\", \"like\", \"like\", \"links\", \"links\", \"links\", \"linux\", \"linux\", \"linux\", \"linux\", \"linux\", \"linux.ie\", \"list\", \"list\", \"list\", \"list\", \"list\", \"list\", \"list\", \"list\", \"list\", \"list1\", \"list2\", \"list=spamassassin-devel\", \"listinfo\", \"listing\", \"listman\", \"listmaster\", \"lists\", \"lists\", \"lists\", \"lists\", \"lists\", \"lists.sourceforge.net\", \"localhost\", \"localhost\", \"localhost\", \"localhost\", \"localhost\", \"localhost\", \"lockergnome\", \"log\", \"log\", \"log\", \"log\", \"log\", \"lookup\", \"louie\", \"lugh\", \"luis\", \"luth\", \"lynch\", \"magnesium\", \"mail\", \"mail\", \"mail\", \"mail\", \"mail\", \"mail\", \"mail\", \"mailer\", \"mailer\", \"mailer\", \"mailer\", \"mailer\", \"mailer\", \"mailer\", \"mailer\", \"maillist\", \"mailman\", \"mailman\", \"mailman\", \"mailto\", \"mailto\", \"mailto\", \"mailto\", \"mailto\", \"mailto\", \"mailto\", \"mailto\", \"maintainer\", \"make\", \"make\", \"make\", \"make\", \"make\", \"make\", \"make\", \"make\", \"mandark\", \"mandark\", \"mar\", \"marc\", \"marc\", \"margin\", \"margin\", \"market\", \"market\", \"market\", \"market\", \"market\", \"marshall\", \"matthias\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"may\", \"maya\", \"maynard\", \"mccready\", \"medication\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"message\", \"messages\", \"messages\", \"messages\", \"messages\", \"messages\", \"messages\", \"messages\", \"method\", \"method\", \"method\", \"method\", \"mgm\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"microsoft\", \"mid\", \"million\", \"million\", \"million\", \"million\", \"million\", \"millionaire\", \"mime\", \"mime\", \"mime\", \"mime\", \"mime\", \"mime\", \"mime\", \"mimeole\", \"mimeole\", \"minder\", \"mindspring\", \"mindspring\", \"mindspring\", \"mindspring\", \"mini\", \"missouri\", \"missouri\", \"mm2\", \"moderator\", \"mon\", \"mon\", \"mon\", \"mon\", \"mon\", \"mon\", \"mon\", \"mon\", \"money\", \"money\", \"money\", \"moon\", \"moon\", \"mortgage\", \"msmail\", \"msmail\", \"msn\", \"msn\", \"msn\", \"much\", \"much\", \"much\", \"much\", \"much\", \"much\", \"much\", \"much\", \"much\", \"much\", \"multipart/alternative\", \"munnari.oz.au\", \"murphy\", \"must\", \"must\", \"must\", \"must\", \"must\", \"must\", \"mx1\", \"n't\", \"n't\", \"n't\", \"n't\", \"n't\", \"n't\", \"n't\", \"n't\", \"n't\", \"naked\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"name\", \"nations\", \"nbsp\", \"net\", \"net\", \"net\", \"net\", \"net\", \"net\", \"net\", \"net\", \"net\", \"netnoteinc\", \"netnoteinc\", \"netnoteinc\", \"netnoteinc\", \"netnoteinc\", \"netnoteinc\", \"new\", \"new\", \"new\", \"new\", \"new\", \"new\", \"new\", \"new\", \"new\", \"new\", \"news\", \"news\", \"news\", \"news\", \"news\", \"news\", \"news\", \"news\", \"news.com\", \"newsisfree\", \"newsletter\", \"newsletter\", \"newsletter\", \"nextpart\", \"nextpart\", \"nigeria\", \"nmh-1.0.4\", \"nnfmp\", \"nonsense\", \"nonsense\", \"normal\", \"normal\", \"normal\", \"normal\", \"normal\", \"normal\", \"normal\", \"normal\", \"normal\", \"noshade\", \"nospam\", \"nov\", \"ns2\", \"oct\", \"oct\", \"oct\", \"oct\", \"oct\", \"one\", \"one\", \"one\", \"one\", \"one\", \"one\", \"one\", \"one\", \"one\", \"oracle\", \"order\", \"order\", \"order\", \"order\", \"order\", \"ordered\", \"ordered\", \"ordering\", \"orders\", \"orders\", \"orders\", \"org\", \"org\", \"org\", \"org\", \"org\", \"org\", \"org\", \"original\", \"original\", \"original\", \"original\", \"original\", \"original\", \"original\", \"original\", \"original\", \"original\", \"originalarrivaltime\", \"originalarrivaltime\", \"osdn\", \"osdn\", \"outblaze\", \"outbound\", \"outgoing\", \"outlook\", \"outlook\", \"outlook\", \"outlook\", \"outlook\", \"outlook\", \"overnight\", \"pacific\", \"page\", \"page\", \"page\", \"page\", \"page\", \"page\", \"page\", \"page\", \"page\", \"palm\", \"palm\", \"panasas\", \"panels\", \"paragraph\", \"paragraph\", \"password\", \"path\", \"path\", \"path\", \"path\", \"path\", \"path\", \"path\", \"path\", \"path\", \"patterns_run_cb\", \"paypal\", \"pcs\", \"pcs\", \"pcs\", \"pda\", \"pdt\", \"pdt\", \"pdt\", \"pdt\", \"pdt\", \"pdt\", \"pdt\", \"penguin\", \"penguin\", \"people\", \"people\", \"people\", \"people\", \"people\", \"people\", \"people\", \"perfectpresence\", \"perkel\", \"perl\", \"perl\", \"perl5\", \"petting\", \"plain\", \"plain\", \"plain\", \"plain\", \"plain\", \"plain\", \"plain\", \"planning\", \"plants\", \"play\", \"play\", \"play\", \"play\", \"play\", \"play\", \"play\", \"plc\", \"please\", \"please\", \"please\", \"please\", \"please\", \"please\", \"please\", \"please\", \"please\", \"please\", \"pop3\", \"porters\", \"posted\", \"posted\", \"posted\", \"posted\", \"posted\", \"posted\", \"postfix\", \"postfix\", \"postfix\", \"postfix\", \"postfix\", \"postfix\", \"preferences\", \"preferences\", \"preferences\", \"preferences\", \"preferences\", \"pregnancy\", \"presented\", \"president\", \"president\", \"press\", \"press\", \"press\", \"press\", \"press\", \"press\", \"pricing\", \"print\", \"print\", \"print\", \"print\", \"print\", \"priority\", \"priority\", \"priority\", \"priority\", \"priority\", \"pro\", \"pro\", \"pro\", \"pro\", \"produced\", \"produced\", \"produced\", \"produced\", \"produced\", \"program\", \"program\", \"program\", \"program\", \"program\", \"promotion\", \"psi\", \"pudge\", \"qmail-ldap-1.03\", \"qmqp\", \"qualify\", \"qualify\", \"quicktopic\", \"quot\", \"rafael\", \"ranging\", \"rates\", \"rates\", \"rates\", \"rates\", \"rating\", \"razor\", \"razor-users\", \"razor-users-admin\", \"razor-users-request\", \"rcvd\", \"receive\", \"receive\", \"receive\", \"receive\", \"receive\", \"received\", \"received\", \"received\", \"received\", \"received\", \"received\", \"received\", \"received\", \"receiving\", \"receiving\", \"receiving\", \"receiving\", \"receiving\", \"recommendation\", \"redhat\", \"redhat\", \"redhat.com\", \"ref\", \"refund\", \"relax\", \"relay1\", \"remember\", \"remember\", \"remember\", \"remember\", \"remember\", \"remember\", \"removal\", \"removal\", \"remove\", \"remove\", \"remove\", \"remove\", \"remove\", \"remove\", \"remove\", \"remove\", \"remove\", \"removed\", \"removed\", \"removed\", \"removed\", \"removed\", \"removed\", \"reply\", \"reply\", \"reply\", \"reply\", \"reply\", \"reply\", \"report\", \"report\", \"report\", \"report\", \"report\", \"report\", \"report\", \"republic\", \"request\", \"request\", \"request\", \"request\", \"request\", \"request\", \"request\", \"required\", \"required\", \"required\", \"required\", \"required\", \"resent\", \"resent\", \"resent\", \"reserved\", \"reserved\", \"reserved\", \"reserved\", \"resource\", \"resources\", \"resources\", \"responded\", \"return\", \"return\", \"return\", \"return\", \"return\", \"return\", \"return\", \"return\", \"return\", \"returns\", \"returns\", \"rights\", \"rights\", \"rights\", \"rights\", \"rights\", \"rights\", \"rights\", \"rights\", \"rod\", \"rohit\", \"rom\", \"root\", \"root\", \"root\", \"root\", \"root\", \"root\", \"root\", \"rpm\", \"rpm-list\", \"rpm-list-request\", \"rpm-zzzlist\", \"rpm-zzzlist-admin\", \"rpm-zzzlist-request\", \"rpm-zzzlist.freshrpms.net\", \"rss\", \"rssfeeds\", \"sadev\", \"sadev\", \"said\", \"said\", \"said\", \"said\", \"said\", \"said\", \"said\", \"said\", \"said\", \"sales\", \"sales\", \"sales\", \"sans\", \"sans-serif\", \"saou\", \"sarun\", \"say\", \"say\", \"say\", \"say\", \"say\", \"say\", \"say\", \"say\", \"scd\", \"scroll\", \"scroll\", \"sean\", \"search\", \"search\", \"search\", \"search\", \"search\", \"search\", \"search\", \"search\", \"secprog\", \"securing\", \"securities\", \"security\", \"security\", \"security\", \"security\", \"securityfocus\", \"see\", \"see\", \"see\", \"see\", \"see\", \"see\", \"see\", \"see\", \"see\", \"see\", \"send\", \"send\", \"send\", \"send\", \"send\", \"send\", \"sentto\", \"sep\", \"sep\", \"sep\", \"sep\", \"sep\", \"sep\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"september\", \"sequences\", \"serenity\", \"servants\", \"services\", \"services\", \"services\", \"services\", \"services\", \"services\", \"sf.net\", \"sha\", \"shopper\", \"shopping\", \"shopping\", \"sid\", \"sid\", \"sincerely\", \"sincerely\", \"single\", \"single\", \"single\", \"single\", \"single\", \"single\", \"single\", \"site\", \"site\", \"site\", \"site\", \"site\", \"site\", \"site\", \"sites\", \"sites\", \"sites\", \"sites\", \"size=\", \"size=3d\", \"skin\", \"skin\", \"skin\", \"skin\", \"skip\", \"slack\", \"slashdot\", \"slashdot\", \"slashdot\", \"slashnull\", \"slashnull\", \"slashnull\", \"slashnull\", \"slashnull\", \"slashnull\", \"slashnull\", \"sleep\", \"smtp\", \"smtp\", \"smtp\", \"smtp\", \"smtp\", \"smtp\", \"smtp\", \"smtp1\", \"smtpd32\", \"social\", \"social\", \"social\", \"socialism\", \"socialism\", \"software\", \"software\", \"software\", \"software\", \"software\", \"software\", \"software\", \"software\", \"sol\", \"sony\", \"sourceforge\", \"spam\", \"spam\", \"spam\", \"spam\", \"spamassassin\", \"spamassassin\", \"spamassassin\", \"spamassassin\", \"spamassassin\", \"spamassassin\", \"spamassassin\", \"spamassassin-devel\", \"spamassassin-devel-admin\", \"spamassassin-devel-request\", \"spamassassin-devel.example.sourceforge.net\", \"spamassassin-talk\", \"spamassassin-talk-request\", \"spamassassin.taint.org\", \"spamassassin.taint.org\", \"spamassassin.taint.org\", \"spamassassin.taint.org\", \"spamassassin.taint.org\", \"spambayes\", \"sponsored\", \"sponsored\", \"sponsorship\", \"src=\", \"src=3d\", \"sslv3\", \"state\", \"state\", \"state\", \"state\", \"state\", \"state\", \"states\", \"states\", \"states\", \"status\", \"status\", \"status\", \"status\", \"status\", \"stop\", \"stop\", \"stop\", \"stop\", \"stop\", \"stop\", \"stop\", \"stop\", \"strategies\", \"streaming\", \"strip\", \"strip\", \"strip\", \"subject\", \"subject\", \"subject\", \"subject\", \"subject\", \"subject\", \"subject\", \"subject\", \"submitting\", \"subscribe\", \"subscribe\", \"subscribe\", \"subscribe\", \"subscribe\", \"subscribe\", \"subscribe\", \"subscribed\", \"subscribed\", \"summaries\", \"supplied\", \"supplied\", \"svanstrom\", \"sync\", \"system\", \"system\", \"system\", \"system\", \"system\", \"system\", \"system\", \"system\", \"system\", \"table\", \"table\", \"table\", \"table\", \"taint\", \"taint\", \"taint\", \"taint\", \"taint\", \"taint\", \"taint\", \"talk\", \"talk\", \"talk\", \"talk\", \"talk\", \"talk\", \"talk\", \"talk\", \"talk\", \"technology\", \"technology\", \"technology\", \"technology\", \"telecommunications\", \"temple\", \"tests\", \"tests\", \"thai\", \"thinkgeek\", \"thu\", \"thu\", \"thu\", \"thu\", \"thu\", \"thu\", \"thu\", \"thu\", \"tim\", \"tim\", \"tim\", \"time\", \"time\", \"time\", \"time\", \"time\", \"time\", \"time\", \"time\", \"time\", \"time\", \"tips\", \"tips\", \"tips\", \"titles\", \"titles\", \"tlsv1\", \"tobacco\", \"toc\", \"tony\", \"tony\", \"tony\", \"tony\", \"tony\", \"total\", \"total\", \"tradesignals\", \"trading\", \"transaction\", \"transaction\", \"transfer\", \"transfer\", \"transfer\", \"transfer\", \"transfer\", \"transfer\", \"transfer\", \"trends\", \"trna\", \"tuatha\", \"tue\", \"tue\", \"tue\", \"tue\", \"tue\", \"tue\", \"txt\", \"type\", \"type\", \"type\", \"type\", \"type\", \"type\", \"type\", \"type\", \"type\", \"u.s.\", \"u.s.\", \"ucsc\", \"ucsc\", \"ulises\", \"ultimate\", \"ultimate\", \"ultimate\", \"underwriting\", \"unison\", \"united\", \"united\", \"unlimited\", \"unseen\", \"unseen\", \"unsubscribe\", \"unsubscribe\", \"unsubscribe\", \"unsubscribe\", \"unsubscribed\", \"updates\", \"updates\", \"updates\", \"updates\", \"updates\", \"updates\", \"updates\", \"url\", \"url\", \"url\", \"usd\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"use\", \"user\", \"user\", \"user\", \"user\", \"user\", \"user\", \"user\", \"user\", \"userland\", \"users\", \"users\", \"users\", \"users\", \"users\", \"users\", \"users\", \"users\", \"using\", \"using\", \"using\", \"using\", \"using\", \"using\", \"using\", \"using\", \"usw\", \"utf\", \"utmost\", \"uucp\", \"var\", \"venture\", \"verdana\", \"version\", \"version\", \"version\", \"version\", \"version\", \"version\", \"version\", \"version\", \"veuve\", \"vip\", \"vircio\", \"virtues\", \"visa\", \"visual\", \"visual\", \"visual\", \"w3c\", \"wake\", \"wake\", \"wake\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"way\", \"way\", \"way\", \"way\", \"way\", \"way\", \"way\", \"way\", \"way\", \"web\", \"web\", \"web\", \"web\", \"web\", \"webdev\", \"weblog\", \"weblog\", \"weblog\", \"webmaster\", \"webmaster\", \"webnote\", \"webnote\", \"webnote\", \"webnote\", \"webnote\", \"wed\", \"wed\", \"wed\", \"wed\", \"wed\", \"wed\", \"wed\", \"weeks\", \"weeks\", \"weeks\", \"weeks\", \"weeks\", \"weight\", \"welch\", \"width=\", \"width=3d\", \"windows\", \"windows\", \"windows\", \"windows\", \"windows\", \"windows\", \"windows\", \"windows\", \"windows\", \"wing\", \"winning\", \"wish\", \"wish\", \"wish\", \"wish\", \"wish\", \"wish\", \"wish\", \"wish\", \"work\", \"work\", \"work\", \"work\", \"work\", \"work\", \"work\", \"work\", \"work\", \"workers\", \"workers\", \"workers\", \"world\", \"world\", \"world\", \"world\", \"world\", \"world\", \"world\", \"world\", \"world\", \"would\", \"would\", \"would\", \"would\", \"would\", \"would\", \"would\", \"would\", \"writing\", \"writing\", \"writing\", \"writing\", \"writing\", \"www\", \"www\", \"www\", \"www\", \"www\", \"x-antiabuse\", \"xent\", \"xent.com\", \"xxx\", \"yahoo\", \"yahoo\", \"yahoo\", \"yahoogroups\", \"yahoogroups.com\", \"yapc\", \"year\", \"year\", \"year\", \"year\", \"year\", \"year\", \"yellow\", \"yyyy\", \"yyyy\", \"yyyy\", \"yyyy\", \"yyyy\", \"yyyy\", \"ziggy\", \"zoo\", \"zzzlist\", \"zzzz\", \"zzzz\", \"zzzz\", \"zzzz\", \"zzzz\", \"zzzzteana\"]}, \"R\": 30, \"lambda.step\": 0.01, \"plot.opts\": {\"xlab\": \"PC1\", \"ylab\": \"PC2\"}, \"topic.order\": [4, 14, 7, 5, 15, 13, 2, 11, 1, 3, 12, 10, 8, 9, 6]};\n",
       "\n",
       "function LDAvis_load_lib(url, callback){\n",
       "  var s = document.createElement('script');\n",
       "  s.src = url;\n",
       "  s.async = true;\n",
       "  s.onreadystatechange = s.onload = callback;\n",
       "  s.onerror = function(){console.warn(\"failed to load library \" + url);};\n",
       "  document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "}\n",
       "\n",
       "if(typeof(LDAvis) !== \"undefined\"){\n",
       "   // already loaded: just create the visualization\n",
       "   !function(LDAvis){\n",
       "       new LDAvis(\"#\" + \"ldavis_el406101403484648204008810904140\", ldavis_el406101403484648204008810904140_data);\n",
       "   }(LDAvis);\n",
       "}else if(typeof define === \"function\" && define.amd){\n",
       "   // require.js is available: use it to load d3/LDAvis\n",
       "   require.config({paths: {d3: \"https://d3js.org/d3.v5\"}});\n",
       "   require([\"d3\"], function(d3){\n",
       "      window.d3 = d3;\n",
       "      LDAvis_load_lib(\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v3.0.0.js\", function(){\n",
       "        new LDAvis(\"#\" + \"ldavis_el406101403484648204008810904140\", ldavis_el406101403484648204008810904140_data);\n",
       "      });\n",
       "    });\n",
       "}else{\n",
       "    // require.js not available: dynamically load d3 & LDAvis\n",
       "    LDAvis_load_lib(\"https://d3js.org/d3.v5.js\", function(){\n",
       "         LDAvis_load_lib(\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v3.0.0.js\", function(){\n",
       "                 new LDAvis(\"#\" + \"ldavis_el406101403484648204008810904140\", ldavis_el406101403484648204008810904140_data);\n",
       "            })\n",
       "         });\n",
       "}\n",
       "</script>"
      ],
      "text/plain": [
       "PreparedData(topic_coordinates=              x         y  topics  cluster       Freq\n",
       "topic                                                \n",
       "3      0.054327 -0.266001       1        1  22.999359\n",
       "13     0.236237  0.108736       2        1  12.349814\n",
       "6      0.216072 -0.006306       3        1  11.366935\n",
       "4      0.221381  0.099358       4        1   9.967681\n",
       "14     0.068512 -0.237734       5        1   8.867432\n",
       "12     0.198047  0.122532       6        1   7.409811\n",
       "1     -0.019707 -0.267274       7        1   6.970847\n",
       "10     0.191727  0.138039       8        1   6.955666\n",
       "0     -0.004467 -0.200983       9        1   5.612169\n",
       "2      0.148325  0.153306      10        1   4.340695\n",
       "11    -0.228310  0.034375      11        1   1.162298\n",
       "9     -0.278319  0.058295      12        1   0.751735\n",
       "7     -0.239746  0.072394      13        1   0.516939\n",
       "8     -0.288558  0.083844      14        1   0.445193\n",
       "5     -0.275521  0.107420      15        1   0.283425, topic_info=              Term          Freq         Total Category  logprob  loglift\n",
       "26            http  89586.000000  89586.000000  Default  30.0000  30.0000\n",
       "3136          font  54670.000000  54670.000000  Default  29.0000  29.0000\n",
       "13667        /font  50294.000000  50294.000000  Default  28.0000  28.0000\n",
       "7              com  92429.000000  92429.000000  Default  27.0000  27.0000\n",
       "275            net  45666.000000  45666.000000  Default  26.0000  26.0000\n",
       "...            ...           ...           ...      ...      ...      ...\n",
       "9745       chinese     24.211231    117.998437  Topic15  -6.1831   4.2821\n",
       "242           list     53.294453  17434.919379  Topic15  -5.3941   0.0756\n",
       "19321       bottle     25.579236    217.545614  Topic15  -6.1282   3.7254\n",
       "542      subscribe     26.821543   1758.156159  Topic15  -6.0808   1.6832\n",
       "771    unsubscribe     26.644350   2205.735568  Topic15  -6.0874   1.4497\n",
       "\n",
       "[947 rows x 6 columns], token_table=       Topic      Freq       Term\n",
       "term                             \n",
       "23902     15  0.922857        +ok\n",
       "23911     15  0.922857     -sarun\n",
       "565        1  0.028281        ...\n",
       "565        2  0.157864        ...\n",
       "565        3  0.092034        ...\n",
       "...      ...       ...        ...\n",
       "2212       3  0.707951       zzzz\n",
       "2212       4  0.143854       zzzz\n",
       "2212       6  0.105582       zzzz\n",
       "2212      10  0.009680       zzzz\n",
       "2582       3  0.998146  zzzzteana\n",
       "\n",
       "[2294 rows x 3 columns], R=30, lambda_step=0.01, plot_opts={'xlab': 'PC1', 'ylab': 'PC2'}, topic_order=[4, 14, 7, 5, 15, 13, 2, 11, 1, 3, 12, 10, 8, 9, 6])"
      ]
     },
     "metadata": {},
     "execution_count": 16
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Clustering\n",
    "\n",
    "Below we attempt to use clustering on our data set to see if we can find any patterns in the data."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "\n",
    "NUM_CLUSTERS = 15\n",
    "km = KMeans(n_clusters=NUM_CLUSTERS, max_iter=10000, n_init=10, random_state=42).fit(tf_matrix)\n",
    "km\n",
    "\n",
    "df['kmeans_cluster'] = km.labels_\n",
    "\n",
    "email_clusters = (df[['directory', 'kmeans_cluster']]\n",
    "                  .sort_values(by=['kmeans_cluster'], \n",
    "                               ascending=False)\n",
    "                  .groupby('kmeans_cluster').head(20))  # top 20 movies for each cluster\n",
    "email_clusters = email_clusters.copy(deep=True)\n",
    "\n",
    "feature_names = tf.get_feature_names()\n",
    "topn_features = 15\n",
    "ordered_centroids = km.cluster_centers_.argsort()[:, ::-1]\n",
    "\n",
    "sample_silhouette_values = silhouette_samples(tf_matrix, km.labels_)\n",
    "\n",
    "# get key features for each cluster\n",
    "for cluster_num in range(NUM_CLUSTERS):\n",
    "\n",
    "    cluster_silhouette_values = sample_silhouette_values[km.labels_ == cluster_num]\n",
    "\n",
    "    key_features = [feature_names[index] \n",
    "                        for index in ordered_centroids[cluster_num, :topn_features]]\n",
    "    print('CLUSTER #'+str(cluster_num+1), \":\", cluster_silhouette_values.mean())\n",
    "    print('Cluster Size', cluster_silhouette_values.shape[0])\n",
    "    print('Key Features:', key_features)\n",
    "    print('-'*80)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "C:\\Users\\dancr\\miniconda3\\envs\\NLP\\lib\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "CLUSTER #1 : 0.08418295404425712\n",
      "Cluster Size 689\n",
      "Key Features: ['oct', 'rssfeeds', 'tue oct', 'oct ist received', 'oct ist', 'tue', 'example com', 'spamassassin taint org', 'spamassassin taint', 'taint org', 'taint', 'example', 'rssfeeds spamassassin', 'rssfeeds spamassassin taint', 'jmason org']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #2 : 0.3515590026855842\n",
      "Cluster Size 658\n",
      "Key Features: ['rpm', 'freshrpms', 'freshrpms net', 'zzzlist', 'rpm zzzlist', 'net', 'rpm list', 'egwn', 'list', 'egwn net', 'zzzlist freshrpms', 'zzzlist freshrpms net', 'rpm zzzlist freshrpms', 'http lists freshrpms', 'lists freshrpms']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #3 : 0.05249950816939316\n",
      "Cluster Size 715\n",
      "Key Features: ['font', 'width', 'td', 'size', 'nbsp', 'height', 'href http', 'src http', 'face', 'color', 'href', 'border', 'img', 'src', 'font face']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #4 : -0.03267184607611213\n",
      "Cluster Size 245\n",
      "Key Features: ['fork', 'xent com', 'xent', 'admin xent', 'admin xent com', 'fork admin', 'fork admin xent', 'sep', 'mailto fork', 'people', 'list', 'report', 'http xent', 'http xent com', 'said']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #5 : 0.17752881393656597\n",
      "Cluster Size 799\n",
      "Key Features: ['ilug', 'linux', 'ie', 'linux ie', 'lugh', 'ilug linux', 'tuatha org', 'tuatha', 'lugh tuatha', 'lugh tuatha org', 'admin linux', 'ilug admin', 'ilug admin linux', 'aug', 'ilug linux ie']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #6 : 0.2620858740274013\n",
      "Cluster Size 203\n",
      "Key Features: ['exmh', 'exmh workers', 'workers', 'redhat com', 'listman', 'redhat', 'deepeddy', 'mx1', 'spamassassin taint org', 'spamassassin taint', 'taint org', 'taint', 'vircio', 'spamassassin', 'vircio com']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #7 : 0.3264143353582599\n",
      "Cluster Size 295\n",
      "Key Features: ['razor', 'razor users', 'sourceforge net', 'sourceforge', 'users', 'net', 'example sourceforge', 'example sourceforge net', 'lists', 'lists sourceforge', 'lists sourceforge net', 'usw', 'example', 'users example sourceforge', 'razor users example']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #8 : -0.013105398992069318\n",
      "Cluster Size 2149\n",
      "Key Features: ['net', 'sep', 'zzzz', 'aug', 'jul', 'netnoteinc', 'netnoteinc com', 'mail', 'mon', 'zzzz localhost', 'webnote net', 'webnote', 'html', 'fri', 'iiu']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #9 : 0.1506731181955181\n",
      "Cluster Size 472\n",
      "Key Features: ['sourceforge net', 'sourceforge', 'spamassassin talk', 'talk', 'net', 'spamassassin', 'example sourceforge', 'example sourceforge net', 'lists', 'usw', 'lists sourceforge', 'lists sourceforge net', 'spamassassin devel', 'example', 'devel']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #10 : 0.24640983160022667\n",
      "Cluster Size 206\n",
      "Key Features: ['exmh users', 'exmh', 'users', 'redhat com', 'listman', 'redhat', 'mx1', 'example com', 'spamassassin taint org', 'spamassassin taint', 'taint org', 'taint', 'spamassassin', 'listman redhat com', 'listman redhat']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #11 : 0.21124597851712237\n",
      "Cluster Size 474\n",
      "Key Features: ['3d', 'font', 'width 3d', 'size 3d', 'width', 'size', 'color 3d', 'td', 'color', 'face 3d', 'face', '3d http', '20', 'font face 3d', 'align']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #12 : 0.24866286706429042\n",
      "Cluster Size 218\n",
      "Key Features: ['yahoo', 'yahoogroups com', 'yahoogroups', 'zzzzteana', 'yahoo com', 'grp scd', 'grp scd yahoo', 'scd', 'scd yahoo com', 'scd yahoo', 'grp', 'zzzzteana yahoogroups', 'zzzzteana yahoogroups com', 'groups', 'forteana']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #13 : 0.4749578394586939\n",
      "Cluster Size 131\n",
      "Key Features: ['perl', 'use perl', 'perl org', 'use', 'pudge', 'use perl org', 'http use perl', 'http use', 'pudge perl', 'pudge perl org', 'cpu59', 'cpu59 osdn', 'cpu59 osdn com', 'osdn com', 'osdn']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #14 : 0.12509472160980775\n",
      "Cluster Size 1591\n",
      "Key Features: ['fork', 'xent com', 'xent', 'fork admin', 'fork admin xent', 'admin xent', 'admin xent com', 'mailto fork', 'sep', 'http xent com', 'http xent', 'list', 'xent com subject', 'fork request', 'fork request xent']\n",
      "--------------------------------------------------------------------------------\n",
      "CLUSTER #15 : 0.05718222656234242\n",
      "Cluster Size 508\n",
      "Key Features: ['sep', 'rssfeeds', 'thu sep', 'thu', 'example com', 'sep ist received', 'sep ist', 'spamassassin taint org', 'spamassassin taint', 'jmason org', 'jmason', 'taint org', 'taint', 'example', 'sep content type']\n",
      "--------------------------------------------------------------------------------\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Output a quick pivot table to see distribution of clusters vs spam/ham\n",
    "\n",
    "Below we do see an interesting case where no clusters have spam, but those do have spam also have a fair amount of non-spam emails as well, with the excpetion of cluster 10."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "source": [
    "df.pivot_table(index='kmeans_cluster', columns='is_spam', values='directory', aggfunc='count')"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "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>is_spam</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>kmeans_cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>689.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>658.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>257.0</td>\n",
       "      <td>458.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>211.0</td>\n",
       "      <td>34.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>691.0</td>\n",
       "      <td>108.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>203.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>295.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>830.0</td>\n",
       "      <td>1319.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>453.0</td>\n",
       "      <td>19.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>206.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>59.0</td>\n",
       "      <td>415.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>218.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>131.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1545.0</td>\n",
       "      <td>46.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>508.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "is_spam              0       1\n",
       "kmeans_cluster                \n",
       "0                689.0     NaN\n",
       "1                658.0     NaN\n",
       "2                257.0   458.0\n",
       "3                211.0    34.0\n",
       "4                691.0   108.0\n",
       "5                203.0     NaN\n",
       "6                295.0     NaN\n",
       "7                830.0  1319.0\n",
       "8                453.0    19.0\n",
       "9                206.0     NaN\n",
       "10                59.0   415.0\n",
       "11               218.0     NaN\n",
       "12               131.0     NaN\n",
       "13              1545.0    46.0\n",
       "14               508.0     NaN"
      ]
     },
     "metadata": {},
     "execution_count": 8
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Elbow to find the best k number\n",
    "\n",
    "Below we do an elbow plot to find the best k number for k-means clustering, confirming our value of 15 above. No need to run the code again below, output is already saved."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [
    "scores = []\n",
    "\n",
    "for k in range(1, 20):\n",
    "    kmeans = KMeans(init=\"random\", n_clusters=k, n_init=10, max_iter=300, random_state=42)\n",
    "    kmeans.fit(tf_matrix)\n",
    "    scores.append(kmeans.inertia_)"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "source": [
    "plt.style.use(\"fivethirtyeight\")\n",
    "plt.figure(figsize=(12,6))\n",
    "plt.plot(range(1, 20), scores)\n",
    "plt.xticks(range(1, 20))\n",
    "plt.xlabel(\"Number of Clusters\")\n",
    "plt.ylabel(\"SSE\")\n",
    "plt.show()"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {}
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Model Building and Evaluation\n",
    "Below we will build two Naive Bayes Classifier models, one with and another without the clusters as features. We use the standard sklearn pipeline approach. Although we do have an inbalanced set, we decided to leave SMOTE out and see how our model performs on the data itself."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "source": [
    "from imblearn.pipeline import Pipeline as imbpipeline\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.calibration import CalibratedClassifierCV\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from imblearn.pipeline import Pipeline as imbpipeline\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.compose import make_column_selector as selector\n",
    "\n",
    "categorical_features = ['in_reply','subj_caps','attachments']\n",
    "numeric_features = ['body_lines']\n",
    "\n",
    "X = df[categorical_features + numeric_features]\n",
    "y = df['is_spam']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                    y,\n",
    "                                                    test_size=0.2,\n",
    "                                                    stratify=y,\n",
    "                                                    random_state=11)\n",
    "\n",
    "numeric_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(missing_values=np.nan, strategy=\"median\")),\n",
    "    ('scaler', RobustScaler(with_centering=False))])\n",
    "\n",
    "categorical_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(missing_values=np.nan, strategy=\"most_frequent\")),\n",
    "    ('onehot', OneHotEncoder(handle_unknown='ignore'))])\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        ('num', numeric_transformer, numeric_features),\n",
    "        ('cat', categorical_transformer, categorical_features)])\n",
    "\n",
    "\n",
    "# pipeline = imbpipeline(steps = [['preprocessor', preprocessor],\n",
    "#                                 #['smote', SMOTE(random_state=11)],\n",
    "#                                 ['classifier', MultinomialNB()]])\n",
    "\n",
    "clfPipeline = Pipeline(steps = [['preprocessor', preprocessor],['classifier', MultinomialNB()]])\n",
    "\n",
    "\n",
    "clfPipeline.fit(X_train, y_train)\n",
    "\n",
    "y_pred = clfPipeline.predict(X_test)\n",
    "\n",
    "print(\"Accurracy:\", accuracy_score(y_test, y_pred))\n",
    "print(\"Recall:\", recall_score(y_test, y_pred))\n",
    "print(\"Precision:\", precision_score(y_test, y_pred))\n",
    "print(\"F1\", f1_score(y_test, y_pred))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Accurracy: 0.7792624265098878\n",
      "Recall: 0.20625\n",
      "Precision: 0.7557251908396947\n",
      "F1 0.32405891980360063\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "source": [
    "from sklearn.metrics import RocCurveDisplay\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "\n",
    "disp = RocCurveDisplay.from_estimator(clfPipeline, X_test, y_test)\n",
    "disp = ConfusionMatrixDisplay.from_estimator(clfPipeline, X_test, y_test)"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "The above results aren't too bad with respect to precision. However, recall is quite low. This means we will show a lot of spam in our inbox, which our clients do not want.\n",
    "\n",
    "Let's create another model but this time use the cluster we created previously to see if it can improve our model."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "source": [
    "categorical_features = ['in_reply','subj_caps','attachments','kmeans_cluster']\n",
    "numeric_features = ['body_lines']\n",
    "\n",
    "X = df[categorical_features + numeric_features]\n",
    "y = df['is_spam']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                    y,\n",
    "                                                    test_size=0.2,\n",
    "                                                    stratify=y,\n",
    "                                                    random_state=11)\n",
    "\n",
    "numeric_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(missing_values=np.nan, strategy=\"median\")),\n",
    "    ('scaler', RobustScaler(with_centering=False))])\n",
    "\n",
    "categorical_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(missing_values=np.nan, strategy=\"most_frequent\")),\n",
    "    ('onehot', OneHotEncoder(handle_unknown='ignore'))])\n",
    "\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        ('num', numeric_transformer, numeric_features),\n",
    "        ('cat', categorical_transformer, categorical_features)])\n",
    "\n",
    "\n",
    "# pipeline = imbpipeline(steps = [['preprocessor', preprocessor],\n",
    "#                                 #['smote', SMOTE(random_state=11)],\n",
    "#                                 ['classifier', MultinomialNB()]])\n",
    "\n",
    "\n",
    "clfPipeline = Pipeline(steps = [['preprocessor', preprocessor],['classifier', MultinomialNB()]])\n",
    "\n",
    "\n",
    "clfPipeline.fit(X_train, y_train)\n",
    "\n",
    "y_pred = clfPipeline.predict(X_test)\n",
    "\n",
    "print(\"Accurracy:\", accuracy_score(y_test, y_pred))\n",
    "print(\"Recall:\", recall_score(y_test, y_pred))\n",
    "print(\"Precision:\", precision_score(y_test, y_pred))\n",
    "print(\"F1\", f1_score(y_test, y_pred))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Accurracy: 0.8792089791555318\n",
      "Recall: 0.9083333333333333\n",
      "Precision: 0.7055016181229773\n",
      "F1 0.7941712204007285\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "source": [
    "disp = RocCurveDisplay.from_estimator(clfPipeline, X_test, y_test)\n",
    "disp = ConfusionMatrixDisplay.from_estimator(clfPipeline, X_test, y_test)"
   ],
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    },
    {
     "output_type": "display_data",
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     }
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "As can be seen above, or model as greatly improved!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Modeling Interpretability and Explainability\n",
    "\n",
    "Above we show 4 different metrics for each model. Which is more imporant here? The model that doesn't use the K-means feature as a feature has a higher precision, but with also a much lower recall. If we include the cluster as a feature, the precision takes a small hit, but the recall is much higher, meaning we won't be showing so many spam emails in our inbox. The F1 score with the cluster as a feature is also much higher, indicating a more balanced model with respect to precision and recall."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Case Conclusions\n",
    "\n",
    "Our recommendation is leverag the clustering found using NLP analysis on our emails to provide a more balanced prediction model. It provides a relatively high accuracy of 70% while maintaining a recall of near 90%"
   ],
   "metadata": {}
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "dc77dc135269c51807c41f729eb5bb02452ccbe9cffcc1fe76852c9e784edfe4"
  },
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.7.11 64-bit ('SMU': conda)"
  },
  "language_info": {
   "name": "python",
   "version": "3.7.11",
   "mimetype": "text/x-python",
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "pygments_lexer": "ipython3",
   "nbconvert_exporter": "python",
   "file_extension": ".py"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}