progress_report_part3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Progress Report #3: [Blog-Sentiment Analysis](https://github.com/Data-Science-for-Linguists-2019/Blog-Sentiment-Analysis)\n",
    "\n",
    "To learn more about the data and see my previous analysis, refer to [Progress Report #2](https://github.com/Data-Science-for-Linguists-2019/Blog-Sentiment-Analysis/blob/master/progress_report_part2.ipynb).\n",
    "\n",
    "# Loading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pretty printing has been turned OFF\n"
     ]
    }
   ],
   "source": [
    "# Import required packages\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import nltk\n",
    "import re\n",
    "%pprint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "dir = '/users/eva/Documents/Data_Science/Blog-Sentiment-Analysis/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "blogdata = pd.read_csv(dir + 'data/blogtext.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>14,May,2004</td>\n",
       "      <td>Info has been found (+/- 100 pages,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>13,May,2004</td>\n",
       "      <td>These are the team members:   Drewe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>12,May,2004</td>\n",
       "      <td>In het kader van kernfusie op aarde...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        id gender  age industry sign         date  \\\n",
       "0  2059027   male   15  Student  Leo  14,May,2004   \n",
       "1  2059027   male   15  Student  Leo  13,May,2004   \n",
       "2  2059027   male   15  Student  Leo  12,May,2004   \n",
       "\n",
       "                                                text  \n",
       "0             Info has been found (+/- 100 pages,...  \n",
       "1             These are the team members:   Drewe...  \n",
       "2             In het kader van kernfusie op aarde...  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "newcolumns = ['id', 'gender', 'age', 'industry', 'sign', 'date', 'text']\n",
    "blogdata.columns = newcolumns\n",
    "blogdata.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data analysis\n",
    "### Goals of my analysis\n",
    "There are three things I would like to investigate in this data. \n",
    "1. Word frequencies\n",
    "2. Blog topics\n",
    "3. Blog sentiment\n",
    "\n",
    "I already looked at [word frequencies](https://github.com/Data-Science-for-Linguists-2019/Blog-Sentiment-Analysis/blob/master/progress_report_part2.ipynb) in Progress Report #2, though I'm not super happy with my results. I also experimented with [topic clustering using LDA (Latent Dirichlet Allocation) and scikit-learn](https://github.com/Data-Science-for-Linguists-2019/Blog-Sentiment-Analysis/blob/master/progress_report_part2.ipynb). \n",
    "\n",
    "For Progress Report #3, I am going to begin by exploring blog sentiment. If I have time, I will revisit the previous two goals of my analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sentiment analysis\n",
    "\n",
    "Ok so, it appears the vast majority of sentiment analysis groups texts into \"positive\" and \"negative\". I would really like to go deeper than that if possible, but I'll try positive and negative classification first.\n",
    "\n",
    "Since I am really not interested in labeling even a portion of the 681,284 blogs in this dataset for sentiment, I think I am going to try using some already trained models or wordlists with sentiment mapping. Many sentiment models were trained on traditional corpus data, using text from news and books. These models don't incorporate things like emoticons :-) and other internet-specific language which would be found in the blog dataset. In order to get the best sentiment judgments, I specifically looked for models trained on more modern corpora.\n",
    "\n",
    "### First try: VADER\n",
    "VADER (Valence Aware Dictionary and sEntiment Reasoner) is \"a lexicon and rule-based sentiment analysis tool that is specifically attuned to sentiments expressed in social media\". It is also part of NLTK, which is convenient. It looks like it can be used on a whole text without tokenization.\n",
    "Credit:\n",
    "> Hutto, C.J. & Gilbert, E.E. (2014). VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text. Eighth International Conference on Weblogs and Social Media (ICWSM-14). Ann Arbor, MI, June 2014."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sia = SentimentIntensityAnalyzer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       well another days gone past and a major highlight of today has been two mass nose bleeds, while the blood is an itresting liquid its rather annoting when its streaning from your nose instead of someones neck. also today was the first full veiwing of the transit of venus across the sun whitnessed by mankind so though it may have been a pathetic black dot moving across the sun it was a once in a life time event and I'm slighty happyer that i've seen what some people can never see.  well with two nosebleeds, a mathes exam, a good roasting at the hands of the sun, an inability to focus properly and a ponding headache coming on I'm not feeling particly good tonight. but i've found my way back into hack this site which is a cool website which basicly gives the basics to the advanced on web hacking if your intrested its  urlLink hack-this-site dot org  aI'm quite happy with myself as im now twice as far as i got last time and i read loads of help articals then  well i dont feel like typing anything else tonight as my headaches getting realy bad but i hope anyone who reads this has a better evening than I'm going to and I'm sure you probably will  signing ooff one again its bald, goodnight people         \n"
     ]
    }
   ],
   "source": [
    "print(blogdata.text[30000])\n",
    "text = blogdata.text[30000]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After doing some poking around I am going to try this blog - it seems pretty negative, but it also uses a lot of sarcasm and understatement, as well as mispellings, which might be an issue. And though it does focus on the negative it also mentions some happier things. Let's see how VADER classifies this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'neg': 0.052, 'neu': 0.798, 'pos': 0.149, 'compound': 0.9709}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sia.polarity_scores(text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "VADER ranks sentiment using a score between -1 and 1. Greater than 0.5 is positive, less than -0.5 is negative, and anything in between is neutral. `.polarity_scores` returns the percentage of the text considered negative, percentage considered neutral, percentage considered positive, and lastly the overall score for the document (`compound`). VADER did a not super great job, since it returned a very very positive score for a negative-seeming document, as well as categorizing most of the statements as \"neutral\". I considered this to be a pretty emotionally charged blog. Let's keep exploring VADER anyway and see how it does."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"           I am pondering the idea of attempting to learn XML again, to make the site a bit more consistiant, in terms of tags, text sizes, and styles... Fear-not, it wont cut into the time I spen Dev'ing, just improve the work I do :)         \""
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blogdata.text[412412]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'neg': 0.0, 'neu': 0.827, 'pos': 0.173, 'compound': 0.7727}"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sia.polarity_scores(blogdata.text[412412])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This statement seems pretty darn neutral to me, though there is a smiley face. The classifier seems to lean positive."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'       Wat my feeling toward Sheffield??? Cold, cool and a bit of disappointment.   Cold – although now is summer, but the weather was quite cold especially the 1st day. It was a gloomy day and rain late of the day. The temperature was below 10 degree Celsius. Then the other days was like the weather in Genting Highlands and sometime even cooler then Genting… but I quite like the weather here, u can walk for hours with out any sweat in the afternoon. Believe it or not….  Cool – the cars, building, scenery and bla blab la were cool. After a few days here, I can see all my dream cars, Ferrari, Lambogini, Lotus, GTO, FTO, Impreza, BMW (the model tat drove by James Bond) and many many more tat I dunno the name. however, there is proton here also. Oh ya, kancil also got. Taken photo somemore… hahaha : p  oh.. bout the building, most of the building may seem a bit old, but inside the building was very nice and class leh…. E.g. Sheffield Hallam Uni… : )  Disappointment – cant get into the apartment type of accommodation as wat the told us earlier as we are the last batch… okie, dun mind bout it… but some of us (include me) was place in Marshall Hall (it is the name of the block and not a hall) for temporary as need to wait the full time student to move out from Broomgrove Hall – the hostel we suppose to stay in. tat mean cant unpack and buy all the necessities. Then there is not internet connection in our room.  Day 1 – reach the Manchester airport at bout 5.30am. but the sky was very bright liao, now summer mar, hav daylight from 5am till 10 or 11 pm. The airport is much smaller than KLIA…. Dun have much activities for the 1st day.. just that in the afternoon, our senior bring us to the Sheffield City and a Chinese groceries store… didn’t buy anything as still in temporary accommodation. … then most of the shop is close on Sunday. So there is very quiet and dun hav much people around the city. It was a tire day and sleep at 10 pm (I cant remember when is the last time I slept so early… \\uf04a) as I dun have any good sleep in the plane and bus…. Hehe…  Day 2 – wake up very early… bout 5 something in the morning… since my orientation will onli start at 9.30, so force myself into sleep again… hehe… I not pig lar, was still very tire leh…. The worst thing is we have to walk for 30-40 min to my uni. As usual the orientation start ed with a welcome speech, then registration, payment of fees and bla  bla bla…. Nothing much to say bout today….  Day 3 – hav a day off… but still hav to wake up early as need to go to the Hallam Union to get the ticket to Manchester and Bath (which it hav a limited seat).. will post up the photo after visiting those places… : )  After tat, as usual, shop for some necessities… we when to he market, I mean wet market. Just like in Malaysia, there is butcher, fish stalls, fruits…. But the diff is that their market r much more cleaner tat Malaysia n no bargaining….Not far from the market is the Peace Garden, a garden built to remember those warrior from Sheffield tat killed in wars. Many were  sun-bath there as today hav nice sunshine (which wont get this everyday)….. I was so happy leh today as Brian, Su Tsheng, Joyce, Rachel, Li Ping, Jo Anne, Cher Ching and Gim May have a simple celebration for my birthday… the 1st time celebrate my birthday at foreign country… hehehe …  \\uf04a  Day 4 – classes started today… so dun hav activities going on… just tat evening cook our own dinner… tis is the 1st time we have rice since we reached Sheffield…. So cham…. Hehehe… : p  Day 5 – finally we can move into Broomgrove Hall (remember??? The place we will stay for the rest of the time in Sheffield).. now can unpack all my belongings…. I was place in room 504,… okie very satisfy wif it, hav a nice scenery…. Hehehe…. Dun worry… will take photo also… hahahaha…  The is a television room… so watched the Euro 2004 match between England n Portugal… eh, wat happen 2 Beckam (missed 2 penalty kick liao), any so call Beckam fan can tell me… disappointed wif England performance (although I m not an England fan)… however the match was exciting leh….  Day 6 – nothing special… just attend lecture….at nite, we hav study grp…eh, wat wrong, 2day is fri nite n 1st week of class… I start doing my tutorial!!!  Something wrong wif me??? Okie.. I m still okie... just feel a bit weird… : ) … but feel good lar, atleast complete something… hehehe \\uf04a … fri nite mar, so after tat lepak at Brian’s room lar…  Day 7 – 9.45am woke up, should b the latest I woke up since reach Sheffield (not like me leh)…. : p  2day didn’t go for the trip to Warwick Castle, so wat the activities going on, okie, we when to Netto, the cheapest supermarket as told by the seniors…. They mayb right, the goods are quite cheap…so all of us buy quite some stuffs… it look like we torturing ourselves… we hav to carry our stuffs (which is heavy) all the way from Netto back to our room… which took bout 45 mins to 1 hour leh…. I think is the 1st n last time…          '"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's try a longer text\n",
    "blogdata.text[300000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'neg': 0.057, 'neu': 0.84, 'pos': 0.103, 'compound': 0.9921}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sia.polarity_scores(blogdata.text[300000])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems to work pretty good on longer texts. I guess this text is positive overall? It's hard to tell. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Running VADER on 10k blog sample"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I want to explore some ways to extract information from sentiment analysis. I'm going to do this on a smaller sample of the data before trying them out on the whole dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = blogdata.sample(10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>222045</th>\n",
       "      <td>3806796</td>\n",
       "      <td>male</td>\n",
       "      <td>27</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Pisces</td>\n",
       "      <td>30,June,2004</td>\n",
       "      <td>Dave can you not use the Afterburners b...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572114</th>\n",
       "      <td>3843153</td>\n",
       "      <td>female</td>\n",
       "      <td>24</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Aries</td>\n",
       "      <td>06,July,2004</td>\n",
       "      <td>Hey Everyone - am off to Darwin in afew...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>422121</th>\n",
       "      <td>1317513</td>\n",
       "      <td>male</td>\n",
       "      <td>17</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Scorpio</td>\n",
       "      <td>29,February,2004</td>\n",
       "      <td>I've got a freaking plunger around my n...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             id  gender  age industry     sign              date  \\\n",
       "222045  3806796    male   27     Arts   Pisces      30,June,2004   \n",
       "572114  3843153  female   24   indUnk    Aries      06,July,2004   \n",
       "422121  1317513    male   17     Arts  Scorpio  29,February,2004   \n",
       "\n",
       "                                                     text  \n",
       "222045         Dave can you not use the Afterburners b...  \n",
       "572114         Hey Everyone - am off to Darwin in afew...  \n",
       "422121         I've got a freaking plunger around my n...  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "222045"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.index[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# I've reloaded a couple times now, so I changed the actual first index number to sample.index[0] \n",
    "score = sia.polarity_scores(sample['text'][sample.index[0]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'neg': 0.0, 'neu': 0.78, 'pos': 0.22, 'compound': 0.9478}"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9478"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score['compound']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def summarize_sentiment(df):\n",
    "    all_scores = []\n",
    "    for x in df.index:\n",
    "        score = sia.polarity_scores(df['text'][x])\n",
    "        if score['compound'] >= 0.5:\n",
    "            all_scores.append('positive')\n",
    "        elif score['compound'] <= -0.5:\n",
    "            all_scores.append('negative')\n",
    "        else:\n",
    "            all_scores.append('neutral')\n",
    "    return all_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample['sentiment'] = summarize_sentiment(sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>222045</th>\n",
       "      <td>3806796</td>\n",
       "      <td>male</td>\n",
       "      <td>27</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Pisces</td>\n",
       "      <td>30,June,2004</td>\n",
       "      <td>Dave can you not use the Afterburners b...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572114</th>\n",
       "      <td>3843153</td>\n",
       "      <td>female</td>\n",
       "      <td>24</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Aries</td>\n",
       "      <td>06,July,2004</td>\n",
       "      <td>Hey Everyone - am off to Darwin in afew...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>422121</th>\n",
       "      <td>1317513</td>\n",
       "      <td>male</td>\n",
       "      <td>17</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Scorpio</td>\n",
       "      <td>29,February,2004</td>\n",
       "      <td>I've got a freaking plunger around my n...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>537733</th>\n",
       "      <td>3607084</td>\n",
       "      <td>female</td>\n",
       "      <td>16</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Virgo</td>\n",
       "      <td>11,June,2004</td>\n",
       "      <td>urlLink    BdC.crew&amp;nbsp; urlLink     ...</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>187169</th>\n",
       "      <td>2324198</td>\n",
       "      <td>male</td>\n",
       "      <td>24</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Scorpio</td>\n",
       "      <td>03,June,2004</td>\n",
       "      <td>Have ordered my car today, well hav...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>527378</th>\n",
       "      <td>3425171</td>\n",
       "      <td>female</td>\n",
       "      <td>41</td>\n",
       "      <td>Non-Profit</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>30,June,2004</td>\n",
       "      <td>They keep saying rain - then we get no rain...</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>389470</th>\n",
       "      <td>619488</td>\n",
       "      <td>female</td>\n",
       "      <td>25</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Cancer</td>\n",
       "      <td>23,May,2003</td>\n",
       "      <td>so i haven't updated much lately.  i've...</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>532406</th>\n",
       "      <td>271835</td>\n",
       "      <td>male</td>\n",
       "      <td>24</td>\n",
       "      <td>Technology</td>\n",
       "      <td>Libra</td>\n",
       "      <td>15,July,2004</td>\n",
       "      <td>Hope you dont have anything to do cos  ...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102139</th>\n",
       "      <td>3226138</td>\n",
       "      <td>female</td>\n",
       "      <td>34</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Sagittarius</td>\n",
       "      <td>03,May,2004</td>\n",
       "      <td>It took awhile for me to look...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135107</th>\n",
       "      <td>1057539</td>\n",
       "      <td>female</td>\n",
       "      <td>33</td>\n",
       "      <td>Telecommunications</td>\n",
       "      <td>Pisces</td>\n",
       "      <td>09,December,2003</td>\n",
       "      <td>Before I get into all of the lous...</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             id  gender  age            industry         sign  \\\n",
       "222045  3806796    male   27                Arts       Pisces   \n",
       "572114  3843153  female   24              indUnk        Aries   \n",
       "422121  1317513    male   17                Arts      Scorpio   \n",
       "537733  3607084  female   16                Arts        Virgo   \n",
       "187169  2324198    male   24              indUnk      Scorpio   \n",
       "527378  3425171  female   41          Non-Profit     Aquarius   \n",
       "389470   619488  female   25              indUnk       Cancer   \n",
       "532406   271835    male   24          Technology        Libra   \n",
       "102139  3226138  female   34              indUnk  Sagittarius   \n",
       "135107  1057539  female   33  Telecommunications       Pisces   \n",
       "\n",
       "                    date                                               text  \\\n",
       "222045      30,June,2004         Dave can you not use the Afterburners b...   \n",
       "572114      06,July,2004         Hey Everyone - am off to Darwin in afew...   \n",
       "422121  29,February,2004         I've got a freaking plunger around my n...   \n",
       "537733      11,June,2004          urlLink    BdC.crew&nbsp; urlLink     ...   \n",
       "187169      03,June,2004             Have ordered my car today, well hav...   \n",
       "527378      30,June,2004     They keep saying rain - then we get no rain...   \n",
       "389470       23,May,2003         so i haven't updated much lately.  i've...   \n",
       "532406      15,July,2004         Hope you dont have anything to do cos  ...   \n",
       "102139       03,May,2004                   It took awhile for me to look...   \n",
       "135107  09,December,2003               Before I get into all of the lous...   \n",
       "\n",
       "       sentiment  \n",
       "222045  positive  \n",
       "572114  positive  \n",
       "422121  positive  \n",
       "537733   neutral  \n",
       "187169  positive  \n",
       "527378  negative  \n",
       "389470  negative  \n",
       "532406  positive  \n",
       "102139  positive  \n",
       "135107  negative  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That worked! Let's see what the sentiments look like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot object at 0x1a1f25d2b0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sample.sentiment.value_counts().plot(kind=\"pie\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.tokenize import RegexpTokenizer\n",
    "tokenizer = RegexpTokenizer(r'\\w+')\n",
    "sample['tokens'] = sample.text.map(tokenizer.tokenize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# stealing my own code from before\n",
    "def Top10CategoryWords(df, column):\n",
    "    \"\"\"\n",
    "    Find top 10 words for each category in a column and plot.\n",
    "    \"\"\"\n",
    "    ax_count = 0\n",
    "    fig, axes = plt.subplots(len(set(df[column].values)), 1)\n",
    "    for category in set(df[column].values):\n",
    "        cnt = Counter()\n",
    "        for blog in df[df[column]==category].tokens:\n",
    "            cnt.update(blog)\n",
    "        pd.DataFrame(cnt.most_common(10)).set_index(0).plot(kind='bar', \n",
    "                                                            figsize=(6,6*len(set(df[column].values))), \n",
    "                                                            ax=axes[ax_count], \n",
    "                                                            title='Top 10 most frequent words for ' + category)\n",
    "        ax_count += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x1296 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Top10CategoryWords(sample, 'sentiment')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I always forget to remove stopwords. I would also like to remove 'urlLink' and 'nbsp' (non-breaking space, I think?). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords\n",
    "def remove_stuff(toks):\n",
    "    \"\"\"\n",
    "    Lowercases tokens and removes stopwords. Also removes 'urlLink' and 'nbsp'.\n",
    "    \"\"\"\n",
    "    lowered = [w.lower() for w in toks]\n",
    "    stuff = stopwords.words('english')\n",
    "    stuff.extend(['urllink','nbsp'])\n",
    "    no_stuff = [w for w in lowered if not w in stuff]\n",
    "    return no_stuff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample['tokens'] = sample.tokens.map(remove_stuff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>tokens</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>222045</th>\n",
       "      <td>3806796</td>\n",
       "      <td>male</td>\n",
       "      <td>27</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Pisces</td>\n",
       "      <td>30,June,2004</td>\n",
       "      <td>Dave can you not use the Afterburners b...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[dave, use, afterburners, blog, business, good...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572114</th>\n",
       "      <td>3843153</td>\n",
       "      <td>female</td>\n",
       "      <td>24</td>\n",
       "      <td>indUnk</td>\n",
       "      <td>Aries</td>\n",
       "      <td>06,July,2004</td>\n",
       "      <td>Hey Everyone - am off to Darwin in afew...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[hey, everyone, darwin, afew, hours, fun, pun,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>422121</th>\n",
       "      <td>1317513</td>\n",
       "      <td>male</td>\n",
       "      <td>17</td>\n",
       "      <td>Arts</td>\n",
       "      <td>Scorpio</td>\n",
       "      <td>29,February,2004</td>\n",
       "      <td>I've got a freaking plunger around my n...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[got, freaking, plunger, around, neck, yay, to...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             id  gender  age industry     sign              date  \\\n",
       "222045  3806796    male   27     Arts   Pisces      30,June,2004   \n",
       "572114  3843153  female   24   indUnk    Aries      06,July,2004   \n",
       "422121  1317513    male   17     Arts  Scorpio  29,February,2004   \n",
       "\n",
       "                                                     text sentiment  \\\n",
       "222045         Dave can you not use the Afterburners b...  positive   \n",
       "572114         Hey Everyone - am off to Darwin in afew...  positive   \n",
       "422121         I've got a freaking plunger around my n...  positive   \n",
       "\n",
       "                                                   tokens  \n",
       "222045  [dave, use, afterburners, blog, business, good...  \n",
       "572114  [hey, everyone, darwin, afew, hours, fun, pun,...  \n",
       "422121  [got, freaking, plunger, around, neck, yay, to...  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x1296 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Top10CategoryWords(sample, 'sentiment')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "These are all very similar... did I do something wrong? Is VADER just not good for my data? Idk."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEKCAYAAADq59mMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzsnXeYG9XVh98ZdWmbtq9772XdG8ZgY0PAQAKmBYhNqKGE0EJJIwHzEXpCDQRsQjXYGAgOLWBTjMG9N1zW9vYurfpo5n5/zFq7Wq1trbe4RO/z7LPSvZo7V6OZe2buOfd3JCEECRIkSJAgQUuQj3UHEiRIkCDBiUfCeCRIkCBBghaTMB4JEiRIkKDFJIxHggQJEiRoMQnjkSBBggQJWkzCeCRIkCBBghaTMB4JEiRIkKDFJIxHggQJEiRoMQnjkSBBggQJWozxWHcgHjIzM0WPHj2OdTcSJEiQ4IRizZo1lUKIrPZo+4QwHj169GD16tXHuhsJEiRIcEIhSdK+9mo7MW2VIEGCBAlaTMJ4JEiQIEGCFpMwHgkSJEiQoMWcED6PBAkSnBwoikJhYSGBQOBYd+Wkwmq10qVLF0wmU4ftM2E8EiRI0GEUFhaSnJxMjx49kCTpWHfnpEAIQVVVFYWFhfTs2bPD9vs/azwUVym4CgEJUjtjSs2N1IUUlUpviL2VHtIdFtIdZnJSrB3Wt7CrFOEqBEBK7YKxUd9ai9uvUOtXKKz20TPTQardhN3cNqeBFg6jeSvQqnYj29LQ7FmYU3PapO0TnUpPkFJXAEXV6OK0k5VsaagUAjxlUFMAliRIygVHZof2rcwdIBTW6OK0kZXcdud6lSeINxhmT4WXXtlJ+Pz+k8dwqApoCmgaGC1giP+uXxMaqqYS0kIYJSMG2YBRPrrrUJIkMjIyqKioOKrtj5b/SeOhuksx/escqNqlF2QPInzFYowp+iC9v8bPBc8vx+0PA3Dm4Bz+cv6QDjEgqrsU4+vnQcUOvSCrP+qVH2JIab0B8QbDvL3qAA/9ZxsARlli3lVjmNQ7E1lu/cUsXAcwvjwNfFUAqH1mEDr3mf95A1LpCTL7lZVsKXYD0DXdxqIbJpJ98HxyFcI/p+kGBKDnFJj1MjjaJTw/pm9Xv7qKDQdcAHRx2lj0q4ltcq57Agqfby3j3sWbEAIkCRZe2g1NgOFEtx2qAjV7IeTV38smyOwHRnNcm4fUEHtde9GEBoDT6iTbnt0qA9LR/E86zNUN7zYYDoDyrWjbPwGg2hPk4Y+3RQwHwKdbyih3Bzumb5s/aDAcABU7ULd82CZt1wXCPPLJ9sj7sCa4e+FGqryt/24hnwuWzo0YDgDDrs+Qave3uu0Tne92VUYMB8CBaj9vrzqApglQAvDNEw2GA2DvV9HnQDuyuqA6YjgACmv8vP79PlRNa3Xb7kCYBz7aysFM10JArU9B1U6C1NeKv8FwgP4E4i0HceTjFtbClHhLIoYDoCZQgyrU9uhpu/E/Zzw0TcNYvTOmXK7+EYBgWKPEFevMK3H5271vAIb6fhyp7GgIhlXCTS7csrogbZHGXgsFMLhiDYXmOtD6xk9w9lZ6Y8p2V3hQhQA1CDV7Yjeqbbe1XVE017c9Fd6Y8+RoCGsCbyh6QNSbPfGNx/p1q/nPF99G3n/42Vc8/PjfiOdiEkKgaEpMuaod3ngsW7aM7777ruWdbSf+54yHLMuoI34RWzHsUgAyky38NL9TVJXNZGBw59SO6B4i//LYsuGxZUeD3WykV6YjquyswbnYzIZWt21JziQ0rEk/jRaMXce2uu0TnXOGdaLprMLl47phMshgTYVRc6IrDSboMblD+nbWkFyazlhePq4bFmPrzwmzQWZE17SoMotRPin8Heu37uY/XzYYj/NmTOGe+34P8pGPm0E2kGaJPi4GyYBJPrzP5HgzHpJoi9vOdmb06NHiaORJAoEAYVUlyRE9YPpdFRgLv8f07aMgGVCm3IuSOwJ7qj7HXO4OsHBtIYvWFJGTYuG+swfSO9uBzdQwHyk0DSXgwWC2YTC2XXhc0F2JXLQK0zd/BUCZfDdal7FYkjOiPlfjDWA2GHBYW7bvolo///efbWwucnF6/2xuPL1PtPO2FSjuCsTm9zCvewUcmYSnz0Wk98FkS2qT9tuCGm8Qs0Fu8XFrDZ5gmI0Hannk0x2Ewho3nd6bU/pmkWqr74OvGrZ+CCtfAGsazJgLOYPAZItqxxsMYzHJGOXYez5FVfGG/CSbbRgMsQOYomooqhYTHOENhtlc5OKvn2zHr6j8akpvpvTLItUe39z9kSiu9fPopztYt7+GMT3SmT3IxJDBg9qk7aPF6/Vy8cUXU1hYiKqq/OEPf6BPnz7cfvvteDweMjMzmT9/Pnl5eZx22mmMGzeOpUuXUltby8svv8y4cePo06cPfr+PzjlZ3Pvra/ELM6s3buWZZ59jzpw52Gw2tm/fzr59+5g3bx6vvvoqK1asYNy4ccyfP5+wFua9j97j/x74P5SQQv++/Zk/b34kEm327Nn8+9//RlEU3n33XaxWK+PHj8dgMJCVlcXTTz/N5MnRNxjbtm1j4MCBUWWSJK0RQoxulwMphDju/0aNGiVaghJSRGFVnfjDe+vFdf9aJVbsLBHVta5IfYXLK3xBRfiri4Wvulj4goqorPNGtREIhUVhtVeUu/0x7Yfc5SK46lUhXp8lgksfFWFXaYv6dzh8QUWUufyipPiAKCk+IEpdfuEPhSP1JbU+sWxHmbj21VXid4s3if1VXhEMKS3aR10gJMrd0e22FeFQUASqi0TAVdbmbbeG0lq/WP5jhbjuX6vFvYs2in2VHuEPtuy4tZYqT0BU1gWEpmmxlWpYiLpyIbxVMVXVnqD49/oicdW8leLxz7aLMlf0OVnmqRAvbZgnfvX5r8S/Nr8pyr2VUfUltX7x0JKt4ur5K8V/t5aKGm+wmb4FD923VlLrDYrCaq+o9QXF1q1b27z9lrJw4UJxzTXXRN7X1taKCRMmiPLyciGEEG+//ba46qqrhBBCTJkyRdx+++1CCCGWLFkipk2bJoQQYt68eeKmm24SIqwIEQ6Jea+8or8XQsyePVtccsklQtM08f7774vk5GSxceNGoaqqGDlypFi3bp2oqKgQkydPFrXuWhFWw+Lhhx8Wf/7zn4UQQnTv3l38/e9/F0II8eyzz4qrr75aCCHEn/70J/Hoo48e8ns1d2yB1aKdxuWTMtqqyhvknGe+o9anzyt+uqWMV2ePYEpqCgDVfpVxD38WcdyZDTKf/mYyGY1ukC0mA52d9pi2Q7465K8fw7TqBX3bHz9D2/MFoVmvYk7JbnXfS91Bpj/xVWTO2ShLfH77FHrWTzet3V/LjW+sjXz+400lfHTLKeSlxf9TJllMJFna587bYDJjcHY68gc7mO2lbmbPWxV5v2RTCR/fOplObRSmHA/pjsM84ckGSIqNrgqFNd5auZ9HPtUd6F9uL+eTzaW8ee14MpMsVHhruP+7P/FN8dcAfFP0DVurNnPX6LtJt6dQURdk1gvfUVij++z+u62chy8YykWju2JoNF+V7mibJ43mSLWbI08yxe22l/gZOnQod955J3fffTczZ87E6XSyefNmpk+fDoCqquTl5UU+f8EFFwAwatQoCgoKohsz1J8/Tabizj33XCRJYujQoeTk5DB06FAABg8eTEFBAYWFhWzdupUpk6cAEAqFmDBhQrP7fO+999rsu7clJ6XPY8P+qojhOMgL3+6n2uVG0zReW1EQFfERUjUWrC6Mq21Z8WBcNy+6bP93GMJt41B/64f9Uc7KsCZ4a6XuiC6u9fHKt3ujPl/lDbG52EWCQ1PmCvDSN9HHzeVXWFlQfYx6FD8uf4h5ywuiynaWeXD59fM7pAUihuMgHxcsIajp52OJyx8xHAd56Zu91PhC7dfp45x+/fqxZs0ahg4dyr333suiRYsYPHgw69evZ/369WzatInPPvss8nmLRTf6BoOBcDh8qGajOLiNLMuR1wffh8NhhBBMnz49ss+tW7fy8ssvt2qfHc1JaTwclti7ySSzAaMsI8syyc3Md6dY47sDFQCmJk8kkgRS6x2MACm22H4cnBs3yFKz3625sgQNHOq4JZ8Qx03Cbok9t8wG/dKVJAmjZGxS1/AU0Zzj22ExIJ8ETuujpbi4GLvdzhVXXMGdd97JDz/8QEVFBStWrAB0CZUtW7Ycto3k5GTq6uqOug/jx49n+fLl7NqlLxnw+Xzs3BkbBdqW+2xrTkrj0S83hX45DXNQZoPMHWf0JiVZL/v5uG6k2RsMSFaShZ+O6BxX25ItHeX0P0WVKflzCJsch9iiZcwa1TVqCiHDYebCkV0AyEmxcfv0fpgarbAalJcSmdJK0DyZyRZ+c0ZfLMaG071PdhKDOqUcw17FR4bDzH1nRztBzxycQ1K94bMZ7Fw24Iqo+muGXE+yWf9umUlmxvV0RupkCe47e2C7TlMd72zatImxY8eSn5/P3Llz+ctf/sLChQu5++67GT58OPn5+UeMajr99NPZunUr+fn5LFiwoMV9yMrKYv78+Vx22WUMGzaM8ePHs3379sNuc+6557J48WLy8/P55ptvWrzPtuakjbaqcHlZU1BJuTvI1EF5pNtk7HZ9kA0GgwSDfgjWW3FLCna7A6MxvjvRoKcaQ10x2u5lSJ1HIjL6Yk5pm9XAmiao9AYjixRTbUYyHJbICvBqb4C6gMrnW8vITbUyqruTvFTb4ZqMQghBpSeEqgnMBon0pLaJtDreqfUGcQfDfL61jMwkC+N6ppPbguN2RFRFXyAphB4lZUs78jZx4g0oVPsUAoqKQZZw2s04Gw3+Fd5qDtQdYG35OsbmjiHP0YksR4PBqPQE2VLkYk+ll9MHZJOdbGmRJE2Z209YAxA4zAbS7Ed/zjQXEdSeKGE1sqrEKEvIzUSqHS0HJUYAZEnG0CRMV9U0Ds5AGySpTVQcDkdHR1udlMajLqDwyeZSgopKZrKF5T9WcOv0/mTWD5RBdwXy8icxrX4JJBll3E2oY2/Amtp6h3drqfaGePHr3ZF57l+e0pNrJ/dqkzvFsKqxtcTNTW+u5UC1n6GdU3nu8pF0TY8NDEjQAkJe2L0MPvo1eCuhzxlw/nOQ3DayLCUuP79+ax2rCmrITbHy1KX5jOiahsXUNlOlh6PMHeDhj7fx4YYSbCYDvzmjL+cMy2vRDUtjOtJ4hMIahTU+PMEwRoNMlzQbdrOMsZlQ5pYS1sK4gi7KfeUIIUi1pJLjyInIi4RVjVJ3gBqvgiRBTooFp92M0dB+kz0dbTxOymmrKk+IuxZu5PcfbOGG19fy2g8HeOP7fShqvRzAgZWYfngW1BCEA5iWP45cfvg5zo5iY2EtL3y1h2BYIxjWeH7ZbjYV1rZJ29W+ELNfWcmBat2BuqnIxa1vr6fG+7/rPG0T/DXwzhW64QDY9V/46q8Q8rW6aXdA4Q/vb2ZVQQ0Ape4AV81bRa0/doVyWxNUFD5YX8TidcWomsATDPPgkm1UnwDnS1hVKXX58QTD9e819lX7aCtllLAWptRbiiY0BILaYC3uoDsSxuoOhKn2hhAINCEocQUIqa2XfDmeOCmNx8ZmBttlOyrwBMJomoblx49i6uWdn3RE147I51vKYso+2xpbdjT4gio1TaLQ1u6vaTCqCY6O6t2xmkZ7lkLI0+qmAyGVH/ZER4X5FZXaDoiWqvWH+ebHypjy73ZXNfPp4wtNgKeJNIoQAkVtG+vhVWJlXeqUOt2YCF29uikHDdnJwklpPAbkxTpCx/RMx2ExIssywe5TYurVHqd2RNeOyKS+sVLcp/RpG3lum9kQcbQeZFBeSlS8f4KjIK1HbFmX0WBufSCD2SgztEu0NI7ZILfZCvDDkWSRGdnNGVM+slvb+XPaCwmwN5nWk5Cigk1ag80YO21nN9qRJRlJgqRmojcdHbimqCM4KY1HdrKFO2f0o0uajd5ZDkZ3d3Lt5J6Y66Nt5N5TCQ+6oD7EViI87DKkrmOOca91xvVK56f5nQ52jZ+N6MzYnult0rbTbuL5K0Yyqlsav5jQjaGdU/n7Zflk/I84zdsNmxPOebJBTiR3GJzx5zYxHml2Mw9fOCyiSZZkMfLUpfmkdECYscNi4edjuzGpjy6NY5Qlrpnck05tGWjQTpiMBvLSrFjrQ5VlSaKz00pb3SaZDWYybA2SQUnmJJxWJ5IkIUkSaTYTKfVLAiQkMpMsUdF+JwPt7jCXJMkArAaKhBAzJUmaD0wBDq5smyOEWH+4No4m2kqtK9eT6/iqIG84hqTshtWggM9VhRz2AhKq0Y4jteFE8Ph91KluNlZsIdueRZ6jE7lJDXf/VZ4yvFqQTZWb6JvWF6c5hayk+PNtVHmCHKjxUVkXYmiXVDIc0Y40t1/BGwyDpN+tpNjabjV4dV0ZNUotu2t+ZFDmUJIMDtIafbcSl5+iGj9l7iD53dJIsRqj1sWUuvzsKK1DEzAgLznacaqGwVsBJRvAngHpPVuW1MhfC55yqNgOecPBng6W5Lb42qh+D3KoForWRvompbThSviQDwIuXZrbZIvJxVHmLWV37R5CWogB6QPIdbQgP4vip9Kj4A8pmE1G0iwSFkfD07U76KY6UM2u2l30d/YnzZpGsrltjhtASa0PRRMYJAmjQSIn5eiNR0dHW4XCaiSXiCzRJs7yg6iaGvF5yJIck4sjrOrRVhIgy2Bow0iv5uhoh3lHPEfdCmwDGs8l3SWEWNhue/RUYHh3Nuyvj9W2pMB1yyCjN6CLtc16fj3F9dLr3dLtvHP9+EjoZpG/mCs//jn++lXj43Mn8MCkh8hNyiQQ9LKqYi2//fpuRH0Q4NVDrubK/peRkXTk6JoqT5Cb31zLivp5bIfZwIc3n0Lv7IZ1KSk2U5sajIPUeip5c8eb/GPLK4B+R/TkqY8w2XwaZrOVEpefO97ZEJnTtpsNLPrVRAbm6X0pdfm5+B/fs79adwR3SrWy6MaJDQaktgBePK0hBLrbRLjkX/ElNQp6YOWLek4Q0K/2WfNgwMwWZWg7FLK3BOmf0/QBHhBdxiAuehUpNb71PUfEbNf/mqHMW8a1n13HXre+yj3LlsUbZ79BXlJes5+PoWgNmf86H7T6OfNRV8EZfwKbE7/iZ/GuxTy2+rHIxx+c9CA/6fmTqMWCrSEv7cSNxjO3gTpwc7zwwgvY7XZ+8YtfMH/+fGbMmEGnTvrNyDXXXMPtt9/OoEHHVgCyvWlXUyhJUhfgHOCf7bmfGKp3NxgOgKAbvnwwkrzl3TUHIoYDYH+1jyWbSgCo8NTwtzVPRAwHwPelKyj36U7rmpCLv656JGI4AOZvmU+Q+BK5HKjxRQwHgDek8sin2/EE2z96JiBC/HPrq5H3AsHcVY/gCtZH8rgCUc5QX0jlkU92UFafy2TJppKI4QAodgVYeFDWJeSFLx5sMByg/wbV0bIghyTo1iOUIp0T8J+7dMXZVqJ5KmHZwxHDASAVroLKjkm49G3RNxHDAVDhr2DBjgXxyU54KmDJHQ2GA2DNvMhxrlPq+Pvav0dt8siqR3AFE5I17ckNN9zAL36hp3aYP38+xcUNql3//Oc/T3rDAe3v83gK+C3QNJxnriRJGyVJelKSpGYn3CVJuk6SpNWSJK1ucW5eTzPRSZ5SUENomkZpM8meDpaFNIWqQGw0SZW/+mDHqA1ER3OpQkVR4xv8K+tio2Qq6oIo4fZfb6NooZhsZTWBmoioW6UnNqNgpSeIUh/fWFLbfJIsTdMgHNKPcVOa+y2aQ1WiB0gAfzVtkThICgeRmutHXTP9bQdKvLH7KfOVEW4mIVAMQtUz1DVF0Q36wTzYjakL1UVlqTuReX9dEZMe/pKe9yxh0sNf8v66ola3WVBQwIABA5g9ezbDhg1j1qxZ+Hw+vvjiC0aMGMHQoUP55S9/STCoXw/33HMPgwYNYtiwYdx5550A3H///Tz22GMsXLiQ1atXc/nll5Ofn4/f7+e0005j9erVPP/88/z2t7+N7Hf+/PnccsstALz++uuRVe7XX389qnpiZRGEdjQekiTNBMqFEGuaVN0LDADGAOnA3c1tL4R4UQgxWggxOiurhau3u4yJdVaOuRZsTmRZ5rKx3aKqZAkuHKVLgGTbnPysz6yo+iRTEgMy+gFgkYyc1eOsqPp+zn5Y45wiGNolFUeT5EtXju8eJZfSXthkC/2c/aLKzul5FiZJ3/fgTqkxek+XjOlCTr1DfdboLlGJgyQJLhvbTV+1a3fqx7gxZgd0HhVf58wOyMuPLht0QZs4naW0zoimCZdMNuhxSqvbjodzep2NoYn22cX9LsZqjsN3YE2DEVdGl6V1i6xgtxqtjMgeEVV9WtfTmo0GOtF4f10R9763iaJaPwI9F829721qEwOyY8cOrrvuOjZu3EhKSgpPPPEEc+bMYcGCBWzatIlwOMzzzz9PdXU1ixcvZsuWLWzcuJHf//73Ue3MmjWL0aNH88Ybb7B+/XpsNltUXWNF3AULFnDJJZewbds2FixYwPLly1m/fj0Gg4E33nij1d+po2lPn8ck4DxJks4GrECKJEmvCyEOCvEEJUmaB9zZ5nt2ZMG1S+GLP+sO3LHXQ6/TI9U5yWb+e/tkhNBHQlmCZKt+cZtMJqZ2nYYsGfhg9yKy7Dnckn8LaUY9XDLdkcXto28jLymP5cXLGZg+kOuHXU9Ok/nrQLAOf8hDkjUDkylaq+rDm0/hkU+3U1EX5Mrx3Tl9QHZMdrWaOn16yJncdvPNmcl5PDf1WVYUr0JRZZIsBsbmjCTNoQcLOG1G3rtxIo9/tpMyd4CLRndh2sAcTPUhjznJVt66djx///JHNA1uOr03ndMaDVJ9psKFL8PKf+i/wbQ/xefvAN2x/vO34evHoWg19D0TxlzTZg5zuk9C/OwfSKtfRtjSYdof0MxJtP8abUg1JjP/rPk8t+E5FFXhqiFX0c0Rp7PeZIWJv9aP4+b3IHsgnHYv1PvXnFYnj095nHmb57GmfA0TO03kioFXkGI5/nW7jsSjn+7Ar0TfkfsVlUc/3RG3Ft2h6Nq1K5MmTUIIwWWX/ZyHHppLz5496ddPv7maPXs2zz77LDfffDNWq5VrrrmGc845h5kzZ8a9j6ysLHr16sX3339P37592bFjB5MmTeLZZ59lzZo1jBkzBhD4/QGys4+9ukVLaTfjIYS4F/0pA0mSTgPuFEJcIUlSnhCiRNJHy58Cm9t85wYTZPWHn/1Dnw6xOaP09iUENd4wzy7bhSxJ/HpqH1KsDcNIdlI6F/Y9n1M6TcYim8hIio5rT1M1ruv2Ey7ucTYOIbBr0Q9w5Z4SXt0yj83VO5jWeTIze59HukM/OYwGmd7ZSTx+8XCUsCDNbooyHLUeP7ur/Dzz5W4Abp7am14ZNpxJbXMnqYVT2bSzFxsLXUwdkM2ozIbB2WYx0TfHxAPnDyakauSmWKKy0jkdZsb1yuCpLAcIyEqxRjduc8KQC6H3VP03aOnAn5wHMx7Q/SeWFDC23VoGDQm5z3ToNh5kI8gW/a8DSHdkke7I4qEJf0ETguzkOB3lB3FkwrgbYfhl+hNTk6exLHsWt468FW/YS7IpGVMbBBgcDxTXNp/m4FDlLUGSJBRVo9oboswdJKRqyM3MkBqNRlauXMkXX3zB22+/zTPPPMOXX34Z934uueQS3nnnHQYMGMDPfvYzJElCCMHsK6/k//70W92nZ7RAUi6RsLAThGOxauUNSZKy0CPY1gM3tNueDjF4VXjDXPziikiu+q92VvDJrZPJanSzZjAY6NSM2KGvrgrzlw9gXfc6kWDLTiMIXPIm1tROVNWV8KulN7OzRpdXXlu+lkJPMb8ZcQt2W8OCqySLCZoZu0rrFC56YUVERuGrneV8fOvkNjEeJbVern1tLZuL3ACsKqhhb6WXe87sQ3pyw4AUYxSakJV8mHpJ0kNsjxaTLSb9alsgF65Cekd/6JUAzEnIN30P1o5TJM5sQTh3DAbDYcOeLUYLFuPJtV6nU5qNomYMRae01p8f+/fv59+fLaPP0JEsXvQOI8dPZtGbr7J9x04G9O/Ha6+9xpQpU/B4PPh8Ps4++2zGjx9Pnz59Yto6nFT6BRdcwNy5c+nevTt//aseEDJt6lTOP/9cbrt8BtmZ6VSXFVHn2073kae3SWRhR9Ehq1aEEMuEEDPrX08VQgwVQgwRQlwhhGi9hkML0DSNN3/YT+PlLaomeGf1gbi2N2kBjBubSDAXr8Oo6s5knxaMGI6DvLfnQ7xqrLO5Od5edSBKf0cT8NbK+Pp2JIJhETEcB3l/XTFB9cS52zkawq4SpFUvRReGPLB3+bHpUIK4uOvM/tiarBK3mQzcdWb/Vrc9YOBA3nrzNWZNn4SrtoYrrr2RPz/2DJdecglDhw5FlmVuuOEG6urqmDlzJsOGDWPKlCk8+eSTMW3NmTOHG264IeIwb4zT6WTQoEHs27ePsWPHAjBoQD8evOsmZlx2I8POuJjpl/2KktISXWvvBOLkWi8fB7Isk5EUOx2SlRzfXZtAgDW1QQQPQJJB1u8YTLIJCSkqlDfVkooU52LMzGb71jbTN0aDjCwRZZxSbEZOAGHl1mG0IOzpsauL4/XHJDgmHPRrPPrpDopr/XRKs3HXmf1b7e8AfRz408NPoTU6+SdMPo05F63B1GjBbl5eHitXrozZ/v7774+8vvDCC7nwwgsj75ctWxb12Y8+aqqlJ3HJBTO55NypTYo7wgPXdpxc6+Xj5NIx3chuZCw6p9k4d1h8DkxhSUeZMTeqTBl3PaH6+XMbRn7e76JInYTEPaPuINUa30rrC0Z2IbfRtFFuipULRnSJa9sjYTEIZk/oEVV239kDcdpPrJO2pRgd6bqT2dwoSX1evu58TnBc89MRnVl+z1T2PnwOy++Z2iaGA/Spy9wm07M5KVYMHeFzMBghtUt9L+qxpOq+uBOIkzKfB4Df78cdVFE1sBshLaVh4FAUhUpfmDX7ajDIEvld01qUn8BfV4U17EEoPjBaUUwOLMkN0RI1daWUBsr5sWoH+bmjSTM6SElqqK8L6AmdVCGwGCGMN//3AAAgAElEQVQ7OSmq/TKXn5AqkACTQSKnhVpClXVBgmENk0EivYn0SanLS7lbYUuxi7E900mxQFZqg28ooIRx+cOomsBqMsTkEalw69LSAjAZZHKO4B9pCaqmUh2oRtEULAYL6db0mCi0o2474MUdUPBrMgZJItkYxp4cK/rXXlS4PQRV/Xcwy4Ls1Jb5Wio9QYKKhtEg4bSb2m3l9NFQUldJWAQxSEZSzSk4LIc+XztanuRwhDWNsCrwh1RsZgNGWWpRvg1NVZHq100JSUY2tGDw1zRdyibk1R3mBnOr/R0nozxJh+Oq87BwbRGP/ncPAUVjcp8MnrhoGFmpethrbaiGZzb9Da/iRRMaq+oyuWHY9WQ74kveYwpUIb19KVLVbkjKRr5gHoolBZNZH0idybk4k3MZmDUsZttKr5f31xXz+Ke78SsqE3qn8+TFw8mt71uZO8Djn//IorX6yu0LR3bhjhn94h6kCyq9XP3qKnZXeMlwmHnm5yMY1d0ZGWxyUx3kpsKwrrHKqG6/wr83FDP3P9vwhVTG9Uzn6ctGkF2/71KXn79/8SMLVhcihOBnIzpz15kDyE1tvQEJq2E2VW3i9mW3U+mvpFtyN56Z9gw9U3u2um2A6iDcuWg7X/1YhcUoc8e0nlw4wkBGWvuHtJa6fLzw1T5e/34fqhDMHJrH72cOjFsjqrDGx7X/Ws22kjrS7CYev2g4E3tnYDsOVFoL64q5bdmtbK/eToo5hT+O/zNjcsaTbk868sbHGKMsY5TBehRJtTRVAW9l/eJTAZZURFpXpHgNgCzr0X4ncJDDSTltVekL88DHuwgo+irbb3ZVMe/b3YRCukNqZelKPtz9IV/s/4KlB5by7s532VwVXzKooKsc43vXQpUeSounHNM7lyP5a+LavtYb5sGPdkbi11fsruYfX+/CW7+adXVBNe+sPoCqiYgjf3VBfBId1d4gty1Yz+4KXYalyhvi2n+ticnhcShqfCF+9/5mfPV5EH7YW81zS3cRqO/rhkIXb67U+6YJWLS2iO/3tE1uh5pgDbd8eQuVft2XtL9uP3d+dSfV/tbLk/gDAV5bsZevftT7GgxrPPTpbqr8HbOq98cyL/O/KyCsCYSAf28s4YttzawabwaXP8TvFm9iW4kezVPrU7jxjbW4A8c+N0S5p5b/++EhtlfrubfdITf3fHsXAa1DY2CODWoI2VPKQQUEKehC+Ko5EWZy2oqT0nhsPhA74KwocOMNBNE0jRUl38fUf99MWXMYCENJExHgQG3cWeO2FMcmqlpV4KIuqBu2b5tJvrN8V2xZc4Q1wboD0e17gmF8ofgGmh/LYy/67/dW6wq/h+mbprVeCiMQDsToMe2s2UlYtH6Q9PqDfLfXHVO+qZnzpD1Yvrv5hErewJEj8IKKxtp90b9pMKx1SDKoI6FoQTZWbogqC2vhBimfk5lQbDIoOVSH0E48mZGj5aQ0HkO7ZsSUTeqVisNqQZZlJnaaGFPfXFlzaBih08joQpvzkIqqTRnSOXaefVyvNFKs+uPrqf1iI4BO7Rff6lOjHJu8J8VqjDsJTb+c2HUxk/pk4qiXLDm1X6zT/9R+Wbo8SSuxGq04LdF9H5g+EKPU+qkZh83CKb1SY8qHd4s9T9qDyYdI8OWwHnm6z2oyMLpH9HGxGGXSOiAZ1JEwGyzkN5FGMckmMm0dc1yPKebYaTnNkoIkHz++qPbmpDQeGXYDD5zbj8F5KeR3TWP6wCzmTOqF2axfcKNzRjOr7yx+N/Z33D3mbn4+4DIGpcfnxDOnZqNe8E/IGqAXpHQifMnburZTHKTZDfzl/AGRjH5T+mVy3eTe2Ov7NrK7k8vHdcMo61nPrhjfjRFxZm5Ld5h56pJ8+tcbgexkC/+cPTpu3Syn3cQjs4YxOC+FEV3TmDogixum9IrMCQ/tnMbsid0xGSSMssRlY7sypkcbJaqyOnlm2jPk2HW/U6/UXjw65VHSba1v32a1csWEnpw1OJv8rqkM7pTCn87pi9PaMad/7ywH153aC7NBxiBLXDCyM6cPiC9MOMVmYu7PhjK2Rzoju6UxOC+Fl34xmtR2kOxvKVmONO4Zcy9DMocA4LQ4efTUJ7DKJ66Ee9wYzOypDvPmYj19tbCmIdmccQd4JCUd/z6hI3HSRltp7lJE2WbwlCP1OAXZkRl5OgiHw1R7gqzaU44sS4zqmdWi6Be3x4sx5EZTAshGM4o5ldTk+E8GTzCI2x9GE2A2iqhoK01VqfAqkakih8VIlsOE3IIkNlWeIIH6aKsMuxlDVLSVn6LaAJsKa5nQO5M0uynKGV9RF2BbSR2l7gATe2WQlWLB0iiyp7wugBLWp6lMBpmsZEvDBRMO6lpie7+BpGw9o15S/Gspan1B3KEagqqCzWghzerUV+K3AXV+BXdAYcXuKtKTzAzOS2lRFFuwrhLZdQCtaB2G7uNRk3KxJMVv2KrqvATC+nEyGwROuwWjMc6nqqAHzVOG2PsNZA1Eyuytn88dQFBRqfSE+HZXBak2E/ldnTEBEiV1lagoyBhIs6RFboSao6XRVoqq4gupqJogyWKsX6sU3wCtqnq0pSeoYDEasBhljG0Ypfbll1/wxOOP8+8P3o+JtgqHw4f9fZOSkvB42tY3lIi2agvcxchvXapntANd7uKaLyBnMKCHPc585jsqPfq8cW6KlQ9uHE9O2pENSCAU4vNt5dyxaGuk7KoJXfj16b1xpsRnQJIsFpIszUdZlNaFOP/Z5VTU6Q707GQL7980qUWSDIdKK1vuDvDs0l289v3+SNkjFw5j5rA87BYjVZ4g17++JjLHbjHKLL5xEoM66RFJZe4A5z+znFK3PlefmWTmo1smNwwm1XvhpdMicuHk5cPlC+MyIHUBheeW7eHFr/dE9e2nIzpH0ge3hpL6vh8MVBiUl8Irc0ZHEoAdjpDXhfzd05hWPBUp037yOOERV2A0H3nqqdzl5YIXfqCwRj8uTruJJbdMpJMzjvNFCNj/HfKbFxNZzTnkQjj7sdbJwMRJUa2f855Zjqf+ZqZPdhKvXz0uyoDkJbePIVNUlYrvXifzh79i8hSjJHVCPeOPyPmXxrW9X9HYW+mLLNhNsRrp4rRReOAAP/nJTzjllFP47rvv6Ny5Mx988AHFxcXcdNNNVFRUYLfbeemllxgwYABz5sxh5syZzJqlq20fHPjvu+93bNu2jRGjxzJ79mycTidLliwhEAjg9Xr58MMPOf/886mpqUFRFB588EHOP//8djlWx4KTctqKih0NhgP0wWzp3EhioQUr90UMB0CpO8B/NhTG1bTLF2TuJ7uiyl79vpBAG/nJFq0pjBgOgPK6IO+tja9vRyKkarz+w/6osr9+sp1avx6NdaDGH+WcDYY1Hv54G+76+o82FkcMB0ClJ8S7a+qlU4J18OVfGgwH6IEFVT/G1TdPMMw/v9kTVfbQx9vaxDFc6fHz1Oc7oxRat5a4IxFMR0JWPJh+eCaqzLz0L4g4I+y+3FocMRwANT6FV5fvJRyO46TxVsAn9xIlA7B5kS6v0s54gwrPLt0VMRwAu8o9rN0f3/duLcF1C8hddjdmTxESArOnCMNHv0HbsOCI24bDKiWuQJTSgzugr18C+PHHH7npppvYsmULaWlpLFq0iOuuu46nn36aNWvW8Nhjj3HjjTcedh8PP/wwkydPZv369dx2220ArFixgldffZUvv/wSq9XK4sWLWbt2LUuXLuWOO+44qaKxTs4nj+Yu6oALVAVN06jyxkbwVHrjzeQn4WkSJqkJCKttc1JUNzNYVnvbJrImrIoYKZK6Rt+lLhB7DFx+hXB9NFWVJ7YflXUhhBBIaljPQd6UOAfYg+G/jfEG2yYcVVHB1cx3i9swaWpsoqqQBynORFXVzZxvVT61PkrtCNMoQkRlQIwQjk8rrTUoqsDljz1ubXU+HgnbV3OR1WitKDnsR3z5Fxh+yWG3FRAxFI05WNSzZ0/y8/X8MaNGjaKgoIDvvvuOiy5qUIc4mAyqJUyfPp30dP2JUAjBfffdx9dff40syxQVFVFWVkZubisEMo8jTs4nj65jdf2pxoy9DpJzkGWZK8Z3jUpqZJQlZo3uGlfTdqPExaOiJbXzu6ZhayMzfNnYrhgadc4gSzHJq44Wi1FmeJfo43LhqM5YTfpp0D8nGWcT5/rVp/TEWR/Zc9Horhgb9U2W4Irx3XSfh90JE2+O3qE1DTo3iUw7BHZzbFTRhSO7RCK9WkNeqo05E3tElSVbjIztFV9UkGa0ofWYHFWmDrkY1RifY3hmfhfMjfxOkgRXTeyK2RyHP8eWpuc1aUxGH/3YtjNpdjNzJvWIKrObDZzWv2M0weS65pM+Sa4jJ4MyyjIZSdHH12yQI+evpdG0scFgoLq6mrS0NNavXx/527Ztm96W0RgJRxdCRNaLNYfD0TD1/cYbb1BRUcGaNWtYv349OTk5BOIIzz5RODmfPBzZcN1XiG8eQ6orQYy7ESmvYbV3nkPiwxsn8PelezDKEr+e2pssa/RdiqqpVPorMcmmqIif5CQHt53Rl77ZSXy6rZKRXVOYPaEHmaktizCpCyiEwhpOuxm50YCc7TDy/o0TeWapPjV28+l9yHa0jZMvL83Gi78Yzbb9pVi1AIo5mX6dMkh36BdSZpKFf99yCs8s3UVxjZ8rJ3RnbI8GiZDcVAuf33YqRSUlSJIgN6cTeWmN5vy7TdR9HD88r+fmOPUu/beIg3SHhecvH8lr3+9jVUENZw7K4dzhndrEeAAM7ZzK4hsnUFQbQAKGdUnFaY0+roqq4fYr2MwG7I3Cm80pWagXvIpHE/hUmSRDGLMBLI44o+DsBlbedzoBJYgmBEkWGyLOtMUYLTD2WkjtCpvegZwhMOFmPSChDSl1+dCEoFMTv9+A3BRev3osL39bQIrVyC3T+jYrLHo4AoqKNxhu+W+Z2hlczUzZph5Z602SJZx2M0ZZpsanYDUZyEwyY2riMNc0gaYJkpOT6dmzJ++++y4XXXQRQgg2btzI8OHD6dGjB2vWrOHiiy/mgw8+QFH03+5wUuwALpeL7OxsTCYTS5cuZd++fS37/sc5J6XxcIfClIQyWJt1K6asMI6Ak5FaMgfFRxRJIWTcy5RR+5EQBA0SIanh7r7CV8Fn+z7jg10fkGnL5LZRt9EtpRsWgz7ICsBuMTJjcA4mCdQWLJJTVJWCSh8PfbyNirogV4zvzpmDcnHWa0gZjSZC4QAT6u+KQ2ENg7HtwjIztSpO3fUocsk61D4zEJ1vRE/0CLIs0cVp5/5zB6OoGsnW6P3aND89PRvoufpBEBqcdg+kjAbqJT5sadB3OnQdr4u/tTAvR1aylVum9sUfUnFYjFFPYK3FajKwt9LHvOUFOO0memclRUWZVXmCvP7Dfj7eVEL/3GR+e+YAOjsb+l+pWfnbf3eytrCOyb3TuO7U3sR7/x0IK+yuCvLk57sJhTWun9KdEV1bIItiz4D8n8OAmfoxbcMkWVVeH3sqAjzx2U4Cisq1p/ZkVPc0clL0m6HMJAun9M1icKcUjLJMcgtDhMvdAZ5Zuosf9lQzoXcGs3rH/5tK0/4E//51tB/NZINpf4xrewEg6eHOB9+LRteqompU1AWp8oZQgmHm/+s1fn3zTTz44IMoisKll17K8OHDufbaazn//PMZO3Ys06ZNizxdDBs2DKPRyPDhw5kzZw5OZ/ST8+WXX865557L6NGjyc/PZ8CAAXF/9xOBkzJUt6jGx7QnvorIkwA8OmsYF9VPTW0t38glH18eqZOQeO/cd+mT3h9VU3l7x9s8vPLhSL3D5OD9898n15FLXZ2bP3z0I+9vKI3U989J5o2r8smMQyep1BVg6uPLIhIgAI/NGsaFo7ogSRLbStz85G/fNPRNgo9/PZkBea3XYFLcZZheOw8qtkfK1CEXo/7kUczx3EWXb4PnJ0Q7b6//GvKGt7pv7c2SjcXc9Oa6yHuLUWbZnaeRl2bDH1L5v4+38a8VDXeGXZw2Ft84iaxkC5W1bq56bSObihp8DzMGZfHX8wfgTD3y77Kr3MWMJ7+N8um8c/1YxvY89pLweyo8TH/y6yj/wBvXjGNSn9ZHUFV7g9zw+lpW7m1Ycf7ahZ2ZMHJY/AKEG9+BL/6iP4GkdtENx7CLj7iZqqlU1IUobxR8YpRl+mQ7MBsNhFWN/dW+qGCAJIuRbun2FokjHk90dKjuiXmUjsCqguoowwHwzuoDlNR60TSNhbsWR9UJBB/u+gCASn8lH9S/PohX8bKtSp//9Ksy/95UFlW/o6wOX5y+3c1FrijDAfDGD/uprdefWrAqOvGTEHqCqLZAUnxRhgPAsPU95HCcaT3Xv0mMx331vNiy4wyXX+H176OjzIJhjVUFujO/LqiweG30PHphjT8SQBBQpSjDAfD5tgqCWnyXz4cbimKCAd78oRB/8NjPfy/ZVBLjWH5r5X48gdY7xQOKFmU4DpY148c+NMMuhts2w/21+v84DAfozvKmmm5hTYsEf2iCKMMB+vsW9e1/nJPSeDQXu5+TbMVskJFlmS6O2GiHLkl6ngBdXiH2rivTrpdJQotJ2GSUpSiH6OFoLulUbpoVU/1ahu4Zsb6THs2UHQ2SwQxN5RMcWYjYNEnN4+weW5be67jPu2w2yOQ1o/x7cK2CLElkp0T/LpLUoLba3O/rtJvjjrbq7IzddyenGeNxIGXRuZn1Q3mp1rjP58NhkKWYTICy1DGny8F0BjF9kvTvJUnE5O6QJel4P5WPK05K49E93c74Xg1O7hSbkdtm9CMjWb9QZvY+n85JDUlleqT0YGr3MwBIt6Vz26jbcJgaHIendDolIpuRnmRj7nkDoqK1fn1aD5KM8fk9ujhtUVpHyRYjd83oH5ErOWdoXpQB6Z5h5+xheTHtHA2qKYnQpLsaCiQZ5ewnkONdBT7wXD3S5yBp3eO+EzyW2MwGfnNG3yhJj3E9nfTK0n/jDIeZB386JCqS7OpTekZ+E7tB467pDdLwkgQPzOxHWpwhdlP65tAvp2FBYF6qlSvG9cBkOvYSIxN7ZzAwr0HTLCfFwpxJPTCbWu8OTbUZ+f050dMoaXZTm/qyDoXJaCQv1Rq1Gt1pN0WuW4MkRQd7AHlpHZQM6iSh3X0ekiQZgNVAkRBipiRJPYG3gXRgLXClEOKwz8hHI09S4vJT4gpQ7Q0xKC8Fp03G1ig8r9JTSoFrL7Ik0y2lO5lJDU8jQTUI3kpEyINssBAy2UhKasj14fV6qQ1JBBQNq0nGZhSkJ8cvb1LlCUb61j83mUxHtIRIRV2QvZW6amfPTEfcKXLjIVRXheyvRq38EWPuEMIWJxZHrCDiIfGUQ/Uefe1DRh9Iji8HSjy4fEE8IY2AomI1Gciwm7C2IGeFyxfCG1IRAhwWQ5R4oKoJKryV+MJ+TAYTVslKVqNkUP5QmBqfwvZSN93S7WQmWaK2r6514QrBnnI3/fLSSDFqpKbGii0eCp+rAkPYD0JDM9mxpcYfLRVWFMo9CoGwhsUok2QSpCY1nG+KqlAbrCUQDmAz2XBanBha8FRT4vKzv8qHX1Hpn5vcosRoR8LtC1LjC7GrrI6+OcnUlR1g8OBBbdb+4VA1XZ4koGgRPbbG0VaqphHWBEFFw2LSw3gNbSDyGX8HFT3wRJL0FLStfBI9GeVJbgW2EQnJ4a/Ak0KItyVJegG4Gni+rXeal2o75EWg1JXiXHI7me5SEGHC6X0InTUXc7J+h2/xVsObl0DpJjCYMJ/+Oxh1lR5NBOxzhblq/irK3EGSLUaevGQ4E3oYcdjjG+QzkiyHlBABfWqrLQ1GY8zJGZCcgTG7L3DEJWqxJGW3eZgo6IZjzf5afvP2etyBMNnJFl6ZM4YhneMboKs8QR78zzbeX1eEEDB9YA4PXzg0cpxr/GXc8fUdbKzciEEy8MuBV3LlgMtxJus3DTazEZvZ2KwMjF8Js7k8xItf/8hZQzrxwoKN/OG8wQxJ1uJSFFbcZZi/+DPGTbrPSO0znfB5z2FMie847q8NMnveSg5U+7GZDDzw08FM7StIT0kirIbZULGBW5feijvkJsOawbPTnmVQxqC4RfoOd620ClUlpXIdKe/OobvJBuEA26a/pfvIOuAOX9WIyJNomh5qnirLkScfgyxjkInSbuswwkGo3q3/R9KDAWzOVhuQjqRdzawkSV2Ac4B/1r+XgKnAwvqPvAr8tD370Bzars8wbF8CxWugZAPGLYtgf30+j5APlj2kGw7Q7w7+ez/49ERCJTVebluwgTK3HsVRFwzzmwUbcIX+d3T82wNvSOPWt9ZHkhyV1wX5zYL1FNXElydlQ2Eti9cWRXz3n28r44vtesIlv7+O17b8i42VGwFQhcpLW+dTEYwvkVWNV+GWt9bx7a4afv/+Flbtr+XWt9dT6o5vBbIo2Yhx4xuRwALDrs9Rt34Y17bltR7uW7yJA9V6UINfUbn3vU3461Pa1gRruH3Z7bhDer6SqkAVd3x1B1WBtknS1Sp8FfDOlVBXoj+tuov166jpav12QNU0imsDBMMqobDuKC+q8Te76rzD0VRwF9UbDgABrgN6+QlEez+jPQX8FjjoEMgAaoWIZPgpBJrNaC9J0nWSJK2WJGl1RUVFm3VI0zQs+1bElBv2fae/ULxQtCZ2wyp90Z6QJHaURS8M8gTD+BPGo1X4FZW6JtEvu5pJTnUomkb1AHy3qwpV0wgodayr2hxTv71ya0xZcwTDaoxMx95KL1qcU77S/thEY5YDy1FDR45yU4TElqLoRFaKKiLReSE1RE0wWgKmyFNEuAMG6COihvRpzsZo4Q6JztME+JToa1JAJNqqPamtreW5556LvC8uLo6IKuod0UBp5qZIPfYJvlpCuxkPSZJmAuVCiMYjcXPPqs2eSUKIF4UQo4UQo7Oy2i4eXpZlQv3PjilX+/9Ef2FJgT7ToyslCbL1uUQDgtHdoxcDZTjM2MwnzuPm8YjNZIiZqhvRNY3mT5lYpg2I9b2cPTQXgyxjtzg5rZlkX8Nz45NOsRgN5DbJIT+kc0r8jt++M2KKgv1mYjAfearIYoAJvaNlVOxmQ2RRqcVooZOjU1R9P2c/zPKxTxaF0apH4zXGYOqQKStZkkhusqJdliRMHbCGo6nx6NSpEwsXLmz4gCTr40wUUpsu/uwI2vNITgLOkySpAN1BPhX9SSRNkiLp4boAxe3Yh2YRXUajnPpbmPkUnPM4yun3odUbB4wWXf5h+M+hzxnQeypc/BrUS5TkpDl44uLhTBuQzWn9spjSL5N5V40hM86ESwmaJ91uYt6cMfTO0qOSRnRN48lL86NWeR+OPjlJ/PbM/jjMBqwmmRum9IokqrJYbPy0zwWc32smRslImiWNuRPuJ9UQn4R+dpLuf+mbrX9+WJdUnrlsZNx+AuHsiTL1Af28mvQblIm3Ifc6Na5tM1Mc3H/eICbWG5AuThuvzB5Nikm/50q3pvPcGc/RN033YQ1KH8STpz3ZJkm0Wo0jCy5bEEmFQO4wsGfqBiROluxZwoyFMxj26jBmLJzBkj1L4trOIEvkplojKgkmg0yPDDsGSaKgoICBAwdy7bXXMnjwYGbMmIHf72f37t2cddZZjBo1ismTJ7N9u74mavfu3YwfP54xY8bwxz/+MZLIyePxMG3aNEaOHMnQoUP54AN9fdg999zD7t27yc/P56677qKgoIAhQ/SEWePGjWPLtu26fI8lhdNmXcuaTTvxWnL45TXXM2bMGEaMGBFp67hGCNHuf8BpwEf1r98FLq1//QJw45G2HzVqlGgpSm2JCKxbIAJfPyXClXuEx10VqavzeUVhtVe8/M1uMW/5HlFU4xN+vz9S71N8Yr+rQMzb+Ir4dM/HotJbEdV2hdsvNhyoEX//Yqf4Zme5KHP5o+pDtaUiuOMLEfzyEREqXCcCrvK4+x0IBIRSWyQCK14SgRUv6a8DgUi9y+8RBbX7xT/Wvyw+2LlElHribzseNFeR0Na/LbSvHhVaxU4RclcceaM2pLDaW//na/G2gVBYlLn8oszlF75gOKrO7/eLKnelKHEXiRJXkXB56lrcfnGNTxTW+ERJrf/IH25EOBwWRTU+8ep3e8VLX+8WB6q9whdUIvWKzy20qj1C++ZJoa15VWi1RTFtlNZ6RGG1VxRVe4SiKDH1Vb4qUeYtE9X+6hZ/r3bHUyGEq1gIT4XYunVr3Jt9tPsjMfq10WLI/CGRv9GvjRYf7f4o7jaUsCpC9X+apgkhhNi7d68wGAxi5eo1oi6giHN/eoF46ZX54vTTp4qdO3cKIYT4/vvvxemnny6EEOKcc84Rb775phBCiOeff144HA69bUURLpdLCCFERUWF6N27t9A0Tezdu1cMHjw40ofG75944gnxxz/+UQghRHHhftG3bx8hwiFx7z33iNdee00IIURNTY3o27ev8Hg8cX9PIUSzxxZYLdppXO8QeRJJkk4D7hR6qG4vGkJ11wFXCCEO63lsaahu2FWC8bXzoHKnXmAwEf7lFxg76zIahdU+zn76G9x+fV443WHmo1sm0SlNX1+xsWIjv/j4F6hCnzPt5+zHi9NfJMOWgT8U5s2V+3ngo22R/V0wsjP3/mQAWclWgu5yDJ/ei3FLw2OqcvbfkPMvw2A+cgSVWluI4R+nNEiZ25yo13+LIU0Xg9tYvpnZn1xJuN5t1DetL89Pe5GcpNbLSWiuYuQ3L4Kyev+AbEDM+Q9St/GtbvtYs724hvOe/Z6Qqs9598p0sODaMWS1IIPk0VJc649K8JVkMbLk16fQPUPftyjbjPTS1AYHanovxJwlSCmdDtXkCUtLMgnOeHcGJb6SmPI8Rx6fzfrsqPtQUFDA9OnT+Xr1Rirqgrzy3FOElTAvP/M4/fr3j0yUBoNBtm3bRkZGBmVlZRiNRtxuN506dcy7SoUAACAASURBVMLj8aAoCrfddltEcn3Hjh3s3buXQCDAzJkz2bx5c2R/B98XFRUxffp0tm7dyt/+9jfKy8uZO3cuo0ePJhAIRLIPVldX8+mnn7Yo6+JJKU8ihFgmhJhZ/3qPEGKsEKKPEOKiIxmOo0Er29ZgOABUBfmrhwi49QiU13/YFzEcoOcneK9ensIddPPU2qcihgNgZ81OCj26umeNT+FvX0QnOFq8rohgfWpWoxqIMhwApq8eQPPFOnSbQ137RnQODH8N6ro3AajyuXhmw98jhgPgx9of2esuiKvtIyHV7GkwHACairR0LmFX7AV8IuH1+njiv7sihgNgT6WXrUUdk9Tok80lUQm+PMEw85YXEFLCqL5a+OrRRpE36JFJB1Z2SN+OZ0p9pc2Xe5svbwlmiyWSEM4gG3DV1pCUksqq1WtjJNkPxdFIrnfu3JmMjAw2btzIggULuPRSPSuiEIJFixZF9r1///4WGY5jwUm5wlw0E8kgK36E0BPwNNWWAiJlmtAIhmPtWaBR8p1gE90sIRolnmku3E7xx+0jlEOxEs9yfdY4VaiH6Fuc2lRHQmmmHcWPJNo/QqU90YSI0ToDojILtifNn2+6jpLQwrrmWFOaK/sfI9vW/ALU3GbkhY6KRrMuScnJdO7ajUUL362vEmzYoGcjHT9+PIsWLQLg7bffjmxzKMn1I0m1X3rppTzyyCO4XC6GDh0KwJlnnsnTTz8dyTS4bt26Q25/vHBSGg9Dl1HgiJ7GUSb+BltqNrIsc+X47lG6NxajzMWj9WmhNGsaVw+9OmrbLFsWvdN6A/qUw+XjopMzje+VjrV+oZFqtKF1jZ7mUUZfjzDH55xl9FVgaBR1YTDDqDkAZDvSmT3ol1Efz7BmMCCjf3xtH4nsQZAcfWGKCbdgSGs2mvqEITnJwY2nRutyOe0mRnRvn9zbTTkvvyHhFujO3Ksm9cRqNmJMykRMvDV6A5vz/9k77zipqvP/v+/0urN9F5beO4gIgogNu2gsMRpjYosp1sSYqmkm5muaRk1UjBpNTNCgsWIBVBDpICC9LuwuW2Zmd3q/9/z+uLMze3cGGGBR9Def18vXS+7dc+8z5557nnue5zmfDwwsLKH+RcZtJ9yOSacN9Vr0Fu6YeMcBWhweXN3o5R+d/Qz/eOZpxo8fz+jRozNJ64ceeog///nPTJ48mebm5gyzwDXXXMPq1auZNGkSzz//fIZyvaKiglNOOYUxY8Zw99130x1XXHEFc+bM4cors9Q+9957L8lkknHjxjFmzBjuvffeHvmNxxJfSEr2eDyGIe7D17oPBR1OZwnCWpqhhGgPhvFGFGYv3oVep+Obpw6i3C5RlubpD8QDbO/Yzpytc+jj7MPVI66mxp79CmoPxEgImURKYNBL2Ex6Sm3ZUs5UoBVl/QuY9q8kMfxLSINmYCxwN3EoFMAWbUFa+jAAyrTbiVprcTjU0j5PxM9u/05e3P4fqq21fG3U16i1VRe00zljf1ilonBajJRYtCWowtcAK55ACjQiTrweuXIEhpKeoyA5FIKxpGqbWd3x3WPXDYXZ6Y3x9Ed7qXaauXH6AGqdFnQ9ubs42KJuKrVVgCnLTxaKJgjEU2xvDSEB/SvslFgMlKd3v6eCbvS+PbDicbXt1FtQHL3RG4+P0s1YIok7HEcnSdSVHl2O6HByHkJReG3Xmzz88cO4o63U2Gq5/YTbmDVkVsH3UxRF3Vsi6dDpteOpU/wrlBaqclmNeUt5I5EIVqsVSZKYM2cO//nPf3qkGkooMkKRQZLQHUYF2oHwRaQn+dQRk2Fnh4nfzk/RHo5z1UmlXDrRSWdhpVMJUuZdyf1VOwCBoWM0KfNEQH0xUikL2/dW0U+5AeHT4fVbKDPLmNITTUII/jx/Fx/t9DCql5N7LhpFaRfiW0NJDcq075JMXI/Jehi8UYBOTpDy7MJUqmqPyN7d6Hplyy4rbS4qbScypmI0Br0e02EOur3tIX7zxmY27w8xfVg535s5jNqSrPFSaV+UM+5BkRPorSWf6gBp8kW57/XNfNLkZ/rQSn5wzjCqnLmMtEeCqICtrX6GVDsQQHMgis2io9TQA7QcyaiaY3vnp6ruxJjL07LH6iounEixpTnIn+dvJ5FSuHnGIGZ0Icc0OKvAWYVcPQpJZ0RnNB8+bcwxQmsgypsbWvjX8r04LQZ+eN4IhtdaqXQc+0IDSafjkqGzmDX4QoDD+kACVLXGUBu6mA/0ZoSrDgxmpDSzLulws8WoR1HEAXacwZo1a7j11lsRQlBaWsrTTz99ND8rY5sIt6GL+kBvQrj6aG37HOALufJoaI9w5p8+IClnf9uvLxnNNZP7otfrEXuXIj1zvqaN+OYHSHUnIMsKzyyt5zdvZpNlFqMqHFTrstISiHDvK5uZvzmr6TGgwsbzN02hruzoqdOTDWswPnWm9tiN72Hse+JRX7vZH+Kq2avY683G088ZXc3vLhtFhf3YTwYHgycY5yuzl7HLHc4cmzmymj9fOSGjBHekkGWZF9bs5acvZ5+pUS+x8K4Z9CsvMJx4MPgb4a9TINFlR/xpP4Lp3wejhS3NAS54+EPNxup/3jCZU4d99mJQh8Kr65q4Y866zL/1OokF35/BwMoj67fDWXkcDRRFhmAzunAXdgpJj6gagWQwkZIV6r0RIols8YnNZGBAxbEXgxKKjAi1oOu6+76LbUeKL2S11aeNVfXtGscB8Oq6/bQGY+oydt1/chtteAEAXzSZqbzqRCypsLVFTYApCizcohWDqvdGMtVWRwvd+n/nHtuQx94jQDSpaBwHwILNbSSOAyaLSFLWOA6AhVvbeiSp7Q5HefVj7TNLyoJlu3uI9sazXes4ADa9DOmJ640N+3MYOf67ppFwDwguHUu4g5EckSxZEby/re0ALQrDp/LBqsjqV73mxjJCUWldFIHGcUC2iOFYQxzItqOgJ/ksFgFfSOfRrzx3BdCv3IbNpFeXvpVDchtVqUlns1GXt32ncJAQ5OwsNul1mA0905Wpylyd41RFz3ypWQz6HIEclUX2s1995uvDGqeFnpB+sBr19CvPDX8NqOih1ZYjT/VPaT+VngMyu+a7YnCVvUcEl44lrEY9/fIIkQ06wlUHgMViwev1HvvJTpK0hSedh3WG7Olug0uv+5TEoA5km/7IgsRCCLxeLxZLz4R4C8UXMudRV2Zl5shqFmxRv5CqHGbunDk0m9Qe+2VVUrVTkrV2HAw7DwCH2ciPLxjBij3ejIzlJeN7U52OvVfZjdx/2Ri++ewaErKCJMHd5w3H2ENBav2oi+HjZ6B1k3qgZjT6URf2yLVtJokfnjeU++dtRwh1wr7/slFUO44BHfdhwmU1cu+FI7n3tU0IoYaV/vDlcZTbj56a3mW1cttZQ1m03ZvRtD5jRGVmk95Rw1au0tl0rhotpXDObyEtsjVtcCUT+payrkH92hxYaefKSX0x9oDg0rGEw2Lm5hmDWLC5lf1+tVR96uAKhtceufPo06cPjY2N9CTZ6YEg5ARSyKMSEQLC7ISOHUg6PUKoOh7t4QQClUGtzG5ih09XMJV9z9rmgI6dSEdIyW6xWOjTp09PmnhIfCFzHgDNvigpRZBIKVhNOmqcZvT67INp9wUIJQWSJGE3CMpLs7oRiiJojXiIpsKY9WYMWKjpIhzkDoZJpCT2eiPUlVkxG6C2y07lSDxFIJYiklCrOCrsJk0cNR7qwJAMI+Q4wmjH6NJ+uSb8LUjpeKhwVGPqdr7ZHyUSlzEbdZgNuYSCB4MnHCYSFzT5wvQrd1Bi0ePs8sWSsS0VQ5gcOba1Bd3ERBwhBBadhRpnz8Xtg7EkgWiS/f4YfcusuKxGTcVVayBGMqUQSynYTPq82hsHgizLeIMRIgkFk0GPWa9Q4cqS06VkdSIJxVPYTAacFgP2LsR6saTKrBtOV+aU2YyZAgqAWMCNUY4jUlG13+yVYMjmapp9UdojCVKyoNppptdh2P5Zo9kXpdkfxWoyUGY1UFvaM7LIhSAe9GJIhkFJIRsdmLqJaHl9AUJJMOgl7HpBaZdnmkzEkaLtKB316BzVJE2lWEuyJJOdlVaBWJISixG7WY/Lml0RhGIpQvEk0YSMw2Kgwm5G13W1EulQw5VKCkyOzMdCJ1rDrURTUQw6A2a9mSpb9rycikPEg9JRj2SvRrGUYHL0fFVjsdrqMNHsi/Doe7v496p9CKGGDZ674aRMQtvtD3P7ixtZtlvdYXz60Ar+eMVYKtMOYH+4mZvn30RDsAEJia+NvI6vj7yOWqda9dTQEeM7/1zDiF4utrUE+PEFw5k+WFDpdBBJpHh3cys/nLuBhKxQajPy75umMKq36pziQQ+6D/+IftXjagysYgjyta9m6EeSgTYMr3wLXVDd1S07e5G89MlMqW9je4Rrn17JHk8YSYKvT+3PLacPobqksCVrpd0OdvImiuNBL9JHD6Ff8YhqW/kg5GtfQ1+mVn61hT3cs+xeljV/BMCkmpN44NQHqLb3jANxWow4Lca8hQct/ijPfFTPkx/uRhHQt9zK8zdOoV+Bqwd3MMa1T69lR5rm/Wsn9eaus4dSVqL2w253mKufXI43nMCol7jvS2OYNa43drOBREphyQ4Pt/5nLbGkgtNs4NkbJjOhbyk6nUTU70G/+kn0S/6gfkmW9iN17esYKgYA0OqPce+rG1m2y4teJzGk2slj10ykJo+u+vEG4W+i15yv0CsehEQYMegM5Jm/Qu869tQpyUAb+rd/hH7zywDoe5+AfNUc9CXqB43bH+b6f65nY5qy/pJxNfz8wuFUuNQKx8ZAiqtmb8JuNuAN7+LWM4Zw1UklOC1GUrLCxiY/Nz23mkhCxmbSM/vaEzl5UAUGvY5gLMl/1zRy/5tbSCmqw59z88kM6gxBhr1qdd2G9KbB2nFwzdyMsmZbpI07P7iTjR6VsWFmv5n8ZMpPqLap77HeYIaSOvQln989VMd30PUI4Q4leH7lvkyScpc7xF/f34k/rO6gXri5OeM4AD7Y4WX5LnUZ3R4J8sjHj9AQbABAIPjnlmcIJtW/b/IF+cGLn9AaTLJou4eWQIKfvLSZhKx+kQSjKX700oYMFYYvkuR7L6zHE1LDJfpQK8aVj2V3t3p3Ij54gERMTRYrOxag2/MBeLaBZxv6PR+g7FwIqF9KDy3YnpGoFQKeXbo3c+2jhT7ixrT84axt7bsR7/+WREQtFli+f0XGcQCsbl3F+/sW98i9D4VgLMUTi3dnEpoN7VF+/842vAX89ng8zhOLdmYcB8C/Vu2nyaeGYtrDce6eux5vWE1YJmXBva9sJJgWpuqIJPjeC+syu9SD8RS3z/kYb1i9tzHpx/ThA5kQBL59SPPvIRpUx8zHDR0s2NJGOCETiKVYu6+DV9Y1qcUbxzFSsQisfEIVRuuoh7Ab6ZMX0XXs/lTuL1o3Y0g7DgD2f4y85jnkVJJUMsF/VtRnHAfAqxtaMxow/miCe1/5hNZAnN3uMP5IkvvnbcmIjbWHE9w+5+PM7v9IQuaOOetoT4+BYCzFfW9sJpUecG3BOPe8shFfJJ3U9mzLOg6Alg2w+mmQUyTlJC9tfynjOAAW7FvAtvZtPd5HnyW+kM5je2suNcDWliChhIKiKKxpDOecX9uoDsJoKspO3/ac83v8+wCQ0LHbo20fTcqZqqBIUs6pvNrRFszMx7JHy4sFYGj7BBFXr2ncvyrnvLFJ5TkKx1NszfPbdrbl/p4jgezNnRQMrZ9k6F42ej/JOb+xfd2nMgnWe3PpOra2BHMqZvIhGk+wsTm3j7a3+AFIKYLtrdpqqaQsCKXFqRIpJUeoqrGLKp3csS/n2vq2TRm6lw0N/pzzG5v8JOTjO2QskiGkllwRLfav/3Tu35x7H1PLWuREjHg8wbr9uWNiY5Pa1/GUwrZuz1QItSQcIKmIDLdVJ7zhRKZK0xtO5FTIbW8Nkuh8t1ty3wWaVkMqSlyOs7V9a87pTd5N+X/o5xRfSOcxqX95TpXOWSNrKLcZ0Ol0XDw2N8xy4Rh1KewylzCj7gzNOb2kZ1RlmgJEUjLaCp2oKTFjTdNP2E16qrrpk88YVpWpJDLUnaCKwXRBYthFGOyqPnpq1KU5tqVGXQ6o7L8zR2rjojoJJvQtTOf7UDD0GpujoZwYdiF6m5rvmdn/7Jw25/Q7/7A3bx0JRtY6MXR7qGeOqKbcfui6eKfdxoWjtM9MkmDSIHUcWI16Th+uHRMuq5ESqxrVtRj19C3X5igm9ivL5Dz0VUNzqmeSQy9AZ1Wfy9mjc2PZ54/thaWnqiyOEfS2CsSIPMUaQ876dO4/bGbOscTIyzDZnNgdDmaNrsg5P2O42tdOs4GzRmjzI1ajPlM1aTHoGFGr3cA7rMaRoZGpdpqxdRN4O3NENQ5LOtI/6PRcg8dcDiYHDpODmf1zbT+tz2l5f+fnFV/IhLk7EGZtQ5D73thMezjB5RP7cOuZg6kpUScAbyDIS2ub+NvifeglidvPGMCssTWUp5NtbWEPL257kXJLKRE5wvDSUYwsG0ulIx1L7Qizfp+P8X3L2NsepsJhYkCpDYvFiKwIdrWFeHzxLhRFYDXquePsYRklukQ0iLRvGca374ZQG6lxV8PpP1F3GQNRvxv95pcxNa8BBIleJyGPvgxribojuS0Q48kPd6ssrRJcPK43E/uVUtoDVUnxaAhd40qMLevBZEMOexEn3YTBqb6EbSEPb9e/w1ObZiMLmetG3sAlgy+mqgfo4A+FjnCcT5oCvLy2ESQot5r41umDqSkw19MeCPO3D3bx79X7KbOZ+NVFwzh5YBmO9ObItkCMhxZuJxKX0UkSN506kGE1Tgx6HUII6r0R7npxHZ80+ZkysJzfXzE+k7CPhgIY2tZjfPNO8Dcij74McdYvM7Qu7mCM97e24rKa0euhoT3MhePqMhV8xzOUQAvSskeR1v4DzCWImb8k1f80jK5jT1kTD3Wgb9mAYeB0AOSm9Shl/TA61fHWEQjxrxX72O1NIBCcOayM04eUUVKiOm13ME69O8D4fuWE4in2eiOM7l2ScfoN7RH+smA7SUVg0EncOXMYfdNl+omUzJbmIHfPXc9eb4RzR9dw70Wjs8UpMT/sXADv3gPxEJz0TZh2i0ovA3giHv699d/M2ToHi8HCrSfcyow+M6i0fjp8ap04lgnzL6TzAJDDXry+AEg6HFYzBqMZk0P9gg5FAihCIppQf7vVLGGQdNis6kTSHo7TEojz8ppG6sqsnD+mltouezu8/iA728K8vbmNKQNcTBpQQaWrSwI67Eap/whp33LEqEvQVY0AW7ZaS5FlUiE3EgJhcmKyZtvGEkm84aQ6SQKXTexDpdOIOV25E44nafLFmLumkT6lVs4bU1twshzUiqUVu72s3tvBuaNrGVrt0LRvDbuZX7+Q/eEmLhlyMWWmaqod2ZVNJBbBn1RDZ6UmB1ZzD+5MD3tg33KoXwIjLlQV6GxZahZ3MMa7m1qp90a4bGIdfcqsGaW4Q8EbihNLyHREkwgh6F1qocKR/d3uQIx6b4R5G5sZW+filCGVOY6pPZwgpSiY9Xpc3ZQjQ7EQ+mgAvQRxnYmkpKc8XaEXi8fxRmReWttIPKlw5Ul9KTMLShyFlby2hTpoCjfw9p55jCgfzbS6qdR0If5MBJsRjasw7PkQeejZiNoxmJ3ZhHaLP8qCLW3scoe49IQ6qp1GTXXgoZCM+NAnQoBA2KrQmz4dp5dKpfCE4rz2cQPBWIorJvWj1KrDle43fyRER0TdcOmyGpk1vhfVdhP6tCaGO+ihOdrCm3vmMcQ1lNP6TtcUd7QFYmxs8rN4h4fpQysZV+fKeZc8oTiKENiMehzdx5osQ9SrxsMsLjB2aZuMEY+2E4h6kQCXoxajtQI+hVV6VxSdx2GLQbVgeOYc8KkUyUg6UtfPx9BP7cN9gUaueP0yomkqc4fRwdxZL1Pn7AXAkh0ern16RSbmOajSzvPfnEIvl5VQJMw/l+/jgXez+YELRldx36wRVJSWQKQdXr0Fts3LGnTu/eqXSQHUA/vaQ5z/0BLC6USew2zgrTtOoW+6Omp1fTtffmJZxrb+FTbmfntqQRxQbYEY97y6kXc3ZXdb33XOMG44ZQB2s5GWkJub5l/P3oDabxIST8z8O1PrJh/y2keNqA/evAs2dtFCOeMemHYbGC24g3Gumr2cXW41ji1J8K8bp3DKkMK+5Nbt6+Dyx5dl8hR1pVbmfmcqvVxW4kmZOasa+MVr2Zj0tMEVPPiVCQWtbFpCbr678Fvs8GXzWQ+f8VfO6Kcy4+73Rbjg4SX40vuGzAYdb91xarZy5yCIJ5O8tvt1fr38F5ljJ1ZP4v9O/QO1jkoigRZM792HYd2/MueT024jOe0ObI4qmn0Rbnx2DZubs4nlx742kfPH9DrkvT9rtPrCXPjo0kxuwqTXMe+2qQypVUO8W1t8zHpkaSZPUeUw89ptU+nlspNMJHi3YSE/XvLDzPXGVIzlodMeosZZTUc4zoMLdvDcsr2Z89dM6cdd5wzrkb1FeHbA46dkdVrslfDtJar87KeIIj3JYUJp3pB1HABCwbD4dySCqhjUC9tezDgOgFAyxGs7XwfUCfYvC7drkmW7PWHq00nycFzhr4u0CdJ5m9zElHRXJkJaxwGw+A9agaeDYM7KvRnHAapw0JzV6v0C0SR/mq+1ba83kqkwORRiKUXjOABmL9qdqUCpD+zNOA5QK81mf/IYzel+O6ZIhLSOA+CjByGmbqxr7IhkHAeoH3t/enc7HeFDUzq4gzEefm9nVnMFlYRxfXrTnjec4K/v79S0WbrLSzSPDkc+tIRbNI4D4IkNf2N/IE1Psr454zhATeY+vWQPiWSSQ8ETa+fJTx7XHFvTtppoSh2PJiWBoRuljXHlbEyyOmm1BRMaxwHwyMKdNHb0TJHFscSirc2apHZCVnh88R5C0SjBaITHF+3R0BC5Q3EWbVfHtyfWzhMbHtNcb6P3E3xJtS8iCYV/r9C+xy+saij4mR8UcgKWPKQV+Ap7YMf8o7/2cYQvpPMgn3iRUBBpGg4lj2BTKq0cKCAvv01m4pEklMNdreUTiDrQn+a7d/oFEYi8tA7yURQ7yUJknJGSp99kIedUnXxq6GJPvj5XhMg804NBOkD7rs4k328s5NoH+jvlELbLBdsu5X0u2rbdrtPlfvnuIX8aBE49gIM9M4FQmXAPeB4U8vRbpi9ze0b9dw/sLhfkn4OU44BErgfxhXQeut7joav+sySRnP5DzOlE21UjvoJZn12aWg1WLh1yMQA1JRZuOUPLfdWnzMqQajVZbjfCzaf01Zw/a3glZn16sJjsMLhbNcr074G1tCDbr57cXyMcZDXq+eoUVcjIZTVxx1nDNH9fV2pleG1htO9mg47TuzG53jB9ICXpCpJBroHUObSblm4a8216l+RWtfQ4jHYYcZH22NRb1Fgy0K/cTv9uPEt3zhxaUIih0mnhltOHaHiLqp1mJvZXcxLldiPfnDFQ02ZS/zJsBeqJ9LL3YkDJAM2xm8Z8i94lal/PGt8bZ5fd6ia9jhunD8RcgF5HubWU60bfpDk2rnI8Nr0a8krojKTGXKE5nzzxOhIGtV9qSiwMqdaGx75z+mB6HUae7LPC6SN7U9olt2TQSXxnxkAcVislVjvfOm2ghp+qzGbkjBFq1WSltZQbRn9Tc73hZcMpNavvocWo54oTtXQel52gFe06YhhMMP0O0HUZP9YyGHb+gdt8DnHMch6SJFmAxYAZdSf7XCHELyRJ+gdwGtBZ/H6dEGJd/quoOJKEuRL20O51o6CjxG5HMtkwp3eIByJ+kEwEYuqE77LqkEQKR1p7wxuK09AR5YWV+6grs3L5iX00ZIj+UAi7HESvJFEkHRFTFU5bl0ks7EbZ9g66fUtRxlyBrvcETeJXvUiT+nViKQNL9uWOJhK4g0n+tVwNH31tan+qHEasJnWiCcWSSMkQllQQRUgk7LXYzYVTlrcFYiza3sbKPR2cP6aGMXWlmiRhLBnDEHKjkyBhcuCXJWqcWceXSMToSG+OK7WZMJu1JayBiCp8JCFRV3aYFBxhj1rBsmcRjPoS9JmUqV4BaAvGeHXdfna7Q3x1cj9VVKlAunZ3MEYkGccT8YHQ0a+0SpMncgdjbG0J8tq6/YzvU8rZo2ty8h3+SJJ4SsZq0uck6r3RDvYF9oIELlMpDkMJ1Q71mYejEcJJHfvaI0hA7zILNr3AVaAmRmuonV3+nbxV/wYjykZzdv+zqO6SME8GWxC7P8C0ZxGJoedAv6mYuihCtvij6X4Lc8WJfehbbqHWlXXEcipJezCCAEqtRkwWrZNOphQ6ogl0SFQ4TJ8K9xNANB6jI5Jizoq9BOIyXz+5P2VWibIStSqyIxzAExb8a3kDJRYDV03uS7XdhNGoPht30MOe0F5e3/0qQ1zDOG/gudR0SZi3h2IosQBWKUFUmNBZSih3FO5UhRB4QwkUBKVWE6auxJ6JKMlEEF8yjISgzFyK3uLKKYU/1vhcJswldYTZhRAhSZKMwBLgDuDbwBtCiLkHvUAXHK7zaAtE2NoSzpTqXnFiH26YPjAzGbT4I7y9qZXZi3aj00nccsZgzhhepXmhQFUhy7eHQfgb4e2fIO1ZBDVjEBc9RKp0IEajEUUR1HvDPLdsL7KiYDXquenUQdkJOuqD5nXw9o8h1Abjr1aTwumXvdkfZtF2D1v2q/H9Ub2dnDasIlMdIwItsOwRpPX/AUc14rwHkKvHYHAUsDoQAtp3w4onEIqMZLTCtFsz946FvJjdm5He/hEEWxBjv4yY/j106fPtfh8Lt7bzlw/qURTBLaf14/zR1ZSnZTlbAzGeXrKHuWsaKbebuPeiUYzpXZJRzDukbR17YPkT6vLeYFb7pSQ3wSiEOOwJUBXIUAAAIABJREFUrCPWwfNbnmfu9rmUmEv46eSfMr5qPFaj6uDcwRhPLdlDOJ5CkiSuPbk/g6ocmS/bfe0RfvLyBj5p9HPyoAp+fcnoTAVeXI6zrX0b9y2/j+ZwMxcOvJBvjf8W5RbVeTT7oyzb5eXBBdtJpgQ3TB/AxRN6U1tyeM71QONRbq9H11GPcPVBat+DUj0yQ3dzqPZ+v4+P9vh5YP5uYkmZb07ry2Un9MqUrXeEEzy/Yi/PLlXFoO65aBSTB5Rn9zt8SshnuzvSzpt73qQp2AAIptROY2L1RMptrkO2BUh17EP35l3oGleg9JmMcuGfMZT1y/m7fAjFUizf7eW387YQiqe4ftoArp7cj7L0viNfqJVXdr/Kc+lS3e9PuJXJ1ZMocRSmKNpT+Fw6D81NJMmG6jy+k/7vmDqPvd4wZ/5pkSa2+9MLRvCNqf0wG418uMPNtU+t1LSZ++2pTBpQ3v1SORD+Jnj9NqQ0ZQgArr6IG95GcvWhLRhj1iNLaA1kk2UaUaOOvfDIRG3886xfwNTbwGBk2S4PVz+5QnPPOTdP4eRBlcjxKLoVf0N679fZkzoD4ra1SGVaje68CLXCk2eq4kWdGDITLn8KrKWIjn1Ij05UpVQ7f+/pPyF18m0YLXY+rndz6ePafnvhphOZMqSWRErmH0vruX9edmetXifx/l2nFcY/FWqDp85WaTA6Meh0+PI/1CX/UUARCnO3z+W+5fdljukkHfMunUeds45QLMlP//cJr61vzpyvsJt4685TqXaqlV5XPrEsQwsDMGVgOY9/7UTK7CbaIm1c8PIFxOXsM79p7E18Z/x3MOlNfNLoY9ajWVoXgCe/fiJnj8pD5X6YSPhbMP3rYnB3ob7oO5n4l/+NueTQnGPb9ndw7sNLNcf+fs1YZo5VJ9GX1zby/RezO70lCd6763QGVn624mEAr++cx08/+pHm2P8ufpUhZYMO2Tbqa8H6wlfUD7lO1I4ldtVcLKWHfi673SHO+vMiTa7s4asmcPEENey7cPdb3PnhDzVtXpk1l8Hlww957Z7E57baSpIkvSRJ64A2YL4QonNW/K0kSRskSXpQkqQeqIvTYlV9e05S8O2NrXhCSRRF4bV1+3PavLmhOedYfgikXe9pD/kbMlQU4biscRwA721tI5ZKJ80bV+Umzra8nhEOen19rm2vd05qUQ/S1te1J5UUNBboWBNhreMA2LUQUirHE/s/1jgOAGnrG0jRdtWOPH300rpWFEXBHYzz9sYWzTlZEazeW1iVGYmw1nEA7P4AkrHC2h8EwUSQN3e/qTmmCIWP2z4GVF6j+Zu1AkfecCLDbRVLyhrHAbBiT3uGhqbeX69xHADz984nmFD3w8z7RNsvoD7TUA+IQenkmNZxADSsRK8Uxnc2f1PueHtpvZtoJEQoluSVj7ViUELARzs9R2xvT8EXDfH23jdzji9uWFJQe6OS0DoOgJZPMCiFPZPFO9w5RRYvf9xEKJ4iEvXx2t63c9os2vd+Qdf+vOCYOg8hhCyEmAD0ASZLkjQG+AkwAjgJKAd+lK+tJEk3S5K0WpKk1YfL/Z9PfGdItQOrSYdOp2Nk75Kc8/mO5YUQUKZNrmKwZIR/rEZ9jsjPoCoHus4wS0UeIarKYWBSQ2ad7LtdMapX2jajHVE5LOc8FYMLs91gzdiZQdnALF1KRe4Xm6gcipIO7Yztlduv43vb0Ol02MwGhuTp90L2Mqi2WcDYjU23tH+PbKqyGCwMK8vttwGuAYC6QhpUpf2S1usk7OmEucmgw96NqqK3y5Ixrcaeu9t6kGsQpjRlyei63LE1otaJtSfoSXQmMHe7vrMWIRV27dF5xv2YWisGowmTQZ/3vRhWU1iBxrGEzWhmiCv3K35YeZ73Iw8UnUGTT1MvWo7QFRaOG1Gb2y+jepVgNugwGiyMcuW+58PLPt1Vx7HGp1JtJYTwAR8A5wkhmoWKOPAMkHcHmhBithBikhBiUlXV4VF+93JZuPSEbNVQ33Irt505hHK7OnFeMKaWCX2zSeCTBpQxY2hhm81kRy/EJX9V+ftNdtAZEec/gEhXb5VYDfzqktGYTRLDqh04zAb+9OXxVHbG/Z21MOFrantnb1Vx7oyfZKqxZo6sZmK/bJhmYr8yzhqZpnG2l8PpPwVXttpLnHAtwl5gHNXiggv/DDoLVI1Qbbj0CUgnEWVrBeLE67N/7+oDZ/4cU5p+ZPrwGqYMyPbbCX1dnDNarWors5m47ayh9OmSJL9sYh01JQUuLK2lMOsv2QoVk121zXZ4zz4fzHozN429ib7Ovqo+i2Rg1qBZmcqyCoeZP1wxLsNlpZPg3gtH4khXSLmsRh64YlxGhdFq1PPnr0ygIl3pVWou5brR12XuV2mt5O5Jd+M0qZPspP5lnDIkO1GN7l3CZRPrNPoyRwxbOclZj4Leoj5Ho5XkxY+jsxc2nsf1KeOMoVnbRvZycuVJ/TEa1QTwDacMZHAXx3rBmFrNvz8rmAxGvjryagaXDlafqc7AaX1OZ1jZ0ILay+ZSkhc/pubWAAxmErP+RtJcWFXk0GoH54/JhrcGVzm4btoAjHodRqOFy4dfydDSrC2n1U1nRA8pgh4vOJYJ8yogKYTwSZJkBd4FHgDWCCGa0wn1B4GYEOLHB7vWkVRb7feFSaRUxluX1UDvbgI2+31RIgm1Kshm0mvEeRRZpjWYwBOO4zQbsBl1VHdJpkfC7YREHG/UTZmlArtkwNlFyCUVaEEvJyDcBs5aZKMDgy07KGOBZgzJCCIRBlsFekctui4TSbM/Qjguq7aZdfTqnsj3NyJFfWCyoxisGX2DQtAa8pAijjfaTpW1Eotko8yeXe0k/c0Y5Jiq3WAtRefSJl49/gChuFoj7zTrqHRpv0L3+6IEokmsJj1mg05D60IqAdF2NfdiqwRLCZi7tE+EVc6gaIf6VWgty77cqNVW0YRMIJqk0mmmxmk+LFLGZn+I9nASm0mPzQg1XYWDZIWOcIL2SAKX1YjTbNDQUXQKfPkiCcrsJkqtRsxdVg6+UCtRkSScCOEyuyi3VqHvIiva7A8RTSjIQuAw6+nl0q7IWkJt+OI+zAYzdoOTans2/xZLJOmIJmgLxii3m7GZJCrs2QlcDrShU+IQagFHrTomuvKNRTsgHlT71l6t7nbuUvXT6gsTTgpSsoLTYqBXt3fFE4zjiyYxG3Q4zIZMUvh4wH5/CH8kicmgx2aW6FXSZTwlYxDxqPk0a7k6lroUYASDfqxKGBH2ItkqiOptOLtUFnYEgoSTEv5InEqnRRUAM2V/e0c4QTCeIpFScFmNOaJs3lAzgWQIg86IQ2+mzNGl+EOR1VB12K1+1JlLCi7nPxx8LhPmkiSNA54F9KgrnBeFEL+WJOk9oAp1N8464NtCiINukT5c59HsD/Lbedt4c0MrBp1ElcPMC9+akqH48PjDXPfc2gwN95jeTmZfc0JGDKreE+bKJ5ZlJEu/fGIffnDOEGpcdlKpFGvd67hl4XeIyTH0kp4fT76Hs/qcTZXDhRxsQ7fpJaR3fqqW4prsiGvmIvWfBkA82IJ+/s8xbHhBNbakN6nr5mEoV0NhnkCYG577OKM9Mbzawd+/fgKVJUf/tecOd7Cw4R1+t/J3KELBarDy6JmPMrFqIgbDoZfr3nArd3xwF9s7VMr6ga6B/PXMh6nME7bJi6Y18OzF6m5ySQfn/g5O+BqYDx3aagvEeHD+dv6zStVZUZ/pyQWHxfZ3hLhy9koaO9Tc1EVjqvnVRcNVSpmjRCLUirT4DxhX/V1dOVlcyNe/jb5S/fJsDXv48eK7Weteiw4dA10DeWzmE5my0f2hFm5853oaQ2o+6oy+M/nZlHuosasrgk+aOvjq7JUE4ykkCb539iCumdKfCrsNon7Y/ja8+t1sldqXn4OBp4HJqqrdLfkzLH1YNdZWATe8A2nbPIEwd76wnqW7O9BJEgMr7Tx/4ySqD4P76rNCsz/CNU+uzEgknDa8gt9fPpaazneldSP846Isu8Mpd8KUb2n3gB0AHYEg/1i6j798UA+oq8//fvMkhvU+uuKNDDw74OlzVDqjTtum33nUxSHd8blMmAshNgghThBCjBNCjBFC/Dp9/EwhxNj0sa8dynEcCZr9cd5Y34oQqi7Dfn+MhxfuoD2sDrJ5G5rY2BQgkVJIpBTW7vPz4XY1qekLRXng7a0ZxwEq8Zo3rCZPWyNefrH0XmKymsiVhcwfVv2OpFAnJZ0cR5p/b3aHaSKM9PrtCJ866emCzVnHARDYj/T+/cQiKlXGgk3NbGj0E03IRBMy6xr9LNxUaDL/4EiKGH9Y9YfMjuVoKsovlv2CtlhhOaWPGhez3rOeqBwlKkfZ3L6ZhfXvFnbzkBte+a7qOEDtn3d/BvHAwdul0R5JZBwHqFQU//fWVtoLEIOKRMI8+t6OjOMAeGNjGw2+o0/GA+iiPowrn1TzYXISwh6kd35GIqImlle3rGF122oUoZASKXb4dvDqjteQZZlIMsZTnzydcRwA7zcsyNDEtARC/Ox/mzJ6IkLAQ/N3Z0g9iQdg3l3ZIoxUHF6/Tf3iBoh1ZB0HQMQLb/1QLRkH1u1rZ8muDhSh6prsaAvx4sq9yHIP0HQcQ8RTKZ5ZulujrbNomzdLxRLYD/Pu1tICfdSNMuQgCCeljOMA8EeT/Oy1LSrZ6tGik8et03F02hbN1X05nvGF3GG+x5Prj3a5o0STMoqisLE1mnO+UywokpSp9+by/jS0q8IzOh00hbQVKAklQaKzSiMVzalYon13JimttO/JubbeuzMjuLSpJVfgJt+xI0FcjmXtTKMp2IRUICXD1o6dOce2+HYgChGDEoraD12hpDJVaodC14m/E/XeMNHUoSe5WCLJTk+uo9jjzhXWOhIogcacY7r2XZlKsZ2+XAGwnYFtxFNJosk4ewO5Ily7feoxRSFnPCoCgp2VWkpSDUl1RagNOqk5QtoqMkB9DunqsG0tuX2wqTVKKnV8U2nEkkl2teWOie2t6cldTuaON8hUNR4KvnDueKn3REj1RKAmFT+AbXme1XGML6TzOGlAZY5w0IXjq6i2W9HpdFwxITfM8qUJ6lK2wm7RJMJApZMYU6fmBSRhYHrv6ZrzdY46TLp0FZPRCt10icWw8xGSGhYy9DkR9NrdyYkxl2JMV35cOiE3f3HpCT3DxGnRW+lt1y7Zp9dNRycVVmFy4aBceoWLB12IVEjewWRTada7wllbUMgKYGRtSUZQqxPnjelFRQHx9xKHg0vHahPIep3ESYN6ZsOWvnK4+ty7IDnyYnTpZ3p2HhGtCwdcgs1soczi5Nz+F2jO6SQdU3tPAcBu1jNzpNbOMpuRUlv6d+tNUN0tEdtvKnQ+09L+avFBV4y6FCxqfL2z4KErvnxCDWZzj1fQ9yhcVisXT9D2iyTBmWl6EswlMLzbeDOXFBSyAqgqsWZoezpx/qhK7OYe2BxpcakMCl1hcqjFM58jfCEp2T2hEDtaI7y3xYNAUOuyMGt8bSYW2uEPsK4pSL+KEoQkaPIGGdfbQVl6p3RbIMpTS+p5+eMmakrM/OKiUYysteOwpneoh9r405o/sqJ5OcPLhvOzk++lztYLo9GISKWgYzfSvB9A60bE4LNg5i+RXKpDiUYDGFs/wfD2jyHcRmr8VxFTvo3RqTo0fzDE4h1e1u9XvzbH97YzY2gFLqd2kpUVBZ0kHdZO62QySVOkiZd2vIQsC8xGI18Z/hVq7VqHJYRAEQJ9N6cQiHhY1bqGmKzuwjYgmFI7GZddWxGVSqXQSZKmCABQE+ULfqXG6KtHwUUPQfmggspxw7EkW1uD/PK1zbQEYlwyvjffnDGocDEof4iNLSGG15aSkmXaAkGGVNhwOg8v55GUkxi7Of9kIozUtgXDW3eDv5HkmMuRTrkTQydrQLCDde5V/G39IyTkBF8fdT0z+51DTZq+xB1u54PG93EYbSTlJFX2WkaWjaI0rfPSEojw7xX7iMRlJEni8hN7M7jSgbEzT9W+B966G5rWQr9pcN7voDRdkScnoW0zvPl98O2D0ZfDjLsyFXbBUIhlezp44F11h/nNp/Tl4vG9KCspsMQ6jYScxKQ/MFVMNB7H2sMOyR0K89KaJp5b2oDNbOBH5w3lxP4uyjuLCQLNsPj36j6q8oFw/h+hckiOM03JCoZu5fWJRILdnjA/e3ULe71hzhtVxZ0zh+bmHhUFEAekHVHSrKW6btcn5IYP/wgbX4Ky/nDhg+pHQPc+lFOgP3KH9blMmPckDtd5JJNJvOEE/1nZQHMgztdP7ktvpyFDueANR/FHZJ75aA86nY7rpw3AZZMos6lVJmG/BykeJBSJIAmZ8spe6J3ZCdIXDOFLJFCkOELRU2EpoVQjLBSno70Vg0iiGO2UlLio7kJF4QnG6PA0o0fB6CjHbLZm6Esi8SStwTj/WFoPwPXTBlLtNGFL81cFY0nqPSr9Sd9yK185qV/BE6hqW5gFm9tYXu/jwjHVnNjPRUWXiaI1EOPF1Q3s9Ua49uT+DKzM8kcl4nFaggn+sXQPioDrpg2gl9OM2WLO9EujL85zyxvoXWLkqikDqC3t8rIF2yC4P5swLx8CzsK//v2RBO5QHAkJi1GH02IsmNuqxR/FHYzzz+V7qXSYuWZKP6odxgwP0qHgDnv5oGERa9pWMqPuTE6qnURVl4oof8yPHHFjBOIGKyazk5L0/otQNEwkKVHvU0Mmdc4KHAaBy6lWBoUDPhqCCv9Yto8Ku4GvnTyAGocZXdq2tkCQLS0R/re2mbF9nFw0rhc1JdmKKF/MhxT2YAZikg7JVoHL0m2/UNirhgktJZpVUjyVIJIIE4wkAAm7xYDRYKbEUljC3B3pYHv7Nt7Y8ypDXSO4cPCFGqGq1kCYRTva+GiHjxnDypg+pCqb0D5K+EIdhJUo+wNBBAqDyipxGkoyqyZfJIEpGcQsh5DREbfW4uwyXtrDCZbs8LBwaytnDK9mxrDKHKJNjz+IrAjsZgMOW5cqNEWGYDOsekqtIJzybXXlkHZM0WiUJn+cZ5buxWHS8fVpA6l1WtAbuziCZBRiAdU5dN9zEvZC/WLY+iYMOVuV/i2w/Loris7jcLmt/GEuenSZJun94k2TmDxE/brf7Q5x/l8+zOwQthr1vHPnqRkajeTWdzDOuTJ7QVdfkjcswOiqJRwO8tjivTy6KKt7cfrQCv50+UgqSl00+6J8/8X1LNud1cC4/awh3HTKAEpsZpr9US7961JaAtmY6rM3TOa0NNttvSfMuQ8tzthmNuh493sz6J+27aOdHq75e5a+pJfLwmu3nlKQGJTHF+CulzazaEfWtttO68/Npw3EabPjDsb50l8/osmXjSU/e8NJnDZMneAbvUFmPrSEWLKLbXecQv8qdZJcsbOFr/x9TaZttdPMG7dOVSt3EmFY9ld4/7dZg/pNhSueyctf1R3uYIzLHltKQ3vWtr9/fRIzRxVW6bVit5ernlye2RWsCgedoiG8PBA84Q7uW/Fr3mtYkDn21eHX8t0J38VlceCJevj6W1+nIZhN6D94+oOc1e8sJElif0eIsx/MCnwZ9RJv334Kg2vUCX5dvZtLn1iZsa3CbuKt26ZRXWonlojzr5X7+M0b2zPXHtfHxexrx1PrchKIB/jN8t/wVv1bmfNXj7iaOybegd146El6f6iFy1+7lFBSzRMaJAMvXPRfhpXn2czaDfFUkpe2/4/frcrSvowoH8GjZz5Gjb2S1kCQ++dt59V12R32V07qzQ/PHUKl8+g3Gm7zbObKeVdnCkBKTCW8PGsuNY5eyIrg5bWN3D13Q+bvh9c4ef6mKVQ6zQRjSe57YzMvrs7mq758Yh9+PmtUYeqUwWb429RsQl7SwbcWQ+1YALY3d3D+I1nxsRKLgXfvnK79mDoQ4iGY/3NY/VT22Ngr4YI/HHY57+ey2uqzxCcN7RrHAfDwB/V4/Woy7Z/L92YmZ1D3gry4Rh1EEV8bxo/+qL2gvwHhVl/ecFLiqaXaBOkHO7wZMaiErGgcB8AzS+oJJdT7bd4f0DgOgMc+2EmLX50U/71yn8a2eCorWuOLJHh4oTb52uxX2WALQUyWNI4D4KlljZnKnV3ukMZxADy8cCcdETU5O3f1vozj6LTtn2n23/ZAkIe7VKcAtAXjbGpMV5REO2D537QG7VumFhgUgL3eiMZxADzy3g7aw4eunmkNRHnsg10aOgl3KM6a+sKoU+JKTOM4AObueIFIWpCpOdSscRwAT2x4go64ev031zdqBL6SsuCpJfXEkwlCwQCPLqrX2OYNJ1i1W02eeiNx/r5YK1q0odFPKK4+h0gqwtv1WiqM/27/L+FkYWJPC+oXZhwHQEqkeG7zs8SSh6bpaI928NyWpzXHtrZvJZDwp3+nxOvrtdQsL61tJiYfPStvKBrk6U3PaLROAokASxsWqbaF4zkCX9tag3jT4mHheIqX1moLX15a20g4XmCV2a4PtJVcQoEP/wTJCIlEgscX7dFQJAViKRZsasq9Tj7Eg7D2We2xjf+FZM8UzvQUvpDOw9g9voiqBdCZH+ieTFfbqMcEkpaHvxOZWKTI277zSL4UhF4vZUScDHni+wadLlPxZMxnWzpRLEmSRr8ga3thjzFffkTf5Vj+a0uZQdLZR11hTh/TIR3wt2WQNy5c2ESSzzaDXldQzkeS1GeQ276we0uQU5Gm7/Jb9Hl+l0FnOPgz1avX1EkS+SQkusbg89mpS19TIjfvpS+QmqTTzpxjkqHgXFq+e+m6TCu6HNsKre07OCQOYLtOfU/VdyW3Y7PdKtH9sXS39aDIl9/RGQEdkpT/XelOW3RQdB9T0vE3VR9/FvUARvYupV95Nj6p10l8/6xBlKd3n147tX+GegJUSpHLJ6o7qe2lVSRPu0f7sCqHIaU38TlMEredoWWwnTW2GmtaDMqkkzinWyjlltOH4LKog2FYrUPDSKqT4I6zhlLjUsNOV03upxEOcpoNXDVJTX66rEZ+cO5wzaAfVGnPy+WVDxa9wkVjtDmG28/oj82s/taB3a6lk+Cuc4bjSlf2XHZi/wyFB6j66l89eQAApSUOvn/WIM0k37/CxvDe6WW2rRJmaFlGGXpOTpXSgdC33MawmqxtkgR3nzucMtuhq62qnVZuP3Ooxun3KbMyvm9hIQCz3sasQZdojl078nocRnU81dhqNNxZEhK3TbiNMou64eu8cX0o71IVZjXquXH6IExGIzaHkzvOHKyZbOpKrUwcoMa3q+xWbj9Ly6U2dVAZNqP693ajnSuGasWgvjH6GxlqlEPhzH5nUGHJxtvNejPfGH0dZsOhQzfVtgq+Pe4WzbGJVSfiNKlhTLNBcPVkbeXhtVP70BNs7nark5vG3IhJl+3XKmsVJ9edAqihv7vP1fJcTexfmhkvTouB66dp+/Ub0wbgLNS4Aadq9cj1JpjxAzBaMBpNfPu0QZrqwCqHmdNHFlg1aXHBydp+5cQb1Iqs4whfyJwHqHmPhZv20xJIcMnEPpRbdZSm46yBaJRATPC/tU3odRKXTOhNmU2PLZ1oCwU6sETbEBv+i1LaD/3wczB0oQBp9/vZ442xYEsbk/qXMqFvGRVd6CZa/FHW7O1gXYOPs0fV0q+b+E6LP8r8za00+2N86YQ6KuwmKtLcV4mkTFsozisfq2ynl57QmyqHGVOaCiOSSNHij/HKuv30KbVyxojqHFqEg8HjD7G+oYNVe32cPbKG/hVWDcWIOxjj/W1uGtojfGlCHbUuS6Y8MZVM4gkleH1dA7IQXHxCX6odZgzpxG4gFMYTSfHq2kZ6uUycOaq3dqdyqFWtDNryBtRNhP5TtS/goWwPxlm03c0eT4gvndCHWpdF8xFw0LaBKP64zMtrGql0mjlvTG1B+Y5OtIW9bPRsYk3rKk7tM4PBrsGahLk36mXp/qXUB+o5f+D59LL3yuQc4rEYHTGZeesbiSUVLpnYF5dFl0nAxsJBWiOCVz5uoNJu5OwxdZp+cwcCNPmTzPukmVG9nUwdVKlJOnfEOtji3cKKlhVMr5vO0NKhlFoKc4yKotAWcfPWnneJpiLMGnwRFZYKbKbCijC8ER/N4WbeqX+bYWXDmdJrskaoqiUQYktzgI92ejl1aCXDa53UHmYl14EQCPsJyH7e3PUaJUYnZw04l3JzRXY8RpM0+aK8vn4/I2qdTBtSmeWYQ02Yb2kO8OEON9OHVDGqd4nGyR8SwRa1kivaAeOuVEvP08SjiXictlCCVz5uwGHWc/7YPlSXWAsrawd1M2fbFlX3fNDpUDsO7Iev6FlMmB+B8yiiiCKK+P8dxYR5EUUUUUQRxxWKzqOIIooooojDRtF5FFFEEUUUcdg4LOchSdLxz9NcRBFFFFHEMUdBzkOSpGmSJG0GtqT/PV6SpL8dolkRRRRRRBFfUBS68ngQOBfwAggh1gMzjpVRRRRRRBFFHN8oOGwlhGjoduj4VospoogiiijimKHQvZ4NkiRNA4QkSSbgdtIhrCKKKKKIIv7/Q6Erj28DtwB1QCMwIf3vIooooogi/j9EQSsPIYQHuOZwLixJkgVYDJjT95krhPiFJEkDgTlAObAWuFYIcWgKzyKKKKKIIo4bFOQ8JEl6OM9hP7BaCPHqAZrFgTOFECFJkozAEkmS3gK+DzwohJgjSdLjwI3AY0dgexFFFFFEEZ8RCg1bWVBDVTvS/41DXTncKEnSQ/kaCBWdQgHG9H8COBOYmz7+LPClPM2LKKKIIoo4jlFownwI6ioiBSBJ0mPAu8DZwCcHaiRJkh5Yk27/V2AX4Ou8Dmr+pO4AzYsooogiijhOUejKow7ourvcDvQWQsio4am8EELIQogJQB9gMjAy35/laytJ0s2SJK2WJGm12+0u0MwiiiiiiCI+DRS68vg9sE6SpA9QRbzjOOiBAAAgAElEQVRmAPen6UoWHKwhgBDCl257MlAqSZIhvfroA+w/QJvZwGxQKdkLtLOIIooooohPAQWtPIQQTwGnAFuB/wH3ANuFEGEhxN352kiSVCVJUmn6/63ATNS9Ie8DndJn3wAOlHAvoogiiijiOEWh1VY3AXegrhTWoa4glqEmvw+EXsCz6byHDnhRCPFGmiNrjiRJvwE+Bp46CvuLKKKIIor4DFBo2OoO4CRguRDiDEmSRgC/OlgDIcQG4IQ8x3ej5j+KKKKIIor4nKLQhHlMCBEDkCTJLITYCgw/dmYVUUQRRRRxPKPQlUdjOn/xCjBfkqQODpDoLqKIIooo4ouPQulJLk3/7y8lSXofcAFvHzOriiiiiCKKOK5R6MojAyHEomNhSBFFFFFEEZ8fFDXMiyiiiCKKOGwUnUcRRRRRRBGHjaLzKKKIIooo4rBRdB5FFFFEEUUcNr7QzqMj2kFrqPWA592BCJ5A9IDn23whfOFI3nOxeBy3z084mr99IqXgjySQlfy0XO2hGM3+A9+7IxiiIxg64PmWQJCOSH7bUGSIdkAqP2dlMhoi4dtPMh7Lez4a9hH2H7jffIEAvkDggOeb/VHaQ/mvLcsy8YCXVCL/+UA0SpM/QDR+AL7NZBSivgPe+1Bw+w/8TFPJJFFfK7FIMP/5RJx4wIucTOY93x6Ost+fvy2ANxjGEwgf8HwgFCJygPGUTCZpDbYQiPrzno/FYnh8fmKJQ/SbODKauEA0SSSRyntOCIE/7ich59d0iyViuIMeEolPX/NNlmVaAh780QM8U1l9T1Oykv98uAPF14Qsy/lvkAhD7MDvwhcZ+l/+8peftQ2HxOzZs3958803F/z3oWiIpkgTD655kDf3vInVYMVpdGI3qcTA7kCYzc0h7p+3jYXb2uhfbsOoE9jMRgBa/REWbmnljwt2smavjxG1Tsx6gdGgFqd5AmFeWLmXRxY3sM8dZHitA7vFlLl/ayDGo+/t4MklewjHU/SvsGE1GdK2RWn0xXngnW387+MmTAYdLqsBh0W9d7s/wPbWIL+et4N3t3gYWGbGpEthNZsBaAmEeH9bC396Zzer6zsYVuvQ2EbYAx8/Bx/8Djw7oGYUmB0Z24R/P7olf8Swaja6aDuibCBS+nw8HgF/I4aFv8K84XkSOgtJWyUGs9pvvkCAna0hfv3WDt7a7GVAmQmLLoXFbFFt80dZvMPDH9/Zxoo97QyvdWLSyZiM6m9LBd0oa/6JecnvkFu3QM1odBZnxrYWf4THF+3i2aVNtEdi9K+w4Ug/ExQFfPtg/s9hzTNgsICjBozWgsaE2x9myU4vD7y7g2W7OxhW48CslzEZ1OsnA22I9XOwLPk/RNsWqBmjsU0OtKAs+QvmFQ+TCnkQFYPRpfsllUrR4Ivwx3e389LqZgQpKh1GHGZ1TLQHw+zyRLh/3nbe2thGTYkZi15kxkwgGGL5Ljf3vbWLJTs9jKix4zDr0ev1ALSF3by++zVmb3yS7b7tDC0bitPk1Py2Jxfv5omPGmkPRBhU9f/Ye+8AOapj7fvXYXLcHJSzkIQkEEKBJHIyySYbMPY1BpNMtHECY2ycwQRjDAaDQQYBAhNNjkJCoLSKKIfNefJMT6fvj+7t2dbsvcj3mvf65f3qH2m6trurT5/TdU7Vc54KEbTvjWlCsgXeuhVW/BlECaKN+9xuqbzKx7v6uO2ljby/tYdJdRHLNtGad/YX+nll5yvcvfpuPu37lIkVEwl7S/2tK9vNIxsf49FNf6Ej183o6BhC3uA+3ft/Kl3ZXl7Z+Sp/WvcHNvZtYmLFJKK+Urv1ZBQe/nAX976zjY5UgfG1YYLeEgDV7N+D+O7tiKseRdAViDQgDJyvKdC7HV69CdYugkg9BCqsfvlvJLfeemv7T37ykwc+j2sL5n9zJvJ/Ug466CBzxYoV+/z37Zl2znjhDLJqaZb3m8N/wwljTgBgY1uSk+9Z4kzCJFHg1e8cxoS6CLqu88yqVr63uFSmJB708I+rD6MhHiCRzvCr17byxIrSHsnpw2M8dMEMauIRejIKFz30MRvbS7ORi+aN4vsnTibglWlL5Dnh9++TKpRmcXeeM4MzDhgOwNaOBMff9SEDCxZRgNe/cwjj6+MAPLt6N9ctWu+cGw3IvHrNoTTGQlDMwZu3wMeD+kr9/nDhcxCqQU+2Ij15PrSvcdTmAReiHnUr3kgVWrId+f551qrFluKp9yPNOBtJktjemeS4uz50VlOCAK9eNZ9JjRUAvNjUxlVPrHbOjfhkXrv2cBrjAYqFLMLbP8Pz8X0l22r3Q73geTzROtqSGa54vInVzaVVxZcPbOSmEydSGwlBugP+OB9yvaXzT7sXZnwVxM9eQL+2vp1LH1/l/A55JV6/9nCGVQQp5JJIH/wWz7JBBTOrJ6Je8AKeeAPFVDfyMxch7lnqqPVpZ6Gf+Fu8oTjtySyn3LOUnkxpZn3LqZM4f/YofB4PO7szHPf791H10lj7++XzmTnSarf3N7Vx0aOldgt6Jd669jAaKkKkCyke2vAwD60vUcCNjIzkz8c9REO4np5Eiquf3sDS7X2O/uRpddx2yiQqYxHIdMKfDrfab0BO+BXMvgRs5/RfydLtPZz/4HLnt98j8vb1C2iMB1B1lYfXP8y9a+519MPCw3j8pMepDlTTle7hx8t+xNL2Dx39ghFHcsvBt1AdrvrMe/9PJFvI8finf+PeprucY/Wheh47YSH14VoSuSI3PN3Em5u6HP2Rk2q485yZxINezGQLwl9OtCYstpgLvo8+53LkQNRyyPfMAm3QCvriV2D0IZ/rc/2zIgjCStM0D/o8rv2FDFut7lrtchwAizYvoiPTgWEYLFy+x7V61w2Tp1Y0A9CZzLNw+R7XuYmcyibbGSg6PLO63aVf25JkwBfkFM3lOAAWfdJMRrH+oKkl4XIcAI8t20NrwrL3iY/3MDjSZZjw5CeWba2JFI8va3Wdm8prrGuxP/ZKGlY96m6MjnXW0hoQtYLLcQAIaxchG9YA0DvWuxwHgHfFAxTS1gB7ZkWzKwxnmrBweTOGYdDan+Pxj3a7zk0rGit3W9cTlBSeNXvZ1rUJ0batqJkuxwHwwpp2tIFoQc9Wt+MAWP4A5Pc6NoR0JTI8usxtW7aos3yn9cGVilk8q//qPqlnC4I60G55l+MAkDY+i6hZ4a9dvRmX4wB4YnkrXRkrBPV8U6vLcQD89aPd5Isq6UyWh5a1uHS5os5H26yPfVrL8Ny251z6Pek9ZFQrDKMYgstxAPxjQydFQ7B+JJrdjgPgkwf3qd2yisZDS3a6jhVUgw+2WvV1EkqCp7Y85dK3ZlpJFKz3qFJ0OQ6A95rfpWgOHfb7V0qimOLZbU+7jnVkO+ixnzuv6i7HAfDO5m7yqt3hsj0uxwEgrPorUsEeH1vfdDsOgGX3OmPt/wX5QjqPSn9l2bGYL4YsyIiiSFXYW6aviVhhIVkSiAU8Zfp40D5mmkT9br0oWOdZ54sIwlDnWgcrguX3rgh58Niz57pw+b1r7WOyKBALlO/rdF0zUOFWCgKI9jUlDwh7vXJ/zImDi3ufCxjBSgTRumfNULZFrDb1SOKQ7VYZsmwzEcptG7DJfjZJdDdcxC9jDtQK80fLzw1UgPjZ+1w9kkBFsNy2Kts2Y+Bae4kg2+0qSuX38UasZwIivqH6i9d5nqqQr0xfHbL0siRRFSx/hqqwHf4wIeaNlel9knVNSRDwSu53GvTKOC05KITkyD62mywKVA9le9g6Jgri0LbJll5AdOwcEL/sR0AoO+dfLQIiUV95nwnY4TpREAh43Csvv0dEHBi8Q4XWgpXOOydUXa4P1e5Tu35R5AvpPEbHRjO1aqrzOyAHuGLmFVTbL/zc2SOpjZQ6dWPMzynTGwGojYX47vGT8MmlpjloVAWNMWswh/0ebj5pgut+/zFvBF7RSriFfTJfmzfa0QkC3HLKVCrtj9eoqiAHjIg7er9H5PpjJ1EbtTr16QeOpD5aipvWR/2cdsAIAOqiEW48YSJ+T8m2A0fGGFFpd/RgtRWSGCyzL3VyHprow5x9iUttHnsbesBytmZ8BPrI+SWl7Mc46icEozUAnDxjuNMOALURH2fOGmX9P+rnuuMmugbkjOExxtVYeQEpXIN63C8Z7FnVA7+BJlu2eyT4xiEjXLbddNJEYgOOOjoMRh82yDYfHHfb0A5pL6mIhLjmmIkEvSXbpg2LMqneil+LwWrU437hsk2bcQGqaD2r4Y2gzr/WdU316J8i2PeuCns5ZHxpwuKVRG46cQKNMev6x06pY2Rl6WNUFfJy0fzReGWZQMDP1UePJ+wrfXSmNESZ0mj1kbpgHdcfdD3iIKd/0piT8IqWY/PLJlctGOVut+PGEfLZfx+qgQnHlZSSB47/BQTLJ1h7i88jccVR44kMsm1SXYTpwy3bqgJV3HTwTUhCqV1PGH2Ck48JSQEumXaZ65qXT7+SsOfzz3k0Rmu5ftZ3XbYdOeJoQpI1FqIBDzccP9F1znXHTnImhqYnhLnfKSWlKGEeextixUjr94iDoWZySe+PwWHXWf3y/xH5QuY8ADKZLiQ1i2loCJ4guhwjHLQ6jqqq9OQ0VuzqRxIFDhgZpyFWSiBm8nlSiolmgCSATxaojpT0uUKB/ryBqpvIkkDIK1ARGnR+tkCyaKAZJh4RasJePJ7S7LQjmWd7d5bOVIGDx1QS9QtEA6XzuzK9qKaFmvEIPmoHxYcz+TyJgsknu/qojfgZVxOifpDtFFKYmU5oXg61UzBiI5DCNY5aS7YjpdugYy2Mmo/qq8AbrXX0yWyajCpgGCayZBD3egkMsi2b6ser26EmKYg3EHGeLZ0rIBbToOYQRAnVEyUWLc3+lGwSKd+DvmsZYt0UzNgIvNGSbelcDl8xgWio6KKXojdOJFi6dz6TsN6LruP3yFRF/K4EZS7TjVcvgAlF2UswXDeo3XKkCpBXDSRRIOgVHYcNUMwlkZUEgpoD2YfuiSBHSrYV0z0I6XaMtiakkQejB2vwhUuOqz2ZYVdvlj19WeaNrSXkFagKhwbp8zQ1J1A0g4NHV1LhF/D7Lds1VaUnW+ST7V1UhHxMbohRHSudm8j1k9Vz5PU8XtFLUA5SHSzNfPvTGdJFgYKqE/BKRD0mscigFUe2x0ruJnbDyLmWQxmUMO9L58hrVj1onwg1sdLHXdcNlGw/FLMIgojhjRCKlFYbeTWPkOvFUHMIkhfdHyM8yKEXFAUx34dgFDFEL4a/goB/35PKPZkUiiZimuCRdOqi7pVOZzJPUTftlYRIZbj08e7PpvBrKYxiFlH2oniixAeNpWS+SGdKYW1zgunD49RFfcQGreKNZCtCYrcVMh19GJo3gmdQnyDTBR3rQUnCiLnWymMf8kjOsyXSKLq1wosGZHe7mKZ1fV0ByWetdMR9v/aAfJ45jy+k89BSbUjrnkZ49xeg5jHHHQWn/QEhaq0uepJZfvP6FlIFHcOE+ojMVUeNpzpmDbj2ZJ5rF63hox19BL0SN504meP2q6U+bg2qbV1pvvXXlezoyVIT9nH3eTOZ0hAmFvSTyGZZ15rj2qfW0JMpMr42zAMXzmJsjXXtjmSOl9d18OTHzXhlkcqgl9+cNd1xAJ3ZLu5r+gPPb7MKLJ42/jQun3EFdaFax7Ybn15Lf66IohqcP2ckJ02vpT5qI396tiMv/oY1i1bzqCf8Bq3xAAKBCJlchs2JDTy99TkOrDuQpW1LufbAqxkVH2vdO5Xg1Q29/OofW8kVdeaOreT350yn3v6QFVNdCO/9Cs/qv1j3mv5VjKN+7DgfLdWJ+PdLEXe8A54gxQU/Rp12NqHYEEv8vSSXz+DpXI9n8desGH3VOLRznkSutWaHXakCb2zs5PZXNpEt6sweXcHd5x5AQ9xqNyXdjvj+7/Cs/AuYBtr0szGPuRVPpN5u9zzfW7yO97Z04/eIXH30BM44oIEG+9n0vl1Ii86Hzg0QrEI97U8Uh80mFI4PbfA/IYVkF9LSu/AktoPoQ5f8GMffhidS+9knA82pVq599zts7t9MzBfj1nk/Y3rVgdSEo+TyBT7tynP5wlV0pAqMqAzwpwtmMaWxPJw0lHQmc/x5yS4eWboLzTA5aVo9N58yhboBx5rthhe+A5tftmbVh90Is78JQdtBJFvgyfOhvclaBZ52H4xdAN4gBaWA3L0J+ekLrL+Lj0Q75wmomYwsf3Z4pyOV4pEPm3l4yR5Uw+D4qbXcetoU6iLWO+tMFbj5+fW8vrETjyjyzcPGcNG8Uc5Y0vubkZ6+ENpWgz+OevJdqKMOIxj97GR9Lpfjg229PPJRC/WxAO2JPHectT+NlUOEAf8b0t6f4dt/a2JNc4KoX+b20yZz2IQqYuGw5Ti6NsIT51kOPzoMzv2bBX75Jx3I/58w/ydFKiQQ3rjZwrYDwva34aP70GyM/PId3Sxa2cY/NnTy2sZOHl3eyrpWK8mXyOW5961tfLTDSkLmijo3P7+BnGqFpVoTOa5ZtIYdPdbsuzujcOnjK8kULSecUwW+vXCVk0Dd1pXhhqebaO+3/j6j6Nz20ia2dmXY0Jbig2093P/eDlI5K/m2pms1z259Ft3U0U2dZ7c+y5ruNbZtBe5/z4JzbmhLsa07w09f2kimYNmWT3Qg//1bVlK8bTV0f4rnqfPxFK3kas5Ic/V71/Hyrpe5bfltvNX8Ft9d8n260xYAIFsUueX5T8kVraThRzv6uPedbSQH9j20rsKz8s/WPhJDR17zV8xdSwAo5DMYy+6zHAeAmsP7xvfxKZ+dmAXwKCk8i84tJXd7tyMv/jr5RKfzHn749/Vkbds+2dXPnW9upT9ntbPQsQ7PJw+CoYFpIDc9ibntLeu5CgX+umw3722xEr0F1eDXr24mkbOAC4VEF9KLV1mOAyDXi+eZC/Hp/8k+mn9SxO5P8Sy/Fzb/Azb9HWn9kxibX9unczvSffzso9vY3L8ZgKSS5LvvX4+G1bf7CwaXPr6SjpTVf5r78lz5t9W09e9b4nZnT44HP9iJqpuYJry8roNX13dY+xp0HdY8YTkOsOCp7/wMknYiuZCEl2+wHAdYYIunLwI7YS7m+5Gf+qrlOAASe5Cf+ip6pnufbGvuU7j/vV0UdQPThFfXd/FiUyuqpqJpOn9f3cprGzoxTSjqBve9u53WhNUO+VQPvHaTNQ4ACgk8z/0HPuM/31s1WFKKyZWL1rFsRx/PrW7lo519/OTFjaQz//OEeCqT5fZXPmWNDRBJFTSufmo92QHMRbYLFl1gOQ6AVCs8ca61gvw3ki+k86B1VdkhYc8yUJIYhsE728o3Wr2/1XIW6YLBit39ZfpPOwY2GQmsb3WjqVJ5jby9gSqr6A6yakBWNyecRPX61vJ7r9jdR0qxP9jtH5XpP2pbZtmmWKG2vWVDm3VNj2BCy14rNCXtIEDyWp5U0W37xt6NmHao/9P2IWzblSRtP4+0460yvWfHmxiGgZ7P4t3zQZlea11bdmwoEbR8GdKLzvV4sNplS2f5Jq+Vu/vJFCzkjrjj3TK9d8c7qJpCoqDx8c6+Mv2aPdbgldCg5RO3Us3/yzZ/CTvLiah9O9+mUPhs56SjsranyXWsaBRLiCbdoDvt3hi4oyeLvo8BhaXby5370u19ZBUN1CwMTAYGS/PH1r9qHlqWu3W6aq1WAEEvWh++wdK/C9HYN7TV8p3lH8tl25OklQLJguZM8Fzn7LCeR9QVpJaP3UpDw0i2l50zlHSl82UIuZV7kuTVoTcT/jOSL+qs2OMea4YJrQMOX1ehb4f7pHS7FcL6N5IvpvMYVr5KM0ceAv44oihy1ITyUMSCidZSNuoTmT2mPAm7X4OVBBQwmT7cHRKIBTzO5qKQT3IlGAEOGFGBYGF62H9Y+b3njK4k6reWo/Ma55Xp5w+zktgRn8zBY8oTndPs5KqKaCXyBosvCvbmyIAcIOp1I1CmVk1FtL3Hfo3lth08JuagifRxx5bp1XHHIYoiUiBEcdQRZXp52IyyY0OJ6QmWJ3Hrp6Nitcuk+kgZim326AoidoLTGHtk2TWV8UfjkX3E/DJzxpa328A+C12QMYbv1W6e4NAIr/+GmEPZNvYY/P7PThzLeJlRM9N1zCf5iPutd+WVROqi7iTtuJoQ0j4Cmg4ZXx5SPGRcFSGfbPWbcUeVnzRijvWvJ2DF+geL5LVi/4ApeSE23K2vHIshlqPThpJ5Y8ttmz8uRsTnJ+aXmTeuPPw0d6x1zJD86CP2GkuijBhr2Kd710YDePZqxINGxQh4/uefzIBX4uBR7rEmiQLDKu08l+SFqnHuk6KN/3bJ+M8NVyYIwgjgr0A9FhryAdM07xIE4SfAJcDA2vUHpmm+8q+8t+6LIh1/O8I7P4diFnP8MTD3MmT7YzBnbDUXzW4kHApgmCbFYpGpdow4Fgpw5ZHj6UjmqQj56M0UOXpyLUGP1ZEa40HuPGcm1z21BkmwkpS3nDKVkNfSh71w/4UHcs2iJrrTCpPqIvzu7OnUx62OEfYJ3HrqFBZ90oxHEqkMebnk8LFEA1aybGbNTM6aeBbPbbWw/V+e8GWmV1sf4HjQz6WHj2VXT5a+bJGiZnDuwSMI2n0qEKtDO/1PyM0rYOpp0LUJVcmg+aLIQEAMc++C33PDkpvoynUxLj6OXx/6C6rtvEDIo/OzM/bjl69sJaNoHDK+issXjCcWtGw3G2eizr4Mz6qHwDTRZlyIMMpybP5AGG3uZRhd6xG3vQ7eEMUjf4Lqq9ynTqb5onDOIjyv3mANHtNAO/1PBOJW0jvokfjFGfvz85c3kVY05o2t4jtHTyBuJzjN+qmoc7+N55M/2zmP8xDsj3bY7+fCuaPY1J7m7U+7CHolrjlmAvGg5Zj8sVr0U+6Bpy6wgAShGtTTH0CVQv+SAWJUT0Sddw2eXe+AIKJVT0GcUO6Ih5K6SAU/mvtjrn/3Wjb2baTSX8mt836GByuuXxGU+dOFs7hi4WpaE3lGVwX5w/kH0lgR+owrWzK6KsBlR4xlydYeRAHG1oY5YVqds7udGedaSeERc6DQD4gQs1Fx/hic9BtQCxAbZs2OZ19iHQcIVKKd84QVukrsgYoxaGc97gJw/FcyrMLHlUeNIaeYhLwSexIZvjRjGB6bFeC0mY2sae7nH+s78Eoilx4xjsa4NY4C0Sr042+39rNoBVAV1MO/hyoG2ResV9Qr8MfzprOpI8vUYXFeaGrje8dPIBLet3b9L68dDvH9kyajGQJHTq5hS2eaA4aFCckDMOAaOGch/P3bFvRXV+HUeyw05b+RfG4Jc0EQGoAG0zRXCYIQAVYCpwNnAxnTNH+7r9f676Ctkuk0spLENA3wBDC8IaJBq9tomkZfRuGTnd2IgsCsMdXUutAtGTJ6imWtH9MQqmN8xQTqw6UXp2T6kHM9sGcp1E9Hj7pRQ8VMLwlFQNU0fB4PcZ+JHCrNfDuSeTZ3pOlMK8wbW0XYV0JrqapKQekjp+UQgIAcxO+rdBBNqaxCUtH5aEcvtVE/k+vDpeQm0J/qpz8vsnx7J1OGVTA87qMqVppBa5ku+vUcmlbEK/uJyUHkQZj13lSOdNFAN0wCHomYzyAULCUJc6l+vEYeMCmKAQKRCgR7SZDJ5xCKGSjmEQQR3RclMgiZ05HMklEMlm7rZb/GKMMr/DQMQvYoqW486WaE9ibMEXNRgnUEIqV2S+dVUgUVw7QQKgPJ8gHJZnrw6VZMWxF9+INVzkcwk8uSLopohokggH8vBF06m0FWM8hGEUOQKXqiRMKDaDYyPXTmOlnXs4GD6g+kwlNBTaQ0821P5tjamWFPX47DJtQQ8ZpUDkI8dSazrGlOUNQMZo+ppMIv47MpZ3S1ALlutB3vIQQrERtmItvgDgAll7bemaEhCRIx2U8wWJq5pvIKmYKOZpjIokDEJ7hQaj3JNNu7MuzuzTJ/Qh1xr0HY9V7yNLUkyRd1Dh5TSWXQg99eSSuKgpzvQdvxAQTiiI0z8cTqnXOTmR5SeoYVbcsYHR/HyMhIqsIlfUHJI+QtBJ0hejH9Mfz+QUivQh87kzvZk9rDwfUHU+GvIDgIytuezLNmT4KMojF/fBVxn0DIRv/pukFXWiGv6ggChLwytYNg7v3pNBkFlu3oZUx1iDFVAQcUA9CTS9Cd72R151pm1k6nNlBLdagUdVBTbdC7DaN3K/KYBRS9EQKDQA5mqg1aVlpoqzFHYASrkbw2gEPV6c0W+XBbDyGfXIbo7E7nyBVNPtjaw+iqEONr90JN5hNWyK/5Yxh2oOWw9wFevbd8ngnzz23lYZpmO9Bu/z8tCMImYNjndb/B0p7Mc8lf1zi5Cb9H5O+XH0LU7pM9GYVT7l1Gd8aKIdZH/Tx/xVzqbAfSkm3mole/imrHZmdUz+S3R9xBfbiGfD6PtOVVpBe+7dzPmH0ZuUNvJBirRku243nvF9QO2ultnvEA2n6nInsDtCfzXPbYSpparJinTxZ59vL5jvMQc91E/nwkkYHEcaQe/ZvvQMz6mDQnC3z5vqUomhUGmzkizh8vOJCGWIC8kmfJthRXPVWiL7no4EauOWYMldE4WrID6YPfULPizyXbTv0Dxf3OwBsI0ZPM8K2Fa1i1p2Tbc5fNYYrdbp2pAqfdu8pJztaEfbx41aHU23s/2hNFTrlvJQU7LjxtWJS/XHQANbEwBVXlk11JF33JWbOGc92x42mIh8im+vB++HvE5RbVhQBIX7qH3NSvEAxY7yUS8BAZYiOide88X/ljE21Jy7aqkJfnrzyE4RWW8Z1pjS/du8zZQTylIcqjF1u2FYtFlu9McMnCJod54MwD6vj+CROpikXpzfTz2KbHeWRjiSLk+7N/xCljvkQkEKI9meO6RWtZZsfbPZLAE5fMpdKmQepI5jjz/uW09FuOrRrry8MAACAASURBVCLo4cWr5jPcdh4k9yA9cCRS0SbCrJ2CdsFi5Ggjuq4j7llK/aJznM2c2tSvoBz3C3yxOnrSBf6ydBd/eGe7Y9uPTt6Ps2cNJxr00ptIcd0zG3h/m5UfkMUtLLrkIGY5tuU5+08fsafPyr/EAh5evOoQRlZanwYp3Yr04OFIip1zqpmEeuELeKKWg9iY2MRlb1+JYVrv/NgRR/HD2d+nKlJPTs2xcPNC7l5don25cfaNnD3xbPyyn/5CPzd/eDPvtVg5IUmQeOi4h5hVP8ux7fwHl7PTBqdEAzIvXnkoA6j4tmSBk+/+wGFsGFsdYtGl85wNv5s7snz1L6scVoRjJ1fxizOmUB2Lki7keG7bs9y9+k7HtitnfIfz9jufqC+IkmxDfuUGpAGwgCAif/UZiBwNWI5D+Otp0LPF0vuiiN961wk3daYKfOneJaTylm2jqoI8+a25jgPZ1pXjgoc+dmw7YkI1v/zK/jTEg9ZKrukJizdrQI78Ecy73AlB/zvI/5GchyAIo4EDgIHs2pWCIKwVBOFhQRA+e5fXPynbujKupHZBNbjzzS102x+9pz7e7TgOgI5UgVfWWIiQrnQfd6+503EcAE09a2jPWok2SUngffNHrvt5VjyA17SuLRlFhL2oLoQ3b0bMWcm/3T05x3EAKJrB717fQmfKGrz6mifddBLpDvSmRZZtqTy/fW2L4zgA1jQn2NNrnZvOqfz0H1td937skzYKmv0hMIsIKx9y6YW3bkFWrER1ayLvOI4B23752lb6ktaxl9e2O44DLKTZ0yst6pT+VJrfvrHVcRwA61tT7OiyPoi9GZXbX9nkuvfTK1vQbBoNn5F3814B3rduRlb2LWn9yvp2x3EA9GaLPP7RHgzDIJnJcMcbW0vUE8DG9hRbOqxrJ7J5bn1li4uy5pnVnRR0a3gopsLjm9zUKveuuYuMZj1bT6boOA4AVTf51auf0p6w9K9vanMcB0B/TuXRpbsoaiqqkkZ479cw4DgAujZi2AgmLd2J540fuNhw5Q2Lke0VVkEzeOB9d3L1929udUAOvXndcRwAmmHys1e20puwEu7vbu52HAdAMq/y8JJd5IoahVwW4YPfWqCLAenejNFsRQG6Um38euUdjuMAeKP5bXI23U1GzXB/0/3udlt9rwPaSCpJx3EA6KbOr1f8mn6bAuTDbb2O4wALmPLA+zvIFhSKmsH97213Uf3s6MmyYpf1rF39SW77x1YXnc4bn/aSUqzfaS3NA2v/6LLtwfX3k7VpXyQ1W3IcAKaB542bUQaQY7uXlRwHgJKCJXegFzLkixoPfLDTcRwAu3tzLNlqfQNa+3P88h+fumx7b2sPyYFnKfTD27e5bOP9X//bsfd+7s5DEIQwsBi4xjTNFPBHYBwwE2tl8rv/5LxvCYKwQhCEFd3d+wbtG5BErhzNkcyrqIaBYRj0ZMuppbuz1jmaqZMa4oOVsgeQAO7BBGAaoNvXNNRy2utC0gntJPLltNTJvIpmdyQpV/6sUtbi4FF1k2S+/Nn67ec1BZHUXnrTBM2wB7c+lG0px7ZUbmjbBjp5T6Yc7dGdVjBNE90wSeTL2zVpP68gUGab9Uy2bTbE1iXFDOI+Jn77MuW29+eK6AZouk6iUG5bv42NtNqtXF+0ibV0U0cz3fqclgObOiUzxLWtdhu4T/lz92U1dMPA1IuIe6PMANPm8RJMLEjs3nrVeheGaZahgnJFbYDUhZwy9FgYsK1viHfely2iGwaGoSEVyhFNJY4xk3SxHAVXsDmfDNOgaBTLdAOh8rxWDptNFVPoptXu/bny/tafK6LqVn/rzZbbPnDMNBlyrBRsVKRpGih7oZcUXSk5QnUIJFwhgTCgH4IbTMj3YxrWWE4M0a4Dthnm0OM4ayMuMc3y++vF8vHxvyyfq/MQBMGD5TgWmqb5LIBpmp2maeqmaRrAg8DBQ51rmuYDpmkeZJrmQTU1+5ZgG5BZoyqI+t0Rua/NG01jPIgoilwwd4TroySLAmcdZFOAhCo5Z+JXXefGfXEmVlqUJKrkR5vh1jNsFppkLUdNKQANboSROfMCdMkKn0wfHi/jgLpo3iiG2Ql188CvuTcCiRLmrIut21QEuWiem4oiFvA46K+Ax+C8gxpd+qmNUQbosAzJb8VPB9s243w0yQo7TWyIl3FAfWPecCqjVozjzFnDkQc1nCjABXNHIQgC1fEo/zHfjayJ+mWmj7TyAn6PwDmz3fQjk+sjDtWKJgZKKB5btP3PRRH2jTr8y7OGu9AxggAXzh2FRxapisX45nz3vcM+mYPGWv0q5BH46mw3Cmd8bZiQ17LNK3g5qNYdNj5h9Il4BCtZP6oq6IRKBuSrc0Y6x06bOdzFPyUI8LX5Iwl4fXiDVahz3BQeeMPIYy3kmumPoB74dbe+ajymDf7wiCKHjHejjk6cVu/crzEeLENjXTynkYqQ9c5P3r/BRcUDcPH80UT8XoLhGOrBV7jv7QkiT7BCN2FvlLPHf9mlHh4ZTtxn90c5wGHDDnPpjx55NAHZeqc1wRrqQ/Uu/fmTzyfus/I5x02tH9K2eMhPwCtxyWFjXDq/R+SoyVZOIhL0cdHB7rEwvCJAtc1r5xMDHDFigUt/+PAFeG1KGjNYDRWjXXp11sUwwJs34QQLkTdIzDmXIQctBODF893n+mSRE6dZz1oZkjl/zkiXviHmpyFmvydPCPY73aVn/DFl9/vfls8zYS4AjwJ9pmleM+h4g50PQRCEa4E5pmme+19d659NmGfyebJFKGgZEDQkM4IsS05Cqi+ZIqOJoPRgmgJSoIqwDPGolUzryvSxqnsVz257irpAA5dMv4Q6X42T4FRTXQjrFyNvfQW98SDMOZciR0uDwEy2wUf3IXQ0YU48EaZ+GcHW5xSF/pxOU2snJhr1kUqGxYOObelUgqDSRapjFwDR+tHkfLVEotaA6kgWaOvPkk/1oCEzbngD1WGvk+DsTWV4ZV0Xr2zs4oDhUS6eP8oFBjBTXSi6gWnqSJKELIBo26YoCl2ZIve+vY22VJELZzdw0KgKi9oba0Zb1HSy9mw27PMgS5IF6wR6k2ma+wu0JFVEwWTm8BjxgETQTnB2pvK8ur6DVzd0Mq0xyjcOGeNKemvJDoRPHkRqWY428STM/c9y00EAyVQKVdeJhUN4PCUqif6sQlrRSGRVTAFifpmwz0O1/QHvS6bI6yJBI4WGBN4IQY9JyM6n9KYybGxLkVDAI5ocODLuareubDdPb3mKlV0rOWL4EZw0+mRqbKCBpmm0pxT+8M52WvrznDlrGPPGVVJnJ9n6sikSOdjYZoWmxtUFqQ4J1NhJ62K2B6mYo9MU8QCVsgyBOJLH7/Q3vXMT2UIRr2AQHDkDaVB/60oWWNvWg2LkCUghpjZWUjcocdyXyqPoRXTTxCPKeETTmRD0ZxU6Uwr3vL2VvGpw6eFjGV0ToC5i2Z5L9eFLN5Pq70TEJNw4CTVY5yS9+7Jd9OT7QPNhCipVkTjVwVJSuTffC4UUIV0jI8mI/hiVgUHgkWwHD69/mJ3JnZw+/nQOGXaI4zwS2RydaY173t5GVtG55PAxjKsOOLnJdEGluT9HX3cnBiJjhjdQF/XjlSX7naZ5f0svT6/uYGJtkEsPH0tDRSlh3p3r4dmtz7G8Yxmz6+Zw5sSvUDMI0aQnm+HDu5G6N6NNPwtz/LEOY4GmZJBSrfD+bxAKCcz5V6NXTUa22Ra6UgV29GR54P0dBD0SVx49nsaon6iNDuxM5VmytZdnV7cwpirEZQvGOfk5wNors3ohbHsTxhwOsy6G8L4xEgyW/ysT5sAhwIXAOkEQBnjAfwCcJwjCTKw1/y7g0n/1jQ1TJaF18vtVd9Kn9HLWhHM5YvjhYMMbg0KOSN86PB/8FgQR9Yib0GsmAVbHMo0Aw+SDuHHmDHRdxCh6UbwmA/O3fsK87z0BaeIRFAUvhxoRFxKgw4iR3f87VMxU6NIC1BBgoEuquk7WbOOljjvpLfTwlfFnM6L6SMc21RRZnohy54eWLdcdHWVSbWn2VSGkqep+Gc/qRzBDteh1P6Woj2TgVVYKac4PLOW06eAXNDxmBCh9BA1DRX73duSONRTHHYcx9zJn+ambOroBX57ZgEeCoiFiDtpckSkobOnMcffb2zAMkyuOGseU+rDjPBAEtnZl+evyFqpCXsbXhJz9KwACAmGfzIlT6xFFoazKooCBEB8B0ToEKeQKsRXyOdpSKr94dTNtSYVzD6zn5On1zkdQ001a+vPc9eZWNMPk8gXjmNZYQpkFhQLhXW/iXfEgRrAS4+hbUaOllZKGzMqWDK9t6GBSXYQpe+3HMUyDmmAtx448Fo/kK7H9ArIsM6JS5gcnTSBf1KmLuSksRB36MwoLl+9B0QyuPGIUdYFSUaKEKbC0Zy2Ltiwk6o1y9QHXMlIKEbGdR4oQTzQ38Mq6TibUhrmxIczgNZ4hpfiw/y+s7VnD7Lo5TB3xdcA6N68USCgqv3p1Cy39BU6dWevUjgEwRQXB28FFCzyISHjkFgRxBNiA1rQZ4PWuah5ZmiHsk7nx+DjDvQIDrsnUQ7zV1MuL61sYVx3guyc0oHtUJBsdWFUsYH54N0Lzcnwj5yMcdt1AVwesGhvXz7qeolF0FbgC8IkCIY/Aj06eiGkKaJqBXyy1e8TMMqX/ffj4Los9uPonEJnMwA2i5DnVt5KjZxTwoSGbUXTNjzRAjaKHiSvHsiB+KL5iAPTSe9M0jTa9go8qL8VTUUQSoswhwgBbWsYQaBNg85SjkEyIySL7iR4GPu+1UT+1UT/ja0PIokA86F79iYKJR4bjp9YjYKIZOpquIw9ApE0TglUw5VSL2+rfkEbqC8lt1Zbp4LS/n0JBLyVQb5t/O6dPsFgytdaVyA8O2vwkCGiXLUGum4aqqjz+cQu3vrjRUYd9Mq/bRY16Ukl++vI2XmgqJbUn1UV45OszaYhH6UqkOe+hlWzvLiX6zj6wgR+cOJF4JExbppPTnz/FFe+9ec6tfHnCaUiSxKbWPk68Z9lg03jlynnsN6wSXdfRVy/E+9JVJdu9IbTLVyDHG1EzXchv3IzQ9ERJXz0R48LnEGPDUZId+BaeDl2lxLU+7WyU435FMFpJZyLLUXd84FCAANx+6kTOmj0aj8dTVkQL4Pkr5jNjRAW6rvPC6j1c+0yp3QIeibevO5SGijDJXJHbX/mURXbdFIDRVUGeuGQuDfEASrIT3xNfseqPDNg25QwKJ/yOULSKzkSGY+780EkEA9x68gTOPXgkPp+fzR0pTrzrA1ctlMXfns+sURYeo9i0GO9z3ygpZT/aFSuQK0bQlynwuze2uuq4jKgM8OS35jIsHqQj3c01713Fht4Njv7IEUfx49k3u+C6/5ns6EpyzJ1LXLY9e+lsDhxjfWpe3fEmN35QYu31il6eP/1FhkcaSedy3PXmNv68tNRujTE/z317DnXxMO3pbm784Fqauku70A9rPIxb5t5GXaSK9mSWk+760MmLAVx37Fi+ecgYgn4/O5O7OeP505w8A8Cfj32YOY2zAXhtQweXPrayZJsk8uZ1hzOyKkQ2l+Ged3Zy/weldquL+njh8jnUxSOQbIVnvwm7B9VCGXsknH6ftentM2RPT5qj7vjAyQcC/O0bBzJ/oh1i3P4OPDYovCN54MqVUDEKJZ9B/PD3eJb8pqQP16Jd8j5yrIFErsgVf1vFh9tKuYt546q47/wDqQh5aUtmOf3eZXQN2r1/2RGj+faCccQCfvakWjjt+VPQjFJ/vOfI+1gw0h2mG0oKapE/vredu94sAR0qQ15evnq+xbWW74fF37RWHQMycq7FbxX854po/f/cVv+krO5Y43IcAM9te4bOVBeGYSDuXTDJNC0OH6Azo/DMSndxnoyiORQgqinx8jp3fe/NnWmnGFReNV2OA+DvTZ0MfI/XdW8oSxT+fftiOrJW0nTRCve9TbN0TEn14F39iNv2Yhat1YK/SrqKsP4Zt75nC4JqI8G0vMtxgFXUyGsz+G5q63c5DoC/rewgYfP5LF7VUjYBevITqxhUXzrDwk/c1A95VWfNbgthkivqPN/kpqrY1ZsjZyOgJC3vchwA0qbnHdu2dSRdjgPgiZUdpLJWWz63upW9FjI88fEeVM0gl+zEa5M5OqIV0HYts+00eGFNm0vd3JcnZycwVVNxOQ6Ad5vfQWPfaDZeWlNu22PLWykUi3Sn+3h2u7ugUtEosqzVAibmFJ3Fa9zFnNqSBSfBr5lFl+MAWNK2xLFtT1/W5TgAFq/qoD9vtevru95wOQ6Ap7cuIqPk6E4VePLjZpeuqBu8vdkCcGQUncVr3GOhM6WUAAKG6nYcYNGdGOUAg6HkrQ1tLscB8OjyNvKFglU185MH3SfoKmyxOMOEQhLP2r+59ZkuTBvJmC/qLscBsGx7r4PI60krLscB8NzqdtI2Hc77zUtcjgPgma2L6MuVAwj2lt6swrOr3GOlL1ukpd9Okqt5t+MA2PORw9X37yJfSOdRGyqPDdYE6vBKHkRRRIuNLNMbceuYLArURsopowfqf5im4RQRGhBZLBXk8cpiGUKoJlJadtYOsUu0JliHx6ZsGBkvpyAYVWEfk7zokXJ6BTFiL6ZNw6rrPVgEEQaKGkneclbOUI3zYase4rnroz5kJ/lanrweVhGwikGJEg3R8kJXA3F/E8raVRAoJUQlj1MYqmRbNaZNnVIZKb93XdSHx7Zt2FC2xQN4ZNGiCo+Uz3RFe/ZrArV7JZUFwaKRAJBF2Xk/A2LF5fcNCjaiorxdR8a9yKKIT/ZQG6gv0w8kkgXMIftj0A4VSoJUVnAp4o04BZeGKj5WF/U7wIeGUHm71AeH4ZVkvB6R+lh5fxxus0uLmK66OAMSGqjBIkjlSV5flH1tt6He6YiYlWdD9MAQ45i4BYwwBAkidWVq0d79LoqCq4YKWKWJBwp4Rfzl+4lqI34EwRosDaHyd1YXbMS3dx8eQv6zb4zzrgSxtEt/QDxBqz3/jeQL6TxGRka40DFRb5QrZl5Bhb17VJp5PsQHoZaqxiHZhV/qYyG+d+IkQoMKBx0xodr5uMT9Hm49bbLLQXz7yNH4ZGvG4pdMLj+sdG1JFPj5qZOoDFsDoSHUyJz6Eh9Q2BPmqplXURu2OsuXZgx3FQ4aVRXk5BlWjDoYrcA86hZ7AFqijz0S06aLMCN1mCf+xlUt0Jx/NZpkdcqi6Kd4yI0lwwWR4om/QwhZS+H6qI8jJ5SWxSGvxE3HT6TCrslx0v4NjKku5U+GVwT4ih0/j0fDXH/cRBfKbf7YCqdQVX3Ux62nTnVVC/zGIWOc4lGKGKJ42PcH2SagnvBb9ICVe6gNyRw7uWRb0CvxwxMmEo9Z7XbMlDqn8BRYH54BdFcgXIG54CZX4Shj5CFQYyHo6iNefnLKVBeS7KK5oxjwaz78fHv6lSXTEPjuQTcRlfeN++rQiXVMqC3F0+ujfs6bOxpZlokGInxr+recJDHAjOoZTKgYbz13RZTbTp3kQpJdNKcRnx3794kBrpzxHdf9bpj1PSIe634xv8RJ+5cmUz5Z5IcnT6TOZh2Y0zCbiRWlokg1gRrO3+88vLKXWMDLtxeMd6pBglXga+ow69yaijg/PWWSC0l23qwGgrINKfWG4ehb3I1x7E/Bt2908QeOrmLqoLxVTdjHNw4ba7EtyB6Yf6VF5TEgDTNhmLXB0B+vRz3+1y4+KG3/89A8Vh+JBzz8+Ev7ue73w5OnELeRkH4PfGVWaaLmlURuPmUyw+KWPdOqpzGlslRwrtJfydenXUzI99m1SuqiYX78JXfBuVNm1BEdgEUGKsuLuh37Uwj8z8sD/CvlC5nzAOjMdNOe7aC/kGC/qolEpbCzUxmsmh9G92YEUUKoGoccLaW884pCf15H0Uw8koBHElwUIH2ZNDlVIqtoBL0Sfo9JzSAqip5khoIukCvqhHwyYY9JbBAnTmcyg2qIqLqBTxao9EtOYSCArkSWnd1pEGBMTcSF+ikU8mQKKvmihixJBGWTWKzUqYqZbjzFrEXLXjMJ3R9HjpYGQTbdQxYFRVfwS35CQpDgoJoVPck0HUmF7nSe/YZVEveLLjqJXLIXScsCJrocJBgrDd58vkBCMfi0I01F0EtDzOsgYwBShSLpvEZBNfB7RHySSPUgVFAmlSCrixQUlYDPQ1iGYKT08ehNpOhIF+lO5Zk8rIKYVyAQLF2/N5lG0QVM08QnQ3WslIBVlDyeQh9CMQuSB13yIw8iyUtl8/jUBKaSRfQGyEtRYtHS+V3pHjJ6hm39O5hcNZGAEKRmEHVKLp2wimTpRXQpgC/unpm2J/M09+UoagbjasIulJlaLNJb7GNrYhsRb4T6YAP1g5A16WyepGKysT3FyMoglUGZ2kG0LtlMFq+WRNAKmHKAoi9GKFDSdybTdGVU2hI5pg2rsPpjqNRfO1JZVE3AxOrvgyljCgWNvoLK1q40YZ/MsHjARaNRLCp48l2gKiD70CS/CyFnpNqsNu/cAPX7Y3pCDroPLARfKq9R1A2CHslBxznPlrIpZ0wDVfLjj9Qg2CWbMU30dKcVivWGoGI00iD6ECWfQVL6oXMdQmwEWqAWX6y0GkkXVPqzRTZ3ZphYF6Yy5HWtODqSGfpyGi39OaY0xAh5RVfRt45MN62ZVrJqjokV46kJlOhwABKFBDkthyiIhDwhFyAgXyzSl9XIFnUCHgmvJFA3qF3z+QwJLUtBL+CXfMSkIMHgP0/U+X8r2up/TfRUK7Uf/5m6hllQMxGeOA/z7MfAdh5GqgPplRuRM51gaJiVYzGO+zmi/ZHtzxtc8teVbGhL4ZEErj56AmceOMyiDgA6UgbfeORjOlIFIj6Z3587k1kjBeKhEP25HCtaktzw1HoyikZjzM/DXz+IAQBOZyLLwk9a+eO721F1k/2HxfjThbNotL+hPcksP31pI80JK946Iu7jJ6dMcSrL9RXg0sfWsq41iUcSuHzBOM492OMUNdqR8XDFwp3UR+tpT3bwvROjHDwqQ0U4TCqbYG3/Om5c8gMyaoa6YB33H3Uv48MDMOAcT63oYNXuPibUhfn9W7u476JZDLNtK6Y68L7+A+QNi612nvwl1JN/58AX00Wdyx5fTX9OpaDqnH9QA1+bO5wKe3XQkVT4+l8+oTWRJ+SV+PWZ0zlsQjXRgJdENs+qNoVrnlxDqqBRG/Hx8MWzmTYw3jJdVK3/G1VNT1qQxff74byFYDuPnmSGX/xjC882dWCacNx+Ndx+xlSq7SJZXqUf4emvWdTroow07wqMgy9DtGlfgrlW5IVftmooeENwyh/IjF5AOFKBquts74K732rj8Akjue/lXfz8y/tTHTYRBIFcsgd5w9PIb98CmoJcOwXtvKeRK4bbz53nt69tZsm2HmRRZEJtmF+dOd2B0yb1FN/94Lt057pRdIVTxp7C16d93WHO7c6r/McjqzBNk1xR56YTJ3D4hDjVkRi5fBZv12o8T19gJVpDNQjnPYVWPx1Zlsmlu6n4dDF1m19l/7ppsOR9tLMfg1CpONkdb2zn6ZXNmCYcNqGa35413ZksJRWVaxetoTWRR9UNTphWz+WHj6XOHguedAvCwjMtCnFPEPnEX6GOPw5PtB5VKUDfbjz/uAHq9of3f4N+8l0YgSpkj4d0QeXlte3c8sIGFM1gXE2IR79xsANZVZKdeN//FfIqq8CXOGYB+hkPIEctB9CTLXL1ol3s6TMpailOmtbF1cfEnZVSp9rHVe9ehWZo5LU8l824jCOlo6ixeeoifg8Rv4eRVeWUH1k1y/Lut7hr1V0E5ADSFok/Hv1HKmxcZX+2yAPvdPFpu02gmt3OvefFqI7YMOF8Lz9Y8gOWti1FFETOnHAmVx5wJRV+a/XbnVb5+iMr2N6dwSeL/OCk/Th+Wi310SCqprElvZPvvHMVvYVeYr4YdxxxF9Pl/fF7y8OQ/1vyhQxbidlehCV3wNNfhftmQ+tK+PAu1JyV9Ba2vYHw6UvWh6RtNcL6xVa9D6A3k+eONzazoc3aZa7qJr97fYuTSG7rz3LtoiaHpiOtaHznyTXkVCuskCuaXPvkOqemR1uywA1Pr6UtYSXSEgWdu9/a5uwKXtea5A9vbyNpJ34/2t7NS+u7aGpJ0tSS5KX1XXy03dp13p/Lc8/bW1ln1wRRdZO73tpGKm/Z1ppIcf1T69jenePD7b3s6MlxzZPryGuWbQUzx/Uf3ERGtfYbdOY6+d6HP6Q7bSWLM4rOHW9s4d0tPTz4wS6a2lLc9eZW+rI2+KB5qeM4AKRPX8KwE3u5XJY/vbuNppYke/pydKUVfv/OLrpzVgijI5nnu8+spTVhPWe2qHPdU01k7KR0ToXvPLHGoZvoSitcs2gNrQNJxEIS3rwFujfBzvegcy28/TOLQA5Y29zP4jUdTkL/9U3dvLXRSkpq+SQsvbdUs8PQED68C8HeuZ9LdCK/eEWp+E4xi/f5S/HbFCD9WZXLHl/Fsh39/Oq1zaxpSXLl31Y7Bb+8Whrv6zdZxZIAujYivHkzubQFgtjYlmLxqlY6UwqtiTzvbunmlXW2bYbGk58+yequ1bRkWujOd/PwhocdOpzOZIofPruRnT1ZdvVa7fq9xRso6ta8T1aSeJ65qFQLJduN/MzFTsElr5rF++pNVqJ62T3Q3oTw+g/J24WFtnRmeGpFs9NuH2zt4fk1bei6Tr6o8PjyPSzf2UdLf57OlMKjS3fTYhdcMlJt8MoNpdoTag7hpWuQ7V3lZr7Psq1zPax9AjrXIy/+GoZ973RB4wfPrXPodrZ3Z7n1xY1OUlro24Zn5UPOzmpx57voa55EVS3Wg2dWtrB0ey8t/Xm60gqPLNvNzh6rb3dluvnlx79kR3IHe9J76M538/PlP6do7ltNjEwxivkMugAAIABJREFUwy1Lb6E7382e9B52Jnfys+U/c9gndvVmefjDXSzd0cfL6zr4aEcfi1Y0o+sGhmnw8o6XWdpmgQUM0+CpLU+xPWHxj3Wmcvzs5U1s77ZsVTSDn7y4gaJmvYTufB/fff8GegtWQn+gAFifUs5E8L8pX0jnsTdqB0BoW4lQTGEYBsKepeXn2M4jVzRY11JOT7K10+YeEgQ271WYKKNo5G3nkitqLg4lwHJEdh7i0/byaze1JhzntGxXORXFwLGMorO2pVw/UKhKFCTH6Q1IXtXJ2ZQMea1g02qUZEv/Fmcvx1AFl9a2JK3CQIC0q7zdPLuXYhgGBaXIypZMmX5zm9XhTcoLYSma4Vw7r+plaKptXYOuN5hHaEDa1zhUMR/tSpSpl+3KoKoqopJGaFtZpqfN2n4kY5Sq4Q2Iplh8RVj8UXvTSezsyaLbtC96766yS0vtqxFsdMzHu8opPqxCVkXyWp5VXeXFyzb2WpBnzRTK2k3VzVIFRb04iC7ElsRuRBtBZSR2l+0RkNqbHNuGKny2cneCtKKSLWhOtbvBsto+JhgaQvsat9LQwXZcoq5adbgHS7IFweaN60gVylBoa1sSzvgxm/cq0AX4WpaiFXMUVJ3lQxSDWm0X+FLNotOGjmmmQVe+q+ycoaSn0FOGQtvYu9GhNGlqKW+Xj3f2kdd0inqRFZ3lYfYBVJyimc4EcEBME1pt/jPD1GjLutF/vYVeF9/ev4N8MZ3HiNllh8yxR4G/AlEUMSeeVH7OxBMBi7nz0L3oHgQBptiJOwHT2TswIFUhr4PMCXll4ntRfMwdU4VgbyqbPqI86XX4+Bon0XzcfuVorOPtY3G/zKF7Fe8RBAYVpzKYu1fRo3jQ4yBzgnKACp/b9gNqD2AgSju1MVZWcOnQCdVOElGfdGKZbdrkkxFFkXAwMGSRrQF6ElkQnEI9AxL1y84Gw4BHKqP4OGBwW9VNdQEBABhzhJMEP2ZyOcLuxCkWlb0RqLDq2O8to6xiQZrgwRh9uFvniziIl4BHpD7qToROGxZ1kF5SzfgyFJs65ijrGuBQZgyWoybXEvZ7CcpBFgxfUKY/oPYAALyiybxx7ncW9EpU2KEZU/KV6msMSN1UdMFqV7FyrFUTYpBoY45wQBdHTCyn/jlqcg3xoJ9YwMuCIfTz/z/2zjtMr6ra/59zztvL9Joy6SG9kEBCj9I7ihQLxUYTRYpiuSreq/6uIteGlAgq7SooIhB6kZJQQgKEkN4myWR6eXs95ffHPm85806SN5ghgXu+z/M+M2fvs/dZu66991p7LdMJk6a4McYOutfg9Oa1nDTFBdVWEyLUH4IuC9pHVHotwnaAoyfV57Wg5AmLSr6dnnQaLk8An0vhpOml2lQ551Ye2cvhTVZzNy7ZRZOvVEtqKNR760u02BY0L8ibix/clwFOmt6Iz+nA4/BwfMvxJfE5R28+p1LiyMqpSHlFGUVyMrFqoiV+VGAULvngObICUG666aYDTcNesXjx4psuu+yyst9X1QxywzSkna+DlsGY/mk49gYUU6vIcFeAJCN1vgcON8ZR30SfdBKyO4jH6WBKcwW7BpJs7YlRG3Dx83NnMabGQ8DjIuhxsXB8De+1hegIpxhf5+eOL8xjRIUDp9OJBBw9qZY3t/YTSmaZN6aaW86fyUjTLEI2m2XayEpWtA6Q0XROn9HMN06YRE1ATE5Bl4xbgffaozgVmauOHcOJ0xrxeVy4nQ4mNwbpCCXZ2hunxufip5+awbi6HG1uDh9XzXttETrCKcbW+rjtC3MZWeHC5XQiaRJHjzqSt7pWEE6HmV0/m58f/TMag+IcN6OqzBxZxVut/aSyGqfMaOK6EydTa9KWdfgwnF6U9ndBcZBdeBXGtLNxuAMoDgcT6v10R1Js6olT7XPx3+dMYWqDF7fbg9/t4PCxNazeFaY9lKKlxsdtnz+UllofTkVGkXSOndTA8m39DCSyzBldxa8vmFM4jzYMGDkXti8T+u5TTheaPGabehSDCo/Cu20RZEniK0eO5sxZI/B5XMgOF9RMgHgPUs8G8FZhnPIL9IapyO4gTo8PreUI5M73TKdFY8medz/Z4EicTjdep8Ixk+t4Y2s//fEMs0ZV8vvPHZoXeqdUA0YfjrJjKWTj6JNPg5P+E7d5gVCRocrnYtXOELIkcdHCMXz60JEE3E5kSWZ0xWjC6TCbBzYTcAX43oLvMbN+Jm7Fjd/tZt6YKtZ2RGkbSDKq2sutn59Nc4UDr8uNihMmnYC84zXh47p5Nupn7sFVI9RY07oBLQtQWpcJ2iadjHHyT3GZDpkkyaAu4GbVzjCSBJ87vIXzDxtFwO1EURRGVXsIJ7Ns7IoS8Dj43mlTmD0yiN/jQnYHYNRh4phuoBUqR2OcezdqcBSKy4eseNAmnIC8Y5kwt9E0E/W8e3HVjsnXy+HjanhtSx/xjMpxh9Tz47Om5x18ZVAwgiNMd7IG6qFfRJp/KQ5PEEmSGFnlJZZWWd8ZIeB28P3TpnLY2Go8TgWfy8vM+plsCW2hLdpGo6+Rnx/7c5o9I/C69m4vzSE5mN84n+Wdy4llYxw54kh+sPAHeTmUU5EZVe1lRWs/umFw4eEtXHLEmPxCrdHXSFpLs75/PT6Hj2vnXcv8pvl4HB58bgezRlWypSfG9v4ETRUefnXBHEZUuvG7nQTdfhY0HcE7PW/Tm+xlUtUkbjnuV4wOjkCW9229/+Mf/7jjpptuWrxPicrEx1bbKpuMoKSFNVtNduIYpPOdjQ/gyMZBAtXhx+m3ru66wglUAzCg0qcQcFtXIR2hOLohIWGUeG1TVZWuWBKQkCSDEZVWswuxZJpISgMJnDLUV1h14aPxOImsaBefUyLot+bfFYkLU+aGUMUMeK20tYej5v0Ig8aAF4fDuvLsibajAwoSdYPujSQyWXHJSwKHXLinkY9PhHGaMpOsw4/Pb91tRKLChpSEQaXXiXuQ6mJXJJU3S9JU4S4ZDDkZhyQJo34WZNOQszqsuGGQR7pYPEHcNAnvc0glXt+0aJfwn20YaJ7KvGfJHJLhHpxGFg0JJVBfUm+90bSwD2V6gLSQlkljJPqRJYMsbryV1pVlKJEknjGQkHA7oDYwyJFVNk4imwBJ3CEZfK+kKxxBM2TAoM7nxjVIcJoKdeJAI4sj730xB1XNose7kAFVceEZZCMplEiRyIh6czmgLmCt9/5InJRpOj/olgh6rbTr4XYkNDBA9zWi7IY2FQXPIC00TTfoj6fRDfA4FCoH7dpTqThKSjC2jOLFF7CO03haFUefElR5XbgGGVLsinWjI46fqlw1eF3lu3LVDZ3+VD+6oeNW3FQOUjHOqLqwkm2A313YReeQyCaIZ8UF20p3JS7FWi+d4aTwNW8Y1PqdeFzWsrdHewAdCZnm4L4Zh81hOLWtPrbMw4YNGzb+r8M2T2LDhg0bNg4q2MzDhg0bNmzsM2zmYcOGDRs29hk287Bhw4YNG/sMm3nYsGHDho19hs08bNiwYcPGPsNmHjZs2LBhY59hMw8bNmzYsLHPsJmHDRs2bNjYZwwb85AkabQkSf+SJGmdJElrJEm6xgyvkSTpOUmSNpl/q/eWlw0bNmzYOLgwnDsPFbjeMIypwELga5IkTQO+A7xgGMYk4AXz2YYNGzZsfIQwbMzDMIwOwzDeNv+PAuuAkcDZwD3ma/cA5wwXDTZs2LBhY3jwocg8JEkaC8wF3gQaDcPoAMFggFJnByLNZZIkrZAkaUVPT8+HQaYNGzZs2CgTw848JEkKAA8D3zQMo9SN3m5gGMZiwzDmG4Yxv77+g5kjtmHDhg0bw4NhZR6SJDkRjOMBwzD+YQZ3SZLUbMY3A+X5hbRhw4YNGwcNhlPbSgLuBtYZhvE/RVGPAZeY/18CPDpcNNiwYcOGjeGBY++vfGAcBVwErJYk6V0z7HvAfwMPSZL0ZWAHcN4w0mDDhg0bNoYBw8Y8DMNYCki7iS71Dm/Dhg0bNj4ysG+Y27Bhw4aNfYbNPGzYsGHDxj7DZh42bNiwYWOfYTMPGzZs2LCxz7CZhw0bNmzY2GfYzMOGDRs2bOwzbOZhw4YNGzb2GTbzsGHDhg0b+4zhvGF+QNERSpJSNRIZjWqfixFVXkt8b6yTSDaKjETQGaA20JSPU1WV7nia/niGgNuB2wHNlcF8fCrej1tNQLQLAvVkFBfuYBPloiMcJZWFeEalxu+k3ufG6XTm4zORXcgpYUNS91TgqhiZj9NUlZ5UHwOpEH6XD4/kpSFYV0iragwksvTG0tT4XQTdTgKeQjP3x9MkMxq9sQz1QTeVHgd+T+HboXCYqCoTT2Wp8buoCXhwOArpM+FO5HQYDDA8lTgrreXujHUTSofwOrz4FC/1gTrKRVckgqor9MZS1Ac9uBWd2kBw7wlz3w4niaZUdAMqvU6aKj2W+I5Qgv5EFp9LweeQaKzyl513Iq0SSWXpi2eoC7ip9DrxOJXyaYt1EVdjaIZG0FlBc8Babz3RNAOJDG6HTIXHSbXflY+LJ5O4MyGMWDeyr4ak7CNQWVtInI5DOgzxHvA3gLcKnIX+3hFKktV0QsksdQE3VV4nPnehTbvCKaJpFVXTqfQ6aR40VtpDSUKJLG6njM+pWOKjySTRNPTG0lT7XDhkg+Z9qNdwMkssJeq21u+ixu/CoRTWtKlIO45MDLQsuqcKV+VIS/qOUIJQUsWlyPhdMk1VvkLadIb+pEpfLEOlz4lHgYbKItqyCUiGId4N/nrwVIGrkJ5UGFIRSIVEvK8elPLbvDeWZiCewaHIVHqd1BS1KboG8V7RZp4KcFeIdvsI4WPJPDpCSX725Doef68DgJFVXv5y2UJaakTH6I1387WXrmFt31oA5tTN5leLbqHO3whAWzjJBXe+SVckDcD580dw7YkTaa4Moqoq7p41SA+cLzqfrOA69WbSU8/CHdi79d+OcIybn9nEP94WtDVXenjw8gW01IgJXI104HjwIuRdKwHQR85DveABHBXNALQnuvjiM5fQlegC4MzxZ/PNudfQYH57TXuEz9/1JomMhizBT86ZwTlzRuJzOwgnszy5uoMfPbYWTTfwOhX+9MXDOHxsNbIs0x+O8MvntvC/KwRtjRVu/n75AkbXBk3aunD8/WLknW+KwoyYi3rhX3FUiImwLdrOl565lI64SH/a2NO5ft4NNJTBQHpjUf61oY/vP7IOTTfwOGXuvuRQDmtx43K59pq+M5zi6395h7daBwCYPqKCuy85LM9A2gYSXLj4DdoGkgCcPrOJH5wxlaZK327zzCGV1XhuXRfXP7QKVTdwO2T+dOlhLBhfiyLvzohCAV3xbr679Dus6FoBwOTqydx2/G00mv2tM5zi/DtfZ0d/AoDTZjTxk0/NoMbvBsA9sAnHvWeIyUyScC36IdE5lxKsrAE1BZufg4e/DLoKDjd89q8w9jhQFDrCSe56dRt3L9sGQI3fxYOXLWRSo2jTrkiK6/72Lss29wnaGgPc86XDaa4UDKI9lOTzd73Jtt44ACdMbeAn58ygyYzfOZDhwj+8QSSpIknw9U9O5LOHjSqLgYSTWf7w6lZufXEzANU+J3+74ggmNgjaMpF2nI9djbL5BZGgbjLqxf/EYS6mOsJJvnD3crb0CNoWTa7nv8+dkW/T1v4UF/7hDUKJLABXHTeBS48cLRiIloWtr8BDF4GWAcUJ590LE08Ah0vU9Rt3wMv/DwwDvNXwxaegYepeywXQHUnxubveZHN3DIBjJ9fzq/NnUxsQbUr/VvjjSZDoF89HXwdHXfORYiAfy2OrjkgqzzgAdoWS/PaFjfTHBTN4vvWZPOMAeLd3FW/sWgZAbzTGL57ekGccAA+taCeUUAGQYh1Ij35dMA4AXUN65ru41FRZtPXGMnnGAdARTvGr5zbSFxcDQN/wRJ5xAMi7VqJvfBKAUCLCre/emmccAI9vfZSeVK9Je5rrH1pFIqOJvAy46bG1RFKC9nha5b+WiMkZIJnV+Pbf36PTLGtPXM0zDoCuSJqbn9lANBoGQNv0fIFxALS/g7b2cQCiySi3v3t7nnEAPNn6BJ1FtO4JaVXipsfW52lLZXVufHgNPYn0XlIKLNvcm2ccIJjoo+/uAiCSSHLri5vzjAPgidWdtA2U12bhZJbv/mM1qklbWtW57qFV9MXKo21l98o84wDYOLCRRzc/iqZpZFSdO1/ekmccAE++38lWc7JOhLpwPHGNmMwADAPXS/+FTzffTw7AY1cLxgGgpuGRKyAh+kQyo+UZB0B/PMN/LVlLZ1jUxTs7BvKMA2BjV4y/Lt9JVtNJpLPc9erWPOMAeH5dd35C3BVK8INH1xBJqjnS+N2Lm8nqZVULkWQ2zzgABhJZfvDoGkKJjAjoWFVgHAC9GzGW30UmkySeyvDnZa15xgHw0sYe1nVEAegKxfnRY2vyjAPgtpe3YJIKiT549ErBOEAwk0evgqQ5maci8NLPRKFy9bzk2sJkvwdous5flu/I1xPAKxt7WL3LbMNkCJ68wZrX0v8R4R8hfCyZx7aiDpXD1p4EyYyGruusDW0siV87sAGAlKqxrTdREr+jX+QpSxKEWq2RalqsAMuhrS9WEralJ0EyKyZ8R9eaknhHpwhLqklaI1tL4ndG2gDQMWjts5Y9o+mkzLyTWY20ah3ZOwcSeQNku/pLadvck8ynUbpWl8Q7u95D13USaort0W0l8dsjO0rChkJa1UkNmnXaBpJIuzWPZsX6zmhJ2IauKFnz6HJrb2mf2NJTWt6hkFH1PEPOoTOSwuQle8WmgdL+tjm8mbSWIaVqbOgqpX2r2YcVNOgb1OaGjm4ea6JlIT0ofawrz0w6w6X9cltfnKwmiB+q3jZ2RcmoOvGMZpmcC/Gi3gwDWgfVq2FgmbD3hL54KfNt7Y2T0UQ/MHrWlcQ7e9ajqUnSqsrmIdovV5dZHQvTy38zx/B1tZQRJAcKTDg5BJPo3yLqey/IagZrO0pdF23I1bWahr4tpQnjHy3vFB9L5nHY2Gocg44TTpvZREPAiSzLnD3u9JI0p487DYBqr4dTZ1qPn1yKzPQR1QBokoIxYZBdx6oxGE7rOfHuMHdUzRC01VPrFdtZbeZnStLkwmq8VZzYcoolziE7mFk/HQCPQ+H4qVbHjM2VHvxucU7rcyqMqrbSedzkepzmGfPUkTW4FGuXOHNGHRV+kcaYce4QtF2ALMuCttGDaJMczG2YXZJmKHicEmNqrUdIx0yqRZLLW8aeMau5JOzsOSNwOhRq/W5Om2mVMSiyxMLxtSVphoLXqTCh3noMc8T4WjzO8obPiWNOLAk7bdzp+Fxegm4Hn5prPceXJVg4rgYA1RFAPWRQf/XVgE/0Rxye0qOUloXgFMd1Y+v8+FzWc/qTpjUSNOVgJ08vldWdPWcEfreD+qCnpF4lSfQZAK9T4oRpjZb4Cq+DutzRzF7QXOkl4LaenJ86o4kKUwanTD5VfLAImZmfweuroSbg48xZI0poO36KoKfCrXDSdCttQbcjf9yGwwMjDh1E0BwRDhBoBE+lNX7qWeDeuwzO41T4zLzRJeHHTzHHpqcSpp1tjXQFoKo0zcEMyTDKXD4dQMyfP99YsWLF3l800R1JsLE7wU+fWEtfLMN580Zx8ZFjaawQHaM/1s37vWsYXTUWDINd0Z1Mq55CTVB0tq5Igrte3co/3+mkocLNj86cwoR6L7WBAABGuA2W/wFJS2NIDph3CWrlWIvQGyCjZnE5rGE90Rhb+xJs70nhccpEkllOnN5IY4WYONPRduSu9cjNcwDQOlZhNEzBbco8euK93L/uf3ls6yPUemr59mHf5ZDKSVT6Ksyyp3jq/U6qfQ4iKY2jJtYxttaHJEnouk5rX4ItPTHq/C52DiQ5bGxNXgAai4XZ0JPmh49voDuS5jNzG/nKMeOorRDlToW7Uba/itL+tqCtcQb6+ONxVzaYtPXxv+v/wqNb/kG1p5pvzb+RaVVTqDBpy0PXQC4VPG7rjfKDf65hTXuEIyfU8P0zpjKisjzha080zVvb+kirQtajGRJHT6qjIegx6yXJfW/s4MG3dlLtc/H906cyvclHbYU1f03XUeRSprCzP8F9b7RiaDoOh8IlR44rEcjvDl2xHlZ2v8Vtq24jo2W4eNrFnDjmZBr9YhLuj2f4y/Id3Pt6K5VeJz88YxpzW6rxmxOrGumCF3+CY+MTUDcJ9bRfk6kci89rToShHfDE9bBrBbQcBaf9AirExDqQSLG9N8lNj6+lbSDBqTOaufqTE/NjoS+W5q3WAVa09mEYMK7ez0nTm/L11hVJ8dCKnTzwxg6CHgffOXUKs0dVURd05+MfeHM78ZSKJEt8au5IWqpdBL3WRUpG1XA5rG2e1XQ2dcX4/iOr2dGf4PRZzVxz/KS8XCAe6cC1awXOF/4TMnGyCy5Dn3VhXjmlO5Lk7yt3cd8b2/G5HNx4ymTmjArSUBkwaUvy2+c38/SaTsbU+vjxWdMZX+fD7zFlaOE2ePJbsON1GL0ATr8FKkeZHUGF3g2w5JtCPjH1LFj0PShDrinqPcNTqzvY0hMFJI6aWM9hY6sJ5pRT4j2w4k9oyRCKBMy6UCwCFOeest1nSJK00jCM+fs101zeH0fmATAQT9MXyyDLEh6HjCIbNJoTUSQepT9hcM9rrcgSXHrkOKp9MgGfiB+IdiEpQWJpDUmCxoDXqnEU6cHZ/hbS2kcwxhyFPvk0lGBhxd+d6Ob57c+zunc1x7ecwKy6mTT4C/Gd8U4e3vgwnYlOzp98Hk3eJuoDIj6RGMCd7Ic37xQvL7ictKcGn1+sNKOJPuJalqSWQTcM6rxVVBatkLoiMd7dGeKp1b3MbQlw8owRNFX4iuKTPLumi5XbQ5w8vZFDWyppHCQ07hsIoRkSQY8Dr68wuaYSYXqSCn9+rRXdgEuPGEODT8frL3w/kU4QyggttqbgIPf0iX7YtRJW/00M1qlnWQZjOp2mP51G1cDtkPDJbgL+vQvLAbKZFHKsE+PNxUh6FhZcjh4cgdMtytYfT6OgE03rSEBtwI3LISObjGIgnuH99jD/eHsXc0dXcdrM5vwEKSq+E33Tc8itr6AfcjpyyxEQbByCklJEkkl0JKJZIVsIKFWoukF9MJB/J6PqhJIZZEkacuUej4ZxqHE0yYmvaogJLBkGNQlOn9DeGYRdAwl0AwIeB9W+Qp1mVJWuSJr73thOIqNxyZFjaKpwEfQUGGM8lWUgkQUJmivcKEUaR12RGJu6EvzjnXamNQc4c/aI/EJIxKd4ZVMPyzb1cezkOo6eWEdDhZXp9sczqLpO0O3EW7RLymZVUlkNOd2FDGRc1cgoBP0i/0RaRdc0wmkVDKgPOFEcTosSQzieJJbRUYCm6iEWIqkwZJNCO23wTgOEbETLCm0o196VK3JIqkk6Yh08sO4BAs4AF065kHpfPQ5ZzCO9sTibuxP8bcUuJjf6OWfuSEu97S/YzGMfmUdHOMnZty6jOyrONyUJHvjyAo6cKLR+tvdEOOk3y/Jn+V6nwvPXHs3IGjGYl+54iSv/9fV8fiP8I7j/lHuoDzQRj0Vxvf5rnMt+mY/XJp5I5ozb8FY10Bnv5vtLv8vyzuX5+CtmXcEl0y8l4PLTFe/k809+wSL0vuOEOzhq5FEir/6tKLcdUZChODxoV72OUjMegDfalvLVF67Mp230NfLX0x6gzt9IKJHgj8ta+e0LBdnDkRNquOX8GTRXBukKJ7jh76t5dVNvPv6qRRO46tixBHx7X0W39UU58dfL8vIZt0Pm2WuOYkx96WRVAjUNb9wGz99UCBtzNFxwL/hqCSUy3PC3VTy/rnDue+WiCXzjk5MsE8ruoA3sRLl9AWTMc27FhXbl6yh1EwFYvq2fCxa/npd/1gfcPP71o2iq9JJRde59vZWfPFE4Y583pprFF8+j1u8WMoQl18H6JYUPLrhcrETL0I5pG0hwyq9fJZYW5+lOReLJbxyT13g6kNgVinPKr5YSNWlzyBJPfOMoDmkaYiIdhFQmzYMr2/jRo+vzYdNHVHDXJXNprgzQG0vzkyVr+ee77fn4Cw4bzY0nH0JNGUdbnaE4J/1maV4gr8gST1x9BFPMI+R1HRHO+N3SvJJFpdfJs9cem99VHUhsDm3mvMfOQzUE7UFnkEfOeYRGXyOapvHIqjZueOj9/PsTGwLc/5X5NFWUr+ZcDoaTeXwsZR6rdobyjAOEEO+2l7bQGRaC8Pve2G4RHCezGn9fsR2AvmgHd7x/lyW/9ng7rWEhtHSpMZzLb7PEK5ufw6mLyT6rpS2MA+D+dfcTzQhh2fr+DRbGAXDX6rvojnUKWlf+2Sp8V1MYK+8BIBzv5s7377ak7Up0sblfCPsTWZ0/Lt1piX9tSz8ZTazE0pphYRwAf36tlWimPLnC31fsyDMOEELu+9/YXlZaUiFY9htr2PalkBZCz0RGszAOgD8t20Y0VZ7wVVv1YIFxgNCiefNOdF2nL5bmtn9tpnid1BNLs2K70M4KJzPc8bJVgLly+wBxU0uNbMrKOABW/hky5QncH1vVnmccIASqdy/dRipbXtmGE0+9355nHACqbrD41a1l0dYbz/CHV6ztv6Y9ktfuS2U1HlvVbol/eGUbyTLVsf61tj3POAA03eCOl7eRSKZIZTVuf2lznnGA0Ip7eWNPWXkPJzJahj+u/mOecQBEs1Fe3vkyAF2xBHe81GpJs7k7RnekPO29gwUfS+bhUEo1dByyhCSJjuYcQjffmRcUS/mtZTEUqWj1O0R8TrAnSaV5K0Xn+0Pl7ZSdea0iQy498zTMc1AJyUpHPn+Rp2Ew5L0DKf93CNokiXL3nkPVq2uIsCFhMKScY7BAdDBt5cIY4qzYkJ1ICAG0MmSfkPOkDVlvue9LUimdkmKm3DsGK0jkvl1+6YYPQ478pD5RAAAgAElEQVQFWRqyH5fCGLJscm4sUDoeZEnavX/RwXQopdOTU5HIZbG7+AMNaTdziFPOjeOh+0Q5d4YOJnwsmcf0EZX5C4EgGuUbx0+k0dwSfn7hWPxFRyEVXgfnHCo0HWqDTVw96wrLRDuuchwtFWMAyLqCZI+63vI9ddqnychCQOiUXXxi9Ccs8V+d+VUqXOIYYGLVRFqCLfk4WZK5Yvbl1AfE+bl86MXifDUHdwXyoRcJOv31fH32FchSodnGVIxhfNUEAIJuha99cpzl2ydNq8eliEnOrYgLaMW4ctEEqtzl3Zo9d14LFd7CoPC7FD63cGxZafHViGOeYkw+RWiZmHkN1jr62icmUuEtT4DomPkZcZErB5cfacFlSLJMtd/NNz45yTJgR1V7mTNatEmNz8V1J0625Hfc5PqCJpDDCzMvsH5w4VXg3vvRDghNsGpfoRwep8yXjxmL27l/haMfBCdOb6a2SK7kdsh89djxuB17vz/cGPDxjePHW8Lmj6kiYPYnr1PhwsOsGkQXHdFiGXt7wjGHNFFfdLzldshccdx4vB4PbqfCVYsmWrQD64NujppQvkWD4YJTcfKlGV/CrRRor/XUcvTIowFoDPr55okTLGlmjqygNlCefO9gwbDJPCRJ+iNwBtBtGMYMM+wm4KtAbm/5PcMwntxbXh9EYN4bS9Jr6nEHXX6cspMGk3nEknEiKYNH396JIkucOWc0lR4Jn1cwnHCsD91wkdUkMHSCPgded4EZJcM9KP2bkTc8iTZ6IdLow3BVFAnM49283f0Oq3vf45OjP8moYEteswaEQP3lna/QGe/gtPGnU+UIUmsKjlPJCM50CO29BwFQZl1A1l2FxysYSiIVoivZy5ItjzPC38RxLZ/M34wH6IzE2NGX5Nm1ncxtqWLemBrLOWpXOMnbOwZYuT3ECVMbGF/no6FIYJ7NZuiJp5AkGY+sUx0sMLJ0KkF/Uufxd3aiGQZnH9pCtcfA6y3z7D4ZJZ1JoKCjGwYun9UcRF8szaq2EG9u7eeEqY1MbAxYhLt7gqaqEOtCXf13JC2LMvt8jEATDqdIH0lm6I1leOSdXTQE3Zw4rcmiLRVLZHFm+nBKGrpukPY2Wsy2EO1Eb38HWpfB5JOQ6w4pW2CeTqdJZXXimSwSkjBrokC1b/8JSAdSA2iGhkN2UOW2ymEMw6A/1Y+BgUfxEHAVBPWqptETS/PEex3EMxqfmjuCuoATn6sw8aXUFLGsOKKrcldZVtXdkSgdkQxLVnUwpTnA0RMbLILf7kiKtR0Rlm3u45hJdUxpCloF5oYhNI8MQ/SFIlVYXdPoiaZ48r02oimNcw4dTZ3fkdcyS6sa/fEMoUQWwxAWEWr8rjJ3TcOLjJZhIBljIJECDBqCfqo9lXna+uIJOsJpHnt3FxMb/Bw3udEWmOczlqRjgRhw7yDmETMM45d7SjsY+8o8wtFu3u57l5++9QtC6RDnjD+DK2dfRa2p8dQf62bJ1ifISjoGBn7ZzUktJ1FrTga9kQS/eHo9/1zVRUOFm5+edQgzRwapqRATaVt/gm29UWaNrGLnQBJFkRlf68Pt2vtqrTvaz+r+93iy9XEkQ6LWW8+XZ3yZBr9YMQ1EYyxZ3UVNIKdWnOKMGY1UVwT2lG0eHQMx/rhsB/VBNwOJDGfPbmJcrQe3e+/3UAaiYd7eFedHj66nO5LmzDlN3Hjy5DzTDUViPLOmhzlja5EkgxXbejl1evm0aaFdSEuuQd7yAjRMRT37Dozq8Ti95aX/d9AdSXHnK1voCKeQkDh1RhNHTqjNC26NcBssuRZpywtQNxnj7N+j109DcXnRdZ2tka0sXrUYVVdxKS6unns1o4Kjyvp2XzjKU+938T8vbiOj6nzlqFFctGAMtZX/frkNw2BHdAffffW7vN/7Poc2HspPj/4pIwNiF5dW06zuXc2PXvsR7bF2ThxzIjcefiO1XnHHJRyN8czaLl7ZHAYMJtf7uOiIFqqDpuZhaoC7Vt/F3zb+Db/Tz7fmf4tjRh1D0JUzWdOJ9PrvUQINkBogO/EU1PoZeE3G2NoX51fPbiCjGTgVmRtOPqRwKpBNQce74mb3wHaYeiac+nNxx6IMdEdSLH5lK/+7fAd+l4MbTp7M8VMarVpyBwj98Qx3L93Kn5e14nEq3HjqFE6Z3lT2Tnp/YTiZx7DZtjIM4xVJksYOV/57QlRPcs3LN2CYZ9IPbnqYkf5mPjflYtxuL+v613LzO/9jSTOl5hBqg43Ek0nuWbaVh0wTIm0DSb503ypeueFYADpDCb719/d4fWvBpEND0M0jXzuSkWUwj5SR4LqXr0E3CkLDClcVl069GL/Hy/rOKD94bL0lzaR6PwvLmKBD4RA3/mM9r2wq0Pa3lW08efVCGsoYT7GswmX3vpsXQj68sp2mChdXHzcOr8fHpp44Nz661pKmpdrL0WXQlg5343ziOuTNz4mArjU47j8b9fLXYJiZRzar8Y93dnH30tZ82JPvd/DCdcdRE3CjRXuQn7oRadOzIrJ7HdJ9n0a+8jVwjaQ72c3Xnv8a7fGC8Lct1sYtx92St0+1J+wKpfmPxzfkn3/9YivTmoKcNOvfL3dfqo8rn7+SnVGhKLGyayXXvXQdd5xwB9WeakLpEJc/dzkZXZjheKr1Kao8VVw/73rcDje7wmm+/Yi1v02o93H6XD+GYfDijhe5d+29gFA/vfHVG3nsnMcIuoIkomHcz/0IZfVf82mdb96B9LUVgI/eaJov//ktyy31Td1R7v/yAnGXIzkA955dUBBZ84gwTnjyz8pSi31xfTd3LRWahYmMxo0Pr+bJb1QdFMxj6aYefv8voYQRzwgzQDNHVn7ozGM4cSBkHldLkvSeJEl/lCSpeu+v7zve63o7zzhyeGHXq8TSA+i6zlM7ni9J89x2ERZPpXlug9U0gaYbrG8Xmjk6koVxAHRHhaXacrC6Z42FcQC8sutFBjLCnMGS1aUmCh5fXZ59qIzh4NXNVtp6YxniZWpTvd/eb9FeAXhxfR8DSaEF8uT7pXQ8urobQ997/goa8pZB9Z7oLzWtMQwYSGZ5fq2VdsMQ6rsAkpYWO45ipEJ5e1JpLW1hHACrelahGeW1+XNrO0vCHn+/l0SqPJM2e0JaTecZRw5r+9aS1kSbtcXa8owjh5d3vkw0K+r9hXWlbfr4ml5SqTTxbJxntj9TEv9mh7BvpqhxlC3PWiMzcYyo6MMptdS8ybqOaEHTMdxWatZn0zOQLjXtMRihRIZn1pTSfjBoWyWzGkuKbOvl8NKGj5b5kb3hw2YetwMTgDlAB3DL7l6UJOkySZJWSJK0oqdn3zrEhJrJJWFTqybjcfiRZZlD62aWxM+unwWA26EwralU13pMndimG4bBhHrritHrVPCWaZ57QtW4krBDqqfic4hjpfktpfKDw1rKuEcByIbG5AZrerdDmNEuB+PrSr8zpTmQTz9vdGn84S1BpCFuZA+GpgON062Bigtc+1evfSj43QrTRwxVNrNNkaBh2iDanHnFBZfsIuC0tvmowKiyz9bnji4VrM8b5d8vAnOX4qLCZS1bk78JhyR2wU2+UvMjh9QcgkcRx6JzRpfeU5k30o/L6cDj8DCrblZJ/JSaKQBoshMaBrWpJCObVpRdimw1Q47YpecVF4KNpVpsDdOFgsJe4HcqzBhZ2qazRpWnxDCccCkyc1tK18WzRn10LOaWgw+VeRiG0WUYhmYYhg78ATh8D+8uNgxjvmEY8+vryzMJkEO9u5aLDvks9Z56JlZOZGLVRL4y6yv4faJjHTf6EyxsWpB/f9HIY5jfJEipDAa54aRDGF8nJjVZgquPHUOVR1TVyGoft5w/iypTe8bjlPn5ubPwu8qbSKpcNXxh6iV5jalxleO4YtYV1JgmPI6Z3Mhxkwo2lxZNquWYyQ1D5jUYddWV/Oq8GfkB63bI/PLTUwk6y1shV3slLjt2DLmxPb7Oz/UnTqYqIGg7YmI9xx9S0GY5ZkINn5hanh8Td3Uz6lm3i/PsYBO4K8ic8TtUZfgvdPndTi4/bgJTmoLU+l0EPQ6+sLAlb3ZcqRqJcdbvBF2BRnBXYJz6SzSHODqpdFfy06N/SrWnminVU6jz1PGzY35Go7e8s/mZo6s5bXo9FR4H1T4nh4+t5ow5oy03tT8oqtxV3HzszXjNCTfoDHLzsTdT4xW2sSrcFVx76LV5Fe8R/hF8+7Bv54Xm00ZUctbMQv+a11LFp+a1ICsKDtnBhVMuZFpNgbGeO+lcxpiah77KetTTfw0VI/P1ljnpF6RkMXaq/S5+99m5NAVdHD2xlqagm1s/N7eg3eWphJN+VlB9rxwFp90M3r0zAKdT4fMLWphdxCzOmTOSiQ3DLz/bGxRZ4rx5ozh0TBXjan2MrPJwzpwRTGk68JdC9yeG9Ya5KfNYUiQwbzYMo8P8/1pggWEYF+4tnw+ibZWOtKMk+9GTIeTqcWieWtzuwkTVF+sipaeRkHBLrrywHCCbzdIbzxLPaLgdCm4Fi0ZSOJ4knhUmzv1uBy7ZoK5IUyKczBJJZtkVSjK21keVz2VxHNQVHUAlRUbL4FG81LqrLD4rBiIx4lnRLn6nZBFIa7pBXyzNjv4ElV4ntQG3ZXWXSScJJTXiaQ2fWyHo0PAFyl+N9UbCpDSFjKrjdUk0B32W+xkDkRgJVezAfE6JmjKF5QCZRAhnJipsBVWOQnP6875A9gd6Y2m6Iik03WBEpddy9m3oOmq0Cym0A8kdRPXW5W1yASTTcSLZMDsjO2j0NxF0+Kkq0pBLRHpwqQn0/u3ItePIKD58FeUZVgQhK+uJZcjqBiMq3BanRaqq0hPLsqM/QcDjoMbvyjO2cpBW00QyERJqAp/DR5W7CmfRvZd4Nk4sEyOtpfE5fdR5reqsXeEEKdVA0w18TqtDJYD+VD+JbAKH7MDn8FFRpEoeTaaIpzW298VpqPAQcMnUFykCpGPdOLIJYTmhZgKq04s7ULQYSsfEMVU2BW6/cGa1D9pSvaEoiayOQ5bwOqG6okBbPJ0lnFTZ2hOjqdJDwK2U5b9lf6E7FKM9lMTlkKkPuKmvKn+shBIJommd1t4YY2oDBDwyNR9AO+8jKTCXJOkvwCKgTpKkNuBHwCJJkuYgble1ApcPx7czkXYcj1+DkhOA+uuQvvwcuIVOenc4zmfvWp83yT2lKch9XwxQb9q+2jGQ4NO3v0k4KW7ZnjqzgZvOnEJjRZCMqrFyZ4TL7l2Z9+9w4ylTOG/eKOqCbmLpLP/75g5+/rQQQroUmXu+dBgLx9ciSRK94Sg//OdGnlknbnpX+5z886qFjKkrMIDqigC7Ewa1DSQ45/fLhK0h4JTpTfzs0wXHQS63tyzh+JBI9FH39PWw9hHx7K2Gr7wAtQWd9D3RtifomobU9R7S/efmfSjox91I8rDL8O6Dt8HdoTea5pI/LWdNuzgvH1fn58HLF+YN/OmhnTj/eALEzHPnCSeQPft2nKaK9YbwJr78zJfz8oErZl3BxdMvJugKkowN4Fj3GI6nrxfCElnBOO9+0u5FuN17H9A94Thfve+dvD+HMbU+/nbZ4Xmvdu2RDJ+5/bW8VYQjJ9Ryy/mzy2Ygboebesfud+d+px+/c+jjwd5YmiseeId3dghfEqOqvTx85ZEWEx81nhpqPDVDpt/ck+Szf3gjb07/i0eO5YrjxtFY6SObDCNvegblsa+jGAZIMvqn7kSdchqOnLqwOyB+HwSRdur+dKKQnQC0HAnn35u3l7a1J84Fi9/Im9P/wsIWrv7EhA+FgXSF4px75/K8D5m5oytZ/IW5+TlmT0hmMryyqY9v/vU9dEPw0p99ahpnzh5BwH3glQFyGLZjK8MwPmsYRrNhGE7DMEYZhnG3YRgXGYYx0zCMWYZhnJXbhexvSKEdBcYBwt3jv/4fqbgQJj/+bpvFl8P6zigvrhNCzYFYjP9+emOecQA8tbqbnqh47olm+I9H3s8zDoD/eW4DaVV00FhK5ZfPFjRrMprOjQ+vpi8mJqWOcDrPOEA4wLnlmY2EonsXHMfTKjc/syHPOACeXtNJe+jfF7wCEOkoMA4Q2jDP37RfhNrZWCfOJdcWnO8ArlduxqUl95CqfLyyqSfPOED4cnjkbeEMKptKYLz8iwLjAJQtz4sdENCf7OfHr//YIlhevHox8awQ9jrVJK7nvldwDKRrOJ+4BilZcD61J7y+ubvgCAjY3pfgweWtZLNZIskMt/1rs8Wczmtb+iyOhIYTK1oH8owDhHbhA29uRytDCaJ9IMFNj62x+GH502utpFRRT1ImivPp7xbqzdBxPvXt/eP0SMvCG7cXGAfAjteE6i/QEU7wo8fWWvyw3P/GjrJNo/w7yGSz3PPaVovzsXd2hnlne98eUhUwkMjww0fX5f3FGAb8ZMkGi6mWgwEfyxvmeqjU3pIS2o6hJtF1nU09pZPt5h5h9yqtakNOxu2mXSxJEo6AipHVjLwGSSqrl2gstYeSee2vjoFSBzXbB1Jk1L13jLSqs7O/1FFVR3j/TMBE2kvDBlqFUcN/EzIShK1aQRg6Rnb/0D6UY6eN3VE0XUfPpnCESh1V6f2t4q+h0x6zll03dFI5TSAtU6oVFO8pe/AM5YhqU28GTVdJZnV2DpS26fa+0rDhwLbeIRyAdccsi6PdQUd46RyMnG0rdLV04ZEcAGM/TOBaFvo2lYabTpY03aB9CNoGynRU9e8gq2ps7c+UhA921LY76EM41IpntLLa5MPEx5J5OFoWCk2eImRmX4gr0Iwsy1wwf0RJmk8fKi5V1fp9nDXHKqD2OGWmjxCaEg5Z4sRBDnDG1/nzMg2/22ExjQJw+qzmvDbWzJZa3A5rtZ9/aCPVgb1v3au8zhInM26HzIyR+0nDpGmGME1djLlfsJr9+IDQnB60aedYA6taMHZznLKvOGv2yJKwzx3egiLLuPxVZGZ/wRqpOHGME5aMfU4fp4471RLd6GvMC5V1xVOiKaZPOAFVKk9b6vTZo0qO8T9/WDMet5fGCg/nHmq9bOiQJY6Z9OGY2Th5elMpbQvG4HbsXZhf4ZY5fabVWVSt35UXiOuKE0bOsyYacxT6ULbh9hUuH8y71BomKzD5ZEGbx8Hps6zytCqfk6aK4T/28Xs9fHae9duyBCdOL513hoLLIXHEIEdls0ZVUsY1sg8VH0uT7IlYP67QFhzP/RDivWQPvRhj5mdwBUVH7wtH2daXwO9xYRiQyqiMqXFTW2k6VIrGeWjFTv6xspPGCg/fP30yE+o8eRMlneEkv3thM69u7mVacwXfO32qhWG0DST46RPrWNMe4fgpDXztExPzwttIPMqO/gz/9eRGemNpPjuvmU8dOjLvcGlv6I9n+Mfbbfxl+Q7qAm5+eOY0JjUEShztfCCoGejdCE9/B6IdcOjFgnn4yhcM7zH7SAfSa79B2fA0euM09BN/gqN2/N4TloFYKsvK7QP88tmNqLrO1z85iaMn1uUvZWWjvRid7+MK1ICuoUou9JpxuDyi3fqSfTy7/VmiqRSKDCeNO8GijqsNtMHzP0Rpfxt17LFIi76LUlnqvXAo9EcG2NidZuWOKAZwSL2HeWMqqKkQTL8rnOLJ9zt44M0dVHqdfOeUKYyv9+edIg0n4uks77WF+fnTG0hmNK5cNIFFh9RTVaZZmK5IiruXbuPZNZ2MrfPzH6dPZUyVF6c502nhNnj+xyhtb6G2LEQ6/gcoFaWM/gMhOQDrn4DXbxUq3yf+FzTNFoJ3hO+at1sHqHdniGQdjKmvZFSNF7dz+Gfh/kiMlzZ0c+fSNrwuhRtPmsC05gCVZSwSQRy73fLsBpZvCzG3pYJvnzKFkVX7vtD6SJon2Z/YV+ahqxqJrIaS6UYyNDRXHZpuUGE2XF8kRldMY/ErW5EliSuOG099wEF1oMAA0pkMffEkDkWioWKQPnmkm5AqEdFdeKUs9R7Ab10pRlNZklmNCo/TommlqipScoBILIYqOah06qiOAL7K8idoVdMZSGRwKnLZg3yfkBgAPQveGlD230DrjqTY1RdCSUfISG5amhtoqChfq6gc9MXTYFBi40iLdCD3bUZ66w8Yvjo46htovmYcpgCyO5LktS19LHmvg+kjKvjs4S0Fl6UmkpF+JDWJ4fLjDZSvsx9NpemOZLnzlS2kVZ3LjhlPc5WLGn8h/3RWozuaRpYlRlbt3zrZE7KaRk80w/rOgqJBQ9CT92JYDmLJNANJFbciWbQSc8gk+jAyMWRXEKdvaMH7B4auC4dNkgx+6xgyIu2wcznS6ocwaifC4ZcjVe4nxlUGNFWjLxoTDr6qyrurVYxwMkUsncXvdlDl/WB9wmYe++qGNhzn5N8ss5xv3nvpXI6dIraNG7uinPabV/NniC5F5tlrj2FsXRmrgkQ/vPT/YPniQtiYI+HTfyi4sNwD1HAHjsVHCyG+icz5f8U17dQ9pProI5bK8vt/beH2Ir8Zs0ZVsvii+WW7c/13YGx+Aen+TxcCvNUYVyxDqhxJIpXmjldb+e0Lm/PR00dUcPfF80rUVj8ItvfFOelXr+TlYrIES75+NNNGHPgLbe2hJMff8nLeT4skwRPfOIZpzfs+2R1M0NJx5Lf/jPRMkSXnmvEYlzyOVMY4/bjAdga1j1i1o69EMHb7qzsIR2PounBgVCx8ymg6D77VNjiboZFNwNv3WsO2vyYEeGVA61htYRwArjd+Qzx04M0qDCciKbXEcdR7bWGLc6nhgh7pRHr9VmtgckC0GxBKqtz3eqlTo8R+0sx5bFW7xfmYbphaSdkDrz3zzJpOSxsYBvzhla1k1OHXShpOyIlepOIFHgjtukFjz8YHx8eSefhdpYJMv0tGkSVkWS74aShC0FPmNt0wSoXKkiS2zWVAcg6xknX5h3aU9DHDUO5kPwwHOIakYAxV70X3C3xDSCOHctjzQTBUfwu4HAfF4BvqeKrC4+Aj5pdoCEjCn/tgOA6eexIfdRwM/Xe/Y3JTBZOKzBS4FJkbTphIwC8ETp9b0JI3LwJQF3BxztzyNCHwVsOi71rDZn+uRLtrd5DqJ1k1dxQn2U/8AH/Ffj4LPshQ63dxw8mHWMJOn9mExzn8XVAJ1sOi71gnjrrJ0DwbgPqgh+tPstpDO2V6Ix7H/plBT5nRRGORlk/Q7eCLR43F9SEIbveGRZPraS46Ngy4HXzlmPE4hvDS91GCVN2CcfwPLbfVjXHHWh2t2fi38LGUeYC41btiWy9dkRQnTB9Bnd+JxyMGSSKdJpTUeX5tJ4os84kpDTQEnDjK8J4GiMt0sU7Y/AKMnC8mosoymQ+QjXShb38dIrtQppxG0lVLMPjx79S90TRd0RT/Wt/NzFFVTGkM0vghyDsAsvEBHOkQrHscAg0w9likojbrjcTpiWu8uL6Lac0VTG0O7rebyJqm0R3N8NKGblKqzknTGqn1O/EMsUM+EOiOpFi2pY9EWuWTUxuoD7g/8swDQIt2IicHYMOTUD8VRsxG2l+aXh8R2ALzD8A8bNiwYeP/OmyBuQ0bNmzYOKhgMw8bNmzYsLHPsJmHDRs2bNjYZ9jMw4YNGzZs7DNs5mHDhg0bNvYZNvOwYcOGDRv7jI+Eqq4kST1AqZOO8lAH7MkmwXDGH8hv27TZtB1M8QczbXuLP5hp2xvGGIaxezeT/w4Mw/hY/4AVByr+QH7bps2m7WCKP5hp+6jTfqB+9rGVDRs2bNjYZ9jMw4YNGzZs7DP+LzCPxQcw/kB+e2/xNm0fLN6m7YPFH8y07S3+YKbtgOEjITC3YcOGDRsHF/4v7Dxs2LBhw8b+xoGW2A+hWVALvGv+OoFdRc+uoveaEOprhvlbBhxuvt9ZlHY1sBNwAYuAJWb6m4AbBn27FaEWpxflkwE2Au8DZwEvAbcBa4D3TLoWAN8EfIPyuxoIATuAN4C7gGlm3M+Av5rfWgs8CVwGLAE+ZYa/C7QBzwAPAlvNMr9rvnMpMAJYCESADUA3cJP5DcUMuxV4zcxzu1kWA/hrEa0OIG7SmwCCwHzzvZj5N1xE1yrgbeBrJs0jgL+b5agyabsVOMes64uBbWY+lxV99yzgO0AKGFEUfilwa9FzxmyPuiH6jGG2y8tmW3eYdKZMWrPAz026rjLL9dvd9D/NTGMAafPvWmCxGV8PbDLr6JfAxUPk8WfgM0XPnzLz+RKw1AybA5xW9L5u/t4HHgeuNevs+4i+FkH0ZQO4D7O/me2WGwNxs952IvrJEjPPnUC7Wa7ngduBI4H/NPM6wczvfAr9c5HZ3rmx12/Wa65uXhvUd3qAdxBj5R1gAnAFhf6S+42lMMbjZh13mOXLIvp4q5n2eeAGIDaofnNlv8Kso1yfa0KMqS0mjRFgHaKfHmmmvQuYhhhrBjDFbItLgc8V98mi7202302avxQQNfPvwjpHfQvwFKV9BggWPedovhRrf/8HcDliPH0euA5YVtQWR36Ic3AVcNVe3zvQzGIvhbiJQRO8GS4Br5sNoZkN+wBwjNnZbjJ/3zI76/tFjVAO88iYHfQmxHnjy8BUxIB8G8E03GaaRsTE2cqgiQ0xqXcB5wxBv2bSHyuaTH5QRF8u/CVgAPhbUfoxwNfNuPkIBvEWgomFMCcA891LMSdhigahWWftQMB8PhUxOYfM8n1h0HttZt0U53EqYsAsGaKNLkUwjz9jTqTAcrMsy4d4XwPm74ZuBcHASuo4Vy7ERJc12+JU8zurzPjtZtqjc32hKK0yOK+iPB8CXkUw/9fN8AvNcnTtod/my2w+5/L5Z1GZisv3Z5P2tPl8D4JpHI3o5x4zvA4x4b5jlqnOLKsBqIh++xLQh2A0Swa1lwPBMJ5l6L7/l6K2WlTcroPaUTXp9Q/qBz3As0PV5xB15DBpXWx++56i+piDGMs/Zmjmke8HufajaE4oar9cPi+HnYMAAA2hSURBVCcjxrBSlMcSsy5vMtviEYbox+a7KbOO/6OoP+aY7E3At4vebQOqypjbXsLa378E/D5XN+XMg8P1o2jO3ON7HxZBH7AQlkoDLkEM3M2ICS2GWNkkzM78d3MAGWZ4BjGJGIN+uVVpbrWXNjuSBvyp6D216P21FBiVQWHXE6OwMi/O30CsTjQzfxWxK9CGeDdUREtxem2IMMMsb2JQWK4cuees+b1cPWiDvmEM8Y3ivHb37uC6zP0yQ7xf/IvvJnxwGWKIFV1xWG7HUw6dqkmnatZR7xDfHKrMxb+1RXmli76V3Q39ufotrp/tiFXq4L5kDApPUVgp5/JLmnnFKUzUu6O1nN9g2ga32+CwVFGaJIIpGObf9KB3dQRz1Qalby8K08ywfyHGhDEobnD5coxQHRSe3U051EF1WkzHUP12T/04WvRddYh3VcRuIltGfm8Peo4X5VHcd4vnrMHlz81fue/l3g0j+nasKJ8dwEWIOXI1MMGcN+uBhxELzLeAo4rm1z8iGNlW4Btm+F/N+nwXuHl38/NHRuYhSdIMxPb/SOC3iIpzIRougVjJRCg05ONmmAvR6TH/rqfQCG+bf9Nm3gCVRZ9VEA0oIXYYsaK4SgoDapcZFgWWIrb4UJgo2s3nGpPWiPmcmyQkxOp0J4UOnItXzf8HgDeLyvY0hQGSMN9xmmEgOqoH0alyA1RCdKxW851+xK4l56vzL1iRC9fMv8X9pXhCBDEBSuZ3VpthetFfb9EzJs1pM02uPiTzvZT5nEWseONm3EBR+j5Eh5dMGt4zw9OINs2aeVVTGMghM13u+8uKyvJcUd6jzb8K8EQRnbmJLzeYbzf/vw9Rl1nEwuZNoAXRR1NmPTUj2hPgh+a7OxC7iIaib8fMOBnBgFREP9aBO7FiC0Mj1zZqUVjxgmadWaalFI69ckdIANPN8kiI/lFlhm8reqcH0Q9B1NfXzO9FEe3kpNDeMfN5FqJtQBzP5fqzA3F0udaM24VozyziyK64XG1Fz0vNv2pRfJpCP8m5Bs31218UvZcbk7nwHF23FIW/iDhOKu63MnA8oh/l+n8WK540/85GjJ1O873cIrPbTLsLUccKom/mFpgpxE5JRkzgiaJyrzHzeNGky4/oo8cCI4FTDcM4HHE893UzzW+AXxmGcRhwrhmXwxTEzuxw4EeSJDkRbbHFMIw5hmF8i93hQO8uyt15II6RcmeLbYjOXbwy6jMbILdqCpmV3EuBqeRWcwaig0XMRsqa39DNCs8xpD4Kq644Qu6QG4QvU1gh5PIMA79CnOHmOlUMMejUIjpeMeMTRd8q3r0M3m3kVp87i8KLV0dxs05yu5HcKjnK7lfMOcYTwjoIcnGDV8c5ene3giumuWeIb+WYWPFqq5jW4km5+DsRCqvdbNF3uhHMI2v+lprhyxETQ+691KD8c+GJonbTi9rMQEzaxWXUit4r3rHk2nMA60o61w5/M8uZMvtXjsYOCqvx9YiJYPDqVjfzH1zng+t0d+HdWOsgURSX2wkPbvdcvzoLMb50rDvGZFGanwMritIW79DiWBntBsSC5RnEKjc1RNl2UNhxhhDjYfDKX6MgN0sDN1MY00Pt0P9/e+caY1dVBeBvtUOntA1IbMVYIyBGmghJWyOxEJPSQEKsv0zjk5hoxKQxGKP8UCFNAz+MgghBE6PxkVBMfD8CYnxk+EEUEQJUWlpK0tbKUNpKp0xLZzq9s/yx1nLvOb2X3tuZdmYy60tO5p599tnPtffae+199hxv/L24ym8tE8PV/d2U0f1OYEt1/wfK2ttYVRbNerjf3aL9Pe5/v+vPn/Q87qcokUP+O9Zov+l+/+npf9HzvMP9PVWVzQ7K2t49LmvrgN/67wNMXHd6CVvT3AzcVvW3zwNvp0uz1ayZeWAa+kequhIzX+3ANHM/NqJfBFyAFdRrmJC2MAUgmIAdxabOBzFhXoyN+qSK41YPNwT3Ln+2jzLDUGxEO4512jHKGgZWeXpq6nIWTj2nZhc2+ngJE4pXPLyfeFzz/O81mIBsrdyG/P0WxXwi/n7Eu48yOtpOGaUPeLzzKaNz9fijIUAZbb5c5b+FjWCUMsXG3c+ndNC1cpjHxNlMKINXKY36MLboOoop9QsoI+zDlBHi3VVYdQO+Cqv78Lfb493mblGH+7CyDhPmt6t8RbhQZip4uYSiP+Dl1QLudffH/Fl0Nscx2VggInswGzxYfR3FlNR5WPkHv/f3RzAZXeDlcowyco7yq9NZo/5ePI8Z8X8p9XCI0qkewso7RtiXYTNroczY1NMTs8SPe1oj7XdgMrAXU5oDlPp9p4e12q+YIW+lyNgm4Hee94OYDIENyE5iZR8z8shjzIgjDaNY2Q75fch8xHOR3/cxsexOUmRwHBuoArzL8xlc4nE+TtlgUg8i4nrFw1/g713RSHfMiCINJxrPw30ck2WwOhnDZrAtbL2rBaiqrvB+cZBSnuNMnHmt8ZnESlVdrqpRltG28fD66JLZpDz+AnxERJZiI7XllFHeq9j0/xg2hY6KU2wHRgvTtMcxoRRM+49iu0IQkc2YEH2ZidPUOFSsRZnKRyM5z5+HOUsxwYrGEZ34xf6838NZ5fdhFnu3+1+KCWiYCcA6EfFrpb9znGKGWgJcie1gOb96L2z9wsSOAS8LsKnqDyr35e5/EUXh4vnB48CfzcOEPsp6nz8bp5iTFrq/PspoL/IUo85o+NGoFmJ1swAQEXkPRXn0UxpWdMRBmO6e8HRFmpdRyilGn2AK/oSnb6FfUc6L3M8otkvtBb8f8nQtxsp6v6f/aqxjuxor210ezocoCv197g7WOc7DBi4XYhsdgj0UBbzHy6rP78O0Ok4x19VKvn62uOG20H/HTqGLqveWYYo5lPvrFNPhLR5WlHt0LvOxeonO52tY/fZhm0vCdImHuxTbMPB3SicLZdCxHpP9BViZDXkcUS9vwQaHI7RnHKvzSyjy3e95XOFhbHD3FmWgBNauI+/bsD4D4K/uHm3nRY/j2ip8MNmKvnQeE9s3FGW1wcO6AqvPfko7irWufqydrcLKdhVlwNNPkaFHKYM+REQoMt/kT9jOT9xvs+00GaaUYWem2zTV44L5J7Bp11asww4zSNgK72HiFPcwNvVsZ3Y4QBH+8H8Iq+C4H6OYQLZ5BTQXnscwm2M9Dd5GGeWdxKbkSllcby4EHmncR1iHKQtmzUXeeqEtrtcqP3+u/IUpT6u/9dQ+psXhdoJittAurhD8N/ITi5q1eWG4TRyHsVFyuJ/AZnfN8GIWEjObzzTS01xgjCuUaiyqR12MNN6v62a4w7OmTNWyN0ZR8pH/vRTzXXP9pF2Z7aSs0TWv7Y37dgvfzTR1qrvm4u/TFNt6fY1W9TzoeYg1jroMht1vcxH9Psx0dczTO9wmjkjDbk5tJ3uxLeJR5nspZqta9jtd0dZDvmvTb10Xtbl4JxNNmu3aYr1ZpJaHfzfCr/3U7XGn+z1IMYVHmAcwGQyz9h89H5+t0vEsJg9DwLe8r1xL2bm5FDO5b3V/3+vQvz4HXOq/f+r3HRfMp11B5KVQtssKtnXxlurZ+4GB6U5jI73zKdtHL6eYV5Zgo6M+7FuYe4Fnusz7mzFl/tbJ+DuDMp+S8Hotq0mGuYf2W5aXVL+/Atw3RXnox7ePAmuadYqZeu+cKeXTJo7rcfv/FId7Whk6i3WyAXhgqvPUy9W1fSs5q2wUkU9ijfRJ3JQkIrdhHzN9bBrT1o5FwIDvzBBgo6qeEJHPAzdj9tkRzHx182nCekhE3oQpnztVdf8k/XXLVIfXibZldZbiWi8iX8WU917s+4Wp4B3Az0UkTCz/r1MR+Q3W6a87w7DPZflMNd3I0JTXiYjcj31b88HJhjWpdLgWS5IkSZKumU0L5kmSJMkMIZVHkiRJ0jOpPJIkSZKeSeWRzDlE5G89+l8rIg+dYVxfFJFFp/eZJLOLVB7JnENVrzmH0cXx4acgIvPbuSfJbCCVRzLnEJGj/netiDwqIr8UkR0i8qB/qYuI3OhujwEfrt7dLCK3VvfPicilIrJYRB4WkWfd7aMi8gXsiPgBERmIuEXkDhH5B3C7b3WNsG4QkV+fm1JIksmR33kkc51V2Cmyg9gpu9eKSHxrsw778v5nXYRzIzCoqusBRORCVT0iIl8CrlPVOM9sMXbo3CZXVM+LyDJVPQh8GvuXAEky48mZRzLXeUJV/6Oq45T/dLcC2K2qu9Q+hNrSRTj/Aq4XkW+IyAdU9UgHfy3sfyvgYT8A3OQfm60BHplcdpLk3JDKI5nrdDpVtNPXsyeZ2G4WAqjqC8B7MSXydRHZ1OH9EVVtVfc/Bm7CTm/9haqebP9akswsUnkkyansAC4Tkcv9vj6Wew92rDgisho7igUReRvwuqpuwY6LX+3+3/CEUlUdxExmt2NH8CfJrCDXPJKkgaqOiMjngIdF5BB2yu+V/vhXwKdE5BnsH/XEce1XAXeJSJxSu9Hdvw88IiIvq+p1HaJ8EFimqts7PE+SGUeebZUk04yIfAd4WlV/ON1pSZJuSeWRJNOIiDyF/T+KG1R19HT+k2SmkMojSZIk6ZlcME+SJEl6JpVHkiRJ0jOpPJIkSZKeSeWRJEmS9EwqjyRJkqRnUnkkSZIkPfM/f+RMZ1ubFXIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.scatterplot(x=\"industry\", y=\"age\",\n",
    "                     hue=\"sentiment\", data=sample)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This isn't informative but like... it kinda looks cool......."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# i should've done this before but whatever\n",
    "def polarity_score(df):\n",
    "    \"\"\"\n",
    "    Returns list of polarity scores for a column of text in a dataframe.\n",
    "    \"\"\"\n",
    "    pol_scores = []\n",
    "    for x in df.index:\n",
    "        score = sia.polarity_scores(df['text'][x])\n",
    "        pol_scores.append(score)\n",
    "    return pol_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample['polarity_score'] = polarity_score(sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9478"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.polarity_score[sample.index[0]]['compound']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Running VADER on all the blogs\n",
    "I was looking into some sentiment lexicons, specifically the [NRC Word-Emotion Association Lexicon](http://www.saifmohammad.com/WebPages/NRC-Emotion-Lexicon.htm), and it looks very rad, but I think I'm just going to stick with VADER because building my own model is intimidating. My next steps will be to run VADER on all the blogs, save the dataframe with the polarity information, and then do some stats stuff with it in R."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "blogdata['sentiment'] = summarize_sentiment(blogdata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the smart thing to do would be to fix my error from before and do this in one step instead of two\n",
    "# but i am not smart\n",
    "blogdata['tokens'] = blogdata.text.map(tokenizer.tokenize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "blogdata['tokens'] = blogdata.tokens.map(remove_stuff)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'm going to edit my polarity score function, because I really just want the `compound` score, and it's a pain to have a dictionary within a dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "def polarity_score(df):\n",
    "    \"\"\"\n",
    "    Returns list of compound polarity scores for a column of text in a dataframe.\n",
    "    \"\"\"\n",
    "    pol_scores = []\n",
    "    for x in df.index:\n",
    "        score = sia.polarity_scores(df['text'][x])\n",
    "        pol_scores.append(score['compound'])\n",
    "    return pol_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "blogdata['polarity_score'] = polarity_score(blogdata)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That took so long :'( I also tried installing something to my computer while this was running and the fan has literally never sounded so angry. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>tokens</th>\n",
       "      <th>polarity_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>14,May,2004</td>\n",
       "      <td>Info has been found (+/- 100 pages,...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>[info, found, 100, pages, 4, 5, mb, pdf, files...</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>13,May,2004</td>\n",
       "      <td>These are the team members:   Drewe...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>[team, members, drewes, van, der, laag, mail, ...</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>12,May,2004</td>\n",
       "      <td>In het kader van kernfusie op aarde...</td>\n",
       "      <td>negative</td>\n",
       "      <td>[het, kader, van, kernfusie, op, aarde, maak, ...</td>\n",
       "      <td>-0.8167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>12,May,2004</td>\n",
       "      <td>testing!!!  testing!!!</td>\n",
       "      <td>neutral</td>\n",
       "      <td>[testing, testing]</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>11,June,2004</td>\n",
       "      <td>Thanks to Yahoo!'s Toolbar I can ...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[thanks, yahoo, toolbar, capture, urls, popups...</td>\n",
       "      <td>0.8805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>10,June,2004</td>\n",
       "      <td>I had an interesting conversation...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[interesting, conversation, dad, morning, talk...</td>\n",
       "      <td>0.9847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>10,June,2004</td>\n",
       "      <td>Somehow Coca-Cola has a way of su...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[somehow, coca, cola, way, summing, things, we...</td>\n",
       "      <td>0.8929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>10,June,2004</td>\n",
       "      <td>If anything, Korea is a country o...</td>\n",
       "      <td>positive</td>\n",
       "      <td>[anything, korea, country, extremes, everythin...</td>\n",
       "      <td>0.7430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>10,June,2004</td>\n",
       "      <td>Take a read of this news article ...</td>\n",
       "      <td>negative</td>\n",
       "      <td>[take, read, news, article, joongang, ilbo, no...</td>\n",
       "      <td>-0.8248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3581210</td>\n",
       "      <td>male</td>\n",
       "      <td>33</td>\n",
       "      <td>InvestmentBanking</td>\n",
       "      <td>Aquarius</td>\n",
       "      <td>09,June,2004</td>\n",
       "      <td>I surf the English news sites a l...</td>\n",
       "      <td>negative</td>\n",
       "      <td>[surf, english, news, sites, lot, looking, tid...</td>\n",
       "      <td>-0.5588</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        id gender  age           industry      sign          date  \\\n",
       "0  2059027   male   15            Student       Leo   14,May,2004   \n",
       "1  2059027   male   15            Student       Leo   13,May,2004   \n",
       "2  2059027   male   15            Student       Leo   12,May,2004   \n",
       "3  2059027   male   15            Student       Leo   12,May,2004   \n",
       "4  3581210   male   33  InvestmentBanking  Aquarius  11,June,2004   \n",
       "5  3581210   male   33  InvestmentBanking  Aquarius  10,June,2004   \n",
       "6  3581210   male   33  InvestmentBanking  Aquarius  10,June,2004   \n",
       "7  3581210   male   33  InvestmentBanking  Aquarius  10,June,2004   \n",
       "8  3581210   male   33  InvestmentBanking  Aquarius  10,June,2004   \n",
       "9  3581210   male   33  InvestmentBanking  Aquarius  09,June,2004   \n",
       "\n",
       "                                                text sentiment  \\\n",
       "0             Info has been found (+/- 100 pages,...   neutral   \n",
       "1             These are the team members:   Drewe...   neutral   \n",
       "2             In het kader van kernfusie op aarde...  negative   \n",
       "3                   testing!!!  testing!!!             neutral   \n",
       "4               Thanks to Yahoo!'s Toolbar I can ...  positive   \n",
       "5               I had an interesting conversation...  positive   \n",
       "6               Somehow Coca-Cola has a way of su...  positive   \n",
       "7               If anything, Korea is a country o...  positive   \n",
       "8               Take a read of this news article ...  negative   \n",
       "9               I surf the English news sites a l...  negative   \n",
       "\n",
       "                                              tokens  polarity_score  \n",
       "0  [info, found, 100, pages, 4, 5, mb, pdf, files...          0.0000  \n",
       "1  [team, members, drewes, van, der, laag, mail, ...          0.0000  \n",
       "2  [het, kader, van, kernfusie, op, aarde, maak, ...         -0.8167  \n",
       "3                                 [testing, testing]          0.0000  \n",
       "4  [thanks, yahoo, toolbar, capture, urls, popups...          0.8805  \n",
       "5  [interesting, conversation, dad, morning, talk...          0.9847  \n",
       "6  [somehow, coca, cola, way, summing, things, we...          0.8929  \n",
       "7  [anything, korea, country, extremes, everythin...          0.7430  \n",
       "8  [take, read, news, article, joongang, ilbo, no...         -0.8248  \n",
       "9  [surf, english, news, sites, lot, looking, tid...         -0.5588  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blogdata.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "blogdata.to_csv(dir + 'blog_sentiment_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now let's head over to R\n",
    "\n",
    "I'm sure there are ways to do fun stats stuff in Python, but I don't know how. I created __[Progress Report #3.5](https://github.com/Data-Science-for-Linguists-2019/Blog-Sentiment-Analysis/blob/master/progress_report_part3b.ipynb)__, a Jupyter notebook with an R kernel. Head over there for the second half of Progress Report #3. \n",
    "\n",
    "*Please ignore everything below this - I'm leaving it in for my own reference.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "blog_sent = pd.read_csv(dir + 'blog_sentiment_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>industry</th>\n",
       "      <th>sign</th>\n",
       "      <th>date</th>\n",
       "      <th>text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>tokens</th>\n",
       "      <th>polarity_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2059027</td>\n",
       "      <td>male</td>\n",
       "      <td>15</td>\n",
       "      <td>Student</td>\n",
       "      <td>Leo</td>\n",
       "      <td>14,May,2004</td>\n",
       "      <td>Info has been found (+/- 100 pages,...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>['info', 'found', '100', 'pages', '4', '5', 'm...</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0       id gender  age industry sign         date  \\\n",
       "0           0  2059027   male   15  Student  Leo  14,May,2004   \n",
       "\n",
       "                                                text sentiment  \\\n",
       "0             Info has been found (+/- 100 pages,...   neutral   \n",
       "\n",
       "                                              tokens  polarity_score  \n",
       "0  ['info', 'found', '100', 'pages', '4', '5', 'm...             0.0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blog_sent.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "19320"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(blog_sent.id.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2059027"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blog_sent.id.unique()[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.0000\n",
       "1    0.0000\n",
       "2   -0.8167\n",
       "3    0.0000\n",
       "Name: polarity_score, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blog_sent[blog_sent.id == 2059027].polarity_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.204175"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "blog_sent[blog_sent.id == 2059027].polarity_score.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "sent_per_blogger = blog_sent.drop_duplicates(\"id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for blogger in blog_sent.id.unique():\n",
    "    blog_sent_mean['id'] =  \n",
    "    sent_per_blogger[] = blog_sent[blog_sent.id == blogger].polarity_score.mean()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}