oddysey_project_deep_learning_w3.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "8171ea9f",
   "metadata": {},
   "source": [
    "# Topic Classification using Deep Learning Methods\n",
    "by [CSpanias](https://cspanias.github.io/aboutme/), 3rd Week's Project for [Solving Business Problems with NLP](https://omdena.com/course/solving-business-problems-with-nlp/) by Omdena, 03/2022"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed7390fa",
   "metadata": {},
   "source": [
    "# CONTENT\n",
    "1. [Text Pre-Processing](#textpre)\n",
    "1. [Topic Classification with Deep Learning](#nns)\n",
    "    1. [Word2Vec](#word2vec)\n",
    "    1. [GloVe](#glove)\n",
    "    1. [FastText](#fastText)\n",
    "    1. [Affinity Propagation](#AffinityPropagation)\n",
    "    1. [CorEx](#corex)\n",
    "5. [Conclusions](#conc)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0051f287",
   "metadata": {},
   "source": [
    "After attempting __topic classification__ using __LDA__, __NMF__, and __SVD__ on [week's 2 project](https://github.com/CSpanias/nlp_resources/blob/main/nlp_omdena/w2/oddysey_project.ipynb), we will now use __Deep Learning__ methods for the same purpose and __compare the results__.\n",
    "\n",
    "Using `LDA` I could infer __some sort of a relationships with some of the topics only from the LDA model__. The results from the other two models include a lot of __overlapping words__, so I couldn't relate any particular characteristic to any of the 5 topics.\n",
    "\n",
    "1. __Topic 0__ seems to be centered around __Penelope__ (the wife of Odysseus) as it contains words such as \"wife\", \"bed\", \"lay\", \"woman\", \"mother\", and \"men\", among others.\n",
    "2. __Topic 1__ hints a relation with __Telemachus__ (the son of Odysseus and Penelope). The words \"away\", \"father\", \"penelope\", and \"kill\" could be related with him travelling away to find his father and being involved in the Trojan war.\n",
    "3. __Topic 4__ must be related with __Odysseus__ as it has words like \"Ulysseus\", \"return\", \"old\", \"home\", \"son\", \"ship\", and \"sea\" which hint towards the return of him back to Ithaca by sea and the return to his old home.\n",
    "\n",
    "For __Topics 2 and 3__ I couldn't infer any kind of relationship.\n",
    "\n",
    "I will experiment with the same text, __Homer's Odyssey__, for the ease of comparison.\n",
    "\n",
    "You can find the __Plain Text UTF-8__ version of the text on the [Project Gutenberg website](https://www.gutenberg.org/ebooks/1727).\n",
    "\n",
    "First, we will __import the text__, we will make sure that is of __class string__, and we will __take a look at the text's first few characters__ to get a feel of it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ae68c992",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Variable 'raw' is of type: <class 'str'>\n",
      "\n",
      "First 1000 characters:\n",
      "\n",
      "The Project Gutenberg eBook of The Odyssey, by Homer\n",
      "\n",
      "This eBook is for the use of anyone anywhere in the United States and\n",
      "most other parts of the world at no cost and with almost no restrictions\n",
      "whatsoever. You may copy it, give it away or re-use it under the terms\n",
      "of the Project Gutenberg License included with this eBook or online at\n",
      "www.gutenberg.org. If you are not located in the United States, you\n",
      "will have to check the laws of the country where you are located before\n",
      "using this eBook.\n",
      "\n",
      "Title: The Odyssey\n",
      "\n",
      "Author: Homer\n",
      "\n",
      "Translator: Samuel Butler\n",
      "\n",
      "Release Date: April, 1999 [eBook #1727]\n",
      "[Most recently updated: January 4, 2022]\n",
      "\n",
      "Language: English\n",
      "\n",
      "\n",
      "Produced by: Jim Tinsley and David Widger\n",
      "\n",
      "*** START OF THE PROJECT GUTENBERG EBOOK THE ODYSSEY ***\n",
      "\n",
      "[Illustration]\n",
      "\n",
      "\n",
      "\n",
      "\n",
      "The Odyssey\n",
      "\n",
      "by Homer\n",
      "\n",
      "rendered into English prose for the use of those who cannot read the\n",
      "original\n",
      "\n",
      "Contents\n",
      "\n",
      " PREFACE TO FIRST EDITION\n",
      " PREFACE TO SECOND EDITION\n",
      " THE ODYSSEY\n",
      " BOOK I.\n",
      " BOOK II.\n",
      " BOOK III.\n",
      " BOOK IV.\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# import text as string\n",
    "with open(r\"odyssey_homer.txt\", encoding=\"utf8\") as f:\n",
    "    raw = f.read()\n",
    "    \n",
    "# check dtype\n",
    "print(f\"Variable 'raw' is of type: {type(raw)}\\n\")\n",
    "\n",
    "# check text\n",
    "print(f\"First 1000 characters:\\n\\n{raw[:1000]}\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37cbf579",
   "metadata": {},
   "source": [
    "<a name=\"textpre\"></a>\n",
    "# 1. Text Pre-Processing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6f3fa56",
   "metadata": {},
   "source": [
    "We will first __sentence-tokenize__ text and then make a main function to perform all the __text preprocessing__ steps.\n",
    "\n",
    "_For a step-by-step explanation of text pre-processing go [here](https://github.com/CSpanias/nlp_resources/blob/main/dphi_nlp_bootcamp/01_Text_Wrangling.ipynb)._"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e30d862d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The text was tokenized in 3855 sents.\n",
      "\n",
      "First sentence: The Project Gutenberg eBook of The Odyssey, by Homer\n",
      "\n",
      "This eBook is for the use of anyone anywhere in the United States and\n",
      "most other parts of the world at no cost and with almost no restrictions\n",
      "whatsoever.\n",
      "\n",
      "Second sentence: You may copy it, give it away or re-use it under the terms\n",
      "of the Project Gutenberg License included with this eBook or online at\n",
      "www.gutenberg.org.\n"
     ]
    }
   ],
   "source": [
    "from nltk.tokenize import sent_tokenize\n",
    "\n",
    "# tokenize the text\n",
    "sents = sent_tokenize(raw)\n",
    "\n",
    "# check the number of sents\n",
    "print(f\"The text was tokenized in {len(sents)} sents.\\n\")\n",
    "\n",
    "# check the first 2 sents\n",
    "print(f\"First sentence: {sents[0]}\\n\")\n",
    "print(f\"Second sentence: {sents[1]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1ed94103",
   "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>Sentences</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Project Gutenberg eBook of The Odyssey, by Homer\\n\\nThis eBook is for the use of anyone anywhere in the United States and\\nmost other parts of the world at no cost and with almost no restricti...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>You may copy it, give it away or re-use it under the terms\\nof the Project Gutenberg License included with this eBook or online at\\nwww.gutenberg.org.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>If you are not located in the United States, you\\nwill have to check the laws of the country where you are located before\\nusing this eBook.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Title: The Odyssey\\n\\nAuthor: Homer\\n\\nTranslator: Samuel Butler\\n\\nRelease Date: April, 1999 [eBook #1727]\\n[Most recently updated: January 4, 2022]\\n\\nLanguage: English\\n\\n\\nProduced by: Jim Tin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BOOK II.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                 Sentences\n",
       "0  The Project Gutenberg eBook of The Odyssey, by Homer\\n\\nThis eBook is for the use of anyone anywhere in the United States and\\nmost other parts of the world at no cost and with almost no restricti...\n",
       "1                                                   You may copy it, give it away or re-use it under the terms\\nof the Project Gutenberg License included with this eBook or online at\\nwww.gutenberg.org.\n",
       "2                                                             If you are not located in the United States, you\\nwill have to check the laws of the country where you are located before\\nusing this eBook.\n",
       "3  Title: The Odyssey\\n\\nAuthor: Homer\\n\\nTranslator: Samuel Butler\\n\\nRelease Date: April, 1999 [eBook #1727]\\n[Most recently updated: January 4, 2022]\\n\\nLanguage: English\\n\\n\\nProduced by: Jim Tin...\n",
       "4                                                                                                                                                                                                 BOOK II."
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "pd.options.display.max_colwidth = 200\n",
    "\n",
    "# create a pandas DataFrame\n",
    "corpus_df = pd.DataFrame(\n",
    "    # col_name: value\n",
    "    {'Sentences': sents})\n",
    "\n",
    "# check DF\n",
    "corpus_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "69602135",
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import wordnet\n",
    "\n",
    "def pos_tag_wordnet(tagged_tokens):\n",
    "    \"\"\"Convert POS-tagged tokens to WordNet form tags.\"\"\"\n",
    "    tag_map = {'j': wordnet.ADJ, 'v': wordnet.VERB, 'n': wordnet.NOUN, 'r': wordnet.ADV}\n",
    "    # if a word does not belong to one of the 4 categories make it a NOUN\n",
    "    new_tagged_tokens = [(word, tag_map.get(tag[0].lower(), wordnet.NOUN)) for word, tag in tagged_tokens]\n",
    "    return new_tagged_tokens"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "eba3185a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import contractions\n",
    "\n",
    "from nltk import word_tokenize, pos_tag\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "\n",
    "def normalize_corpus(raw):\n",
    "    \n",
    "    # remove special chars\n",
    "    text = re.sub(r'[^a-zA-Z\\s]', ' ', raw)\n",
    "    # expand contracted words\n",
    "    text = [contractions.fix(w) for w in raw.split()]\n",
    "    # join text back\n",
    "    text = \" \".join(text)\n",
    "    # tokenize text\n",
    "    tokens = word_tokenize(text)\n",
    "    # lower-case words\n",
    "    tokens = [w.lower() for w in tokens]\n",
    "    # define stopwords list\n",
    "    sw = set(stopwords.words('english'))\n",
    "    # remove stopwords\n",
    "    tokens = [w for w in tokens if w not in sw]\n",
    "    # remove non-alphabetical chars\n",
    "    tokens = [w for w in tokens if w.isalpha()]\n",
    "    # POS-tag tokens\n",
    "    tagged_tokens = pos_tag(tokens)\n",
    "    # convert tags to WordNet form\n",
    "    wd_tagged_tokens = pos_tag_wordnet(tagged_tokens)\n",
    "    # instantiate lemmatizer\n",
    "    wnl = WordNetLemmatizer()\n",
    "    # lemmatize tokens\n",
    "    tokens = [wnl.lemmatize(w, tag) for w, tag in wd_tagged_tokens]\n",
    "    # join back text as a string\n",
    "    text = \" \".join(tokens)\n",
    "    \n",
    "    return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "e6b5ebe0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['project gutenberg ebook odyssey homer ebook use anyone anywhere united state part world cost almost restriction whatsoever',\n",
       "       'may copy give away term project gutenberg license include ebook online',\n",
       "       'locate united state check law country locate use ebook', ...,\n",
       "       'project ebooks often create several printed edition confirm protected copyright unless copyright notice include',\n",
       "       'thus necessarily keep ebooks compliance particular paper edition',\n",
       "       'people start website main pg search facility website include information project include make donation project gutenberg literary archive foundation help produce new ebooks subscribe email newsletter hear new ebooks'],\n",
       "      dtype='<U724')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# vectorize function\n",
    "normalize_corpus = np.vectorize(normalize_corpus)\n",
    "\n",
    "# apply the vectorized function\n",
    "norm_corpus = normalize_corpus(sents)\n",
    "\n",
    "# show normalized corpus\n",
    "norm_corpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "42002851",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['project', 'gutenberg', 'ebook', 'odyssey', 'homer', 'ebook', 'use', 'anyone', 'anywhere', 'united', 'state', 'part', 'world', 'cost', 'almost', 'restriction', 'whatsoever']\n"
     ]
    }
   ],
   "source": [
    "# tokenize corpus\n",
    "tokenized_corpus = [word_tokenize(sent) for sent in norm_corpus]\n",
    "\n",
    "# check first sent\n",
    "print(tokenized_corpus[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cfc51e77",
   "metadata": {},
   "source": [
    "<a name=\"nns\"></a>\n",
    "# 2. Topic Classification with Deep Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df09984c",
   "metadata": {},
   "source": [
    "We will try various deep learning models:\n",
    "1. Word2Vec\n",
    "2. GloVe\n",
    "3. FastText\n",
    "4. AffinationPropagation\n",
    "5. CorEx"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "868aa9cc",
   "metadata": {},
   "source": [
    "<a name=\"word2vec\"></a>\n",
    "## 2.1 Word2Vec"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bec499ff",
   "metadata": {},
   "source": [
    "In our workflow, we will tokenize our normalized corpus and then focus on the following parameters in the Word2Vec model to build it.\n",
    "\n",
    "* `vector_size` The word embedding dimensionality\n",
    "* `window` The context window size\n",
    "* `min_count` The minimum word count\n",
    "* `sample` The downsample setting for frequent words\n",
    "* `sg` Training model, 1 for skip-gram otherwise CBOW\n",
    "\n",
    "More info about `gensim.models.word2vec` [here](https://radimrehurek.com/gensim/models/word2vec.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "ea33fd71",
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models import word2vec\n",
    "\n",
    "# Set values for various parameters\n",
    "feature_size = 15    # Word vector dimensionality  \n",
    "window_context = 20  # Context window size                                                                                    \n",
    "min_word_count = 1   # Minimum word count                        \n",
    "sample = 1e-3        # Downsample setting for frequent words\n",
    "sg = 1               # skip-gram model\n",
    "sorted_vocab = 1     # sort by descending frequency\n",
    "compute_loss = True  # computes and stores loss values\n",
    "\n",
    "# instantiate and train word2vec model\n",
    "w2v_model = word2vec.Word2Vec(tokenized_corpus, vector_size=feature_size, \n",
    "                              window=window_context,\n",
    "                              min_count = min_word_count,\n",
    "                              sg=sg, sample=sample,\n",
    "                              epochs=5000,\n",
    "                              sorted_vocab=sorted_vocab,\n",
    "                              compute_loss=compute_loss)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71c856cb",
   "metadata": {},
   "source": [
    "We can __visualize the results with__ `matplotlib` using __t-distributed stochastic neighbor embedding__ (t-SNE), and see how the tokens are clustering.\n",
    "\n",
    "Click [here](https://www.youtube.com/watch?v=NEaUSP4YerM&t=79s) for a brief t-SNE introduction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "2124b120",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from sklearn.manifold import TSNE\n",
    "%matplotlib inline\n",
    "\n",
    "# extract labels\n",
    "words = w2v_model.wv.index_to_key\n",
    "wvs = w2v_model.wv[words]\n",
    "\n",
    "# instantiate TSNE\n",
    "tsne = TSNE(n_components=2, random_state=42, n_iter=5000, perplexity=5)\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "# fit TSNE\n",
    "T = tsne.fit_transform(wvs)\n",
    "labels = words\n",
    "\n",
    "# plot visualization\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.scatter(T[:, 0], T[:, 1], c='orange', edgecolors='r')\n",
    "for label, x, y in zip(labels, T[:, 0], T[:, 1]):\n",
    "    plt.annotate(label, xy=(x+1, y+1), xytext=(0, 0), textcoords='offset points')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9c6595b7",
   "metadata": {},
   "source": [
    "We can __visualize the vector reprentation__ of each word via a pandas DataFrame. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "e035b267",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>go</th>\n",
       "      <td>-0.046563</td>\n",
       "      <td>0.113128</td>\n",
       "      <td>0.095247</td>\n",
       "      <td>-0.220213</td>\n",
       "      <td>0.323735</td>\n",
       "      <td>-0.175819</td>\n",
       "      <td>-0.183081</td>\n",
       "      <td>-0.251394</td>\n",
       "      <td>-0.053384</td>\n",
       "      <td>-0.327375</td>\n",
       "      <td>0.799953</td>\n",
       "      <td>0.200113</td>\n",
       "      <td>0.010235</td>\n",
       "      <td>0.069672</td>\n",
       "      <td>-0.230568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>say</th>\n",
       "      <td>-0.188229</td>\n",
       "      <td>-0.059395</td>\n",
       "      <td>0.246310</td>\n",
       "      <td>-0.270955</td>\n",
       "      <td>0.544837</td>\n",
       "      <td>-0.417879</td>\n",
       "      <td>-0.111734</td>\n",
       "      <td>-0.252836</td>\n",
       "      <td>0.070907</td>\n",
       "      <td>-0.098581</td>\n",
       "      <td>0.488377</td>\n",
       "      <td>-0.089023</td>\n",
       "      <td>-0.175871</td>\n",
       "      <td>0.118060</td>\n",
       "      <td>-0.171658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ulysses</th>\n",
       "      <td>-0.059593</td>\n",
       "      <td>-0.007981</td>\n",
       "      <td>0.222293</td>\n",
       "      <td>-0.265131</td>\n",
       "      <td>0.430034</td>\n",
       "      <td>-0.507800</td>\n",
       "      <td>0.001793</td>\n",
       "      <td>-0.352427</td>\n",
       "      <td>-0.228065</td>\n",
       "      <td>-0.130676</td>\n",
       "      <td>0.363253</td>\n",
       "      <td>-0.132570</td>\n",
       "      <td>-0.122735</td>\n",
       "      <td>0.042868</td>\n",
       "      <td>-0.059485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>come</th>\n",
       "      <td>-0.126199</td>\n",
       "      <td>0.050187</td>\n",
       "      <td>0.246202</td>\n",
       "      <td>-0.221790</td>\n",
       "      <td>0.525650</td>\n",
       "      <td>-0.276142</td>\n",
       "      <td>-0.149167</td>\n",
       "      <td>-0.127144</td>\n",
       "      <td>-0.112732</td>\n",
       "      <td>-0.401638</td>\n",
       "      <td>0.352934</td>\n",
       "      <td>0.041375</td>\n",
       "      <td>-0.009032</td>\n",
       "      <td>0.120529</td>\n",
       "      <td>-0.071357</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>one</th>\n",
       "      <td>-0.007612</td>\n",
       "      <td>0.306861</td>\n",
       "      <td>0.246842</td>\n",
       "      <td>-0.123052</td>\n",
       "      <td>0.508135</td>\n",
       "      <td>-0.287940</td>\n",
       "      <td>0.034621</td>\n",
       "      <td>0.010167</td>\n",
       "      <td>-0.229990</td>\n",
       "      <td>-0.262424</td>\n",
       "      <td>0.189677</td>\n",
       "      <td>-0.180511</td>\n",
       "      <td>-0.203855</td>\n",
       "      <td>0.318198</td>\n",
       "      <td>0.056860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>murderess</th>\n",
       "      <td>0.344259</td>\n",
       "      <td>-0.705999</td>\n",
       "      <td>3.223940</td>\n",
       "      <td>-1.727517</td>\n",
       "      <td>0.698907</td>\n",
       "      <td>2.139905</td>\n",
       "      <td>-3.864358</td>\n",
       "      <td>1.507824</td>\n",
       "      <td>-0.927818</td>\n",
       "      <td>-1.379901</td>\n",
       "      <td>0.491853</td>\n",
       "      <td>-0.204213</td>\n",
       "      <td>-4.311629</td>\n",
       "      <td>0.042543</td>\n",
       "      <td>-3.628473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cassandra</th>\n",
       "      <td>0.082976</td>\n",
       "      <td>-3.871452</td>\n",
       "      <td>2.122782</td>\n",
       "      <td>-2.230313</td>\n",
       "      <td>2.002559</td>\n",
       "      <td>-0.680998</td>\n",
       "      <td>-4.929260</td>\n",
       "      <td>0.935890</td>\n",
       "      <td>-1.525238</td>\n",
       "      <td>-2.046451</td>\n",
       "      <td>2.080466</td>\n",
       "      <td>0.167836</td>\n",
       "      <td>-1.040072</td>\n",
       "      <td>-2.692603</td>\n",
       "      <td>-0.301813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nobleman</th>\n",
       "      <td>0.928609</td>\n",
       "      <td>0.652556</td>\n",
       "      <td>-2.598921</td>\n",
       "      <td>-6.757182</td>\n",
       "      <td>-0.276411</td>\n",
       "      <td>3.097641</td>\n",
       "      <td>-1.418341</td>\n",
       "      <td>0.642705</td>\n",
       "      <td>-3.842592</td>\n",
       "      <td>2.460270</td>\n",
       "      <td>3.250930</td>\n",
       "      <td>1.906017</td>\n",
       "      <td>-2.563366</td>\n",
       "      <td>0.701659</td>\n",
       "      <td>-1.294039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gorgeous</th>\n",
       "      <td>0.778151</td>\n",
       "      <td>0.413311</td>\n",
       "      <td>-2.083295</td>\n",
       "      <td>-6.204207</td>\n",
       "      <td>-0.362479</td>\n",
       "      <td>2.591512</td>\n",
       "      <td>-1.123152</td>\n",
       "      <td>0.334635</td>\n",
       "      <td>-3.302349</td>\n",
       "      <td>1.961974</td>\n",
       "      <td>2.820362</td>\n",
       "      <td>1.784599</td>\n",
       "      <td>-2.230946</td>\n",
       "      <td>0.477338</td>\n",
       "      <td>-1.017646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>afflicted</th>\n",
       "      <td>-0.512977</td>\n",
       "      <td>0.094830</td>\n",
       "      <td>0.639893</td>\n",
       "      <td>-1.455328</td>\n",
       "      <td>0.686708</td>\n",
       "      <td>-0.782136</td>\n",
       "      <td>-3.302348</td>\n",
       "      <td>3.395071</td>\n",
       "      <td>-0.406889</td>\n",
       "      <td>-0.359497</td>\n",
       "      <td>-0.653650</td>\n",
       "      <td>-1.261872</td>\n",
       "      <td>-5.853262</td>\n",
       "      <td>1.171912</td>\n",
       "      <td>-1.508409</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5902 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 0         1         2         3         4         5   \\\n",
       "go        -0.046563  0.113128  0.095247 -0.220213  0.323735 -0.175819   \n",
       "say       -0.188229 -0.059395  0.246310 -0.270955  0.544837 -0.417879   \n",
       "ulysses   -0.059593 -0.007981  0.222293 -0.265131  0.430034 -0.507800   \n",
       "come      -0.126199  0.050187  0.246202 -0.221790  0.525650 -0.276142   \n",
       "one       -0.007612  0.306861  0.246842 -0.123052  0.508135 -0.287940   \n",
       "...             ...       ...       ...       ...       ...       ...   \n",
       "murderess  0.344259 -0.705999  3.223940 -1.727517  0.698907  2.139905   \n",
       "cassandra  0.082976 -3.871452  2.122782 -2.230313  2.002559 -0.680998   \n",
       "nobleman   0.928609  0.652556 -2.598921 -6.757182 -0.276411  3.097641   \n",
       "gorgeous   0.778151  0.413311 -2.083295 -6.204207 -0.362479  2.591512   \n",
       "afflicted -0.512977  0.094830  0.639893 -1.455328  0.686708 -0.782136   \n",
       "\n",
       "                 6         7         8         9         10        11  \\\n",
       "go        -0.183081 -0.251394 -0.053384 -0.327375  0.799953  0.200113   \n",
       "say       -0.111734 -0.252836  0.070907 -0.098581  0.488377 -0.089023   \n",
       "ulysses    0.001793 -0.352427 -0.228065 -0.130676  0.363253 -0.132570   \n",
       "come      -0.149167 -0.127144 -0.112732 -0.401638  0.352934  0.041375   \n",
       "one        0.034621  0.010167 -0.229990 -0.262424  0.189677 -0.180511   \n",
       "...             ...       ...       ...       ...       ...       ...   \n",
       "murderess -3.864358  1.507824 -0.927818 -1.379901  0.491853 -0.204213   \n",
       "cassandra -4.929260  0.935890 -1.525238 -2.046451  2.080466  0.167836   \n",
       "nobleman  -1.418341  0.642705 -3.842592  2.460270  3.250930  1.906017   \n",
       "gorgeous  -1.123152  0.334635 -3.302349  1.961974  2.820362  1.784599   \n",
       "afflicted -3.302348  3.395071 -0.406889 -0.359497 -0.653650 -1.261872   \n",
       "\n",
       "                 12        13        14  \n",
       "go         0.010235  0.069672 -0.230568  \n",
       "say       -0.175871  0.118060 -0.171658  \n",
       "ulysses   -0.122735  0.042868 -0.059485  \n",
       "come      -0.009032  0.120529 -0.071357  \n",
       "one       -0.203855  0.318198  0.056860  \n",
       "...             ...       ...       ...  \n",
       "murderess -4.311629  0.042543 -3.628473  \n",
       "cassandra -1.040072 -2.692603 -0.301813  \n",
       "nobleman  -2.563366  0.701659 -1.294039  \n",
       "gorgeous  -2.230946  0.477338 -1.017646  \n",
       "afflicted -5.853262  1.171912 -1.508409  \n",
       "\n",
       "[5902 rows x 15 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# create pandas DataFrame to visualize the vector reprentation of each word\n",
    "vec_df = pd.DataFrame(wvs, index=words)\n",
    "\n",
    "# show dataframe\n",
    "vec_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2cb96e28",
   "metadata": {},
   "source": [
    "In addition, we can __visualize the similarity between each word__."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "39fff9f2",
   "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>go</th>\n",
       "      <th>say</th>\n",
       "      <th>ulysses</th>\n",
       "      <th>come</th>\n",
       "      <th>one</th>\n",
       "      <th>take</th>\n",
       "      <th>make</th>\n",
       "      <th>house</th>\n",
       "      <th>would</th>\n",
       "      <th>tell</th>\n",
       "      <th>...</th>\n",
       "      <th>adore</th>\n",
       "      <th>champion</th>\n",
       "      <th>aeacus</th>\n",
       "      <th>march</th>\n",
       "      <th>joyful</th>\n",
       "      <th>murderess</th>\n",
       "      <th>cassandra</th>\n",
       "      <th>nobleman</th>\n",
       "      <th>gorgeous</th>\n",
       "      <th>afflicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>go</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.784994</td>\n",
       "      <td>0.707739</td>\n",
       "      <td>0.828199</td>\n",
       "      <td>0.518138</td>\n",
       "      <td>0.823628</td>\n",
       "      <td>0.663033</td>\n",
       "      <td>0.800744</td>\n",
       "      <td>0.494337</td>\n",
       "      <td>0.711212</td>\n",
       "      <td>...</td>\n",
       "      <td>0.034568</td>\n",
       "      <td>0.200013</td>\n",
       "      <td>0.236061</td>\n",
       "      <td>0.463483</td>\n",
       "      <td>0.106786</td>\n",
       "      <td>0.267245</td>\n",
       "      <td>0.435536</td>\n",
       "      <td>0.305865</td>\n",
       "      <td>0.317004</td>\n",
       "      <td>0.029740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>say</th>\n",
       "      <td>0.784994</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.904715</td>\n",
       "      <td>0.871355</td>\n",
       "      <td>0.686315</td>\n",
       "      <td>0.751056</td>\n",
       "      <td>0.639545</td>\n",
       "      <td>0.816861</td>\n",
       "      <td>0.644854</td>\n",
       "      <td>0.752593</td>\n",
       "      <td>...</td>\n",
       "      <td>0.207467</td>\n",
       "      <td>0.344065</td>\n",
       "      <td>0.420303</td>\n",
       "      <td>0.391003</td>\n",
       "      <td>0.299565</td>\n",
       "      <td>0.287494</td>\n",
       "      <td>0.459055</td>\n",
       "      <td>0.110327</td>\n",
       "      <td>0.124409</td>\n",
       "      <td>0.248187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ulysses</th>\n",
       "      <td>0.707739</td>\n",
       "      <td>0.904715</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.841191</td>\n",
       "      <td>0.744740</td>\n",
       "      <td>0.711776</td>\n",
       "      <td>0.741211</td>\n",
       "      <td>0.785834</td>\n",
       "      <td>0.654935</td>\n",
       "      <td>0.672801</td>\n",
       "      <td>...</td>\n",
       "      <td>0.073960</td>\n",
       "      <td>0.240226</td>\n",
       "      <td>0.341695</td>\n",
       "      <td>0.293718</td>\n",
       "      <td>0.306982</td>\n",
       "      <td>0.145173</td>\n",
       "      <td>0.403910</td>\n",
       "      <td>0.112693</td>\n",
       "      <td>0.133564</td>\n",
       "      <td>0.124105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>come</th>\n",
       "      <td>0.828199</td>\n",
       "      <td>0.871355</td>\n",
       "      <td>0.841191</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.795513</td>\n",
       "      <td>0.829481</td>\n",
       "      <td>0.677157</td>\n",
       "      <td>0.766050</td>\n",
       "      <td>0.763894</td>\n",
       "      <td>0.815549</td>\n",
       "      <td>...</td>\n",
       "      <td>0.260289</td>\n",
       "      <td>0.307498</td>\n",
       "      <td>0.324712</td>\n",
       "      <td>0.325134</td>\n",
       "      <td>0.106066</td>\n",
       "      <td>0.308956</td>\n",
       "      <td>0.523733</td>\n",
       "      <td>0.080812</td>\n",
       "      <td>0.096292</td>\n",
       "      <td>0.186922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>one</th>\n",
       "      <td>0.518138</td>\n",
       "      <td>0.686315</td>\n",
       "      <td>0.744740</td>\n",
       "      <td>0.795513</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.742679</td>\n",
       "      <td>0.734730</td>\n",
       "      <td>0.574217</td>\n",
       "      <td>0.880302</td>\n",
       "      <td>0.497193</td>\n",
       "      <td>...</td>\n",
       "      <td>0.284244</td>\n",
       "      <td>0.193289</td>\n",
       "      <td>0.208818</td>\n",
       "      <td>0.085373</td>\n",
       "      <td>0.269487</td>\n",
       "      <td>0.231896</td>\n",
       "      <td>0.158645</td>\n",
       "      <td>0.050920</td>\n",
       "      <td>0.048618</td>\n",
       "      <td>0.350760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>murderess</th>\n",
       "      <td>0.267245</td>\n",
       "      <td>0.287494</td>\n",
       "      <td>0.145173</td>\n",
       "      <td>0.308956</td>\n",
       "      <td>0.231896</td>\n",
       "      <td>0.207217</td>\n",
       "      <td>-0.004421</td>\n",
       "      <td>0.125063</td>\n",
       "      <td>0.344188</td>\n",
       "      <td>0.236123</td>\n",
       "      <td>...</td>\n",
       "      <td>0.659391</td>\n",
       "      <td>0.554918</td>\n",
       "      <td>0.568004</td>\n",
       "      <td>0.416076</td>\n",
       "      <td>0.446460</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.616768</td>\n",
       "      <td>0.376194</td>\n",
       "      <td>0.373142</td>\n",
       "      <td>0.764730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cassandra</th>\n",
       "      <td>0.435536</td>\n",
       "      <td>0.459055</td>\n",
       "      <td>0.403910</td>\n",
       "      <td>0.523733</td>\n",
       "      <td>0.158645</td>\n",
       "      <td>0.326019</td>\n",
       "      <td>0.165562</td>\n",
       "      <td>0.400755</td>\n",
       "      <td>0.202012</td>\n",
       "      <td>0.473930</td>\n",
       "      <td>...</td>\n",
       "      <td>0.417086</td>\n",
       "      <td>0.562209</td>\n",
       "      <td>0.605828</td>\n",
       "      <td>0.283106</td>\n",
       "      <td>0.288026</td>\n",
       "      <td>0.616768</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.235393</td>\n",
       "      <td>0.251796</td>\n",
       "      <td>0.411056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nobleman</th>\n",
       "      <td>0.305865</td>\n",
       "      <td>0.110327</td>\n",
       "      <td>0.112693</td>\n",
       "      <td>0.080812</td>\n",
       "      <td>0.050920</td>\n",
       "      <td>0.062895</td>\n",
       "      <td>0.173074</td>\n",
       "      <td>0.411694</td>\n",
       "      <td>0.156485</td>\n",
       "      <td>0.187258</td>\n",
       "      <td>...</td>\n",
       "      <td>0.485417</td>\n",
       "      <td>0.534199</td>\n",
       "      <td>0.546470</td>\n",
       "      <td>0.473069</td>\n",
       "      <td>0.628878</td>\n",
       "      <td>0.376194</td>\n",
       "      <td>0.235393</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998045</td>\n",
       "      <td>0.303783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gorgeous</th>\n",
       "      <td>0.317004</td>\n",
       "      <td>0.124409</td>\n",
       "      <td>0.133564</td>\n",
       "      <td>0.096292</td>\n",
       "      <td>0.048618</td>\n",
       "      <td>0.057805</td>\n",
       "      <td>0.170680</td>\n",
       "      <td>0.430620</td>\n",
       "      <td>0.158440</td>\n",
       "      <td>0.201928</td>\n",
       "      <td>...</td>\n",
       "      <td>0.469576</td>\n",
       "      <td>0.545374</td>\n",
       "      <td>0.561290</td>\n",
       "      <td>0.473539</td>\n",
       "      <td>0.630321</td>\n",
       "      <td>0.373142</td>\n",
       "      <td>0.251796</td>\n",
       "      <td>0.998045</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.288995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>afflicted</th>\n",
       "      <td>0.029740</td>\n",
       "      <td>0.248187</td>\n",
       "      <td>0.124105</td>\n",
       "      <td>0.186922</td>\n",
       "      <td>0.350760</td>\n",
       "      <td>0.240076</td>\n",
       "      <td>0.131633</td>\n",
       "      <td>0.116788</td>\n",
       "      <td>0.517387</td>\n",
       "      <td>0.085687</td>\n",
       "      <td>...</td>\n",
       "      <td>0.692487</td>\n",
       "      <td>0.517817</td>\n",
       "      <td>0.558451</td>\n",
       "      <td>0.319628</td>\n",
       "      <td>0.516759</td>\n",
       "      <td>0.764730</td>\n",
       "      <td>0.411056</td>\n",
       "      <td>0.303783</td>\n",
       "      <td>0.288995</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5902 rows × 5902 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 go       say   ulysses      come       one      take  \\\n",
       "go         1.000000  0.784994  0.707739  0.828199  0.518138  0.823628   \n",
       "say        0.784994  1.000000  0.904715  0.871355  0.686315  0.751056   \n",
       "ulysses    0.707739  0.904715  1.000000  0.841191  0.744740  0.711776   \n",
       "come       0.828199  0.871355  0.841191  1.000000  0.795513  0.829481   \n",
       "one        0.518138  0.686315  0.744740  0.795513  1.000000  0.742679   \n",
       "...             ...       ...       ...       ...       ...       ...   \n",
       "murderess  0.267245  0.287494  0.145173  0.308956  0.231896  0.207217   \n",
       "cassandra  0.435536  0.459055  0.403910  0.523733  0.158645  0.326019   \n",
       "nobleman   0.305865  0.110327  0.112693  0.080812  0.050920  0.062895   \n",
       "gorgeous   0.317004  0.124409  0.133564  0.096292  0.048618  0.057805   \n",
       "afflicted  0.029740  0.248187  0.124105  0.186922  0.350760  0.240076   \n",
       "\n",
       "               make     house     would      tell  ...     adore  champion  \\\n",
       "go         0.663033  0.800744  0.494337  0.711212  ...  0.034568  0.200013   \n",
       "say        0.639545  0.816861  0.644854  0.752593  ...  0.207467  0.344065   \n",
       "ulysses    0.741211  0.785834  0.654935  0.672801  ...  0.073960  0.240226   \n",
       "come       0.677157  0.766050  0.763894  0.815549  ...  0.260289  0.307498   \n",
       "one        0.734730  0.574217  0.880302  0.497193  ...  0.284244  0.193289   \n",
       "...             ...       ...       ...       ...  ...       ...       ...   \n",
       "murderess -0.004421  0.125063  0.344188  0.236123  ...  0.659391  0.554918   \n",
       "cassandra  0.165562  0.400755  0.202012  0.473930  ...  0.417086  0.562209   \n",
       "nobleman   0.173074  0.411694  0.156485  0.187258  ...  0.485417  0.534199   \n",
       "gorgeous   0.170680  0.430620  0.158440  0.201928  ...  0.469576  0.545374   \n",
       "afflicted  0.131633  0.116788  0.517387  0.085687  ...  0.692487  0.517817   \n",
       "\n",
       "             aeacus     march    joyful  murderess  cassandra  nobleman  \\\n",
       "go         0.236061  0.463483  0.106786   0.267245   0.435536  0.305865   \n",
       "say        0.420303  0.391003  0.299565   0.287494   0.459055  0.110327   \n",
       "ulysses    0.341695  0.293718  0.306982   0.145173   0.403910  0.112693   \n",
       "come       0.324712  0.325134  0.106066   0.308956   0.523733  0.080812   \n",
       "one        0.208818  0.085373  0.269487   0.231896   0.158645  0.050920   \n",
       "...             ...       ...       ...        ...        ...       ...   \n",
       "murderess  0.568004  0.416076  0.446460   1.000000   0.616768  0.376194   \n",
       "cassandra  0.605828  0.283106  0.288026   0.616768   1.000000  0.235393   \n",
       "nobleman   0.546470  0.473069  0.628878   0.376194   0.235393  1.000000   \n",
       "gorgeous   0.561290  0.473539  0.630321   0.373142   0.251796  0.998045   \n",
       "afflicted  0.558451  0.319628  0.516759   0.764730   0.411056  0.303783   \n",
       "\n",
       "           gorgeous  afflicted  \n",
       "go         0.317004   0.029740  \n",
       "say        0.124409   0.248187  \n",
       "ulysses    0.133564   0.124105  \n",
       "come       0.096292   0.186922  \n",
       "one        0.048618   0.350760  \n",
       "...             ...        ...  \n",
       "murderess  0.373142   0.764730  \n",
       "cassandra  0.251796   0.411056  \n",
       "nobleman   0.998045   0.303783  \n",
       "gorgeous   1.000000   0.288995  \n",
       "afflicted  0.288995   1.000000  \n",
       "\n",
       "[5902 rows x 5902 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "\n",
    "# create similarity matrix\n",
    "similarity_matrix = cosine_similarity(vec_df.values)\n",
    "\n",
    "# create pandas DataFrame for the similarity matrix\n",
    "similarity_df = pd.DataFrame(similarity_matrix, index=words, columns=words)\n",
    "\n",
    "# show dataframe\n",
    "similarity_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "cf1bdb64",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "go                           [leave, get, back]\n",
       "say                  [ulysses, therefore, come]\n",
       "ulysses                       [say, come, back]\n",
       "come                       [say, ulysses, back]\n",
       "one                       [see, would, another]\n",
       "                            ...                \n",
       "murderess             [slut, shameless, smirch]\n",
       "cassandra    [encouragingly, desolate, defence]\n",
       "nobleman          [gorgeous, picnic, miserably]\n",
       "gorgeous          [picnic, nobleman, miserably]\n",
       "afflicted          [reveller, shudder, elysian]\n",
       "Length: 5902, dtype: object"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# assign feature names to a variable\n",
    "feature_names = np.array(words)\n",
    "\n",
    "# create df with each feature (word) and the three most similar words to it\n",
    "similarity_df.apply(lambda row: feature_names[np.argsort(-row.values)[1:4]], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fa0ba5f",
   "metadata": {},
   "source": [
    "<a name=\"Glove\"></a>\n",
    "## 2.2 GloVe\n",
    "\n",
    "**GloVe** stands for **Global Vectors** which is an **unsupervised learning model** which can be used to **obtain dense word vectors similar to Word2Vec**. \n",
    "\n",
    "However the **technique is different** and training is performed on an aggregated global word-word co-occurrence matrix, giving us **a vector space with meaningful sub-structures**. \n",
    "\n",
    "The **basic methodology** of the GloVe model is:\n",
    "1. create a huge **word-context co-occurence matrix** consisting of `(word, context)` pairs such that **each element in this matrix represents how often a word occurs with the context**. \n",
    "2. apply **matrix factorization to approximate this matrix**.\n",
    "\n",
    "We aim to reconstruct __WC__ from __WF__ and __FC__ by multiplying them. For this, we typically initialize __WF__ and __FC__ with some random weights and attempt to multiply them to get an approximation of __WC__ and measure how close it is to __WC__. \n",
    "\n",
    "We do this multiple times using **Stochastic Gradient Descent** (SGD) to minimize the error. \n",
    "\n",
    "Finally, the __Word-Feature matrix__ (WF) **gives us the word embeddings for each word** where __F__ can be preset to a specific number of dimensions.\n",
    "\n",
    "<img src=\"https://cdn-images-1.medium.com/max/800/1*UNtsSilztKXjLG99VXxSQw.png\" alt=\"Drawing\" align=\"Left\" style=\"width: 60%;\"/>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec80d110",
   "metadata": {},
   "source": [
    "The **spaCy framework** comes with capabilities to leverage GloVe embeddings based on different language models. You can also get **pre-trained word vectors and load them up as needed** using `gensim` or `spaCy`.\n",
    "\n",
    "We will be using the __[`en_core_web_lg`](https://spacy.io/models/en#en_core_web_lg)__ model which consists of **300-dimensional word vectors** trained on [Common Crawl](http://commoncrawl.org) with GloVe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "d49fc312",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total word vectors: 684830\n"
     ]
    }
   ],
   "source": [
    "import spacy\n",
    "\n",
    "# load spaCy model\n",
    "nlp = spacy.load('en_core_web_lg')\n",
    "\n",
    "# check the length of vectors\n",
    "total_vectors = len(nlp.vocab.vectors)\n",
    "\n",
    "print('Total word vectors:', total_vectors)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12958dcd",
   "metadata": {},
   "source": [
    "Now we can get the __GloVe embeddings__ for each of our words in our corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "2b9f93b8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>290</th>\n",
       "      <th>291</th>\n",
       "      <th>292</th>\n",
       "      <th>293</th>\n",
       "      <th>294</th>\n",
       "      <th>295</th>\n",
       "      <th>296</th>\n",
       "      <th>297</th>\n",
       "      <th>298</th>\n",
       "      <th>299</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sire</th>\n",
       "      <td>-0.356530</td>\n",
       "      <td>0.306260</td>\n",
       "      <td>-0.574830</td>\n",
       "      <td>-0.158250</td>\n",
       "      <td>0.453700</td>\n",
       "      <td>-0.070828</td>\n",
       "      <td>0.263960</td>\n",
       "      <td>0.357110</td>\n",
       "      <td>-0.252050</td>\n",
       "      <td>0.62315</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.423440</td>\n",
       "      <td>0.158120</td>\n",
       "      <td>-0.894090</td>\n",
       "      <td>-0.556880</td>\n",
       "      <td>0.325100</td>\n",
       "      <td>0.320650</td>\n",
       "      <td>0.583920</td>\n",
       "      <td>0.086472</td>\n",
       "      <td>0.192270</td>\n",
       "      <td>0.015134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agreeable</th>\n",
       "      <td>-0.106050</td>\n",
       "      <td>0.181300</td>\n",
       "      <td>-0.501130</td>\n",
       "      <td>-0.151170</td>\n",
       "      <td>-0.032616</td>\n",
       "      <td>0.074835</td>\n",
       "      <td>0.057952</td>\n",
       "      <td>0.176440</td>\n",
       "      <td>-0.024474</td>\n",
       "      <td>1.29580</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.067799</td>\n",
       "      <td>0.144830</td>\n",
       "      <td>-0.102080</td>\n",
       "      <td>0.031814</td>\n",
       "      <td>0.503610</td>\n",
       "      <td>-0.053432</td>\n",
       "      <td>0.332430</td>\n",
       "      <td>-0.405190</td>\n",
       "      <td>-0.094457</td>\n",
       "      <td>-0.034074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poor</th>\n",
       "      <td>-1.172800</td>\n",
       "      <td>0.245140</td>\n",
       "      <td>-0.380370</td>\n",
       "      <td>-0.053600</td>\n",
       "      <td>-0.640900</td>\n",
       "      <td>-0.307810</td>\n",
       "      <td>0.452380</td>\n",
       "      <td>-0.205380</td>\n",
       "      <td>-0.335610</td>\n",
       "      <td>2.66580</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.576840</td>\n",
       "      <td>0.263030</td>\n",
       "      <td>-0.021811</td>\n",
       "      <td>0.175080</td>\n",
       "      <td>0.009650</td>\n",
       "      <td>0.420790</td>\n",
       "      <td>0.055763</td>\n",
       "      <td>0.058612</td>\n",
       "      <td>-0.465920</td>\n",
       "      <td>-0.174160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spell</th>\n",
       "      <td>-0.143630</td>\n",
       "      <td>0.047154</td>\n",
       "      <td>-0.306080</td>\n",
       "      <td>0.125900</td>\n",
       "      <td>-0.248360</td>\n",
       "      <td>-0.261990</td>\n",
       "      <td>-0.215640</td>\n",
       "      <td>-0.006327</td>\n",
       "      <td>-0.576460</td>\n",
       "      <td>1.58310</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.056274</td>\n",
       "      <td>0.381410</td>\n",
       "      <td>-0.225300</td>\n",
       "      <td>0.667710</td>\n",
       "      <td>-0.127370</td>\n",
       "      <td>0.158590</td>\n",
       "      <td>-0.547710</td>\n",
       "      <td>0.015922</td>\n",
       "      <td>0.444380</td>\n",
       "      <td>-0.440930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conceal</th>\n",
       "      <td>0.071534</td>\n",
       "      <td>-0.244040</td>\n",
       "      <td>-0.122910</td>\n",
       "      <td>0.162130</td>\n",
       "      <td>-0.302430</td>\n",
       "      <td>-0.394810</td>\n",
       "      <td>-0.301080</td>\n",
       "      <td>0.285360</td>\n",
       "      <td>-0.802920</td>\n",
       "      <td>1.35020</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.408850</td>\n",
       "      <td>0.181410</td>\n",
       "      <td>-0.426560</td>\n",
       "      <td>0.171560</td>\n",
       "      <td>0.160380</td>\n",
       "      <td>-0.192060</td>\n",
       "      <td>0.615800</td>\n",
       "      <td>0.282470</td>\n",
       "      <td>0.369020</td>\n",
       "      <td>0.210130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>borrow</th>\n",
       "      <td>-0.094431</td>\n",
       "      <td>-0.553700</td>\n",
       "      <td>-0.003532</td>\n",
       "      <td>-0.083429</td>\n",
       "      <td>-0.250680</td>\n",
       "      <td>-0.184440</td>\n",
       "      <td>-0.153040</td>\n",
       "      <td>-0.486350</td>\n",
       "      <td>-0.140010</td>\n",
       "      <td>1.82520</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.495000</td>\n",
       "      <td>0.048966</td>\n",
       "      <td>0.142280</td>\n",
       "      <td>0.140880</td>\n",
       "      <td>0.447370</td>\n",
       "      <td>-0.150510</td>\n",
       "      <td>0.140760</td>\n",
       "      <td>-0.256790</td>\n",
       "      <td>0.011806</td>\n",
       "      <td>-0.314910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reverberate</th>\n",
       "      <td>0.289160</td>\n",
       "      <td>0.503070</td>\n",
       "      <td>0.034871</td>\n",
       "      <td>0.211800</td>\n",
       "      <td>0.529000</td>\n",
       "      <td>-0.032653</td>\n",
       "      <td>-0.486030</td>\n",
       "      <td>0.240300</td>\n",
       "      <td>-0.494350</td>\n",
       "      <td>0.70125</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.322180</td>\n",
       "      <td>-0.035876</td>\n",
       "      <td>0.243560</td>\n",
       "      <td>0.418760</td>\n",
       "      <td>-0.042285</td>\n",
       "      <td>0.504090</td>\n",
       "      <td>-0.144760</td>\n",
       "      <td>0.255230</td>\n",
       "      <td>-0.236810</td>\n",
       "      <td>-0.204200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>overhung</th>\n",
       "      <td>0.846040</td>\n",
       "      <td>-0.195260</td>\n",
       "      <td>-0.291340</td>\n",
       "      <td>-0.137640</td>\n",
       "      <td>-0.041836</td>\n",
       "      <td>0.206930</td>\n",
       "      <td>0.186900</td>\n",
       "      <td>0.446240</td>\n",
       "      <td>-0.350470</td>\n",
       "      <td>-1.06740</td>\n",
       "      <td>...</td>\n",
       "      <td>0.225750</td>\n",
       "      <td>-0.995840</td>\n",
       "      <td>0.544220</td>\n",
       "      <td>-0.638280</td>\n",
       "      <td>-0.578730</td>\n",
       "      <td>-0.184290</td>\n",
       "      <td>0.745720</td>\n",
       "      <td>-0.337410</td>\n",
       "      <td>-0.218750</td>\n",
       "      <td>-0.436880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>loathe</th>\n",
       "      <td>0.003006</td>\n",
       "      <td>-0.164040</td>\n",
       "      <td>0.061765</td>\n",
       "      <td>-0.124250</td>\n",
       "      <td>-0.166020</td>\n",
       "      <td>0.173580</td>\n",
       "      <td>0.019472</td>\n",
       "      <td>-0.267190</td>\n",
       "      <td>0.029233</td>\n",
       "      <td>1.20070</td>\n",
       "      <td>...</td>\n",
       "      <td>0.020546</td>\n",
       "      <td>0.199140</td>\n",
       "      <td>-0.350530</td>\n",
       "      <td>-0.131970</td>\n",
       "      <td>-0.240000</td>\n",
       "      <td>0.427640</td>\n",
       "      <td>-0.280870</td>\n",
       "      <td>-0.074931</td>\n",
       "      <td>0.140110</td>\n",
       "      <td>-0.067065</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perished</th>\n",
       "      <td>0.507110</td>\n",
       "      <td>-0.155330</td>\n",
       "      <td>0.372450</td>\n",
       "      <td>0.142940</td>\n",
       "      <td>-0.075347</td>\n",
       "      <td>0.030305</td>\n",
       "      <td>-0.283890</td>\n",
       "      <td>0.307210</td>\n",
       "      <td>-0.560370</td>\n",
       "      <td>1.94870</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.453200</td>\n",
       "      <td>0.163960</td>\n",
       "      <td>0.014067</td>\n",
       "      <td>-0.403030</td>\n",
       "      <td>-0.294060</td>\n",
       "      <td>-0.012077</td>\n",
       "      <td>0.026874</td>\n",
       "      <td>0.134930</td>\n",
       "      <td>0.301100</td>\n",
       "      <td>-0.750070</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5902 rows × 300 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0         1         2         3         4         5    \\\n",
       "sire        -0.356530  0.306260 -0.574830 -0.158250  0.453700 -0.070828   \n",
       "agreeable   -0.106050  0.181300 -0.501130 -0.151170 -0.032616  0.074835   \n",
       "poor        -1.172800  0.245140 -0.380370 -0.053600 -0.640900 -0.307810   \n",
       "spell       -0.143630  0.047154 -0.306080  0.125900 -0.248360 -0.261990   \n",
       "conceal      0.071534 -0.244040 -0.122910  0.162130 -0.302430 -0.394810   \n",
       "...               ...       ...       ...       ...       ...       ...   \n",
       "borrow      -0.094431 -0.553700 -0.003532 -0.083429 -0.250680 -0.184440   \n",
       "reverberate  0.289160  0.503070  0.034871  0.211800  0.529000 -0.032653   \n",
       "overhung     0.846040 -0.195260 -0.291340 -0.137640 -0.041836  0.206930   \n",
       "loathe       0.003006 -0.164040  0.061765 -0.124250 -0.166020  0.173580   \n",
       "perished     0.507110 -0.155330  0.372450  0.142940 -0.075347  0.030305   \n",
       "\n",
       "                  6         7         8        9    ...       290       291  \\\n",
       "sire         0.263960  0.357110 -0.252050  0.62315  ... -0.423440  0.158120   \n",
       "agreeable    0.057952  0.176440 -0.024474  1.29580  ... -0.067799  0.144830   \n",
       "poor         0.452380 -0.205380 -0.335610  2.66580  ... -0.576840  0.263030   \n",
       "spell       -0.215640 -0.006327 -0.576460  1.58310  ... -0.056274  0.381410   \n",
       "conceal     -0.301080  0.285360 -0.802920  1.35020  ... -0.408850  0.181410   \n",
       "...               ...       ...       ...      ...  ...       ...       ...   \n",
       "borrow      -0.153040 -0.486350 -0.140010  1.82520  ... -0.495000  0.048966   \n",
       "reverberate -0.486030  0.240300 -0.494350  0.70125  ... -0.322180 -0.035876   \n",
       "overhung     0.186900  0.446240 -0.350470 -1.06740  ...  0.225750 -0.995840   \n",
       "loathe       0.019472 -0.267190  0.029233  1.20070  ...  0.020546  0.199140   \n",
       "perished    -0.283890  0.307210 -0.560370  1.94870  ... -0.453200  0.163960   \n",
       "\n",
       "                  292       293       294       295       296       297  \\\n",
       "sire        -0.894090 -0.556880  0.325100  0.320650  0.583920  0.086472   \n",
       "agreeable   -0.102080  0.031814  0.503610 -0.053432  0.332430 -0.405190   \n",
       "poor        -0.021811  0.175080  0.009650  0.420790  0.055763  0.058612   \n",
       "spell       -0.225300  0.667710 -0.127370  0.158590 -0.547710  0.015922   \n",
       "conceal     -0.426560  0.171560  0.160380 -0.192060  0.615800  0.282470   \n",
       "...               ...       ...       ...       ...       ...       ...   \n",
       "borrow       0.142280  0.140880  0.447370 -0.150510  0.140760 -0.256790   \n",
       "reverberate  0.243560  0.418760 -0.042285  0.504090 -0.144760  0.255230   \n",
       "overhung     0.544220 -0.638280 -0.578730 -0.184290  0.745720 -0.337410   \n",
       "loathe      -0.350530 -0.131970 -0.240000  0.427640 -0.280870 -0.074931   \n",
       "perished     0.014067 -0.403030 -0.294060 -0.012077  0.026874  0.134930   \n",
       "\n",
       "                  298       299  \n",
       "sire         0.192270  0.015134  \n",
       "agreeable   -0.094457 -0.034074  \n",
       "poor        -0.465920 -0.174160  \n",
       "spell        0.444380 -0.440930  \n",
       "conceal      0.369020  0.210130  \n",
       "...               ...       ...  \n",
       "borrow       0.011806 -0.314910  \n",
       "reverberate -0.236810 -0.204200  \n",
       "overhung    -0.218750 -0.436880  \n",
       "loathe       0.140110 -0.067065  \n",
       "perished     0.301100 -0.750070  \n",
       "\n",
       "[5902 rows x 300 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the tokenized vocabulary of the corpus as a list of words\n",
    "unique_words = list(set([word for sublist in tokenized_corpus for word in sublist]))\n",
    "\n",
    "# get the vector representation of each word\n",
    "word_glove_vectors = np.array([nlp(word).vector for word in unique_words])\n",
    "\n",
    "# create a pandas DataFrame for each token and its vectors\n",
    "vec_df = pd.DataFrame(word_glove_vectors, index=unique_words)\n",
    "\n",
    "# print DataFrame\n",
    "vec_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1990e9b9",
   "metadata": {},
   "source": [
    "We can now use **t-SNE** to visualize these embeddings as well as check __pairwise similarity__ between words, similar to what we did using our Word2Vec embeddings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "62420fb8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# instantiate tSNE\n",
    "tsne = TSNE(n_components=2, random_state=42, n_iter=5000, perplexity=3)\n",
    "\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "# fit tsne to GloVe vectors\n",
    "T = tsne.fit_transform(word_glove_vectors)\n",
    "\n",
    "# create labels\n",
    "labels = unique_words\n",
    "\n",
    "# create visualization\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.scatter(T[:, 0], T[:, 1], c='red', edgecolors='r')\n",
    "for label, x, y in zip(labels, T[:, 0], T[:, 1]):\n",
    "    plt.annotate(label, xy=(x+1, y+1), xytext=(0, 0), textcoords='offset points')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "06f85486",
   "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>sire</th>\n",
       "      <th>agreeable</th>\n",
       "      <th>poor</th>\n",
       "      <th>spell</th>\n",
       "      <th>conceal</th>\n",
       "      <th>post</th>\n",
       "      <th>phantom</th>\n",
       "      <th>boyhood</th>\n",
       "      <th>guest</th>\n",
       "      <th>add</th>\n",
       "      <th>...</th>\n",
       "      <th>unblended</th>\n",
       "      <th>trill</th>\n",
       "      <th>minstrel</th>\n",
       "      <th>address</th>\n",
       "      <th>periplus</th>\n",
       "      <th>borrow</th>\n",
       "      <th>reverberate</th>\n",
       "      <th>overhung</th>\n",
       "      <th>loathe</th>\n",
       "      <th>perished</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>sire</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.098265</td>\n",
       "      <td>0.148598</td>\n",
       "      <td>0.097415</td>\n",
       "      <td>0.117951</td>\n",
       "      <td>-0.018772</td>\n",
       "      <td>0.099848</td>\n",
       "      <td>0.162351</td>\n",
       "      <td>0.037072</td>\n",
       "      <td>0.012147</td>\n",
       "      <td>...</td>\n",
       "      <td>0.115568</td>\n",
       "      <td>0.058784</td>\n",
       "      <td>0.117152</td>\n",
       "      <td>-0.011183</td>\n",
       "      <td>-0.073837</td>\n",
       "      <td>0.044701</td>\n",
       "      <td>0.015654</td>\n",
       "      <td>0.003029</td>\n",
       "      <td>0.010817</td>\n",
       "      <td>0.191841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>agreeable</th>\n",
       "      <td>0.098265</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.213928</td>\n",
       "      <td>0.141108</td>\n",
       "      <td>0.302527</td>\n",
       "      <td>0.065201</td>\n",
       "      <td>0.072885</td>\n",
       "      <td>0.230690</td>\n",
       "      <td>0.166938</td>\n",
       "      <td>0.118935</td>\n",
       "      <td>...</td>\n",
       "      <td>0.157312</td>\n",
       "      <td>0.169259</td>\n",
       "      <td>0.138573</td>\n",
       "      <td>0.096961</td>\n",
       "      <td>0.043787</td>\n",
       "      <td>0.196874</td>\n",
       "      <td>0.145783</td>\n",
       "      <td>0.165006</td>\n",
       "      <td>0.276481</td>\n",
       "      <td>0.143282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>poor</th>\n",
       "      <td>0.148598</td>\n",
       "      <td>0.213928</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.325400</td>\n",
       "      <td>0.193021</td>\n",
       "      <td>0.216357</td>\n",
       "      <td>0.207787</td>\n",
       "      <td>0.140590</td>\n",
       "      <td>0.175145</td>\n",
       "      <td>0.149904</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.034030</td>\n",
       "      <td>0.012772</td>\n",
       "      <td>0.119715</td>\n",
       "      <td>0.211686</td>\n",
       "      <td>-0.099329</td>\n",
       "      <td>0.310593</td>\n",
       "      <td>0.048996</td>\n",
       "      <td>-0.002018</td>\n",
       "      <td>0.261208</td>\n",
       "      <td>0.270684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spell</th>\n",
       "      <td>0.097415</td>\n",
       "      <td>0.141108</td>\n",
       "      <td>0.325400</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.202184</td>\n",
       "      <td>0.180775</td>\n",
       "      <td>0.195759</td>\n",
       "      <td>0.103995</td>\n",
       "      <td>0.110342</td>\n",
       "      <td>0.308786</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.094417</td>\n",
       "      <td>0.137915</td>\n",
       "      <td>0.114855</td>\n",
       "      <td>0.157787</td>\n",
       "      <td>-0.012213</td>\n",
       "      <td>0.224752</td>\n",
       "      <td>0.109754</td>\n",
       "      <td>-0.051527</td>\n",
       "      <td>0.204420</td>\n",
       "      <td>0.157312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>conceal</th>\n",
       "      <td>0.117951</td>\n",
       "      <td>0.302527</td>\n",
       "      <td>0.193021</td>\n",
       "      <td>0.202184</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.001453</td>\n",
       "      <td>0.227936</td>\n",
       "      <td>0.167946</td>\n",
       "      <td>0.032916</td>\n",
       "      <td>0.206740</td>\n",
       "      <td>...</td>\n",
       "      <td>0.053394</td>\n",
       "      <td>0.039907</td>\n",
       "      <td>0.109532</td>\n",
       "      <td>0.190963</td>\n",
       "      <td>-0.009526</td>\n",
       "      <td>0.276654</td>\n",
       "      <td>0.171657</td>\n",
       "      <td>0.167810</td>\n",
       "      <td>0.304583</td>\n",
       "      <td>0.181789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>borrow</th>\n",
       "      <td>0.044701</td>\n",
       "      <td>0.196874</td>\n",
       "      <td>0.310593</td>\n",
       "      <td>0.224752</td>\n",
       "      <td>0.276654</td>\n",
       "      <td>0.165534</td>\n",
       "      <td>0.161821</td>\n",
       "      <td>0.147624</td>\n",
       "      <td>0.126247</td>\n",
       "      <td>0.256923</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.038976</td>\n",
       "      <td>0.014234</td>\n",
       "      <td>-0.000767</td>\n",
       "      <td>0.139318</td>\n",
       "      <td>-0.036999</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.095376</td>\n",
       "      <td>0.015482</td>\n",
       "      <td>0.291598</td>\n",
       "      <td>0.045249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>reverberate</th>\n",
       "      <td>0.015654</td>\n",
       "      <td>0.145783</td>\n",
       "      <td>0.048996</td>\n",
       "      <td>0.109754</td>\n",
       "      <td>0.171657</td>\n",
       "      <td>-0.040103</td>\n",
       "      <td>0.092368</td>\n",
       "      <td>0.236488</td>\n",
       "      <td>-0.028263</td>\n",
       "      <td>0.025591</td>\n",
       "      <td>...</td>\n",
       "      <td>0.086396</td>\n",
       "      <td>0.229439</td>\n",
       "      <td>0.189568</td>\n",
       "      <td>0.060316</td>\n",
       "      <td>-0.003194</td>\n",
       "      <td>0.095376</td>\n",
       "      <td>0.999999</td>\n",
       "      <td>0.188640</td>\n",
       "      <td>0.158033</td>\n",
       "      <td>0.181134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>overhung</th>\n",
       "      <td>0.003029</td>\n",
       "      <td>0.165006</td>\n",
       "      <td>-0.002018</td>\n",
       "      <td>-0.051527</td>\n",
       "      <td>0.167810</td>\n",
       "      <td>-0.105686</td>\n",
       "      <td>0.062333</td>\n",
       "      <td>0.120863</td>\n",
       "      <td>-0.075627</td>\n",
       "      <td>-0.125493</td>\n",
       "      <td>...</td>\n",
       "      <td>0.252357</td>\n",
       "      <td>0.210364</td>\n",
       "      <td>0.094980</td>\n",
       "      <td>-0.113433</td>\n",
       "      <td>0.083908</td>\n",
       "      <td>0.015482</td>\n",
       "      <td>0.188640</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.103201</td>\n",
       "      <td>0.144359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>loathe</th>\n",
       "      <td>0.010817</td>\n",
       "      <td>0.276481</td>\n",
       "      <td>0.261208</td>\n",
       "      <td>0.204420</td>\n",
       "      <td>0.304583</td>\n",
       "      <td>0.087498</td>\n",
       "      <td>0.053988</td>\n",
       "      <td>0.141764</td>\n",
       "      <td>0.067585</td>\n",
       "      <td>0.131941</td>\n",
       "      <td>...</td>\n",
       "      <td>0.052317</td>\n",
       "      <td>0.084430</td>\n",
       "      <td>0.105463</td>\n",
       "      <td>0.000383</td>\n",
       "      <td>0.006808</td>\n",
       "      <td>0.291598</td>\n",
       "      <td>0.158033</td>\n",
       "      <td>0.103201</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.157121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>perished</th>\n",
       "      <td>0.191841</td>\n",
       "      <td>0.143282</td>\n",
       "      <td>0.270684</td>\n",
       "      <td>0.157312</td>\n",
       "      <td>0.181789</td>\n",
       "      <td>-0.020072</td>\n",
       "      <td>0.083334</td>\n",
       "      <td>0.285736</td>\n",
       "      <td>0.002715</td>\n",
       "      <td>-0.079315</td>\n",
       "      <td>...</td>\n",
       "      <td>0.088118</td>\n",
       "      <td>0.026818</td>\n",
       "      <td>0.096128</td>\n",
       "      <td>-0.044937</td>\n",
       "      <td>0.096010</td>\n",
       "      <td>0.045249</td>\n",
       "      <td>0.181134</td>\n",
       "      <td>0.144359</td>\n",
       "      <td>0.157121</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5902 rows × 5902 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 sire  agreeable      poor     spell   conceal      post  \\\n",
       "sire         1.000000   0.098265  0.148598  0.097415  0.117951 -0.018772   \n",
       "agreeable    0.098265   1.000000  0.213928  0.141108  0.302527  0.065201   \n",
       "poor         0.148598   0.213928  1.000000  0.325400  0.193021  0.216357   \n",
       "spell        0.097415   0.141108  0.325400  1.000000  0.202184  0.180775   \n",
       "conceal      0.117951   0.302527  0.193021  0.202184  1.000000  0.001453   \n",
       "...               ...        ...       ...       ...       ...       ...   \n",
       "borrow       0.044701   0.196874  0.310593  0.224752  0.276654  0.165534   \n",
       "reverberate  0.015654   0.145783  0.048996  0.109754  0.171657 -0.040103   \n",
       "overhung     0.003029   0.165006 -0.002018 -0.051527  0.167810 -0.105686   \n",
       "loathe       0.010817   0.276481  0.261208  0.204420  0.304583  0.087498   \n",
       "perished     0.191841   0.143282  0.270684  0.157312  0.181789 -0.020072   \n",
       "\n",
       "              phantom   boyhood     guest       add  ...  unblended     trill  \\\n",
       "sire         0.099848  0.162351  0.037072  0.012147  ...   0.115568  0.058784   \n",
       "agreeable    0.072885  0.230690  0.166938  0.118935  ...   0.157312  0.169259   \n",
       "poor         0.207787  0.140590  0.175145  0.149904  ...  -0.034030  0.012772   \n",
       "spell        0.195759  0.103995  0.110342  0.308786  ...  -0.094417  0.137915   \n",
       "conceal      0.227936  0.167946  0.032916  0.206740  ...   0.053394  0.039907   \n",
       "...               ...       ...       ...       ...  ...        ...       ...   \n",
       "borrow       0.161821  0.147624  0.126247  0.256923  ...  -0.038976  0.014234   \n",
       "reverberate  0.092368  0.236488 -0.028263  0.025591  ...   0.086396  0.229439   \n",
       "overhung     0.062333  0.120863 -0.075627 -0.125493  ...   0.252357  0.210364   \n",
       "loathe       0.053988  0.141764  0.067585  0.131941  ...   0.052317  0.084430   \n",
       "perished     0.083334  0.285736  0.002715 -0.079315  ...   0.088118  0.026818   \n",
       "\n",
       "             minstrel   address  periplus    borrow  reverberate  overhung  \\\n",
       "sire         0.117152 -0.011183 -0.073837  0.044701     0.015654  0.003029   \n",
       "agreeable    0.138573  0.096961  0.043787  0.196874     0.145783  0.165006   \n",
       "poor         0.119715  0.211686 -0.099329  0.310593     0.048996 -0.002018   \n",
       "spell        0.114855  0.157787 -0.012213  0.224752     0.109754 -0.051527   \n",
       "conceal      0.109532  0.190963 -0.009526  0.276654     0.171657  0.167810   \n",
       "...               ...       ...       ...       ...          ...       ...   \n",
       "borrow      -0.000767  0.139318 -0.036999  1.000000     0.095376  0.015482   \n",
       "reverberate  0.189568  0.060316 -0.003194  0.095376     0.999999  0.188640   \n",
       "overhung     0.094980 -0.113433  0.083908  0.015482     0.188640  1.000000   \n",
       "loathe       0.105463  0.000383  0.006808  0.291598     0.158033  0.103201   \n",
       "perished     0.096128 -0.044937  0.096010  0.045249     0.181134  0.144359   \n",
       "\n",
       "               loathe  perished  \n",
       "sire         0.010817  0.191841  \n",
       "agreeable    0.276481  0.143282  \n",
       "poor         0.261208  0.270684  \n",
       "spell        0.204420  0.157312  \n",
       "conceal      0.304583  0.181789  \n",
       "...               ...       ...  \n",
       "borrow       0.291598  0.045249  \n",
       "reverberate  0.158033  0.181134  \n",
       "overhung     0.103201  0.144359  \n",
       "loathe       1.000000  0.157121  \n",
       "perished     0.157121  1.000000  \n",
       "\n",
       "[5902 rows x 5902 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create similarity matrix\n",
    "similarity_matrix = cosine_similarity(vec_df.values)\n",
    "\n",
    "# create a pandas DataFrame\n",
    "similarity_df = pd.DataFrame(similarity_matrix, index=unique_words, columns=unique_words)\n",
    "\n",
    "# show DataFrame\n",
    "similarity_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "8869276d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sire                          [stallion, mare, foal]\n",
       "agreeable      [pleasant, advantageous, exceedingly]\n",
       "poor                          [bad, worse, terrible]\n",
       "spell                    [enchantment, summon, word]\n",
       "conceal                [hide, disguise, concealment]\n",
       "                               ...                  \n",
       "borrow                      [borrowing, lend, repay]\n",
       "reverberate              [resound, resounds, rankle]\n",
       "overhung                       [thickly, crag, hewn]\n",
       "loathe                      [detest, loath, dislike]\n",
       "perished                     [perish, fled, escaped]\n",
       "Length: 5902, dtype: object"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# assign feature names (words) to a variable\n",
    "feature_names = np.array(unique_words)\n",
    "\n",
    "# sort DataFrame so that each feature (word) has the 3 most similar words next to it\n",
    "similarity_df.apply(lambda row: feature_names[np.argsort(-row.values)[1:4]], \n",
    "                    axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "343e22d9",
   "metadata": {},
   "source": [
    "<a name=\"fastText\"></a>\n",
    "## 2.3 FastText"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39e7bf53",
   "metadata": {},
   "source": [
    "FastText is **a framework for learning word representations** and also performing robust, fast and accurate **text classification**. \n",
    "\n",
    "The framework is **open-sourced** by Facebook on [GitHub](https://github.com/facebookresearch/fastText) and claims to have the following:\n",
    "\n",
    "1. Recent **state-of-the-art English word vectors**.\n",
    "2. Word vectors for **157 languages** trained on Wikipedia and Crawl.\n",
    "3. Models for **language identification** and various supervised tasks.\n",
    "\n",
    "*A **high-level overview of the FastText Model** can be found [here](https://www.youtube.com/watch?v=4l_At3oalzk) as well as the official **FastText tutorial** [here](https://fasttext.cc/docs/en/supervised-tutorial.html).*\n",
    "\n",
    "**Predictive models**, lile the **Word2Vec model**, typically ignores the morphological structure of each word; **considers a word as a single entity** and **generate a dense embedding for the word**.\n",
    "\n",
    "However this poses to be a serious limitation with **languages having massive vocabularies** and many **rare words** which may not occur a lot in different corpora. \n",
    "\n",
    "The FastText model considers each word as a Bag of Character n-grams (**subword model**). We add special **boundary symbols**, **`<`** and **`>`**, at the beginning and end of words, so we can **distinguish prefixes and suffixes** from other character sequences. We also **include the word itself in the set of its n-grams**, to learn a representation for each word (in addition to its character n-grams). \n",
    "\n",
    "Taking the word **`where`** and **n=3** (tri-grams) as an example, it will be represented by the character n-grams:\n",
    "* **<wh, whe, her, ere, re>** <br>\n",
    "and <br>\n",
    "* the special sequence **< where >** representing the whole word. \n",
    "\n",
    "Note that the sequence , corresponding to the word **`< her >`** is different from the tri-gram **`her`** from the word **`where`**.\n",
    "\n",
    "In practice, it is recommended to extract all the n-grams for **n ≥ 3** and **n ≤ 6**. \n",
    "\n",
    "This is a very simple approach, and different sets of n-grams could be considered, for example taking all prefixes and suffixes. We typically **associate a vector representation (embedding) to each n-gram** for a word. \n",
    "\n",
    "Thus, we can **represent a word by the sum of the vector representations of its n-grams or the average of the embedding of these n-grams**. \n",
    "\n",
    "Thus, due to this effect of leveraging n-grams from individual words based on their characters, there is a **higher chance for rare words to get a good representation** since their character based n-grams should occur across other words of the corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "5fa5a051",
   "metadata": {},
   "outputs": [],
   "source": [
    "from gensim.models.fasttext import FastText\n",
    "\n",
    "# Set values for various parameters\n",
    "feature_size = 15    # Word vector dimensionality  \n",
    "window_context = 20  # Context window size                                                                                    \n",
    "min_word_count = 1   # Minimum word count                        \n",
    "sample = 1e-3        # Downsample setting for frequent words\n",
    "sg = 1               # skip-gram model\n",
    "\n",
    "# instantiate and train the model\n",
    "ft_model = FastText(tokenized_corpus, vector_size=feature_size, \n",
    "                     window=window_context, min_count = min_word_count,\n",
    "                     sg=sg, sample=sample, epochs=5000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "530a8f47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuQAAAFlCAYAAACjoP2FAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOxdd5gUVfY9VdXVcbpnpifnnJhhAkNOQxQBRRQRBAFFUQEFXdHFAJh1xRXxt6yuq2IADJjdVXcVUWExgBKUHAYkShpgcug+vz+q6033BBgwbajzfe/r7uoKr169qjrvvnvPlUjCgAEDBgwYMGDAgAEDvw3k37oCBgwYMGDAgAEDBgz8L8Mg5AYMGDBgwIABAwYM/IYwCLkBAwYMGDBgwIABA78hDEJuwIABAwYMGDBgwMBvCIOQGzBgwIABAwYMGDDwG8Ig5AYMGDBgwIABAwYM/IYw/dYVaAvCw8OZnJz8W1fDgAEDBgwYMGDAwH85vvnmm6MkI37NY/5HEPLk5GSsWbPmt66GAQMGDBgwYMCAgf9ySJK059c+puGyYsCAAQMGDBgwYMDAbwiDkBswYMCAAQMGDBgw8BvCIOQGDBgwYMCAAQMGDPyGMAi5AQMGDBgwYMCAAQO/IQxCbsCAAQMGfjEEBQWd87bXXHMNNm3aBAB48MEHxfLdu3cjMTFR/AcAs2fPxscff3zuFTVgwICB3xAGITdgwICB/3J0794dgEZk8/LyAACffvopLrjggp9l/++++y4efvjhgGVLly5FTk4OKisrT7ttcnIy3nvvPeTm5qKwsBCbN28WdXzqqafQrl07AIGEHADKy8sDCPm9996LAQMG/BynY8CAAQO/OgxCbsCAAQP/5Vi1alWzZYMHD25x3dYszacj8MOGDcPMmTMDlj377LP461//CofDgUWLFqFjx45wOp0IDw9HXl4ebDYbAKChoQGjRo3CkSNHUFtbi4EDB2L//v3o0qULkpOTkZiYiGnTpqG6uhoFBQVwOp348ssvcerUKVx11VWwWq3o2bMnxo0bh9dffx1//etfkZWVBZvNhuDgYIwfPx61tbU4cuQIHA4HYmNj4XA4kJaWhi1btpxtUxowYMDAL4L/CB1yAwYMGDDQNphMJmzcuBFZWVkANKt4amoq2rVrh127dkFVVVRVVYEkvvjiCxQXF+PkyZMYM2YM7r33XiQkJODCCy9EcnIykpOTsXPnTnTv3h3/+Mc/cOLECaSlpeHQoUMIDQ3Fm2++iW3btmH27Nk4deoURowYgerqamzbtg2rV6/GJ598gvr6ejz66KNITEzEsWPHcPDgQWzatAkk4fV6UVlZierqani9XtTW1oplLpcLJLF371689tpr8Hq9OH78OCwWC/bv3w+v14uQkBD8+OOPWLNmDb799lt88cUX8Hg82L9/P9LS0nD06FG88847+PTTT1FZWQmr1Yrbb78dF110ETp37oxHH30UzzzzzG98xQwYMGDAsJAbMGDAwH89SKK+vh7x8fGor6/HggULUF9fD4fDgZqaGpw6dQrvvfceAOCTTz5BeHg4PvjgA5hMJmzfvh1fffUVHA4HAKCkpAQHDhxAeno6BgwYgKlTp2LIkCEYPXo0Tp48ibfeegtbtmyBzWbD3LlzAWiDgmXLlmH37t3weDxQVRUA8PDDD8PhcMBsNqOwsBCVlZXIyMiAxWJBaGgoTp48iQULFqCyshIWiwWVlZW4/fbb8dxzzwEAvF4vtmzZAovFAqfTiYcffhiqqsLr9cJkMqGqqgrZ2dno2LEjAOD48eN46qmnMGzYMHg8HuzcufPXvhQGDBgw0CIMC7kBAwYM/IeitLQU2dnZqK+vBwAMGjQIJFFUVISamhrIsozc3FwAgMPhwN69e9HQ0IDZs2fD6/Vi7969AABJknDs2DHk5ubi6NGjOHLkCLKyslBXVweTyQSz2YwdO3YgKCgIn332GTIzM5Gamgqn0wmPx4O//OUvcLlc8Hq9sFgsgAk4+eNJPDT3IQBAjx49sGbNGtTU1KBdu3YoLS1FXV0d1q9fH3A+KSkpCAoKgiRJKC4uxldffYWnnnoKFRUVAIDa2lq89NJLOHHiBEwmE3Jzc3HddddBVVUxYDh48CBI4oorrsB9992HzZs3Y+vWraioqICiKPj73/8usj/PmDHjV7lOBgwYMHAmGBZyAwYMGPgPxU033QSbzQav1wuv14u5c+fC6/WiV69e8Hg8iI2Nxfbt2wEAmzdvRllZGRoaGsT2vXv3xjvvvAMAiIqKwsMPP4zDhw/D5XLhjTfeAAAoioL169fDbDajpqYGDzzwABISErB582YcOnQIJOFyuTBo0CCMGTMGx44dw7EjxwAAPx75EQCwcuVK1NfXIyQkBMuXL0diYiIAiM+6ujqUl5fDYrFg7969MJvNUBQFJ0+eRGhoKEJCQgBoA4d169bhj3/8IyRJwuWXX477778fHo8He/fuRXl5OSRJgslkwsmTJ1FZWQmHw4F77rkHl112GVwul1B92bp16y98dQwYMGCg7TAIuQEDBgz8h2LYsGE4efIkYmNjcdNNNwmy/c/P/glZlrF3/15UV1cDAOLj4+F2uyHLsnAZKS8vx7Bhw0ASP/74I66++moAwMmTJ3HjjTcC0EgwAERHR6O2thYzZ87Epk2b4HA4EBwcDJfLhTlz5sDtduP5l57X3iop0OZfi7R6VtdUo6KiAsePH0dMTAy2bt0KVVXxzDPP4PDhwzCZTPjxxx+xfft21NXVwel0AtBcUmJiYpCdnQ1ZlkESY8aMwZ49e2AymfDAAw/g+uuvh9PphKIoOHr0KC699FKQxPz58yHLMk6dOoXrr78eTzzxBKqrq9G7d2+0a9cOb7755q9yjQwYMGCgLTAIuQEDBgz8gnj++edxww034NNPP21R7USHv3Tg3XffjUcffRRAoLrJlKlToNgVSJIEySJh+ozpuO+++5CTk4M//elP6N27t7azwQBmAchs3P/Bgwdx7NgxeDweYSXesmUL0tLSAAAulwsejweyLEOWZbGOTt4dDgdkWYbX64XT6UR+fj5iYmJw//3348EHH8Qrr7yCqvIqIBpABwB2ABqXRz3q0bt3b5jNZlgsFpjNZgQFBeH2229HZGQk7HY77r33XmRkZOBf//oXgoODAQATJkzAJ598gkOHDkFVVciyjHXr1mHLli3o168f8vPz0dDQgGPHjmH48OGYMWMGTCYTLBYLMjIyUFJSgoEDB6JTp07o3bs3JkyYgE2bNmHTpk1444038Omnn/6ka2vAgAEDPxckkr91Hc6Ijh07cs2aNb91NQwYMGAAJEESstxoz2hoaIDJ1BiSM3v2bPTu3RsDBgzA888/jzVr1iA8PBxBQUE4deqU+O/xxx/HtddeC7PZDJPJhKCgIFRUVODuu+9GUFAQZsyYgU8//VRTKUlKxJMvPgmMAJAI4AcArwEgsOjZRXjv3ffw6quvahW4AkA6gL8COKCtA2gKLD169MDGjRtx9OhRTJw4ER999BH27t2Lw4cP45133sGkSZNw++2348EHH4TVakV8fDxcLhfatWuHt956C6WlpXA6nbj++uvx5ZdfwmQyISUlBX/72980a3o4AAVAELS6WgDcp7VbS9i9ezcuuOACfP/99y3+//DDD+PJJ5/E4sWL0bNnz3O+bgYMGDDQVkiS9A3Jjr/qMQ1CbsCAAQOnx8qVK9GnTx8oioL6+nqQxNixY1FTU4O//e1vAICuXbti7969OHDgADweDxRFQUZGBg4ePIja2lpUVFSAJOx2Oz788EPcfPPN+PbbbwEANpsN7dq1w9q1a9HQ0IArrrgC//znPxEfHw+n04lvv/0WFZUVQBKAK/0q9hIAn1CIbr0GAIQAsEKzUpdCEPKgoCAkJiZi7969UFUV33zzDWJiYmC1WpGdnY2srCx8//33woq+adMmbNu2TVjRX331VTz00EPwer1QVRULFixA165dRXUUuwLvZV7NZUVHKSC/JsNT5fk5L8l/FM406DhbXHnllaiqqkLnzp3bFJh64MABTJs2Da+//jrWrVuHAwcOYMiQIT9LXQwY+G/Eb0HIDZcVAwYMGDgDvvrqK3g8HrRv3x5ZWVmQJAmlpaX44IMPEBUVBYfDgT179qCsrAw7duyAJEnIycnBhg0b4PV6Icsy3G638I1eunQpdu/eDZLIysqC1WpFUFAQSOL3v/89Xn75ZZhMJrz//vv4+uuvUVxcrJHqCgAvADjhq5gmLAJJluDx+AjvBABXAzADKINmrTYDSRlJKC8vx8aNG3Hq1CkcO3YMycnJsFgsIInNmzfj7bffxo4dO1BaWorvvvsOHo9HkHEAGDVqFNatW4cNGzbgm2++CSDjAHDdVdcBb0AbBHh8n2/4lv+P4sSJE3jppZda/O/555/HgQMHxO/HH38cVVVVLa47ZMgQnDhx4pzqEBsbi1deeQUAsG7dOrz//vtn3KZpZlQDBgz8sjAs5AYMGDBwBkycOBELFy7EypUr0aNHDyiKgqioKBw5cgR/+ctf8PnnnyM5ORn33XcfbDYbqqurRdDi9OnTkZKSgoMHD+KCCy7A3//+dyQlJeGbb75BbW0trFYrvF4vbDYbTp48iccffxyPPfYYUlJSMGDAALz55pvYtGkTahtqgVHQSPlWAJdDs5AfApJik7B7226oqooGawOQDc1lxALgY0BSJbz07EsYO2bsL95WU6ZOwdPPPw1PlQeKXcG1V16LPy/48y9+3H9X7N69GwMHDsTBgweFlnpmZqZQkGloaEBNTQ1OnDgBu92OCy+8EK+++iry8vKwY8cO7N+/H3v27EFiYiLcbjfCwsJw6tQpOJ1O1NTUwGaz4bbbbsMbb7yBI0eO4Mcff0RISAhkWYbD4UBBQQG+/PJL7Nu3T+jHV1dXIy4uDrfffjtGjRrVYr119ykDBv4XYVjIDRgwYOA02L17N/Ly8s64nh5I+fzzz+Phhx/Giy++2Oq6V155JV5//fWAZYuXLIY72i2CJxcvWQxA09MGNCnAU6dOoaGhAffccw/effddPPbYY1BVFcuWLUNmZibq6urwr3/9CydOnEBZWRlycnKQkZEBSZLg9XqRnJwMs9mMvXv3ora2Fo8//jhkWRaBmfHx8di9ezdmz56tWefbtQdeBuCE5j9eqn3KDTIeuPsBAIDFYsHk8ZMhbZCATwF8DFht1l+NjAPAnxf8GQ2VDSCJhsqG/2kyDgAzZ87E7t27UVlZiZ49e4qsoz/88APWr1+PU6dOoaKiApWVldi/fz+eeeYZkdE0KSkJANCrVy9YLBZUV1fj3XffxeHDh7Fr1y78+OOPkGUZ1157LeLj43H48GHU1tbCZrMhKioKX3/9NZYvX4433ngDUVFRuPPOO4VEpk7GDx48iN69e6OwsBB5eXlYsWIFZs6cierqahQWFmLs2F+n3xgw8L8Og5AbMGDgvxbPP/88BgwYgPHjx7d5m8VLFmPcpHEoqyvTVEIcQF1DHQDgqb88BQCwWq0ANAWS5ORkDBs2DLGxsejWrRs6deoESZKgKAomTpwIr9eL+vp6DB48GOXl5QCA9PR0HDlyBA0NDdi1axf279+PI0eOCFeXEydOoLKyEvX19XjuuecgSRK+/PJL2G124HUA1QBeBkySCVOunYKxY8bC6/Wirq4Of17wZ3hrvdi/fz+efvJpZKVlwdPwv+u//Vvj4YcfRnR0NAAtsPXll19GUlKSkJN888030alTJyQlJcFkMuHaa69FcXEx7rzzTnz11VeIjo5Gz549QRJutxs2mw0AkJCQgIceekgkfnrvvfdQXl6O48eP45tvvsGhQ4dgtVqxb98+KIqC8vJyrFu3Dvfeey8uuugi3HrrrTh48CCWLFmCQYMGYd26dVi/fj0KCwvx8MMPw2azYd26dVi8ePFv03AGDPyPwSDkBgwY+I/BW2+9ha1bt0JRFCiKgvz8fDz99NOwWCwICgqCy+VCVlYWXnnlFaxfvx4rVqzAoEGDEBsbiy1btkBVVaFZHR4ejq+//loEvDkcDlgsFoy7YhxoItAPWlCkDM0fGsCUyVNgsVhQU1ODmpoaeL1erF27FosWLcK+ffvw+eefw263Y8uWLfB4PLjyyisBaLrezzzzDB555BEcOHAAqqoK3+zevXsjMzMTjz32GEjCarXiwgsvxCeffIK//e1vWLNmDbxeLzIzMzFr1iwsfGohhg8fDtYSc2bN0Ug6NNlEPWPnnj17EBUVhUmTJuGaa64RwaM/N3bv3o2UlBQxQDlXDB06NID4PfXUU3jxxRdx4MABJCcn4/XXX0dBQQHef//9gFmSNWvWYNq0aQH70mdHWoIu5dgU/jKTPxWLlyxGcmYyZEVGcmYy3n7nbaiqCrPZjAEDBuDvf/87Nm7cKAj50aNHsX79elG3b7/9Fjt27MC0adPg8Xhw/PhxvPnmm6ivrxdKPpIkCet5QUEBAOCll15C//79MX36dFx99dX47rvvMHLkSNhsNpw6dQqVlZW4/PLLIcsy7HY7SkpKsHr1anTq1AkLFy7E3Xffje+++07EORgwYODXhUHIDRgw8G+Fvn37IicnBwAwbdo0mEwmjB49Gunp6fjDH/6AhoYG5OXlITg4GDt37sSNN96Iuro63HTTTZgzZw6OHTuGf/zjH9izZw/MZjMGDhyI/v37Iy8vDw0NDXC73YiLi0NCQgImTpyIHTt2oKamBh9++CHWrl2ryfOlQiPhHgBjofliQ7NwJicnw2QyYdOmTfjoo4+EVTspKQnDhw/HHXfcgfHjx+O1117Dxo0bkZSUhJ07d2Lz5s2orq7GuHHjMGTIEHz99dd4/PHHkZSUhM6dO+PgwYPC+rlo0SI8++yzuPDCC3HgwAHs3LkTmZmZWLJkCV566SXMnz8fADBp0iR89tlnKCgowBdffCHSx3/66acoKChAUVERXn31VUyfPv3XvYhniU6dOuHgwYPi9/XXX4/x48cjNjYWffr0AQDcfPPNuPvuuzFs2DD88MMP+Pjjj9GxY0c88cQTZ3Wstrg9tTXwsSly2uXgiquuwJ5de0ATsWf7Htw842bs3bsXAHDdddehrq4Oe/fuhclkgslkwsKFC1FbW4ujR4/CZDJh9+7dmDRpElRVhcViQWxsLLp06QJZlmGz2VBTUwOSWLNmDUiipqYGALBr1y4MGjQIy5cvx8mTJwEAp06dgiRJAdlZnU6nmKkBtAHh559/jri4OFx55ZWnde8yYMDALwhdU/ffuRQXF9OAAQNnj7KyMi5YsCBg2bx581hZWSl+Dx48mGVlZS2u6w+LxcIjR46wpKSEq1evFssXLlzIqVOnsqSkhDNnzuQLL7xw2jrp67eE+++/nzt37mRubi5JsrS0lKqqctSoUYyOjiYAyrJMh8PBxMREut1uWiwWAmBBQQHbtWtHRVEIgFOnTmVycjI7derEoqIiqqpKm83GQYMGsV+/fszOzmZcXBwdDgdHjhzJzp07c/78+QRAqCA6gJBBBENbBjA0IpQkef7553P58uV87733GBYWxoKCAjocDqakpHDixImnvyi/IlauXMkbb7zxF9v/ihUraDKZKMsyzWYzJUmi1Wql0+lkr169KMsyJUmiJEnctWsXn332WdrtdtGeNpuNd955JwFQkiSqqsqgoCDm5+czPj6e0dHRlCSJzz//PCVJoizLYluLxcLo6Gi6XC5269aNo0aNotVqpaqqtFgs7NGjB9PT0zlw4EB27NiR7du3p6qq7NChA6Ojo+lwOCjLMu12O6Ojo/nAAw+I8zpdH20Jb731Fjdu3EhZkQkFhAnEBBCzQIxq7D+pqamivXJyckTftVqtlGWZMTExTE5O5ty5cxkREUEAdDqdLCkpoclk4nXXXcfk5GQCYHJyMuPi4hgeHk6bzca+ffsyPz+fQUFBtNlszM3NZVRUFMPCwvjNN98wISGB5513Hg8fPsyCggKqqsqnnnqKu3fvZkNDA0ny//7v/zh9+nSSZEhICOvq6n7W/mLAwH8KAKzhr8x1f3Oy3ZZiEHIDBs4NpaWlgtzqSEpK4pEjR1pct127dq3uS9/udITcf3lraEp2SktLmZmZycTEREFcfq0iSdJpfzctNpuNsixTVVXGxsby3Xff5ejRo0myzef/a2Hbtm1MSkripk2bzmn7srIyPv3006ddZ8WKFaJtXnnlFebn5wtSLMsyH3/8cS5ZsoSSJLGwsJDjxo0jAO7evZuHDx+mw+Hgk08+yeDgYFqtVs6aNYvJyckMCgrirFmz2KVLlxavgT85l2WZKSkplCSJycnJYkAWGRnJyMhIAqCqqly4cCElSWJYWJjYzmazMS0tjQDocDhYUFDAoqIiWq1Wut1uFhQUcNKkSezQoQNDQ0MZGhrK/Px8vv3225wwYQIXL17MUaNG0eVyafWRfIM4FUQUiEE+Ym4DIUHUW5IkXnDBBRw8eLCot9vtZrdu3Wi322m323nRRRcxPz+fYWFhjI6OpqqqfOmll8QgVb+H586dy+DgYPG76f2l37der5czZsxgbm4u8/Ly+Morr5Akn3/+eebm5rKwsJA9e/bkrl27SJK33XYbs7OzOWbMmHPqPwYM/CfjtyDkhuyhAQP/xRgyZAg++OADWK1WOBwOBAUFYc+ePZAkCW63G6mpqfjmm2+QnZ2NEydO4MCBA3A6nSCJuro6mM1mOJ1OREREYMOGDZg9e7bwdW5o0JQ0OnTogOLiYnz//ffIzMxERkYGNm7cCFmWsWbNGsiyjIEDB+KDDz7A999/j+effx7vvvsuqqqqsHPnTvTr1w9PP/00TCZTwNR6AELQqL39EyFJEpo+99LS0kQCH6HnrcMEwFetAQMG4OOPP4bJZEJISAhcLhfKysrw9ddf4+KLLxZydP369cOuXbtE0iB/NM3q+e+ImpoaTJo0CXPmzEF6enqr661cuRJ9+/aFJEmoq6vD0KFD8eWXX+LkyZNCzWPTpk1o3749LBYL2rVrhw0bNiAyMhIXXngh3nnnHaxevRq5ublwu90YNmwYNm3ahI0bN6Kurg7Hjh0DSRQVFeG7775DQ0ODuH5utxvHjx8HoAXXNjQ0CP9+s9kMq9UKSZJQUVEBSZIQEhKCo0ePtnj9dVgsFuTn5+Obb76BLMvIysqCx+PBvn374HK5cODAASQkJAh1EwBwOBw4fvx4832GQwsKPub77QXsdrv4u76+Hh6PB9nZ2Zg9ezYmTJiAyy+/HGvWrIHNZsOSJUuQnp4usrZefPHFGDlypIgH2L59O0aNGvWLxQcYMPC/jP9I2UNJkhIkSVouSdImSZI2SpI03bfcLUnSR5Ikbfd9hvqWS5IkPSFJ0g5JkjZIktThp9bBgAEDLaN79+6QZRnV1dX4y1/+AkVR4HK5UFxcjLq6OrzzzjtwuVzYsmULXnnlFQQHB6OyshINDQ3o168fPB4PnnnmGaxfvx6SJCE8PBwpKSlQFAWjR4/GW2+9JUjRl19+iaqqKixZsgQZGRn4+uuvYTKZ8Mwzz0BRlIB6rVu3Dq+++iq+++47/O1vf4PVaj19YGB563+dLVoiYzt37kRVVVVzMg4IMq6vB2ikXs/MWV9fjwsuuAAbN24ESURHR2Px4sVYsWIF3nnnHQBaoOGwYcPQr18/9O/f/2c7l5UrVwa02wUXXIA+ffpg+vTpsFgssNvtCA0NBaCRTX9f8szMTEyfPh0NDQ3Iz8+HzWaDy+XCPffcg06dOuGll15CbGwsUlJSYDabYbPZMGDAAEhmCZIqQZIk9OrVCw0NDfB6vVi6dCk+//xz0YYk8eGHHzYbZJ133nn43e9+h6+//hoHDx4MuB4NDQ0YPXo0Dh48iNjYWMTGxgIAXC6XJkHpG0ja7faAgUJOTg5CQkJQUlICQCO71dXVqK2thaqqcLlcQtlEP57JZIIsy5g5c6Yg17W1taisrITX64UkSVi3bh0OHjyIiooKHD16FACwb98+REZGIiQkBDU1NXjmmWeQn58vgjRhg0bEawGMBODVfidlJKFHjx6w2Wx4/vnnMW7cOERGRuKaa65BamoqSCI9PR3fffcdbrjhBtx0000B7ZaWlobg4GCsW7cOALBw4UJcddVVZ9VfDBgw8G+Mn2piBxADoIPvuxPANgDtADwCYKZv+UwAf/B9HwLgA2iPrK4AvjrTMQyXFQMGzg2LFy8mAN52220cOXIkR44cyeDgYHbv3p2DBg3iO++8w/DwcFqtVpaWltJkMjE2NpYzZsxgQUEBhw4dynnz5pEkAXDfvn0MDg6my+ViVFQU+/Xrx7CwME6dOpUWi4Xh4eE8//zzWVZWxsTEROHGsX79eubm5nLR4kUMi9JcBiCDMLfRvUQ6d7cUk8l0RpcURVEoy3KAj3JcXBxdLhezs7PFslGjRlGWZQYFBbFTp07Mzc3lwIED2adPHwYHB/PKK6/kSy+9xHfeeYfnnXceMzIyWFFRwYULFzIuLo7Hjh37Wa/vihUrqCgKs7KyGB4eLs5RkiT+4Q9/4I4dO1hUVMRt27bx/fffpyRJJMk9e/ZQlmX26tWLYWFhVFWVK1eupMPhoCRJjIqKYlFREUNCQpiens65c+cyPiFeaweHzyUjFUR4Yxvm5ORQlmXhsgKfn3eHDh2Ez/MVV1zB2NhY1tfXs7KykgD47bffMjQ0lA6Hg3Hxcdq+ffsMCgoiAJaUlAh/6+DgYNrtdrZv316s1759eyqKwn79+lFVVZrNZrrdbpb0KQm4xqqqiv4gSRJtNhvz8vIC1nn77beZmZlJVVVJkoqi0GQyccuWLcLlZe3atczJySEAvvvuu8zPzxfbB/StkMbvo0aPYs+ePRkXF8d27doJX/fMzEzedtttdLvd3LlzJ0myrq6ObrebJDlnzhzOnTuXJLlo0SJOmzaNDQ0NTE1N5dGjR3/W/mTAgAEN+A1cVn6yhZzkQZLf+r6XA9gMIA7ARdCSPMP3Odz3/SIAL/rO+UsAIZIkxfzUehgwYEDLkihbZEgWzZp49fVXw2QyoX379li1ahW+++47sa5uFdQte7q19MCBA3jsscewadMmrF27Fn/961+hqioAiAyDp06dwuHDh/HZZ5/h2LFjeO6551BbW4tTp05h+fLlSEtLww8//ICVK1fizjvvxPz587F582ZcccUVOHbUN4fvBdAJmoKJdIYT+wmedWazuXE3rbgqeDweYQXWsX//flRUVASkjo+MjIQkSXA4HEhPT4fX60VISIhIf/7ll1/i4YcfxvTp0/HFF1+gpqYGP/zwAwBg4MCBcLvd534iLWDDhg3weDyw2WyIiIiAxWJBfHw8rFYrfv/73yM9PR0kkZGRgX/84x8giezsbJSUlMBms6GqqgpVVVUwm824+eabERISApJQVRVerxcVFRXo2rUr9u3bh33792kHPR9APYDO0BRofG8Rh8OBtLQ0oQJis9lQW1srXCreffdddOvWDYcOHYLZbEZQUBBMJhMSEhJEXfbv26+5CPkumZ4psqamRsyynDx5EtXV1QEp52trawEAn3zyCerr61FfX4/jx4/js08/E/XzeDyor68XFvvQ0FBUV1fj+++/hyRJkGVZ9A+9z+jtGxYWhscff1z8v3XrVtjtdkiShKVLlwrVEpLoXaIldoIK4AQgyZoKT/du3bFmzRocPnwYK1euxKFDh9CpUyfs2bMHTz31FBwOR6OVHQj4rmPEiBH44IMP8Le//Q3FxcUICws7yx5jwICBf1f8rLKHkiQlAygC8BWAKJK6jtUhAFG+73EA9vptts+3zIABAz8BU6ZOwZPPPgnaqKVVnwXUlNSgoaEBFZUVGDt2LHbs2AGz2Yzq6mqsXbsWnTt3FtuPGzcOgJb0Jj8/H+Hh4Th48CC2bduG119/XWSY1EESHo8HkiQJ6bW6ujrU1tbi+PHjiI2NhaIo+Oc//4nnnntO21YB4PKr9L+gTe1LAEJ/mXapqqpq9T99UAJopK+yshKKooh04l6vF3//+9/FOqGhofB4PPB4PPB6vdi3b59IAlRTU4O6ujq88cYb6N69Ozp16oSHH35YSDh+/vnnAfrY3bt3b7VeycnJwkXCH800rt9+G7IsY+3atfjmm2/g8Xiwd+9eVFdXQ5ZlpKam4r333kNycrKQsysrK8PRo0dRW1uLSZMmgSQURcGaNWuEG8fRo0exb98+4VO9Y8cObQAFaHOgAPAWgKXQ5jwBvPbaa9iwYQMAzYVk9OjRuPfeezFv3jwMHDgQOTk5mDJlimg7PWFSeHi4Vg+ZwAQAvwdwB4C7AYzQ9r127VoAmo54amoqevfujXvvvVckxYmNjUV6ejoiIiJgMpmE/CNkaPV2AAgC4OcVpZNrp9OJnJwcuN1ueL1e3HHHHdi0aRNIoqSkBJIkoaysTGRzPXbsGBYuXIiysjIA2kBDlzUMDw/HVVdqbiQRIRG46667EBcbh7CwMEybNk0MyCorK1FeXo5t27ahuLgYTqcTsizj1VdfBQC8+uqr6NatW7Prb7VaMWjQIEyePNlwVzFg4L8NP5epHdrj7hsAl/h+n2jyf5nv828AevotXwagYwv7uxbAGgBrEhMTz2HCwYCB/y0odoUI9ak63O0rV0CTYfMpTaSmpjIoKIiSJDElJYUkGRERQavVyjlz5gi1CYvFIlwf0Ab1Eb2YzeZzdi2BzfdpaeKqYgURB+I83zI/lwZ/pY2QkBDhotK+fXvxX2hoqKj/3LlzaTKZhPuDJEk0mUxUVZUOh0O4WqiqyvT0dLFvs9nM3NxcfvLJJ0xNTaXT6WR+fj5TU1N52WWXMTc3l7m5uXS5XIyIiOC1117L/v37s2fPniQ15QtFUdospdeSEs6ixYuoOHzXWIL2adLaISsrK8CFA03ccMxmM0ePHt3MjUeW5QA3C6vVSgC02+1CDUR3aRL7HuRzNzIFuhN16NBB1MFut9NisTAoKIi9evViVFQUH3300VbPV+97mOXXd+/2/QbO6j4oLy8nSa1eCoiOTfbpux+Sk5OFfGZcXBxNJhPNZjNJ8q677qLD4SCgqbVcd911LC4uZlRUFC0WC202G10uF5OSkkiSy5Yto91up6IodLlcVFWVbreb2dnZVBSFERERLCgoYGxsLHv06CHkNoODg2mz2bhgwQImJSXxtttuY/v27dmxY0du375dtI3uskKSX3zxBePi4oRUoQEDBn5+4D9V9hDa5Nw/APzOb9lWADG+7zEAtvq+/wXA5S2t11oxfMgNGDgzBDlqgdRIstRs/clTJlO2yYIEq2aNjE6cOJELFy7kRRddxISEBMqyzOrqap44cYIxMTHNiLROYgGwsLCQF1988emJt8PvuwzC7SNJagvr5oDIAJHr569sQjOfckmS6HQ6BRFNSUmhqqpUFEWQSwDMzc0VxN3fXxw+P+TNmzczKiqKkiRRURRBVnVyqp+roii88MILmZWVxUGDBhHQ9KJjYmKEL7c/+dWl9ZxOJy0WixjwrFy5ksuXL2f37t0ZGRlJVVWpqioTEhL40UcfiWN06NCBJrNJay9/Qi5rx3C73YGkuUlJTEwUcn9NB067d++m1WptNujSBy0tFkvzZaGhoezYsaO4HrqWtizLdDqdLCws5KlTp/jVV1/RbrezoKCAubm5/Pzzz7X+a0HgYPJu328ZLCsrO23f95ecfOyxxzSdbV9/xqU/jeRPmDCBS5cubfuN2EaUlZUxIyODl1566VltN3fuXN51110/e30MGDDQiN+CkP8cKisSgGcBbCb5mN9f70KbgITv8x2/5eN9aitdAZxko2uLAQMGzhGKXdHkAX9o8scPQGJaYsCiKVOn4MkXn4T3Mi8wC8BooF7S0q6vXr0aALB582YoigIvvbDZbCjoVNDsmJIkITs7W/wuLS3FP/7xj9NXtNrvezg0f2RAcy8wI9Cf/AiAo9BcDUIBREAb/rNxFVVV0aFDB3z66acgNTnGXbt2oa6uDldccQVeeOEFeDwekMT333+PsrIyVFZWIjk5GSdOnEBSUhLsdjs2bNiA7OxsHDp0CF6vFw0NDVi2bBkAzXXlgQceQGhoKMLCwtChQwesWrUKW7duRYcOHWC1WpGdnY3KykocPXoUNpsNY8eOhclkwuDBg5GdnQ2LxQKv14u3334bJ06cAABceeWVALQU8FdccQW2bNkCSZKwd+9eZGdnY+3atfjggw/wzTffoKG+QWujYQDu8n1qrv0oryhHcHBwgAvOLbfcgoKCAsiyjL179+K8884TfskJCQlCgq9Lly4ICwtDSEgIzGYz3G43bDab8LU2mUyw2+1CMQZAgPJMr969oCgK6uvrMWvWLKiqClVV8dFHH+H48eOIiYnB0KFDcdFFF8Fms+Gdd95BUFAQ1q1bh/Xr16OwsBAA0L93f+ANAKXQMqSWAngD6N+/P2666SbMmDEDmzZtCuhK11xzTbNlN998M9atW4eFCxdq/ujfIRA/QGRe/S0REhKCbdu2YenSpW3e5uKLL8aLL774b5951YABA+eAn8roAfSE9nrcAGCdrwwBEAbNHWU7gI8BuH3rSwAWANgJ7VHZzF2laTEs5AYMnBmTp0zWrIzBaMwSOAE0hZi4aPGigHUVu9LcGjmi0dpsMpk0i2sL1mi96JZmf8uqrlzh/1u3Lp/WnUVuZbkLRBKIEp8FfYTPOhvSaFGXJIkPPvhgm9qooqKCqamp4hyLi4uFpdzhcHDMmDG02+0MDQ2l2+0W2UFVVWVYWBjj4+OpKAqtVisVRaHFYmFwcDDNZjMfeeQR4fYSGhrKxYsX0263c+DAgVQURRyzoKAgQJUjKSmJFouFO3fuZGlpKUNDQ2m32/nYY48JpQ2S2rWIb3LNzgtsQ12VxG63i4yQurLIVVddJc7VYrFQVVUxs+B0OoWLkl50dxaLxUJFUbh8+XJGRUURABctWiT6yIIFC4RrS0FBgZhJKC8vZ2RkJIODgzl//nzu3buXJDly5EgCYEJCAnv16sXY2Fj+/ve/Z2RkZKBKiQWMiYlhUFAQ3W434+PjOXToUNrtdjocDoaHh/OWW24hSUZFRbFDhw50OBw0mUy89957WVtbS8Xk25/i60sFIILA/gP7/0x3Xduxe/dujho16lc/rgEDBs4e+E91Wfmli0HIDRjQsH79esbExLT6/+QpkymZJSEnGBQW1IyMk9RISgSIoSCG+8jdgEZyt3r1aoZFh2mk1+LnJmFtXMflcglyl5qaKlw7TCYTJUliz549+dBDDwW4QOj/62ReX05qPtIWm+YmYVJ9UoVO3/HsIGJB9PK5rPjqVVJSQofD0eb2e/zxx5mQkCBSgk+cOJHh4eG02+0kyf4D+zcSQhXMztEkD2NiYjhlyhQ6nU4GBQWxc+fOBLSskRMnTqSiKDz//PNZWFjI2NhYms1mLlmyhHa7nampqXS73WJAsnPnTr7++usEwI8//phvvPEGJUniF198wdLSUrrd7pYJOaC57/gT8iG+uvoGX/r1ufTSS0WsQHJyMiMiIhgXFydcaPLz8xkcHMwbbriBubm5NJlMDA4OZnh4OCVJYlZWFseMGSMGJBaLhcuXL2dSUhIVRRGDruDgYD777LMEwNmzZ5MkVVVlly5dOHfuXLZr147du3cX+/GPS9D7j14nRVEYFxfHhIQEMVix2Ww0mUzs16+fGNg5HA5GRUUJ96Ho6Gixz9DQUOEfX1RU1Mx9SK/Db4EVK1aIvk6S8+bNY2VlZYvrLl++nAMHDmTnzp2Zn5/P4uJiFhQUiOyaZ4v9+/dzxIgR57StAQP/i/gtCPnPqrJiwICBXxanTp0SiXhawp8X/BneWi8aqrQsmuVHyzF2zNhm68lWWXMV6QSg0LdwnfZhsWnz+ceO++QJrWh0kyiG5gYATcYvLCwMdrsdsiyjuLgYGRkZcDgckGUZ//rXv3DnnXciPT0diYmJcLlccLlcaGhoQGFhISZMmABVVWGz2VBYWIi5j8xFYnwiVq9ejUfnPgpJkqDW+nwy6gEcALASmmpGA+B2uNG5c2cEBwe3uf2OHTuGAwcOIC4uDunp6Vi5ciXq6zWfmQHnDcCyL5Y1Kr6MAbbs2AJAk8zTE/pUVlYKN4nq6mr8+OOPIInq6mokJSXhxIkTIIlVq1bB6/UiJiYGYWFhIAmn04l77rkHK1euhCRJKC0tFYlu/u///k871fp6VFVVIS4uDlVVVdi9ezcAwGQ2Bbr7eKC58IRAU69J8V0fAG+//Ta8Xi+ys7MRHx+PiooKuFwukSHUbrcjIyMDL7/8MkJCQjB06FCYzWb84Q9/AElccMEFuOOOO5CWlgZJkpCcnIw+ffpg7969iIiIQHV1NZxOJyRJwh/+8AckJCRg4cKFyM/Ph9frRU5ODrxeL+rq6lBWVoaKigpERERgwoQJuOSSSwBoijwdOnRAcHCwyMBZVlYmlGsuu+wyAJoKSkhISIAsoZ7B02Qy4dprrxWuOH/961+FZOHw4cPRvn17SJIEm80Gu90ustD+FhgzZgxqa2ths9nQqVMnzJw5E2FhYbBarbj00ksBALt370ZkZCQuuugifPzxx9i+fTuuvvpq7NixA5WVlUhJSQGg9cGJEyeic+fOKCoqEu5EGzduROfOnVFYWIj8/Hxs374dADBlyhSUlpYiNzcXTz/99G9y/gYMGDgDfu0RwLkUw0JuwICGCy64oJmlUVVVpqSkCLcL3QrdoUMHSpIklD/gszoOGDBAuDZAdxfR3VIkzVqZlZXV+H+IzzqdByIFwvoeHR3NW2+9lV26dKHL5WJBQQEfe+wxERA6YMAAJiUl8f/+7//4xz/+kYWFhezSpYtIjrNjxw4OGjSIHTp0YM+ePbl582aSWhDdjTfeyG7dujElJUUE1OnHad++Pf/4xz+SJF9++WUOGzas1faaPGWy5p4DULEr7NmzJ6dPn86XXnqJvXv35j333MOkpCTa7fbGoELVN3twN4h8iHabMWOGcKmIj48XCia5ubkENIWS2NhY9urVi6qq8pZbbhGuHz179mRWVhZVVWV4eLgIsLzwwgu5fPlyxsTEsKCggOnp6bRYLHS73bzwwgvFsSVJ0uros9wjCpp6iAlakOzdIDr7ZhJ81y09PZ3BwcE0mUwMCgriM888w4qKCgYFBYl6SJLExMREXnLJJfzoo48CLMn6uerrREZGNnNRUlWVGRkZzM7ObrY8KSmJkVGRzSzU/sGnevvAbwZFVVXKssxOnToJq3dxcbGwoi9cuJBjx46loig0m80iCBcAvV6vsMZfffXVbNeuHQEtyVN4eLiw5v/aSExMFPUCGgOK9bawWq3s06eP+L9Xr15iRkV3K4qKiqLL5WJRURGzsrL417/+lSSZkJDA0NBQ4VZ133338bzzzmNKSgrnz59Pkly7di1zc3NZVVXFzMxMdunShUVFRSwqKuK//vWvX709DBj4dwcMlxWDkBswcDqsWLFCvLSXLVvG7t27C3k7AExNTWWPHj0EKUpMTBTk3WKxcOTIkXQ6nSwoKGjRZ/uuWXexe/fuXLVqFWXFp0IyCMRMHzmNhcg+mJKSwsTERL766qscOnSoqOPChQuZlpbGU6dO8fDhw3S5XHzyySdJkjfddJPI/NmvXz9u27aNJPnll1+yb9++JDVCfumll9Lj8XDjxo1MS0sjqU3jDx06lJ9//jnz8/PZvn179urVS8jDNcXkKZOJIAT408OquUr8+OOPfO+99zh48GBmZWUJhRbMakLIr9HO9Y477mBFRQUVReHHH3/Mbdu20W63MyQkRNRZHziUlpZSVVVBBgFNBUW/LomJiTSbzZQkiddffz0PHz4coGgiSRI/+OADBgcHi+vY0rWC7htt9nNd8Ssul4uyLNPlcgkynZaWJjKPOhwOKopCt9tNt9vdomuH7v5hNpuFm0phYSGTkpIIaL7qhYWFghxPmDChsd/4FyeIZN/Azzf4mzt3LiVJEn3R5XLRbDYzMTGRnTp14qxZswRh1/35JUni+PHjeeWVV4r65eXlCTepESNGiHPp3bu3uC+efvpp5ubmioHpggULWuwzCxcuFNKUTz75JF944YVmsoPngpCQENrtdkqSxOzsbEG29X4QFBTEK664Qiy3Wq3Cz3/IkCEcOHAgi4uLmZeXR5KMi4tjZGSkpibji1mIjY1lv379aLFYeM899/DLL79kZGQkSXL69Om0WCzMz8+ny+Xip59+SpLctm0bjferAQPNYRByg5AbMHBarFixQpCUzMzMZlJ1TqdTBLalp6dz/PjxNJlMTE5OZteuXRkfH0+TySTk6RwOhyCjkiTxtddeY0FBAbOyshqDMDN95LQARLHmx2x1WRkUFMTzzz+fTzzxRDNCfs0114jfCQkJ3LdvH0ny2Wef5fTp01leXk6r1cqCggJRsrOzSWrkdtGiRr/3oKAgko2EvK1oMXB1gEYIrVYrbTYbs7Ky+Lvf/Y6hoaHauaY0IeQTtPUjIiKYm5vLSy+9lMnJyczPz2dISEirhFy3nv/ud7/jiBEjBHkFwKeeeoper5dms1lcG7vdzh9++IE7d+6kxWJhYWGhSOlusVgYEhIiyLHVauV5553XaDFvKfBWarRwx8TECDJosVg4adIkyrLMxMREoYN9Jp15/dj+lnPd0ttWjfqWij7Y8N+/Xuff//73QmbT7Xbz/PPPZ/v27YUkpd5/Z8+eLWYRli5dKrZJSkpiXFycuBYAhE95bm5us/7S0NAQQMh1nA0hf+GFF+hyuWiz2Wiz2RgSFtLqOftbyjt27Chmv/TicDiYkJBAi8Ui7kWTyST00WVZ5sCBA6mqKvft28fExER27tyZwcHBwmKuKAoHDRrE4uJi5uTkkCR79OjBAQMGMC8vjwUFBbTZbG2+p86EsrKyVgc7bcE777zDhx566GerjwED54rfgpD7vEENGDDw74jFSxbj6uuuRm1dLVAHIfnndDrRqVMn1NTUIC4uDlu2bEFZWRlKSkpQUlKCP/zhD/B6vSgqKsKLL76IvXv3ory8HBUVFZBlGQcPHoQsyxgyZAj27duH1atXIywsDCkpKdiyZQvuuecevPvuu7BYLVj++XJNgo4AKgDsBp55/hlcd+11UBQFHo+nWb0tlkZdOVmWxW9ZltHQ0CBSzq9bt67F8/bfXns2nj08VR4gscnCbgA+1ny//fHHP/6x0Yd8DLTtSqHJ7g3oj4//+fFpj/X888+L78nJycjLy8PmzZuFbKLFYkFcXBy2b9+OJ554Ao899hgkSYLD4cCIESPw5JNPIikpSZzrunXrEBWtJTeeNWsWjh07hs8++wzffvstzGYz1q9fL9LN+0sQAoAapEL1ar75FRUVsFqtOHToEBwOByorK0FqmTn379/f4rXToftlkxQSiICWLfPkyZMwmUwtbx8EwA7gODQ/dwuAmuarBQcH4+TJkwC0DJQulwsTJ07EQw89hEOHDmHevHlITk5GWFgYgoKC0KNHDyxatAihoaGoqqqCx+NBZWUl7r33XrHPcePGob6+HmazGT/88IMm2+mXXba2thYAsGXLFjgcDnH/HD9+HBERESgvL0d9fT1WrFiB48ePo66uDnFxcSgtLcWDDz6I4OBg3HzzzbDZbHj66adRV1eH9PR0vPTSSygtLcX999+PNWvWYPLUyVj26TJUH/P1Mwna/QMtRsBms6G6uhphYWE4fvw41qxZA1VVYbVaUVNTg7y8PJSWlsLj8aC2thZutxvdu3cXyzp16gS73Y433ngDJLF161aUl5fD7XZj8ODBOHToEL755hshR1lVVQVZlrFlyxZ8+eWXGDFiBP7xj3/A6/XCavVLX/oTceLECfz5z3/G5MmTMWfOHDidTpSXl6N3794YMGBAs/XffvtteDweTVa1oADPP/88CgsL8fDDD2PmzJlnffzu3btj1apV2L17N1atWoUxY8b8HKdlwMCvAiOo04CBf1MsXrIY4yaNQ61SC1wOTS98kPbfqVOn8PXXX6OyshI1NTUiMBHQyJ8kSfjxxx+xbNkySJIEs9mMBx98EADQ0NCAo0ePwmKxoLS0FDt37oQkSZAkCeXl5fB4PMjLy8PBgwdRUV6BXt16IWFlArABCDoYhIF9BgYEitpsNpSXl5/VublcLqSkpAgNZpJYv379abfRX+4AMGTIEKHlHRQUBAB4//33kZycLNaXzTLw9yY7+cGn194CPv7nx+jfrT/wCoD7ALwC9O+mkfE+ffpgzZo1rdZtwHkDIFm1NpSsEnbu3ImoqCisW7cO69atQ2FhIdLT0wEA1113HUaMGIExY8Zg+PDhIAlVVUGVQD60NO8hwI+HftSq/MMP+Nvf/oZDhw4B0Ihs7969MXbsWKiqipCQkIC6TLxiotBmr62tRZ8+fRASEgKXywVJkrBjxw54PB6NTJsgtMybgmSArrlO0HUS3dDQ0DjV6o9KAIehkfGmJh8p8GdISAjCw8NRUFCA1NRUXHbZZWhoaMATTzyByZMnY+3ataiqqsKHH36Iu+66Cx06dEBCQgJCQ0MBAGvXrsW1116L4OBgFBcXIzg4GHFxcVi7di1UVUVqaiqWLFmCrKwsAMAFF1wAp9OJ4OBgVFRUICEhAR07dsSVV16Jr7/+WhDgnj17wuPx4MiRI4iIiEBycjL69u2L2267DdOnT8fx48exevVqrF+/Hjk5OXj22WfxySefYOTIkcjLy8Oyj5Zp5w9oAcJNmqi6uhqyLOPo0aPweDyQZRn19fViwPD999+jqqoKR44cgdlsRkpKCr755htUV1fj4MGD+Pbbb/HKK6/g2LFjaGhowMMPP4yamhqh875y5UpUVVWhoqICu3fvRlVVFbZv346ZM2ciOjoabrcbsizjpZdeOu2g7Gywe/du5OXlYePGjTCZTLj//vsxc+ZMPPLII5g4cSIGDRqEmTNnoqCgAIqiwGw24+KLL8bo0aNx4YUXIj4+Hv/4xz+wfft25ObmIi0tDSaTCYqiwOFw4G9/+xvuvvtuBAcHQ1VVyLKMoKAgPPHEE6isrMTQoUNRWVmJvLw8PPfcc1iyZEmL9SQZMEgzYODfBQYhN2Dg3xR33n0naCEwHJqChgIgDYLkbN++HceOHcP69evRvn17AMBnn32GBx98EA0NDejQoQM+//xzAFrynFmzZqG+vh6yLKO2thYmkwlerxdlZWWor69HfX09rr/+esiyjKuvvhpWqxU//PADvv/+e8SGx2LMmDEYf8V4ZGZmBtQzLS0NiqKgoKAA8+bNa/P5LV68GM8++ywKCgqQm5sbmHimBeTn54vjDBw4UBBxHZs2bQoYGFx3zXXANjRLNDPh8gn485//DAD49NNPccEFFwAAHn/8cbz79rtgjUYyr77iajzx+BNnPI+U1BQsW7EM6A7gegCjgcr6Suzbtw9btmzBqFGjsGbNGsTExECSJCxYsACvv/46unbtiuPHj2PQoEEaye0H4BJfXc3QFFQAPPfcc9izZ49Q1+ncuTPefvttPPPMM5BlGU6nEyaTCYMGDYIkSXjh+RdQX1+PnJwcAMCLL76I6upqHDhwACaTCRs2bNAISSQ0xRqdjzV5G/irmgBo1t6q2sjkFcVvkENoVnErgAwAtX4bhWnH6du3L+rr61FdXQ2S2L9/PyZOnIgBAwagqKgIr776KoqLi5GTk4Mbb7xRKOnY7XYsWbIEXbt2hcViQXx8PL7++mt069YN27dvh8vlQnl5OeLj48U9sHXrVqE2snLlSvTr1w+KomDQoEE4cOAAPvvsM7z88svo1KkT6urqUFdXh8rKSsiyDJfLhR9//BHZ2dno3Lkzjh07JhIt9erVC2FhYXjgwQcwbdo0TL9lOpYuXYq6ujrtHh3na9Oh0H77DU5UVYXZbEZoaChkWYbX64UkSdq0tcmEnj17Ij4+HkOHDoXFYoHVasWiRYsgSRJOnTolrmuHDh2gKIogn8XFxejYsSNmzJiBbt26ISIiAl999RU+/fRTZGRk4O2338by5cvxr3/9CwUFBWKm4OdCZWUlAAg1H0VRMGzYMOzduxeffPIJnnzySdH/goODIUmSmDGLiooCSbz//vt45513EBERIch3XV0dRo4ciZMnT4IkzGYzxo0bh8rKSkyfPh3Tpk1DbGwsduzYgaioKDz66KP44IMPkJKSgnnz5qFz585ITk7G+PHjkZeXh+TkZIwd22hU+Otf/4qbb775Z2sHAwbOCb+2j8y5FMOH3MD/IiRZ0nyDZzXxg77R52c6AQyLCmNubi7z8/M5YcKEZmm+V61axYyMDBYWFvLOO+9kUlISSS1BzogRI5iTk8OLL76YBQUFIsDyt0BTP91HHnmEffr0odlsZkxMDCMjIzl06FBGRkbSZrNRlmXGxsYyLi6OkiSxXbt2wjfX4XAwPz9faFjDLwBSD4A1m800m810uVyMj49nZGQkZVlmQkKC0OO+9NJLWV5eTrIxNXuHDh141VVXiXqOGjWq0fe8AMRI3zVKa+4v3b9/fyqKwqCgIKGlbbVaSTJwXRVEDIgu2m/df/y6664jAN58880iFkBRFHbo0IG33Xab8ClPSEjgwYMHSZLjxo0TwYO5ubm02W2NQZc2aAotppb9u5OSkgJ8xP19xxVFETriISEhzYNCZd/+hwf6t6sWlXFxcZw/fz4dDgfDwsIoSRJvvvlmlpSUiPZYsmQJ8/LymJuby9tuu40k2adfn2Z13LhxIwsLC3n++efT5XIxODiYwcHBPHnypFCmac3HvWmwbGvrttQGZ0x01VKJbO6Xb7fbWVRUxOzsbKHnr6oqX3/9dX7//fdcvnw53W43V69eTZJ8++236XQ6+eWXX3Lz5s1UFIXJyckkydDQUI4ZM4YbN25kYmIi4+PjWVNTw2PHjnH37t2t3nu7d+/mq6++es73bl5eHkNCQrhw4cIA9Rjd1z8zM7PF9ti8ebNox4iICC5btkzEs+g6/3o76e2fnJzMkJCQgLgA/X5OSkqiyWTiBx98IHTc09LS6PV6OXfuXALgF198wa1bt7KwsJCpqakiH0G3bt24YcOGc24DA/99gBHUaRByAwZ0JGUkaQl5JqB5Rk0fUZfkc0ty0tDQwOrqapKa/GBycjJra2vPej+lpaXMzc2lw+Hg/PnzmZ2dzdjY2GaBcTpaC5CbMGGCCOokteBPp9PJdu3aiaC22NhY2u12BgcH0+Vyce/evQwLC6MsyyQplEDq6+vpdDpps9k4depUXnbZZbRYLLzwwgsZFxfXjIg1zTRqtVoZHh5ORVHocDiYm5srpOf0dVRVDSRkfRpJP9BIQvVzU1WVaWlpdDqdvOqqq5idnc1hw4YxKiqqUYKyAEQ6iN+D6Ne4D6fTycGDB5NkQBKkUaNG0WQyMSwsjFarlXa7nbm5uczIyODdd98tkuTY7XZecsklXLR4kVY/XRrRhsZETwpaJE3+xT+Zk3/Rr8HptnU6nc0IvU6mgOZJgmw2G9esWUMAIsGRyOYaB5GlVR9smM1mQaj1JEf6tfrd737Xar1aq7feJ/xJX3JycrPgUzH40NuzaXCty2+ZHLh/XXmlU6dOTE5O5muvvcZevXqJug8cOLAZIa+pqWHXrl1psVjocrkYFxcnAlQTExMDAr2Dg4PZvn17dujQgV988UWze668vJz9+vVjZGQkx44de9b3/qxZszjz9pkMiw1r1iZ6O0mSxCFDhgjiLcsyMzIyqCgKv//+e9E3IiIiuHz5cqHk49/Oc+bMEW2v33smk4kvvPCCuAdVVeWxY8fEcyI0NJQul4tWq5UHDx7kpk2baDKZWFdXx9///vf8v//7P15zzTV88803uXnzZnbs2PGsz9/AfzcMQm4QcgMGBBYtXkTJLhHBCJTuc0CzwE4AkzKSzmnfp06dYnFxsZAPfP/9989pP/6EPDU1lWlpaSIdvSzL/Oijj9i9e3cmJCTwscceY0JCAlNTU9m+fXt26dKF69evJ6mRVqfTycTERKanpzMpKYk2m43Z2dm0Wq3My8tjeHg4VVVlVlYWo6Oj+fHHH7NLly6UJInXXXddABFpiWS1SKbaUPwlCdtUghq/+2eK1NUx9BIREUFFUWi1WgNTxqdBu+Zm0B3mpqqqjI6O5vDhw6mqakDbp6amMi8vj++//z7tdjvT09P59ddfMz4+nnPmzOGYMWMYFRXF+Ph4xibHalbaIt8xdGKe6zueAk3S0o9UyrIsrJ16+9rtdoaFhYmBTWpqKqOjoxkVFcU+ffowNTWVNpuNS5cu5V/+8heqqkqn08m0tDRBcKdOnUpZlkV2TkVRaDKZqKoqExISGB0dzdLSUgIQEoso9tUrFgEZY5uW8PBwjh8//oyk+2yLfxsEFL29zC1sF+FHxBXt+7CLhnHAgAEkNQWh66+/Xqi7tG/fngUFBYJItgR91oYkH3roIU6bNo2klo3TZrPx4MGDAd9/KUyeMlnT7geazbLExMaI+613794B96BOyJ966ilxv+pa87rij39bx8XFMSQkRPQ3vT89/vjjAfdodXU1zWYze/bsye7du3Po0KFMSkpiaWkpS0tLGRoayqVLlzIlJYXHjx/nl19+yWHDhvG22277ScowBv47YRByg5AbMBCARYsX0RJkaXzZm6G5R0wA7ZF2Llq86Mw7+QVRWlrKrKysAKLiTy6DgoJoNpupKEog6URjIh19yrkpmdGl7PT96C4Z+sva7XaLBDskhXUtPDxc6G5bLBZBDFoj5P51b4sbQlOCfjqdcKFvDoiU8E2L/5R80yLLskjkc/nll7Ndu3bs0qULc3NzmZOTQ5PJRKfTyeXLlzMxMZFhYWGMjIyk2+1mRkYGQ0NDAwl/GIhsaMTcnyjafJ+pgYRcl9nTLdyqqnL69OncsWOHaM/U1FQOGTKERUVFzM3NFW2+dOlSDh8+nIBmAc/PzxcW7LS0tAAC7d83dMuxTsgvu+wy7b8Bvnr1hqZp7lv/vvvu4w033nBO8ot6Uif9+Pr1amr1lySJsbGxnDBhAiVJEtKKsKNxdsHRClHXr6UiMyYmhnfffbeQ73Q4HJw6dSqvvvpqXnbZZcKFQieSLeGVV15hQUEBc3NzOWTIEB4+fJgkefPNNzM5OZm5ubnMzc1lQkICb775Zt5www0Bg6ljx47x0KFD4jrpOujffPMNS0pKKEkS09PTOX/+fEZGRvKtt97ihAkTRN+PioqizW5r7D9yy23b9Hro7mIJCQlCdlNfbrPZxAyXbk1v6XpGRESIe7S4uFjci2azWUhiRkZG8sUXX2R+fj4BCEKemprKmJgYXnbZZaIti4qKGB8fz+PHj/+CT0kD/4kwCLlByA0YaBWLFi9iUkYSJVliUkbSL0bGf/zxRzqdTkF88/Ly6HK56HA4aLVa6XA4OGvWLE6cOFEQYv8yYMCAZqTS/7fD4Qh42Qb8ryDAHQHQMjX6r6OqKqOiojh+/HhhdSYpSJXJZBK64roV9myJ2tkQ8tOV0yb1QeMgQM+wqtffn9To+xg+fHig24jc2H7Jycm0Wq288847ec899zAzM5MffvghJ0yYwIyMDD7yyCOMSYzRSGIqiDEgwn37ucRHcENAXBt4nex2O3v16sWioqKAOjf1GTeZTJRlmeeff764Js0IlRK479baIzIyUmim6yQMQOOg1IQAi6yqqi3qsIt1LC0fR6/j6Yi8f78zmUyCDEZERDS3vMuNxNNkMjE7O5v9+vUTPuhXXXUVs7KyWiTkl19+OW+44QaS5CeffCKI5NmgZ8+eon/qbde9e3dKksRHHnmEJJmamsphw4ZxxIgRlCSJd911FxsaGlhcXMzOnTvzyJEjtNvtfOCBBwThz8vLY35+PhVFYUNDAy+//HLtfMOhZYdt2vZ+/VpRFOEfP2TIENpsNmZkZDAkJISFhYWCQOvb6PkPoqOj+bvf/U64c+nXac+ePZw+fToVRWF8fDxLSkoIQGip2+12du3albm5uYyOjqbFYuFdd90lZvKysrL4wQcfiDZ76KGHOGrUqJ/6yDTwXwgYhNwg5AYM/BbQ/b/HjBnDq666ig6HQ1jrrrjiCpHxsa6ujk899RRDQkL41VdfsX///s38qZuSUJ0U6L+dTqdwT2iVDLXwkvcPrNOtrP6WVv+iWz5/idKUTMqy3IywttYWZ1vi4+Nb/9/PRSIjI4MWi4WKojAmJoaKorBXr14cPXo0ZVlmt27dmJDos9DHQnNb8d+X6iPMeYHLQ0NDRRCd3vYZGRkBZNVut9Ptdgv3Fn0ApK/XUrn22mvFwKMpsY2LiwsYoFitViqmJoOqpoGosq89Ivz6Tit+8SKhUgt9q6UAVv/ESf7LW+q7/gRfJ+KqqtJisZyWkD/xxBPs2rUr8/LyeOWVVzI7O/usCXm7du1EgCepWdnbtWtHSWqMM3n00UcZExPD0NBQms1mkbDrT3/6EyMiIjhnzhw6HA7abDa6XC4OGjSIKSkpDA4OpizL/OCDDyjbZG2mJdTXlyLRzF3HYrWIoGr9GP/85z+pKArDw8P5ww8/cPjw4WJGQpZl3n777WImbf369cKNJzo6mqqqMiIigiTZtWtXjh8/noMHD2bv3r25cOFC9uzZk5WVladtn/379zMjI4Mej0csGzp0KD/++OOzamcD/xuAQcgNQm7AwM+B3bt3N7P8nC7jYFZWFvfu3UuS7N69uyAucXFxTEpKoqIoLCws5NSpU+lyu05LeJoSrKlTpwaQl4yMjNb9uZv6BksQrikZGRksKiri1KlTuWjxIjqCHZq/dgcQA9GYll3yq5vJb59qy/VtiZid7j99KlwvuspNS9ucq996q8WfiLbgKhAWFsbMzExBBvUSGRnJ3//+9y2fo73147VERJu6cuiqMfqxUlNTG/cR0ZystaWNoqKiBHG2Wq2B/7e0P6uvbYpb+K9JO02cOJHt27enoij84osvGB4eLnzii4uLmZKSwgceeIBut1tYxcvKypicnMyoqKiAAONFixYJ4qhnlB06dCiXL1/Offv2MSkpicePH2d9fT179+4tgp2TkpJ45MiRZvdhWzJV6sfrP7B/o/Xforn6hIeHi/XCwsLOSMj1OixYsIAul4s9evTgn/70J1qtVqanp9NkMvGmm27ilClTaDKZeNFFF2nHc0OzkMeicaYFjfefHiTpf56rV69mSUkJSS1mZP78+SQ1tzdd/ck/G69/1tTly5cL16xOnTqdtn1awgsvvMD4+Hi+9tprJLWMohkZGbz00kvPel8G/jfwWxByQ4fcgIH/Quzduxdvv/32GddraGhAbGwstm7dioSEBEiShFWrVglN5IMHD+LYsWMgiXXr1mHBggU4dVzTQUYruTVsNlvAbz35j47t27eLzI96shkBPaOjvpjA8ePHcezYMezZswfr16/Hiy++iEnTJ6GyphLoAeAAgGV++yAatbU9ANL99hl9xiYRkCQpIGNoWFgYAGDDhg0B9d64cSM2bNjQ4vn4Z7g8FzTTiPbP4RIXWFdA03/et28fvF4vCgsLAWhGl8OHD+ORRx4RvwPqWeW3zyaJfLT3EgKSx/gnVSGJyspKkQGUJHbt2tW4gyPQMsw2gcvlCqi7JElISUlBTk4Ohg0bhgsuuADV1dVYvnw5Lr30Um1Fu2+DCGjJjGStmFST1m9MvvYJ9/2nAOjg+24CIGna6l9//TV2794Nt9sNs9mMqKgofPzxx/B4PNi+fTt2796NRx55BMnJySgoKACgZRJ9++23ERERITKENu0felvpiIuLwx133IHMzEw4nU7s3btX6Km3hmHDhrUpQ+U7776jZZUdDS1h2GjgRM0JHD16FBs2bMDatWtx7NgxjB07FiTx2GOPAQCefPJJFBcXo3///qivr8euXbvg8XjwzjvvwGazITQ0FCNGjIDH48GOHTsgyzJuuukmdOvWDaqq4oYbbtDuozJfOycAOKq1Lcy+NifOmHjn5MmTiIvTOrB/ltvW0KdPH8yZMwdjxozBVVdddcb1m2L8+PHYu3cvRo4cCUBLSLVt27ZmzyYDBn5T/NojgHMphoXcwH8T5s2bx/POO49lZWWtrtNUT5wk8/PzOWvWLJKNutitQQ8gNJlMwqdU96eOj4/n9OnTmZ+fH2jNtgdaH88YJNeKhfxsgut0lZHW/tcDNQEtiGvgwIGNFlGAiIcmA5kGzd1Ct4LP8vu826/M8tu/VbP2zpkzhzfccAMtFguvvvpqkpqEXEJCAi+99FI++OCDLCkpYWVlJUNCQjh79mw+8cQTdDqdDA4OFjJsQNuCQn+rol9rs9l8zq40V1xxRYBEob8V3v9aBey/Ffcjl8tFu90utissLGRkZCSnTZvG9PR0AuCf/vSnxmDY3mh0TTFBBHkuWryIqkMlMn3X9yLfcfqAGOU7vhmUbTKfefYZbt26laSmn71r1y5xz/hLSi5dupQTJkwgSV544YX8y1/+wuLiYsbHx9PlcvH9998PuEfr6+sDttdRXl7O9PT0gDiIK6+8koqisHfv3rTZbLRYLPz73/9Oknz88ccpSRLj4uKEFn5xcTHNZjNvvPFG4VICCUSUry3yQSSCuDCwjePj47lixYqA4GhJkvjHP/5RBHXq/cJms7F9+/bMy8sLcCXLy8vjunXrWFRURJfLRUmStPNQfffcCGhSnb5jOhwOXnXVVaypqSHZuoW8tfwIrVnISfLgwYO0Wq2nfW4aMPBzAYbLikHIDfx3QA8iagmtTVeTmj44qSV08SfkJSUlHDp0KJcuXcqkpCR27969RULu9Xrp8Xi4YsUKqqrKvLw8vvjii+zZsyfT0tI4Y8YMAuD48eP56quvai9St++FakZzpQgEqp20pcTExDSTHvQn3kOGDAlQVvEPXPQnbLFxsVy9ejXnz59PWZbp9XpJUiMjhT5SrpMzJ4ib0OiiMgHagKHYj4xPgpZwByAsoKRq5+R2uxkfH8+UlBQ6nU6uXr2aSUlJnDhxIh0OB1esWMGSkhI+/fTTQo+8U6dOAed4Lgofv0TRtcdtNhujoqKEtKI/IWtTXS1odPU4TVBkWwYAUNq2D111w78IDXL41UcfNPqI/u9+9zuqZr/+4++iIjVen4SEBMbFxXH48OEkySeeeILR0dGMi4sjSZE8KTc3VyTZKi0tZUZGBvv27cv27dszPT2dN910E0kyKiqKQ4YMYXFxMR999NEWCbm/K1OPHj14wQUXMDs7mwCE9GFubi6dTie7d+/OqKgoyrLM4uJiulwuyrLMzz//nKNHjxbuM+JeudHXrk4QJb5PgGvXrmViYiLDw8P57LPPisDK3NxcPvLII8zNzeW3334rEvbEx8fzqaeeEgHbWVlZ7N+/P8PDwzlo0KAWn1OTp0ymYtfcmBS7wslTJvOuu+5ifHx8i+v/HFi6dCmvuOKKX2z/Bgz4wyDkBiE38F+CmTNnMjo6mhUVFQwPD6csyyITn05Qe/ToQZIENOk3s9nMkJAQ4QOckJDAdu3aCX/P8PBw9urVSxDZoqIiIdtlNpvpdrsZFBTEiy++mDk5OQQ0ybhp06YxOzs7MPiyleyM/iTmjP7Pso8EpLZMDHWrfFZWFkeMGBEQDKlb52w2m1CXIDVf9pEjR3LcuHFiFmD+/Pk0mUxinfDocM0619nvmHHQSDpAZIEIAlNSUzRL3ihovq5uP7ImgXfceQcVReGzzz5LkkJJQifkLVn2nlv4HK02qyB5v7RlvC0a2j9VRaap0gUAIhqNeup+BHf+/PlCEnDo0KG0WCxUVZVDhgzhqFGjxMCrb9++TE9P17KqWhrPwe6wi4BNs9nMvn37UlEU5uTk8NFHHyUAXnXVVRw/fjwjIyOFzJ5JNTWSe0djnQYMGMB27drxhRdeYGVlJQcPHsznn3+eycnJnDdvHqdOncqqqirm5uby6NGjPHz4MNPS0khqMRVZWVlcsWIF16xZw7y8PFZUVLC8vJzt2rXjt99+22xgPXfuXM6ZM4ekNkiePHmy+G/w4MEsKytjWVmZ0LVesWJFs/besmWLsEqrqiqu3+DBg5mYmEhJkgKCWiMiIvjMM880v975fver3zWaPn16wIwYAN5+++1CocfhcPDNN98U61dVVTE6Olr0t4suuogjR45kamoqBw0axGXLlvGiiy4S5/nPf/5TDGr8ER4ezhkzZpzrI/O0uOGGG5iWliZmNwwY+KVhEHKDkBv4L8Gbb75Jp9PJrl27EtCURU6ePMm4uDiaTCa63W5u2bKFsbGxBLRp7O7du1OWZXbo0IFjx44V+t6RkZEsLCykoijMy8tjUlISzWYzU1JSePPNNzeSQT/XAKvTKghjv379mJycTIfDwY4dO2rrtAcxDIEBkBKIpMb9tCpf518c0IIqfRbVTp07EdDcS/QMibIsc8aMGYIgWCwW2u12Wq1WTdPYZuPJkye5ceNGOp1OPvjgg8KSa7fbhSbxK6+8QofD0UhIhqFVybW2FP8Bh55BsOk6NpuNkydPZlJSUqP2dAuZF89YWgls1OUZ20rC9f/8Ld5RUVGNx7BDmyFQTtM2sQhwN4qIiAjQSxfn5mq+j6FDhzI1NZWSJLGgoEBkXywuLmZQUJDIkul2u1lSUkKHw8HXXntN9NHo6GgRrKlrYEuSxODgYI4cOTLgHPVMra2ltAfAW265RQxmVVWlqqqMi4vj+++/LwIQ8/Pz6XK5RLbKgQMH8osvvuCtt97K0NBQer1ePv7448IdzOFw8K677uL8+fMZFxfHrKwscV9fcsklTEhI4NKlS1lSUsJPP/202b0/7/F5VM0qJVliVJx2be644w4hEdq5c2dxPnv27OGgQYNE2+qzBDabjaGhoWK9W2+9VbiLDBs2rPGeV6ENijMb+7TVaqWqqrz11lvZrVs3SpLELl26cPz48YyIiGBaWhpffvllyrJMVVX50ksvMTg4mDExMRw3bhx3797Nhx9+mHFxcRw0aBC9Xi+zsrKE3vnll1/Od99991d4ihow8NvBIOQGITfwX4Jt27YR0FxDFEVhaGgoJ0yYwJCQEFosFkHIdT3nZ599luHh4ezRowdff/11oVphNps5bdo0KopCu91Op9MppqzDw8Obp/OWfITM1UjgWiV4phZIWxMlkqbW1zNZY/39i8+mCN9UBOp8++td6+vo1v+mpamVOsA3XWpSoClBmM1mBgcH02QyCf90VVU5bdo0jh07loqiMD09vTG9vQLiCh/5taJFF58WSysuGy0R8nO2tjcdKLRAqM9Ummmsn2b70w3U4uPjhVZ3QkLCaX3WLRaLUPbR2yQyMlL0d73PSZLE1157rfFaQCOgPXr04NatWzl//nxGREQIEh0VFSVmVkpKSvjyyy8zMjKSUVFRjI2NZXx8PBMTE9m5c2e63e6AWAB/LW/9d1xcXIDftW7Jtlqtje5YQ4e0eI5FRUUtJobq0KGDGAj6J4jyz5AKgO3btxfbnE7DXd9PbGwsc3NzOWPGDJpMJnbo0IFXX3013W43FUXhyy+/LBI97dq1S7jDBAcH89ChQ3z44Ydpt9uFy8r999/Pxx57TKjN1NfX/5aPVwMGfnHAIOQGITfwnw1/30pAc5vo3Lkzu3fvzpCQEJGVTifkOiFbuHAhzz//fJ5//vmMjIxkbGwsk5KS6HK5eNttt7GoqIh2u539+/cXFvKpU6cGJmhxQXMzuAGaz7QfWWyRPEW2Triart/S9npK95CQEEFk/ImXCMD0LfcnUrpfs8lkYkJCAm02G7OzsxkaGioC4Ox2uwie1ElZQkICFUURswn+hMf/2M0svkAzK7U+ha8TfV3TXCfhujyjf72hQgumU30ku6QNBNmvtDY48id//sGR/m4x/oMWfV/it/9ASoGmE+0C0Q+BlvwBIIKhDcauhpYgCGBdXR0vuOCC05K9/v3788CBA8JHvVu3boJAZ2RkCHK6dOlSOhwO/u53v6PD4WB2TrbWVmG+tnOeflBwpr7YtA8CCMjs2HR9/R7RgxX9B43+2V/bUhe9vf0TYvXu3Vu4fDQrSvPz8B9w6RlnAY1IT5o0KWB7va661ru+zH/wWlBQQKfTKRLu6H1o3rx5/OabbyjLMtu3by+0xDt06BBgISfJr7/+mmFhYSJIOT4+nm63WxDy/fv3s0OHDvzzn//MW2+99bd8xBow8KsAhuyhAQP/mZgydQpkm4wn//wkPBYPcB4AK1C6qxTff/89unfvjtraWiiKAqvVirq6OtTW1gr5uYqKClitVtTW1gLQ5Nk6dOiAuro6LF68GLW1taitrcXGjRu1kbRvm0suuaSxEtMB5AP4DNrr/HfaYpPJhKioqMb1FN/nUQBBgHgKXN64CkkEhwQD0OTbdMk/RVHgdDphMpngdrtxySWXwGq1QlVVOBwOdOvWTexj69atAW0UHh4uvnu9XqiqClKTVqyursaWLVtQVlaGsrIyeL1eVFVV4cknn0RlZSWsVisAYN++ffB4PFi1ahW8Xq9ov3Xr1qG+vl7sv7y8HIqiBMrM+cnvDR48GA0NDTCbzejWrRvCwsKEtF9ERARefvllrFq1ClarFZVVlY1tlgxgMoDu0GTfVvqWywAKfd8l338RvuZWFSQmJmLcuHHweDy49dZboSgKsrKyhJxcTU0N4uLikJWVhaqqKnzyySeQJEm0EQDcd999CAoKAqBJMEZERCA4OBiSLAFJ0GTnAE2O8pjvswsAp2+5CmANgJO++v4NwPvaX+3atcPf//53VFZWojUsW7YMsbGxaGhoQHV1Nb744gsAwKlTp1BaWorKykrU1NRg1KhRqKysxOOPPw6v14stm7docoQStD5X0bjPZrKXLYAkVFUVv/W+AEC0R11dnWgnSZIwdepUKIp20bxeLzIyMlBfXw+SePfdd4XUYnFxMQDtfmsL9GMcO3ZMLPv8889x6NChljfwk6nU78H6+nqYzWaxP/2eb2howKpVq8T6kiShokJrrB49egipSKvViuHDh0OWZTgcDkydOhUWiwUVFRUIDw+HLMsICQnBTTfdhCNHjoj2sdlsUFUVX331FSwWC7xeLx555BFs3boVDQ0NiI6OxsKFC6EoCqZPn46EhARcc801AIDY2FjExsbi/vvvPyfZQQMGDLQBv/YI4FyKYSE38O+MyVMma5bpCdBk1yb4rIB+CVd0K2dwcDAvvPBCYbHTLWCxsbF0uVyMj4/nnXfeKaaPs7OzhcU0PDycffr0EVbBlJQUJiUlNVpAg6EFLqb5rN/BjVa9GTNmMCgoqDHNuAxNHm6cnzXvFt+nC5qMmp8VUZ8y79Onj7AqRkZGcsqUKQwPDxfWxwALsJ9F2mw2nzlINB8BgZeKovD7779vJqknJN/8ZgD0mQbdkq4oCnv27CmW65kH9WyW4eHhIjV8bm4uzWazUMSwWq1cuXIlN27c2OiiY/NZO+NAzAHR0Wd9ngDN+qu7//hLQZq1tp82bVpAf1m4cCGdTqcIGnU4HBx20TAR/KjYFcYnxNNms3Hu3Lm89tprxTlKksRrrrmGo0eP5pw5c2g2m7lo8SLaI+1aPQrRKAkZ66uffk11txkZRF/tP1OQSSjJ+F+L01mMA65FC+vrVtqAvqCixRkZSZKEtbmp2o7eX/XZkpaOVVtbK+6viIgIduqkxTD0798/sI5+xyUZ0C9P2yfbUAJco5q6JvkdVz9Pu90eUAf/snfv3oD7bsKECQTAuXPn8sknnxT/9e3bly6XS7jA6f1UdytJT0+nxaJly/zjH/9IkvR4PGJ2TVdCmjdvHl999VXabDbhenPPPfeQbC69+vLLL7NLly6/+PPUgIF/B8BwWTEIuYGfB3pWylmzZvGjjz76RY+l2BWN+NztV0b4yJlP91pXZ9i8eTPbt2/fTH+3qebu5s2b6XK5mJ2dzYsvvpidO3fm6tWr+cADDwQogLzyyivai14n2QVolDH0y3qZnJwsiFSAL6s/QbrWtw8LiJmNy4cNGybIRG5uriBKw4YN45QpU9ipUyeN8Aa7NOKlk+qRjcQ6Ly8vgKApihLoCgIf2U2CFnDqR5gkSaIkNyGIKoiM5oSmqY+7P3kzmUzif1VVm6VQLygoCCBXuhKIJEuNhNz/+DZf++kSgSOgBdcpjcQ1PDycd9xxB5977jlKksROnTrxmmuuEYR8zpw5jbrOQSC6QXPt8COocXFxAfWfPn06ExISeMMNNwgZRn//45ZcZawOK6+7/rqAZapV5d333N1inIGu0nEuco5ZWVnNl7eiWX8uxb9OH3/8ccB/en8JcOUyg+jZMukVg8TTqQ6doQ4BpWmm2ZDG77rMoKqq/Oqrr8Ty8847T3zfuXOn+D5v3jyWlpYyNDSUd9xxBw8cOMCQkBBedtll7N27NwHNNWvu3Lm8/fbbA9RgTpeV90xYvXo1b7zxxoBlJSUlHDlyJJ955plz2qcBA/9pgEHIDUJu4NxRWlrK6OhoZmdn02KxcPDgwSTPnETn6quv5iWXXMKNGzeS1BL3VFZWkgxMVOEP/+UAWk5AI4GYoFk8z+VcsrOzhXyaDlmWW1x/8pTJGil0+IhhBBqtoz7iIcsy8/LymJKSIpYHkOJJzYmJJEl86KGHBJENDw/neeedx759+9LpdNLtdjMtLa3RaprrI6oytODH05AYQZ51RZB4iCQu+jrx8fGBBEe3iqu+7Xzr3nbbbQwJCQnwCXa73UJK0l+e7pVXXmFqaiqTk5PZv39/TpgwgSUlJZw0aRLnzp3L8vJyFhQUsKioiBaLRVh8z1j8rKP+snWpqalCXzojI0OQXVmWedVVVzUSOcV3XmGBBFaWZQYFBYmkNKqqMiUlJSDYsVk7AcIf+FyJ7xlnNH7hEh4e3uJyXZmoaTmtRORZ+q2fcwlGs8BovW66So/JZOIjjzzS4vb+92NiYiIzMjIoSRKjo6Pp8Xg4dOhQut1uWiyWAJ/5wsJCSpLEjz76iPX19Rw7dixjY2OZn5/Pfv36kSS/+uordu3alYWFhezWrRsffPBBTp06lQsXLuTFF1/MQYMGMT09Xaiz6M+hxYsXMygoiEVFRSLhz9nA/3lqwMB/CmAQcoOQGzh33HLLLQTATp06cfTo0Zw7dy4nTJjAdu3aCW3pW2+9lXl5eezUqRO3b9/e4n78LdBtIeQtWsgngAgFEQROnjK52fZnwsaNGwMk7pKTk4USha6+0rNnTyYmJjIsLKwxqYodGhlWQAxpfNHr7iR5eXlimj4iIiIwgU+T6XZ9Gz3JCACR4dNsNtNmszEyMpKZmZmNpFXxHb8EAQGPuqxa586dhYydsGJ2g+YiEwGNkOdq21itVprNZlqtVtqD7I2kVa9jLAQhN5vNLCgoEFZI/+JyuRgcHMzIyEgRVJuSksKcnJwAGT6n08nc3NxmSY1aK02tyPpn00RKYWFhgtxefvnlHDZsGAHNUiqCA91aX0EmiEHQtNQBvvfee+zQoQMBMCoqShDsyMhI4c6gE76mLkN9+vQJIHpvvfVWY/3llolja0WSpOYKLD93OY2ii79rif85ttmCb2/cv9Pp/MmJnNq8vdTY/4uLi0V/fOKJJ2ixWJiZmcmpU6cyMjKSDoeDpaWl4vxSUlLYo0cPDhs2jFFRUayrqxME3b8P+h8vPDxcDM4sFgsLCwvZqVMn4VKmz0zp+4iOjg5QjtFnGFRV5fDhw5sF+SYkJPDdd98VzxtFURgREcEuXbowPz+fHTt2pKqqDA0NFTKYus681WqloijMyMhgXl4e33777bN+Lhow8GsBBiE3CLmBc8OaNWvEi8VflWLUqFGC1MmyLF4QLpeLTqdTkENVVfnRRx818yktLCxk586d2bVrV6alpTEoKIj5+fm88cYbmZaWxpycnEbVhlBovs1joFly1bMj4/Pnz2d2djZDQkKEhrKqqmzfvj1VVeXFF18syJWuN9wSSQkgN0pzIiFJUoCkWmuEQ7fWderUKUBFxZ+A6u4Sghjox7VDc0HxI1S6hdf/uAFuFmZo1nUfabHZbDSbzSwuLuaAgQMo21q2gPbq1YtJSUkBadhbK7rl2l/xQr/mTV1qWtpeUZRWLbGn26a16yT+c0IbEOlJb3wzFfPmzRNqNcePHw9Q9/C/DmcaRFit1kCLt7v1dX+t0kyZpAkh16U/e/TowfDwcKGcY7fbOWXKFHGvh4aGMiwsjKtWrWL//v0bz1NG42xLEYiUJseRQFwOLXtrG7OR6veOP1FdsGBBqzMRubm5BDSfb31m6p///Cc3btwoEizZ7Xb26tVLaPPr56ijtLSUkiTx+++/Z2RkZICijN6H/H/rg+eYmJiAmRR/8u5/D+gDFFmW6XA4GBERQQDN3Mx0Nxk9j4Isy7z++uuZmZnJ/Px8BgUFccSIESLZ0VNPPSWejcHBwezduzdTUjTVqSNHjjAtLU1k39XR1G/dgIHfCjAIuUHIDZwb5s2bxyFDNA3gRx55hCEhIZQkiTExMZRlWbhXAOBdd90lrKLXX3893W43rVarkHGTZZlvvfUW3W63IHg7d+5kfHw8Z8yYwbvvvpuZmZk0m830er284IILOGToECF3KKkSu3TVgp8WLlzI/fv3t+kcsrKy+P3333PBggVctmyZeNmqqkqLxSIISVOLrG55EySkyI8URDa+gDMyMgT59D83PbmLnnUR0Ky6559/vniR6yQnJiaGkZGRIjV7YmJiM0LQEiE9W4ukThjuvvtuBgcHN7qAtGBFjY6OpqIoHD16dIA7Tls00duSCfNcttfJj8ViYXp6+jkfryn5+tnLWVjJf0p7nEs5nbuNTogVRRH+8/rgsMVAzSg0T+RkhzaLJEELcG6DH7l+H+gkV5ZlTp06VZy37pYSHx/PXbt2iWtvNpvF/aXjlVdeYUFBgYgp+dOf/iSynw4bNkyst379egIaqdXPLSoqSiRSOt018JcX1QeT/nX376v6+vryyy67TDwj9EBofXv/a6MHXeuuM3obhYSEMDk5WbTH6NGjGRERIZJKWa1WHjx4kGQjEW8rIW+aQdUf+/fv54gRI1rd1uFwtOl5bOB/GzAIuUHIDbQNixYvYlJGEiVZYlJGEq+44gqOGzeOALhr1y663W6RREeWZd54440BLyyd5LjdbjocDqGprb98srKyxEtMURRmZmbS4XAwJyeHV155JS+77DJaLBZeddVVHDt2LDt27MjZs2czMzMzoJ5paWkcOnSo8E/3d4fRXwyLFi+iw+UIqJ+u+GGz2ZpZXsPDwxkUFBSgvhLwUvYnrXIgmfB/WTedjvafytbrIcsyMzMzRTZIi8XCoKAgUXQVE1VVW/X5TU5OFn6/+fn5QtVElmX27NmTLpeLiqIwNzdXzGTcddddTExMpN2upVrXgzqt9sZZgTvuuCPgHJzBTvHC19uuNVKlKIrYtmfPni22R2vZNXUS0hpZa/r/uaS2b01T2yg/Y1ERmNiphcGeqqqCbAKa5VqWZdHPdDeikJAQMQsxcuRIjho1igkJCUIzPCIiQtzzgwcPZllZmSCVEyZM4L333svU1FQ6nU726tWL27dvF8+4lupuMplEf2163+qzYGazWcy0tbS93l91bX+z2UyXyyWs9C6XK2BQ29pg68iRIyJQvF27dqK/6zNXervpajBWq5VPP/00zWYzs7Ozeckll3Ds2LHNCPldd93FCRMmsKGhgY888gg7duzI9u3bc/bs2SwtLWVGRgazs7N5zTXXsF27dhw4cCBPnTp1xneHQcgNtAUwCLlByA2cGYsWL6LZbSZ6odHvWG20qM2ePZt2u53h4eHCenbJJZeIF4h/0hh/stRUkcP/U1dC0F9c+sspOjpaWOX0dYOCghgWFiZeUhaLhVarVfh2BgUFiZdeenq6RjbPQoXC7XYH1FV/6Z2OvMmy3CwYrmlSFH93EpvNxuuvv54AmJOTI5YHBQVx1KhRwt80MzMzgKDoFrwPPviAJSUlwkLX1O9X35/eDrrbkG7xnD17Nm+55ZZANwsZwqUFgJhab6201h56P9EJzZncXE63j9aKv2W7qXvRv3v5bxsEBPhZ+5NuBzT5SqWF/1ppD73v6veOTlj9VYN04qm73Oi+23rMQVJSEgcOHCjuFYvFwrS0NKamplJVVZE5VlKkc57B8O97QKARQu+f+j2ZnZ0tBsL64BiAGGi0RMT9VWz870P/tvYn8/4DAJPJxCVLlgTUU1VV3nfffRwxYgTDwsJEfWJiYvjGG29w+PDh7NOnD9u3by+MFGlpaSIL8oUXXki320273c68vDza7XZu3ryZpaWl7Nixo3AHmjBhAh0OB5cvX86SkhKOGDGCWVlZHDNmTDP3GQP/24BByA1CbuDMCIsN08h4CAK1v30vVkVRhNKK/nKJiYkRLwT/F8WIESPEi0HP1Ki/NPwVD3T3DQB88803hTV98uTJgqzHxcW1GHBlMpmYnZ0ttsnOzg4I2Gyr/2pbi78V+ecuuvX+TK4J/m4a/u2hB3bpv7OyssTAqaSkhBEREQwNDeXs2bM57CIt+BFmaO4FaQgYuJjUJiog5uZ1PRNpaTP5bOLu0FImUP14uiqK7jZ1unr4a6y3NCDUv4v/TGgkaacJgjzr8jO7rpyu+PvBn1UJanm57lKVmpra4gA7QAIxFI3+85KvPU3Q1FF8vvv6bIw+YPN3g2pNVrPpYDcoKIgffvihGIy15Lal5xpoul+n0xl4bfU+r57bNddnCVv6T/cb978m+mCz6aCzLbM2epZdAExLSxPtcv755wt/ekCTGAU067wu6xkbG8vhw4cT0LL8ktqMYk5OjjA6pKenMyQkRORsiIuLY5cuXRgXF8fk5GTedNNNLC0tZUpKCvv27cvKykoOHTqUL7zwArdt28aEhARByF0uF/fu3UuPx8OuXbtyxYoVv/GbzcC/E2Bk6jRg4Mw4duAYsAXARQBSoGVRTAEwUPvf4/Fg27ZtUBQtQ6LJZILZbBaZGPXsdwDw7rvvAgBI4uTJk2J5Q0MDKioqRCbBr776Svw3Y8YMkS1wyZIl+OyzzyDLMvbv36/dVLIMRVFgs9kAaJkCDxw4gPLycgDAzp074fV6xXFR63dyZ05cGAA9458/aqpq2pQB0WQynd3BoNX3u+++O+P+/TMZ6lkQa2tr4XQ6RdbF2tpabN26FUePHkVDQwPmz5+P6upq1NVpKTXf+/A9rT08ANzQsooWNx6job5B+xINbb1636cvqaPexqc7F5JQFAWSJEFWmjwO/U/Rb1eRkZHweDzN2iAkJASyLMPr9aJ79+4IDw8HyZbrYWusQ3V1NQCgoKCg8dwaGsR3k8mk9RMAaEBj9kf/y/cTn+SKVznzSj8T9KyvZ42KlhfX1taioaEBu3btEvc4oGXxlCQJI0eOhM1m067XKWj9BGjsWw0AagBEaplo01LToKqqyCzrf485HA7xXZZlcV3KysoAAC6XCwkJCaioqMD48ePR0NCAmpoaeDyegP0oioK6ujrxzPGvd3l5uUZbAcCCxuvdmIi2GWw2G+x2u/gdHBwMWZYRGhqKuLg47XQlCZGRkQC0vrpw4UJ06NABHo8H9Q31CHYHa3XxesQ+Lr30UlFv/Xmmn3NeXp5WRYtF3Avl5eWiv7tcLrHNhx9+iE2bNkGSJHTq1AmlpaUwmUw4cuQIfvzxRwBAVVUVTpw4AUVRcOzYMRw/fhwDBw7Ejz/+iOnTp8NiseDdd99F+/btcccdd6C6uhr19fXIzMxEnz59sG/fPrz88ssYMmQIDh48iIMHD6K+vh4fffQRHnnkEYwcOVJkLgWAzp07Iz4+HrIso7CwELt37269gQ0Y+BVgEHID/3kwQ0vBndhkeYb2QRIejweVlZU4f/D58Kpe7NmzR7xIvIr2wlAUBf/85z9hs9mQnp4uSNCQIUMQFBQEWZaxfv16SJKE8847D7IsIy4uDjt37kRRUREsFgvOO+88jBw5ErIsw+l0CjIuSRLcbjcA7UWYmZkp0pL7p3hvBrb+lw49JTjQnHTqL0Y9zXZT+JNIf9LXVthsNtG+p4NOUJrW8ciRI2hoaGiR0A8fPhxVVVWoqanBkiVLwFpqTyidkG/1ffog0p0fAtAOQJzvD52Yo22p2T0eD8aNGwdZavI4JAS5hx9fPXr0KCwWi+hP+fn5iI6OxokTJ1BcXAxJkrBs2TLs2LFD26DpmCkcQHXzeuzbt6/F+jUbOOnN6d+NfsKT3GQyiWO0pb1+KrZt23bGdc5lsNgUFRUVIIm1a9ciODhYGyxThlypNZYECRkZvodGLYBDgKfWgx07dqC+vh579+4FoA2gAa1toqOjW2wjfVBVWVmJw4cPA4D41AegUVFRYn39ntD3lZSUFLhfJxrrpS9u4dLo7eTxeBASEgKTyQRJknDq1ClIkgSbzYZly5YhIyMDVqsVZWVlkCQJ1dXVeOCBB3D48GE0NDSgVqrFSZ4E7IDX7gXMWj9//c3XxcDP6/XCYrGguLgYQUFBKC0tBQAkJiaKc9TbCACeeOIJnDhxQpy7PkDV28vr9WL//v3o06cPFEXBAw88AEVRoCgKZs6ciaFDh6KhoQEOhwOLFy+G2WzGBx98AK/Xi8zMTFRXV6OqqgoZGRmw2+2w2Wy4/vrr8f777yMtLQ2bN2/GvHnzIMsyvv32W6xZs0YM9gFtIKFDUZRzeh4aMPBzwiDkBv7jEOYO0yxHPzT54wACXlpTpk7Bky8+Ce9lXmAWgFG+P3yk1+PxoK6uDiRx8OBBYelZs2YNqqqqEBERgdGjR4v9ORwOlJWVoaGhAatWrUJtbS2Wvr4UixYtEg/zoKAgeDweKIqC9PR0AEBsbCxyc3MbK/YTjZH+ZFi3QDVFa8RKWFrPEVVVVW1az59QxcbGIjo6GldffTW6desGh8Mh6iHLMlJSUgAAu3fvBknYbDacOnVK29gDwAHgCDQr5j8aj+E/MMFGAPugXdseEISmref74osvtvxC1kmv3/jD6/UGDDg2bNiAQ4cOwev1YvXq1c0HLI0cQMPRlutQXV2NxMRE2Gy2xsEGtOtttVohSVIAqQuwkFvQ2PdzAFh93/37mtr4VR9AAtrATN8vycb9OHBaMhhwzJ8ZZ0WOmk8SAWgcsH399dc4dOgQVFXF8OHDUVJSgoiICFw76VoosgJVVQP7EgBIjaTZvy7bt28Xfcpms4k21I9FErW1tQG70gfgP/zwg7gv/ft/i9C7rRVaO/vqJIg6gPSMdGG9lmVZPJumT58OVVXh9XpRXV2NmJgYJCUlISoqCk6nE3a7HSdPnsT27dtx7NgxjZCjFugHzahxHoDLobEDL1BXpXVg1ax1oLVr1wY8B/bu3SuIbn19vbB4/+lPfxIzAFdddVVA342KioLX60VBQQFMJhM8Hg/27dsHp9MJj8eDoqIiTJo0Ca+//jqysrIwdOhQlJeX47bbbsP69esxa9YseDwe1NbWIj4+HmazGW63Gxs3bgQA1NXVIT09HSdPnoTL5cJrr72Gl156SbTJ1q1bW253AwZ+S/zaPjLnUgwfcgP+WLR4keYvHIRAH3IH2H9gf7Feiwl7pEA/zLi4OKEkEhkZKZbbbDaqqiqkEEeNGsWJEydSluVA30oLtLoEB/qGAmeR6bA1H/IWtKJ1n2S9Dk31p/0DstDUf7Zp0ZMItaGOrfqNtpIBUV9f9yft1q0bJ02axJ49ewrFlpb2GxERIQLmhN+r/Sx8ZvVrG3H2vrZNS2vX79cKemzJ7zcuLo45OTmBy1vyLdZ/x0LL/NlkP88880yAdNycOXO0vlAAojeI86FJZvqngpd9pemxWghIlhWZcXFx2rWW0BiM6wDhQmN21lbaV+8Dejvo0n16329VuaZJf9aDmf0DDwcNGsSMjAzecccdLSc7aqVe/veS5Mtge8Z+GHnm66xLIjbbXj+XNt6jrfVT/dz1NtT/1yUaxXVJ9jt2iO+7rfV9nylGY+DAgcL/3uFwCL/x0NBQRkRECIlFPUZCV5mx2WwsLCwUCdzuvPNOlpaWiueGqqq89dZbGRoayoyMDM6YMYNTp07lK6+8QofDIWIKXC4Xt23bRgAiyF/3p3/qqacCEr7pGUtJsqysrFmGZAP/e4AR1GkQcgNtw6LFi2gJsjQG8lkCyThJbXlLKe39Xphni/Lyck2KzAJtQHCLb7/TG19euhRgW17y4mXYwnKTpTkhtNlsAYoJ/fv3F3JlqqqyoLCg7S/tc3zRG+XXL7qEZ//+/RkXF8chQ4Y0HxiYQHSCNkjSyfMQtJgcKjY2Vsh9Ll26VGQPFaTMBi1bqLXJ/v37jQSRAKtpfa+44gohVQeASPD9538/DgTRU1t+6623cu7cuSSbp1rXZQJJLXmWHnwdoFzjH+jqG8DZbDb2799fBO/27duXQ4YMEQGBffr0YUREhEaG/e+F+Cbn4/tPlmX269dPyHtecsklQklpw4YNjcHidmjPBhnEKIhnlD+h13Mj6N/1AcYjcx9pPO6AxvbRiz9x97/+sbGxIsmZvkx/DnXq1Ekss1qt4loHKK9EoHGQ1A5EP98xk6AN0Hzbf/jhhwS0wFyS4pybJnnSg1hjYmKEyoxu/LBarXQ6nezXrx/vueceXn755QEDrAEDBjA7O5upqakBgaT+CchkWWZkZCTj4uICgp71JFH+yY4AjcQXFBSQJEtKSrh69WqS5D/+8Q927dqVRUVFvPTSS1leXn5ajXMD/zuAQcgNQm7g50OrKe2tIEJBxa6c9T5LSkoaXzzD0CLZ7z+wf3PrfUua1jpJ909IEgQtc6Dfei0l8NBfQLpCiU7U25Lc5NcsP9Wa3Ewb/D+o7j+ltGad14lLi3VzoVWLZtNzaqqTLgjRabTXmxUzWh1oqqrK1NRUtm/fvnG5boGfBSLb99unnHLxxRdz4MCBnDt3boBWP9mo3b98+XI6HA6uWrWKNputsY3M0Ai/X110BR5decXhcDAhIYGhoaFCC1u3zgJoWdWkyW99IKOqKlVV5fTp0/l///d/BDT9baHK1BvarIQMIhpigJCWlsaMjAyxL53Mtm/fnueddx4VRWFwcHDgDFyT9vXPuOqfPXfy5MkiwVnTvnLTTTcF/Nb336NHj0ClIf28M1vvB7169SKgPZPCw8PbfI/ExMQIku6/TXh4OO12e4DcIoAA5aG0tDQxE6irMDW10Ou5BPzPD9As4/pv3fouyzLnzJnDI0eOsEOHDiwuLmZBQQGjo6M5bdo09uzZU1jsZ8yY8bO9j/wHAgb+/QFDZcWAgZ8P1155LfAGgFJoPsCl0H57AFT5/j9LfPrpp0jKSNJ8O0Ob/PkDoNgVLPt8GTACgQow/RpXa6aMoruoWn3F5348evRoREdHIyIiAs8++ywyMzNhMpmQnZ2Nbt26wW63o2/fvggNDUVNTU1j8GQcgAG+Y+uHaslv/Rz8f3VlhbZCe66dO/RA2DNBDyprBkvLi9uC1ur+SwY+6v70rflQ677pLdbtFFoMFvWHvp3/9rrfLoDm/u6nQx002tNKPXft2oXvvvuucaEEYD2A+wBsgxac6lNOeeutt7Bq1Srcf//92LNnD+Lj46GqKm677Tbs2bMHERER6Nu3LyorK9G9e3dUV1c3xinUAdgbWBe7TVMc2bNnDzweD1RVhclkEjEjQUFB8Hq9je3sfx5N45UlwKSa4HQ6UVJSgoaGBpDEggUL8Pnnn8PhcMDr9SI0NFQ7xxRowZiEFoNgbdxVaWkpgoOD4fV6MXfuXK36dXUiZqK+vh713nqt3w4A0CuwKjU1NSIoum/fvsIH/a9//SsqKiqET7uqqpg1axYATeEE0NRtnE6n8JfXn0Nms1lrS/28d6AxdiII4jlhtVqxZcsWAFqMSO/evdES/JVodPz444/weDyNlkAAoaGhOP/881FVVYVTp06JPqgoCn74ITBASH8OlJWVoaysTOxDD2LdtWuXeAa43W5xjlVVVeJ+NZvNqKqqQnR0NB5//HFMmTIFGzZsEMorISEhOHHiBJ5//nm0a9cO48aNE37vs2fPxscffyzqs27dOrz//vstnr8/du/ejSVLlpxxPQMGAOA3t363pRgWcgMkuWLFClosloBlraVa1tMnT54ymbLNZ0mxaBYnySxx8pTJYl1/i1xL/oVNsWjxIm2fTa3gQeDkKZO1YzV1lbnRZ7WyK6ysrGRKSgqjE6K1OrVgxW9qvS8vLxffH3roIU6bNi1galWxK5o1zf+4QKPF3OSz3F3dxPLXgp86mliZWvSz/Y3KWadm/ykzBj+nzvcvUFqyTv5SCYj8U6qf1prsKx06dGhTnfyvp/96ur+w3dHoS95WFyvd+q3XW19+2223saioiMAZ4jtauO4Wi4WSJAlt+aZtP3LkSJpMJu2+8Xfp8XP5mTt3rrhuTqdTzB7oScUURaHZam48/hnyE/jHj+j1KSwsJKBZ7PX2DLh2fm2SnZ0tlulJjE53vKYZQVvri02fF3rSLz0uQFVV9u3blwMHDuSUKVMIgMuXLxcxPJmZmSJ2R99ed1fRn0sWi4Xh4eHC6v7MM88IPfMZM2YwMjJSzID4z7L555rQ9637s7tcLmZlZXHs2LE0m80cMWIEExISxGyG3W5namoqN2/ezIsuuiggg3FISAjvvvtuDh06VLjBZGdnMy8vj4MGDeLgwYPZo0cPrl69mmazmdOnTxfvqKeffppJSUlMTk7mY4891up7Z/ny5ad9Lxn4eQHDZcUg5AZax9kQ8rNBWwl5fX29+L5o8SI6gh3ipSnbZEHyWyTZI7T1wsLCmJWVxQcffJAkNQLfCrH3xyuvvMKCggLm5uZyyJAhPHz4cAAhB6D5gfof1+o7bgQCXXZ0f1s/snDd9ddxwIABtFqtLCkp4YQJEwhoiTv69Onzs5C6cyrBaBOpbuvUeUvrBSRp8v9PgTZ1Hw7NHUQnSv/Fvvc/l5tO08ys+qeerKu1gZUeEN1SIp1zKacj3q0GhrbQHl26dKEsy7TZbMzKyhKBp7q7T1sGsHl5eYKcXn755cKnPDo6mna7PbBN/H33z2JgeKYBWUhISMD5t3Sd/BOi6e49P+Ua2O12tmvXjkAr8TWn6Qdnk7xL38Y/8ZueUVj/PXbsWHFeOlH375Mmk4kFBQWUJIkpKSnC/91utzMlJYVhYWG84IILAgZ8TzzxBIODg0WmY1mWee+99/KZZ55hUFAQLRYLCwoKmJaWxi+//FIMLvLy8vjUU0+xY8eOdLvdnDp16mnfUwYh/3UBg5AbhNxA61ixYoVIXS/50rSPGjWKUVFRdDgcNJvNTE9PZ2xsLEtLS5mRkSGyZzqdTubk5HD16tXs3r07s7KyRDS+JEksKirili1bBCHPz88XFqyQkBCGhIQwMzOTERERlGWZCQkJzM3NpcPh4Oeff87f//73zMnJYfv27ZmYlKgROZ1kR4GwgbHxsQHns3btWv7973/n5CmTNQs3NMt4UzLeGgYOHEiLxdJoKQuD5q+qk0cJGqm8CJpfqMv38lKhkdyEll9sDodDBK4FkJZoNA94OwfS8O9a9NTmLZ7bOVraZVn+WWYYWiKW/kokpy1+1+ZM5KalLJpNMza2hcjGx8c3W2az2Wiz2U5bB913uFlpRc2naR3dbnczMtfSPs1msyCa/kGVTYt/XcPCwhgUFMS4uDiWlJQIcq0reqiqKjK06ttERUWJ47/22mtC8cTtdjMrK4uAppxTWFgoslGe8V4K/2l9ST8nnbg2jStIT08XbZiZmUlJkviXv/xFZDUGQKRDGzScYXCqK+M0VXgRxXXu56E/o5qWdu3anbGPnu6a+6/j32ZN+5Esy/zXv/5FVVVpNpuFb7v/NroqUHh4OLt06cL8/HxKkkS3281JkyYJhR2d+KelpTEnJ4fx8fE8fPgwjxw5wqSkJKalpfGtt94yCPmvCBiE3CDkBlrHCy+8QAB8+umnSZKxsbHMzMwUShEbN25kdHQ0rVYrS0tLhaLAY489JiwjH3zwAV0uF5OSknjgwAG+8847BDRZw0suuUQQ8uLiYtpsNiqKwqKiIoaEhPC5554T1rEXXniBW7duZWFhIY8ePcrMzEx6vV6SmmyWoiiNU84S2Klzp2bns3DhwmZWkTlz5nDu3LmcNWsWP/roo9O2x9q1awmAy5YtY6g7tPHlqLsRmEHV4kem9CAzGUTfwBd/dHQ0u3XrxpiYGGZmZjI1NVV76UX7vfRSoKlq9G7hRazg7OQJ0cI+WioqiDzfd2ejMoTFYhEKF6qqcubMmc23zUSgfBw0MulPDPxfoMOGDRMkLiIiQjsn3Uquq5bYESBx2ep5+YIrVVUNkCn0P3bTdOttJSptIROtlabkuq2lVRLdQhBpgHrHOR7D36oZ0Kfb0GdaS28vpPzOsK1/G7W1rf3JW2vW/4SEhGak1D848azdsdpYmtZNr8O8efMC6qMrpQQFBTU759DQUKEKg9Am1+E0Az7/320eQLZQ9NmE1vqVPsDSDSwBgcrn0Af1wZW+zOVy0eFwMDg4mFarlSkpKZRlmXfddZfYpunAOy0tjZ06deLQoUOZlJTEuLg4tmvXTgSM6s+a8ePHMy4uTrwLSkpKOGnSJM6bN49HjhxhVFQUL7nkEsNC/isDBiE3CLmB1jF9+nTKsix+n3/++YyOjqYkSczIyBBTjSaTiaWlpeKh2K5dO+F/eOONNzI4OJihoaFs166dsBIFBwczKytLEPKUlJRmL0+r1cpBgwYRAAcPHsx+/foJS70uwxUTEyOsjCaTiampqezVqxdvu+02YUVJS0tjUVERo6KihHXl6quvJtlIyNvaHqqqMj8/ny6Xq9nUsizLrauUNJFaPHr0KCMjI6mqKq1Wq/BFFaUpEfo13Takxk9/pQbdQqZfg59KVvyn6lsllG0575a0uv1e7PoxY2JiAvtYnN/6ltbrqfuz3nnnnY3rOHFWCin+ZEW3YrZGTlp0G/iV1HzOdQBx2n50mvJzuMmciVi35AJyzmS8Defkf631GRtdj9t/ua4THhISwnvvvVeoyuzdu1c8bwBos34zoT1DghFgsW/pPjwtMW7hfjqbtmj6jNblQQOOeRb3haqqlCSJDocj4HkwatQo5uXlNVs/Ly9P9FGXyyV822VZ5lVXXcXk5GSWlJQwKSmJDodD+OoXFBSI2ZV169YJQl5aWkq73c733nuPZrOZM2fOpN1uF+4tQ4cObdGQczr4uzY2lRSVJClA0ai17ZrC4XC0+fjk2b3X/l0AQ2XFgIFGDDhvACSrBEmSIFklvPHmGwEKFz/88APq6+uhKAoWLVqEdevWwWq1Ii4uTqyzYMECREREICcnB2lpaXjqqadgs9nQo0cPHDlyBPfddx9CQ0PxySefoKamRmxXW1sr1AIURUFERASCgoLw/fffQ1EU1NbW4pNPPgEArF+/Hv369cPx48ehqipOnDgBSZJQUFCA+Ph4HDhwAA0NDYiJicHll1+O5ORklJSUQFVVXHnllbjlllvw/P+z9+VhUlTX229V73v3TM++9Oz7AsOwycCwCAqoiKCIgOOu4L4vEVEToxEjLvncfknQCNEEjUvURI3RaDRxX6IxGnXUCC4oIjBsw8z7/VF9b1f1MtMzDDBAn+c5T1dX3aq6devWueeee8577r4bLS0tMoPccccdhwceeAAAcMkllyAtLQ2KqrWF2WXGwQcfjHvuuUebVQMoLS1Feno6TCYT0tK0/PLd3d1wu90ANISEnigzMxPr16/HmWeeiW3btuHtt9/u+eV0J9gfP3FoQvL7/QAiaAkxpEAb+gBkZmWiqzuC1LBu3ToAwNSpUzF37lxjxsUoIJvsnGyJwiDQOcxmMw477LDIrcKpxgENTSYmeyOQ+Lmjy4TrDCskwg1JbNiwAVarFQ6HA19++WUE5cMMLdOs/rn1l+yO3FhVVWzevBnXXnttpMBWRJ75iPD1XOF94f2KqsDlcsFms2HDhg1oamqC2+2G2Ww2Iv9YNKZKWW9xX5lZcke4jlYYMoYWFRUBAGbNmmXMIlsQ/s1EBAEoitLS0hAMBg37EqLn6Ci63yREwTkm/u6cnBw4nRoqi3hWQTF9wIJe81vr31U80suZZM9JSOy9iEAvEe9OVVV0dnaitLRUlrFYLPjpT38KAKisrMTPf/5z+Y2VlZXhjDPOwO23364V/jGA6wB0APDDkHlWn5o++v5xKc5j96Ut9GhEiqJg06ZN6OzsNN6zM86JcchsNkvZsmXLFsO1f/e73+Hdd9+NOef999+XmVhFdtTvvvsO3d3duPfee7FmzRp8+22kgebNmwdVVfHqq6/i3HPPhd1uj+njiqIgMzMTqqrirrvuwrZt26RM11NXV1ffMtoCuOmmm5LOtiyor/fY2fP2a9rdM4D+cMpCvv/RpMmTNHeFYWGLjAq5bP3tt9+SpFyi9Hq9POmkk0iSFouFOTk50mVl1KhRbG1t5fHHHy+t5/Pnz5dLjwKj98QTT2QoFKKqqrzxxhsZCAQMPorHHnssJ0+eTLvdLv3YhUtLe3s729ra6HK5mJaWJv0G09PTuXDhQgaDQWZnZzM/P59vvvkmbTYbf/WrX3HcuHE88sgjpV+px+OhqqrMyMhgc3Mzr7jiioh/uAJiPohWxLgzLF261OCjuauWvncH92rpzoIWBBtlGRRWqkTnxwte66nMgHIcJBJFUThmzJjEy/g9IGwktDj2YrXWWxNFHxUJVAxl66B9b8WJz++Ne7Q0x8ksKlYPNm3aZMAuF/Ef4r9Ab0l0/R4DEPXuNQlWOiorKw19qLCw0Pgeo1CJ+mLBF+86oZ/8LuIeM5vqWLRzdJ/wer2cM2dOpK/qM3km+84HOSdyc4r3fvX9Ub8v+lu22WysqqrihAkTaLFYOHbsWC5atIhms5l1dXW02+202Wx8++23mZ+fH5N1OZqzsrJoMpkMWWdF3SsqKlhdXW3o/y6XizU1NbRYLAYUHYF3L7KohkIhGT9TVlbG+vp6jho1iunp6TJ4uaioiGlpaQwGg/L7zMvL44MPPsimpiaqqkq3282xY8dy/vz5POCAA1hfX0+73c7GxkYeffTRXLp0KceMGcO33nprT6oWSRNSLisphTxFGsEGLYnIZGjoICqk767FYjGkXxfLgXqF6uSTT5ZBnUBEUfN4PPzggw84bNgwg+DVR+YLNqSWjsPiXD3Mmj5hht1uZ1ZWFu12u3STmTBhAm02GzMzM1lcXMwhQ4ZIF5epU6cyLS1NJhlJS0vj1KlTtfs5QZRAU0itkMpXtMKyqwaogea9efDe31j4xvb2zkSQY9yy+kmJEvULzUc5NzeXLS0tMf1ZuFfk5+cnRsTR1UHfd3emn8VM7kQcQT+vF53QZ1dxvOvrFcvoNjGbzfT7/XHliCgrgjv3ZD+Mxzsjp3qagLtcLjocDqngir4t2kD44gvltLCwUMI2OhwOlpeXs62tjUOHDmVjYyODwSA///xznnLKKbJvZWRksLa2lu3t7fJc8Z4KCwvle3Q6nczMzJRjnohreOaZZyRqzaxZs2R/ramp4ejRo2mz2fizn/1MPtMzzzxDh8NBi8XCM844g2+99RYBzaBzxhln8IADDmBxcTG3bdvG2bNnS3Stiy++WLpeTp48Wfq/P/vss7TZbBw9ejRbW1s5Y8YMGYP097//nXV1dbzzzju5cOFClpaW8qKLLuLepMshpZCnFPL9nc4//3yjoFSgBRH1IIj1ikBPQlY/oOhhqwwDTT8G3OjI+ngDochEF69+Yt+sWbM4YsQIQxpqaU1LRyQzos5H+fzzz9cQLRRovpwCQQURZT3a8pIInWB/4kT9RJ/OfCCu15PVWlXVhDBwon/3tmIwoD7WA8A9ZhKNxwqoqNrk9eyzz4581wpikFVifO6BhP76e6p9enruhNb7PsqcniZI+oDa6D6Znp4uFTL9+xIY4QJHPdoCG5f7GsA9QG0YXcbr9XLChAkx/Q9AxHfcH/4NhvubLjZCGFD0BhWHwyFhKG+77TYZ+2MymZienk6TycT58+dLZByLxcLbb7+djz76KA855BCGQiFedNFFfP7555Me9y6//HIp681mM2+99VaJ1iOgE0X8iYjzycvLM9TbYrHQZrPx1ltv5fz582m1Wg3QjiRlrNQBBxzAH374gYCmwBcWFsrrfvjhhzzssMNYXl5Oi8UigROefPJJVlZWykDWr7/+miaTSaKcCUShww47jMuWLePixYvZ0dHB0tJSnnXWWZwwYQJvvfXWXaU6DDgh5UOeov2ZXn/9dTz44IMRH0IVWsa672PLTpgwQW6TlH54Pfkf6v2FRTmRPU6SGX3OYKk/33Ct8H3MZjM2btyIrq6uuPUT5zzxxBP417/+JX2jzWYztmwJp178DpHMiDof5ZtvvhlffPGF9v9baP6SOyLX3bhxI3744QfD/fQ+jfsrJeonGzdu7PF4X68ns7DGkbTd3d0Rn2xRLPxf+MDG88nVk/BfHSzUYybReESA3cTWrVtx8803R3x/CWCjsajB516Q3l8/Du3u9unpueP5jgNILi5BR9u2bYt7L5PJhO7ubrk/Ly/P4AMvsvk6HA5DzIDH40F+fj66urpgMpnw/fdxBG409c0NuU/UUxuKeosydrsd7e3t8rjBd1zIcdGPvgVAYPz48QAAp9OJ7u5udHZ2QlVVKIqC7u5uNDQ0IBAIoLu7GzfddBPy8vLQ1dUFv9+Po48+GqWlpXC5XMjLy8M111wDALj66qvR2dmJt956Cw899BDq6+txySWX4Oqrr074LAdOORCKRYFiU/CTn/wE27u249DDDkV3dzfOOeccrF+/HqqqYty4cSgsLJQZZwGguroaF198MQ466CAUFxdj2LBhyMjIQFdXF1paWvDkk08iJycHzc3NxtgPRPqJoL/97W/47LPPcOONN6KhoQHTpk3DCy+8gIkTJyInJwd33nknSGqxXIoCmy2S/lhVVSxYsADNzc047bTT0NLSgrFjI+llnU4nJk+ejI8//hhvv/025s2bl7A9UtRreEqKUrT76IUXXkBhYSFcLpcmeHoYbP/+97/LbSEoAMDn8xmEjz6Nswgg7JUITZgnoZgXFRVBVVXk5ubKfSKYTdQrJycn9sQqyK9PDC7btm1DZ2cnzGYzVFXVZsxqz59oKnBm11DCwMD+UgKl67vvvjMW629wX5LUW3/qLw14e6UoIekVIkGKoqC2thYZGRlQFAVOpxNffPGFDFbVK5xOpxNLly6F2WxGZmYmLr30UjhdTpDaxGiXypQ4wbw9UXS/yszMNPxft24dPv3sU02WBgE4dQdVaDK8IbLL6XTKsWPLli1wOBwyGNNqtSIrKwuKoqC4uBg2mw2dnZ3YvHkz3G43XC4XfvGLX6CwsBB/+MMf8PHHH2PmzJmwWCwYOXIkPv/8c1x77bU47rjjsHTpUnR0dOCNN96I+1y19bV45vlntMDruQAWAF2mLvzxqT+iu7sb2dnZALSxITMzU8oJMS6sXr0ara2teO6552A2mzF79mw5YW1vb8fmzZvx5ZdfYvz48eju7sb27dvx7bffwmq1Ys2aNQiFQvB6vQCARx55BJ988gkyMzORlZWFGTNmYNu2bfD5fPj888+xdOlS+Hw+XHPNNSgoKMAbb7wBRVHw+uuvw2Qy4d5775UToa1bt2LTpk0YN24cHn74YWzZsgVz587Fn//8ZxQUFCAQCPStA+xvtLtN8v3hlMvKvknR2S4Vi8KcnBx6vV42DmmMv1wZXp5W1Fg/SMRZ7pw+fbrcLioqilnylC4BYtl7Ut+WUcUy9Ny5cw3LpjabzZA5Tp4v7mOO80xh1xsRgNOv5XYFSUGhpbhn3l2uDjsTSDoYfXr3R9a/w93hDib8lvvy/uNhw4sEa4aycQKQk2J9ZtE+wAwKTpQbANBcbaJdt04//fTIOdHyLoELkHDNqaioMMQiXHrppdJNR7SJqqpUVZVVVVXSd9tsNtNsNrOyspJLlizhsGHDpC95RUUFjznmGOneIXDQp06dSpL85ptveMQRR7C5uZnFJcWaa2EAkczK843tJqBGowOxXS4Xhw4dKrN9ut1uCa8r6qKPbZo4cSJNJhMDgQDr6urocrloNptZVFREkgTACRMmMDs7m2azWUL7trS0MD09nTabTQaLiqDOSZMm0W63s7y8nKqqcsKECXQ6nczIyKDX6+Ull1xCkvzJT37C8vJyjhkzhh6PR0L77i2ElA95SiHfX2jFyhVUHaqGmDEWmr+oXkkV2/4EQjwBykY0i+DNaNb7VTucjp1WYvVJIerr6w3XP/jggyODthC6FmjPrrtGRkYGq6qqYtsARl/iuH7Frv7Xfb/h/iobPXECH/GEytIA+t7Gy4YZw7sTL34/5V0xMdIntkkUBFpTU2PoAyJNvJjMl5SU0G63y0Rahx12mCwnlEwpNxVoAeP9lYP686K/M/0xHXqQ8D1WVZU///nPDe0YCASYnp4u8zS4XC5eeeWV/NWvfkVAk7FtbW1aeSuInPB9h4M4VLvG2LFjSZJz586lyWTivHnzePfdd7OwsJCFhYWsra1lS0sL//SnP9FsNnPs2LEMhUL0er1saWnhxo0beeSRR9JutzM7O5v/+c9/mJmZySuuuMKgkP/617+m3W7n8uXL+e6777KkpIQZGRk85ZRT+N1338k6vPDCCyTJvKK8SHsshqaQC16sGZv01NbWxlWrVvU4nq5evZpZWVlctGjRLhitd45Wr17N8vJydnV17emq9ImQUshTCvn+QqHykGZVGQtNSXEgbvAmAKP1JQHrk8To919zzTUxZaOT5fh8vliFXj+IZMYOMsKCEY36IJJuCGsKoCnaUokWCvkVIKqM17Tb7QkVPL0VLmnrbbTyt7uVsx6g++JyTg/n9KHuCaEElfjXURSFfr/fCHGnL59kEpyESZiS5BjFTsD0xen/ejQRj8cTSbsezSVIrZjsRk5aOU8wOTSZTDH9NxgMyuBzPSRhY2OjtOwq4dTuIg27oig88MAD2dLSIjPFigDiKVOmkGRsUio/4n9nNQmewZ34+WRwr1WJPwlVIyuHOTk5/PDDDzl69GharVaGQiFeeuml/NnPfiYRqxRFYUNDA2trawloclYq5EOhKbMKiBNAlEGW0Vu7bTYb8/PzpRXZ6XSyubmZZ555JoEImg+gZS1ta2szGD8yMjLodDqpKApHjx4tVx0qKirkM6elpTEzM5N2u50ej4cul0ue19jYyIaGBkMboALaWHAlNCNNPWixWjhy5Eh+9dVXfPHFFxkIBFhUVMTGxkZ+9NFHMWPpPffcw/z8fC5atKhPCYN2B4m6/f73v9/TVekzIaWQpxTy/YWky0kGIhnf2kCcGR4YXGGBVZtA6FfCgAkcbyBMT0+XKZeBxO4BCQfRaHg2W+Qaw4cPl+fqMx0+//zzBLTBxmKxcPiI4fGv3RIZlA0pl3chekHMc+0KNse5vhOJldpkM0xaYvf1yd1DCbd3D5M7MbAC0DC447VTD3Xd3fjvgw1hZW/lXr//HjjZd56sou52uxMiNQGQFu941xbwq+K/3++XEzWbzcaTTjqJqqrS4/HIrJwIQJOzGdDyG7jR4zcSCoUikLPhzLlWq5UWa6SOK1au4DnnnGMwWAAwTLa9Xq/E4pbHrZFrpqen86qrruK5554rEU9effVVbt++nS6XSyrkqqpqsqEt/L6mgzBr6FQ5OTlUFIWZmZkG9JV4bQxoE9to+FuPxyMnLsLQAoC5ubnyGiI7qB5fXkATPvfccwZUF7vdTpPZZDQOOEHMjrRBWVkZMzMzWVtby40bNyZlIU/RwBNSKCsp2ltp/fr1uO2223otJzJHFpYWAnZEkEEOB1AMLRDHAmAstEA4T/hET9SFPgPQBRl4qaoqbDYbzGYzvF4vTCYTvvvuO2zYsEFmZiRpyE6nqip8Ph9MJhMmTpwoy0nKARBAJPTZqQXdKYqCUCiEyspKkBrCy3PPPQdFUWQE/5YtWxAIBPDqW6/Gb4i/h+uPKCSIXoKe7Ha7ITAvYTBdvN1iH5MoGyYR1e/z+QBEggJFYJnYL+l4AEOiLrIZWjs2QWtLv+5YOrR2cOvqYQpv50ddxxQuFyYDGs84ALWRY3a7XQa0mUwm7Zm7AGyLfTZBerQGDEdsOwEa0k0UiXdA0pjxMh4NYOxjQgSRPgbO7XHaA6OQHn2EZNxAybjv3wJDFlhDEG4WDEGFBx98MEaMGAEACfuFoiq4+JKLYTKZcOaZZ+JHP/oRRo8eDZPJhPfeew/Nzc2wWCxwuVywWq3Yun0r6KVWhzIAdq3+AHDBBRfITMXBYBDz58/H1KlTYTKZUFxcjN///vcgicmTJ+OZZ57RKrAeWn9ZCw0RyAetj5oAqNo38oc//AGzZ88GANx3332YNWsWioqK8PRTTyM3Nxdr1qzB9m3b4XK54HK5cNSRR+Hdd9/F9OnTkZWVBY/HA9WqAtsAxazAbDajvLwcn3/+uYY8MwLAYmjBjVYNKaWzsxOjR4/GI488goKCAiiKgpdffhn33HMPVFWVwZ1ZWVmwwALTKu0bV55WMHniZHR1dWHLli0wm80YP368QU499dRThm9VL0M3bNigZUUO73M6nTLI9Ze//KVE9/rqq69kduG0tDSZvVm8a/FOZs6cKftXeXk5rrrqKtht9ojKD2hZdrWkzKgoqsBbb72FW265BQ6HAzfeeGPcfpOifZR29wygP5yykA9eam9vl8kNamtrE5ZbvHgxb7jhBtrtdoZCId5+x+2aD7mwmuh96Y4GMTq8vxjGZVRH+L+wVIYtryNGjGB5ebkMbhFW55ycHBn4grCVQ5+FTSxJOp1Oo6UaOkuRsO4WGi1FPeGPl5WVGc8Vrg9RFnCLxcKFCxfGt0iJJeE41uFeOZ51T0H8oM94y+eK5spz6aWX0m63S1/PQw45hLm5uXS73QQQsbSJZ7XFsbD1x3c7enVCNVrYes3oOZCcZJCaqqr8yU9+Iv/H9KcoVhTF8Bx98UXe5wI6xTe+K/z8FSM+tbDa6gPgEranCUa3pfC3qJpUbfVNjeyDCZq8you8I2FJFVmF09LSaLVF3vmFF15IknQ6nczJySFJLl++nKeffjrffPNNTpo0SVpHTWaT9g20he83F1owoE4+qKrK4uJiaQX2+Xy0Wq0sLCxkfn6+TDhWVlYWkQXVSLgaoCgKr732Wo4ePVq6edjtdgYCAWZnZ9PtdsugQYFlPnz4cJ5xxhlsa2tjZWUlDzzwQA4fPpz19fW85ZZbqChKRG64oa2SuUDUgxhhvL/FYmF9fT2Li4ulxT07O5uzZ8/mxIkTZcBnXV0dAfD2228nSZloBwCHDh1qeL9jxowx9AeRIyK6fwBGN7S8vDyZxA2AwWUskZucSC6nKArdbjc9Ho+8RlZBFhVVYag8xBUrV9ButzM9PZ2NjY0MhUL0+XycN28eR44cKfvAkiVLuHTpUjmmPv300yTJZ599ltOnTydJPvLII7z22msHfIzvrczKlSsH7J6DgZByWUkp5HsbiY/1kEMOkUJKLDmWl5cT0HwXX375Zc6cOZOApkz5/X5DAhyoIEJh4SyU7qiBoadBV0SYCwVH/EYvKZeWlhqUc5vNxpycHObk5MjlYHmOOTxgTI4M7PHq5Ha7taVInVCX/sgqjL9Rrhs2my1+trkgIgGtVuO9xaQiehAQ9TeZTNog7YFMhhGtZBiWZuMkV2loaJDtGQgE2NbWxpaWFtbU1BjehSFDah00NyKdMu1yuThx4kQCmoKvqipvueWWnlOc6xSfaLZarbTZbKypqYlRuvQuTCNGjNglSmv0NefNm2dA2NGjbAglQbLX2I9cLpchKYk+IYtY8hb/xfvSK5F7DYuAwXhuS3ZdGSAhytGu5KSzPXrRsxuLDREFHpFgzHjXj56M9dTX+vq+a2trDfecPHky/X6/9EM3mUysqanhSSefpJXJhFHeKmBraytNJhODwSArKiro8Xh49tln85xzzjEohaFQiGPHjmVlZSUVReErr7zCmTNnsqysjGeddZYM5rv33ns5fPhwFhQUsKamhjt27NDuZYEm74OI+LHr3kttba3BbzoUCvHCCy9kXl4eL774YjqdTppMpsQxFGGOpzAnal+9n77+uldffbXhnYlvcvTo0bzpppvk/tzcXImIIpL7AJpLnJgMqarKf/7zn/zuu+/46aefkiTtdjuPPvpokuSqVavY1tbG9vZ2pqWl8de//jVJo0KuJ71CvqvG+J5oV95/TxFSCnlKId/bqL29ndnZ2VLgZWVl8e677yYQ8X3Myclhenq6QREWFgyfzxexQuSAKNIJTZ21LNHgpR94ysvLDRYMwdE+hPF4xIgRscgDPXEv/qWXX3557P5E1r94yroCozU9fL+0tDQ5EFgsFublaRH7hYWFMnhLllfibCfJArrLarXSZDLR5/PJ4KTosvFSbgsuKyuLGfDiQbDtKd6Z1NvxOAZOU88Ze/55Bz0nGbxrQCPStfvo0aN7fKeJ+t2e7I8DGXugn+gWFBRQURTOnz9fyrZgMMjS0lI6nU65yqdYjc9eUlIS8ROHpsxaLBZ6vV4ZVCkg7Gw2Gx0OR0zmZP31hPyoq6szrE4m+0wmk4nFxcVSzgzUuxJ1jpZp+mcXK4EAOGnSJEMmWlHO4XBI4xMA/v3vf497P+Ezr392h8PBxx9/nLNnz5b7zGYzMzIyWF9fb5CtQt6bzWZecMEF9Pl89Pl8vP7666WSX1RUxOLiYpnKvr6+nhaLhbm5uSwpKeGQIUNYUlLCqVOn8sADD2ROTg5DoRCzs7Pp8/k4bNgwfvfdd3zttdfY0NDAhoYGXnDBBVIhb29vZ0tLC4cOHcqhQ4fyxRdfJEmOHDlSgytubOSNN97IHTt28IILLmBzczPr6+t5xx137DEdpb+ElEKeUsj3NnryyScJgDNmzCAAaYkBIlHuZrOZTqdTKt764Ji0tDTa7fZYBUaBFugZlT5bURQee+yx8v+IESMMAlZcV48gEJ2eXC8Q9QLPIOgTBTbpl1R1yoN+UE04wI6Goa6yXHp4vwm9KiSKonDBggUSOUAMGjk5OWxpaUl8rg7rvLq6Wr4jgTCgZ5fLRbfbTYvFwmAwyGOPPZYjRoxgVVVVLBJJnPqJNM4AZLrnRINob8qI3iKsD5pKdrDdbZwEElCid7I/sl7RScjR34IZrKisoKIorKys5BtvvCGVqaOOOoqqqrKkpETKmfT0dB566KGGa+jljLAYAz2kte+B9d/gznBPCmYiyMMBb+so1rdTaWmpoc3Es3/yySfye9c/g0gnH/1sCWXkHvgOoicE0SsW0e0ejXIVLYuESxLQw5gS5srKStpsNno8HrkSNnbsWM6cOZOLFi1iXV0dzWYzTSYTzznnHNpsNo4bN44rV65kbm4uAQ1UYO7cubRarRJHfMiQIWxsbORtt93GCRMmsLS0lDU1NXz77bfp9Xr585//nCQ5efJkBoNBbtiwgdXV1XS5XLz99tu5ePFiDhkyhMuWLWN9fT3/9re/kaRBIe/o6OCWLVtIkh9++CGFbhZtIb/zzjv54x//mCS5detWDhs2jJ988slu00sGgpAK6kzR3kIHTjkQil3BQQcdBAB4/Q0ta9eDDz4IUkuzGwwGZdCLxWKR+3Nzc1FUVAS73Y78/Hxs27YNHk84alMERZmhBXWGg4tEICBJPPfcczKop7KyEhaLBW63G+Xl5QgGgwC0YKTGxkaUlJTA4XAYsmdmZ2cjEAjA5/Oho6MDJpMJNptNBpwC4Xt6oAVPReJAgdd027o4sPT0dBnwKLLiAeGsnSJm6B/h02w2dHV1RQLC1kErYwv/6uO/MrR9ZqcZw4cPh6qqeOedd2A2m1FcXIyhQ4eitbUVmZmZMrsbAC0DnCN8PTcAb6T9vv32W5lZ7a677kI0dXR0YNOmTbLsb37zG7zyyiv4z3/+g88//zymvJ5IynTUAFBbW2tI5x1NvWWmFIIK0AJlDdSD9DKk0N4dFCcrutPp7DmDZfwm2S9o8+Ykcq9HdY2cjBx89+13UBQFX375JT799FN5rLa2FsFgEOeffz6am5thtVrR0dGBP/7xj4ZrWCwW+U66urpk30qY1j6KROC3uEaP7zdJSvRtALHfR9zg0zikr9emTZv6XKft2yPRyx9//LHcvvjii2W9xo0bJ7/3srIyWeazzz6T37/+2fTPYmi3gfwOLL0XcTgcyM7ORkODlsKzuroaJOFwOGR2ZYfDIdvabDajpaUFfr8fqqri7rvvRlpaGqxWK9ra2qCqqhzD1q5diw0bNmD58uVQFAWqqqKqqgoAMHz4cPh8Pqxbtw6dnZ3o7OyEqqpwuVy444478Kc//Ql/+9vfYDKZoKoqZs6cCY/HA6/Xi2OPPRavvfYa5syZAwDwer145JFHAABlZWVyXP3Pf/4jAQYmTJiAsWPHYty4cTCZTJgyZQoAYO3atTCZTHj00Ufx6aefwm63429/+xseeughvPfee/jlL3+Jf//736ipqcGnn36Khx56SLZdZ2cnTj75ZNTX16OhoQFvvPEGrrvuOgDAhx9+iBtuuAGAFjh7xRVXYMiQIRg6dCjeeecd/Pe//x2AF7yP0+6eAfSHUxbywUMrVq6gyWrSfKvbQIyD5gcYDrh74IEHpKXg6quvZmlpKe+77z56PB5pqSksLGRLSwtVVeWaNWvocrkimLLCTaMMWoCn8G/OjFgqTjvtNOm7nJOTw8LCQlosFra3t/OMM84gAF5//fVGK4fOr7O+vp4VFRUsLS3lKaecQo/HwzFjxkgrPxB+tgxoAU+NOkuOgPWzQcO/RcQqLHzihVUkNzfXAKUXE+QnLENORHymE1iLXC4XfT4fFUXhvffeSyC81CrKm7RMpDFQY+Fj0FnKhHWosbExJrAtIyPDYD2K14a98S61TutxyndFAOAAsv49DISlc69nnUtWXyEbRdyE8NEfMmSIDBAEMDAJbQYBR/eTnXHPEP0v0ffY12urqmpYkUzEidxX9PEV+hUJUSbeasOuCN4eSPek3mSd1WqVboXi3iaTScb7CNcr4V708MMPy/gY4boSDAZ5zjnnMBAIMCMjQ65yBgIBulwunnjiibRarZw+fTotFguPOuooTpw4kQsXLiRJ/vOf/6TP52NBQQG//fZbDh06lFlZWZw7dy5tNhszMjK4du1ajhgxwpBldO3atWxvb2dpaam0kC9ZsoTnn38+u7q6ZMZYUrOQl5eXS//2I444gjabjWRyPuiDkbC3uqwA+DWAbwC8q9uXBuBpAP8N/wbC+xUAtwD4CMA7AJp6u35KIR8ctGLlClq8lkik/5XQovwB4oCI0BZCKhgMsqCgQH7kws/OYrHQbrfT4XDQ5/PRbrczKyuLikmJKH1uaAE+Svg37NYhUv4uWLBApjH+/e9/LwMQHQ4HXS4XD2g5wOibrVPkzjvvPOlWUVdXx+zsbHo8Hs6YMSOCj34iNIW8EsQERALy9MqzPYJpq/cd1rvm2Gw26ddeXFysnSeCv/TKQJzMo36/nxkZGVRVlVOmTJGDmP5+8RQKVVXp8/sM+/R+9MkMqineOY5WLuT70gerJplwaJ/hAXhePXpGwgRQg433oWyp1153beS/3n1PNznWK7x6hVX/7vSxJSaTSUtr7459n8IdZqB9/Pt6PfEc+gyq0c8n33WC3Bl6Nx4xFjY2NkrkGqvVypKSElnO6XRy1qxZEg3sxhtvlOhdwmVGuBb2Vn+r1dpjuUQTn7S0tJgkeNHPHQwGWVxcLN1sFEWR7qqXX345Z86cKdtNZLG2Wq10u9286KKLOG7cOFZVVTEQCLC+vp5TpkzhmjVr+PXXX7OpqYkk+dZbbxEAP/vsM5JkSUkJOzo6dqm+g71YIR8HDWVYr5BfD+CS8PYlAH4W3p4G4E/QFPNRAF7u7fophXxwUKg8FMmmKWAKz4YGWRVWeAsLC/nZZ5/R6XSypqaG1dXVPPXUU9nW1sabb75ZBpkUFxdTVVX6/X5WVlayurqaQ4YMMQoEa/jadlAxKZw2fRpbW1tJajP1KVOmsKysjF1dXRJ68c9//rM2cxdW7GpoKClHQ/qjNzc3c8qUKUxPT2dOTg5tNhubmpo4Y8YMbZA3Q4v6F4lkZoJoiBVWJ5x4Al977TX6fBHlV/iVCouHvrwMXlV1zycGM51inR5M5913323IAqqHMIsRnAcjLjSfqIs+EGkgB7XdxXtjvZOqc+aer+fexnqlQo9gAaB/8KADwYPMyr7bWL9ClUQbRH8T0Ul49gQnuyIhMoYCPUwERRxJguRu8RRiv98fg6ikZ/2x3Nxczp07V1rOe4qtsVgsmpErDjSvyWRiRkaGAdJRrN6KAF1RVkwAVFXlsGHDZHkJ6wttUnHKKafI9jz55JPlanF6errMgPq3v/2NFouFFouFq1at4iGHHEKbzcZ//OMfHD16NO+66y4ef/zxvP/++3n88ceTJGtqavjDDz/w1ltvZXNzM1esWMFPP/2Uo0aN2uX6DvZWhVyrO4pgVMg/AJAT3s4B8EF4+04Ac+OVS8QphXxwkKIqmuC1I2IhF9wGwoYez9cvXbW1tcmlsmXLlvGqq65iYWEhzzzrTKqm8Ew8rGS6A26uWLmCnZ2dMnjknnvuoclk4hlnnBEXcxWANmlohaaQXwmZUrk3sjqtEZechvA1XKDL52JHRwdra2sTws8J4agXpGJpL6a8mDSE/4vyYhVADBbi/LjsQWSSJAbFKOVcBLXGCP2oQXR3Z5pM8V7IvSheu7MP6RUEAw+gRXqnJ4Nx2stqtUbgSaHJDBH8rr9vX++9025RfW03V/z9aWlpbG1tjdlfVlZmkI8G44JtYN/bvsDRgcbBYJAul4tHHHGEoW+I/jRz5sy4mVz1fUNRFA4ZMoRHHHGEnCDo83ZE89tvvy3vEQwGmZaWRovFwlGjRhkQc0RdBWxjXl4eTSYT58yZw6amJubl5bGrq4uqqkqEH7FynJ+fT1VVabfb6Xa7WVdXx8mTJ5MkTzrpJD7xxBM88sgj+Yc//IGnnHIK7733XonfvysJ+5hCvl63rYj/AB4D0KI79gyA5p6unVLIBwdJC3klIgrr4vCvG5w0eZKh/M0338yqqirm5uZy/PjxdLlc0tI7btw4ejwe5ufnEwCzs7O5dOlSbty4MUY46FPTC8ESbT0WeLEGYaS3ROtQMNLS0gzXLSgo4P3338/Ozk7a7XYOGTJES1oUtcSuKArT0tIilgU7iFIYLEXCFcbr9cq66LHGG4c0GtPLeyLXdrlczMzMpNls5qZNm+Q1s7OzOXLkSAaDQeOga0V8N4CogdKA965nW5x9+wInUBp7S9QTjwfELSKBxWzQ8QD45O+Lk7qk+k0SFuKEbky7gaORpgbkPSmI22eEe4TNZjOgMimKwkWLFjEjI4NNTU0RA4EDGgLVldCSrx0KY0I1D/Yr965otxgxxvn9fo6fMD7heYWFhXLcUVWVbrebhx9+OIFIDIaqqiwsLGR6erpUoltbW+WYlpeXJydNiqJIhdxkMtHhcLCqqoqKovDmm2+my+Vieno6c3NzOWTIEFqtVl5++eUkjQp5c3Mz8/LySJImk4kHHHAAV61axdNOO42FhYV85513Elq8f/Ob3/DSSy/liBEj2NXVxZEjR/LUU0/lo48+umuVHe7DCnn4//fsg0IO4BRomBavFRYWDlATp2hnSPqQ+6Ap5ULJtYE1dTWy3PXXX8958+bRarVKBdZqtca4UIiBQT8wOZ1OHnbYYTHCJlnrT1+sRGKQEoKquLhY4m7ry7k8LhYUFBDQ/CClZccF4ngQ2bHXFpOPzMxMw4AeCATiDvBxrec6Ab0zAj7hwJ9gkBMTlsHGSSkRPQzc0djVu5r3VCBnnwPh9leXi4Hi9OTKCUtmDAxosu3fSzm9nOgz1GHUd9MbvGkMR63KiQDOyZMnS4g/MSaoqsr6+vqIHBwH4mJoCrkNxBG6awWi7jPIA7kHmk0mE4uKinoeA3T9Qow7qqqyrq5OYqWbzWbpjpKZmWlwZ9HDdwaDQUOs0tNPPy2PWa1W6WZ07733ylWdnJwcnnnmmfT7/czPz+eTTz4pM9Sed955PPzww+l2u9nV1SUna/fddx8vv/xy+nw+/vnPf2ZpaSmff/55vvvuu9y+fTvfffddktqqekFBAefNm0eSnDp1KgsKCrhu3bpdru9gH1PIUy4r+yCtWLmC6dnpUgCn56ZzxcoVhjIrV640BLAoimJIWa8XNkIg6Pf3VSHUX7ev6A2JrhOtNOutBn0RpkIYRh8TExOHwxEfX1uXdbKoqCimPv2x9u4XnFr2TvH+xP20HHs8HunPG+/4s88+G3d/wkyxOkW5vr5eGjCysrIM8k1kcfZ4PJHMtDYQPwKxANr3O193vTpoVvME7jFxJ+lJyoCBWCUQcjjRuJBov8vl4rhx4+T/I4880nCOqqoMBoM8//zzDTjw8tmGhX8TxE3EG3MqKytpMpmkRVvsdzqdLCgokO2hfya9W4uol3A7EVmTzWYzq6qqaLfbZTmPx0Or1colS5bwkksuYSAQkPcPhUIsKipiRkYG8/PzOXr0aJaXl9PpdDI/P581NTW86667pD6Rn5/PO++8kyR5zTXXsL6+frfoOtjHFPKlMAZ1Xh/eng5jUOcrvV07pZDvXbRs2bKkrHS7woLYHyGrD8rUsxxIElmm4ghD1aQarudyuQwBOHoostmzZ3PUqFG77Tl3KevgF6OP6WEVo/f1yPuict0Dsg6AuBPXvr7r6G9PP5gaygplbiJ26n77E++MzIr3HSTNSVqGe7S874Kg196eJZ5lVyh8RUVFsQmCBgrWdJCv+iRKRCRQTQRKSXl5eaRcqe4adhA1IEqgrVgfqe1ftmwZSXLHjh1cv359DOTg0qVLuWTJEpLk6NGj+fvf/54k2d3dzbfeeivueP7DDz+wsbGRH374Ia+99lqeddZZfdYJNm7cSJL89ttvWVJSwi+//LLP19jdhL01MZCiKPdBS3tSqSjKF4qinAjgOgCTFUX5L4ADw/8B4AkAn0CDPfw/AIsGog4pSo4MyW8AXHnllRLMv7+06PRFMLvMUBQFZpcZP7v+Z9jeuV1L6mOOLS+SAO3YsSP24E72SJ/P1+dzfvjhh7j7ZSIaIn69OmN3dXd1G56rs7MTa9euFZNUmRyjq6sLDzzwAP75z3/KhEIAtDZLgnpLqLPbieHfOPl4xLOL3+htSdHPPsgecUCICbbD1NnZaWgbs9mcdDIYQfqkLgBkYqaYPiO63V8RUz5F8SmuzEqS4n0HSVOCPFci+Ywgi0WXGUfcxhH+jSOvZNIy3bBgt9ulHI2X+CgrKwsWi8WQAE0kTHK73bDZbDLBTllZGbKyspCRkQGTSfvAx48fD6fTifLycjidTuTm5kbqq2+aUwA0IJLgrC+kv06cMWhPE0n5naqqKvvVunXrQBI7duyAqqpQVTWS5Gw1tIRvTmiJyD4A8CVkUrn84nw8++yzqK+vx7Bhw/Dvf/+7xzqsXLkSv/rVr9DY2Ija2lqZaEhP//73v1FWVoY1a9Zg5syZeOGFF3D55Zf3+XkPOeQQDBkyBGPHjsXixYuNSexSFKHdPQPoD6cs5ANHLpfL8H/JkiUSzF9QZ2dn0tdbuGhhbICnRWeNaUWsH+Ag5WhrT0Lrj3i2OBYcr9fLYcOG0Ww20263G/zhbTZbjPXRYrEY4R5zEtRvoC0+0dayQW5RSnGKU6yx3l0tWkaJIPmkOPqbT1IGRLvv6a3gNpuNoVDIgMChr6vZbOZll11mcNVzunSB05mIuKeIfA0WGDHPASm/9KsOsh4KiNmIrLAFQRQhZsUtxk8+AdKLPqFRVlYW6+rqpC98d3e3hPVTFIVer5eBQIAkOWnSJNpsNjocDln+2WefNVitxXhcWVnJP/3pTzHja1dXFxsbG6naVW1snQXNOq6COBYGUIVo19EU7Rxhb3ZZ2ZWcUsgHjhwOB4PBID0ej0ymIwINfT4fq6qq2NLSIgOQrFYrR40axVdeeYXZ2dnMzc2Vwqe2tlYTfnXQfJ5FVk0hBIWQG8DByGQyGQRkzBL/TkTjD8RSvd/vZ2ZmZv8RFBINijvrvmHt4dr7EPfY7nsJ7rfeLcLj8ST3bnemXVK8z/DOxNDoua8+0YJtNhurq6tl7JCgJUuWMBAI8IADDuAVV1xBs9nMuro6mkwm5pfkR+T2JSBmIAIaYIdmpJgVPm6BlqStWPufn59Pu93OmhoNVGDhooUROaeEuQ1atmUriJbwf0BLzqNAg7b1QQsqrQZxoO7e0FzADjzwQHZ3dxvG0jlz5tBut9NmszE3N5cffvih4XhHRwe7u7vZ1NTEqqoqTp8+PWY8drlcvOaaa1heXs6uri7ZVkuXLuV7773H4uJinnfeeTy27VhtctIC7VdF5Ls3gyedfJK85jXXXDNg+sL+TEgp5CmFfFeT8Ee7+eabeeyxx9JqtbK5uZmBQIAjRoxga2srq6ur2d7ezg0bNrCwsJBVVVU88cQTCWgBIFdccQUXLFgQgf8TSCMZ0JRzFYkVY51SKAcPBRr0lSiTpgkdoSDrExjoUVrS0tI4d+5c7ZiwpHix5xKE7Erez9AFUpzi/ZV3ZvLUJ6OCXkYrkBCsgIYGFQ3dKnGxc+Nfz+Vysbm5mUcffTQBcObMmSTJ7du30+Fw0O/302az0WKxMBgMGmN3hOEmGK6HI+r6CuKOKS+88AKHDRtGVVXZ1tbGmpoamXimublZW02wIpL12Q7CDXq8Ho4aNYrBjKA2XjmgWZtHQ5O1FVq5ucfM5aZNm/o11j7//PNsaGhgfX09x44dy//+978xZW644QaazWaeeOKJcl+iVeuFixZqcLyiPQCGykMMBoNcu3atLBu9Cp6i/hFSCnlKId/VZLFYmJaWxvPPP5+hUIg2m41+v592u50nn3wyc3NzedZZZ3HixIl0uVwSE3XcuHFMS0tjWloaP/nkEz7zzDORVMhViFgjVEQs5Z7wMb11Nyx4TSZTRCAnsP4KDGiPx2OI+BZscAEREIy70gqcjJV6P7BC77VctnvvlyhYOBHvKZjEvYp3Ax61RP+I5l34bQu4OiCSSl4eV6N+vdjpFbNklf7o5DTIS3xvfYIY8RylpaWsqamhyWQyTBZ8Ph8ffPDByPnCTUXRcRqIrPj38vq8zMnJ4QUXXMDvv/+eGRkZVFWVDoeDDQ0NLC0t5ZgxYyLBzLp+4/P56Pf7+eCDD5Ikpx8yPXJta6QeFouF48eP5/PPP8/HH3+cJLl8+XKuXr1ajqcnnngi33vvvX6PxyNHjjS8CyED3G43S0pK5Cq2QDQRri8i8DMtLU2uNowfP54XX3wxVVVlY2MjjznmGJLkYYcdxqamJtbU1Ei0khT1TthbgzpTtHeRyWQCSVx66aUoLS3FoYceipEjR+KQQw5Bd3c3nnvuOXz22Wf47W9/i4ULF8LpdGLbtm0AjEE+iqJoQUT/AeAD0AKgElowXj20YBoTAKvu5nnaeaNHj0Z+fj5ycnK08m5Egm/Cv/fffz8A4Mknn0R1dTU6OzsRCARkQJDX68V5550HxawAMwGkAciAtp1EEFC8gCUReCS27XZ75GAysW59DT7aDeT1evd0FfYc6SXc57v31omChRPRzgQM7je0i5pILwvWrl0bOaCXXeznxZMYZTs6OjQLGTS509mpi8CMljub4uzrI8WTfXry+XxQVRVbt241Hlgd/942mw3HH3+8oXxXVxe++OIL3Hbbbejq6jIEC3d0dODoo4+OXKAj/GtCRO1eB+Dr2HqbTCYE04P48ssvsXHjRrz77rtYv369lN2LFi3C119/jbfeektTclQVM6bPQF5eHsxmM0pKSvDDDz9g3rx5cDgcePavz6K4uBgkseTSJTj++OPROq4V+fn5mDlzJj7++GM88cQTAIC7774ba9askfX55S9/iZqamh7bMhG9//77cLlcMmjb6XTC7XZDURRs27YNn3zyCfx+P5588kkAwLZt21BcXAyz2Qyz2YzrrrsON9xwA3bs2IHt27ejuLgYV111FRwOB9avX4/8/Hw0NTXh0EMPxQUXXAAAOOuss3DWWWf1q74p2g20u2cA/eGUhbz/NGnypMhyoA1yZv3oo49y2LBh9Hg8XLBgAUeNGsV77rmHpaWl9Hq9DIVCfOGFF5iRkUGHw8GRI0cSAEtKSnjdddfxxBNPlFiziGc9UqG5jljD2zocWVVVaTabZfau8oryuFaQvLw8ApEgJp/Px4kTJ9Lj8UiLS0lJidGaJKz00RaPKI5OUx2vjNvTe3KNGFivaNeS6KXXPcA9BqcKy1FPfsiVe/4ZUpziXcVCJor/CVcq+mshV+KfK5KkRO9PlP4cViQVL7CzbdGnc3SyNhmo20QuNT3dN9E5qqoyJyfHsE/v3tjbNUTCIiHDXS4XZ8+ezeHDh7O1tVVmWHY6nUxPT5cryhaLhbW1tdy8eTNbW1uZl5fHb775hr/97W9ZV1fH2tpaXnTRRXIMdrlcvOyyy9jQ0MCRI0fyq6++IkneeuutEhJXZOIcNWoU09PTZWInq9XK3Nxc2SdJMjs7mwUFBTzssMOYn59Pq9XK5557jgsWLOCyZcvocrkYCoX4s5/9jCR53nnn0WKxsKamhl6vl01NTXzooYd2ryKyFxJSFvIUDSQdOOVAPPOPZ4CjASwGcDRAavBn559/PtatW4etW7fij3/8I9577z1s3boVdrsdbW1tWL16NcaOHYutW7dKKC0BVXT11Vdj1apVIIlAIIAhQ4doIk5PCjSYLYHAFraAKIoCu92OHTt2YPXq1VAUBf/98L/ytGAwKK03X331FQCgsrISgGZxfPnll7Fp0ybU19ejqqoKn376qWZNEgYfwmiRMiLASdJbIwXEl8fjgc/nk/83bdxkPMmOGJKQVHJHVIEtxr9WqxUDQkl8uQKKjIx+OWHqRMTimKCdAGjwWnsLOaL+D1Bzp2jfJZJyBRDoYaVC9xkZoEp7vQFi5SM0eEqDJTxMW7dujf/NbkfP32k/KNpSnlBWJCKdtTwabjNu8QSQmj3dV5wTXdfu7m58+eWXCAQCcl9HR4fcFuOWxWJBd3e34Z0pioLu7m5s3LgRzc3NyMzMRHFxMf785z9j3Lhx+N///oeysjKUl5ejqqoKI0eOxKeffopRo0bhuOOOw0knnQSHw4Hvv/8eFRUV6OzsxMUXX4y//vWveOutt/Dqq6/i4YcflnUaNWoULrr4Irz/4fvIzs6GYlFw5jlnSkhci8WC4447Dl6vFxs2bJDP4Xa7ccwxx6CqqgqKouDiiy/G9u3boaoqNm3ahOLiYtkubW1teP755+UzzpkzB8899xz+9Kc/Yfbs2XjvvfcwdOhQTJw40VAuRYOIdvcMoD+cspD3j2CDFlF+pY7bQNgQkzCgNxLlt27dKmERX3rpJTY2NpIMR7dHW4cVxOxTFEXzMxSWFWus1SI6c6XeuiEsWQIFBoBm6bXrrpOkFUmcL3zVAXDixIkMBAK71AqV4j6yHRHozLClsampSfYLQ38ZBCsSexXb9/D9ByGL9OApHly8qxNWOZ1OKooSN3Oyx+NhY2MjnU4nr7vuOg4dOpRtbW00mUzMysriCSecwAULFvDkk09mVVUV8/LyWFpaSpfLRavVysqqSpodZm1FUqzi+mBYYZh7zFzZ9+rq6ghocu7ZZ59lYWEhVVXlDTfcQIfDQUVReOGFF3LYsGE0mUycP38+//KXv3DmzJn0+/0sLCzk2rVr+fDDD7O5uZkLFizg+++/T5vNxgsuuIDnnnvurlA59ilCKqgzpZAPJAHQIsevBHE2NLeRdO3jFwp2a2srm5qa+OqrrxrOfeihhwzBKiNHjmRJSQn//e9/c8iQIWxoaKDL5WJubi5JctlNy6ioiZcd42UgBMICyQbCGXtOdGa76PMNinOSy8kWi4UzZsyQ1/J4PIbl6vT0dGP2TDVcNzW5e9TW1e70wLArIOp2O+xdVFvFG+R2luMO0L1kwtxjbZAX+y4G6p3sN8Gg+2Lm1hTvMhbuKzGBqWEWhpjp06ezrq6OZrOZiqJw9uzZ0o2pvFxzpwwGgyTJcePG8bnnnuMBBxzApqYm2u12/vOf/+TKlSu5YMEC1tfX87nnnuMvf/lLjhw5kqqq0uVy0eawaWOvQCBTQUyChpceJ1DZ4/EwEAjQYrEwKyuLDoeDXq+XQ4YMkc9jtVqlMl5RUcHs7GwuW7aMF154Ic1mM2fNmsWtW7dywoQJNJvNnDp1KseNG8empiY+/PDDu08R2UsJKYU8pZAPJBks5GdDs1bPBmEDzz33XFosFumLXVpaypdeeol1dXV0uVzST27NmjVsbW2lz+ej1WplTk4OJ0+ezHHjxsWmPY4SKnq0gBjlQ1jOA7Hnms1mOhwOmkwmGUlusVhos9l4/fXXy/sWFhbSYrVoqAMFOmFnib1ma2sr/X6/bJvS0lJWVFTwgQce4EknaRiuLpeL69ev562/uFW7lmi7K0GMg4ZlG+cZ9e0ghLrJZIpVlPqqUCQJ3xgvBXPcNu9lf5/q6oz6jWaH8ZgYHPWTn6TqFowqE36vA4W3POCcxCQglZq+Z07YP80wrrilFPR9nnelIUHISqvVyosuukj+1yvwfZEzSSeWi8dK4mvp5UVlZWVMGeEDX1xcTK/XK63sVquVJSUlvOKKK3jEEUfQ7/fLiUZeXh7r6ur4/vvvkyQ3bdrE448/nsOHD+eQIUOkwt7e3s6WlhYOHTqUQ4cO5YsvvkiSfPbZZ9na2spZs2axsrKSxxxzTAxO+95OSCnk+69C3t7ezqqqKv6///f/SGodPl4iAUEnnngihw8fLi3b8bBLJ02eFMmiWRdfEAjhIwJLAMjkP4KFwicgEAEtAU5PAiWaxbVijpliz00kBBsaGnjXXXdJy7jT6dSuqUJT7PVWbNV479zcXGZnZ/OFF14gSaanp7OkpIQNDQ20WCzMzMykzWbjipUrItdYDC0hRUb4v9l4zUAgwCuuuEJmx1NVlfX19TSbzRw5ciTT09Mj9beGn7XEWDcRvGO322X7xASKliDWipLAYi8wgKMFebKQf8kEZsXUI9H712VoPf300wlogbl6WDkRuJuQk8BfN9R5N1vF98qEO8lg2u+LWP77y/tNlvfzdyzk4zHHHMPrrrtO7p85cyadTqcMhNyl/cMT/5pybEPEkm82m3nVVVcZxmOxSg1orlaqqjI9PZ1FRUXMzs6WAAqANsbk5+dz0aJFNJvNzM/PZ15eHtesWcNLL72U9957L1evXs3DDjuM5eXl3LRpEzs6OrhlyxaS5IgRI1hVVSX1E5PJxMcee4xdXV30+/28/vrrWVNTw8bGRm7evDlGL1m8eDGffvppkuSyZcvY0dExUOoT29rauGrVqpj9r776Ks8880ySGmTl6aefTpK8/fbbec8998j9eihLQUgp5Pu3Ql5eXs7a2lp+//33PPvss3tUyEmytbWVr776Kjs7O+Mq5GQUykqYhw0bJgVSVlYWLRaL3AcY8WQtFgtHjRpFk8lERVHY2NgYf4k8gaVU76oghFdMJHxYicrNy5X7nE6nxBkX7HQ62dTUxKamJiqKwsmTJ/Omm27qdcne7/fT5XIxMzOTVquVpaWl9Pv9DAaDDAaDrKuro81m01I7W3TXEpOFBAOXfuIg2kvvjw4gorAKX12xXBmlOKanp0tF3GazxV5HtFN6+Foia1syA4UCYlrPg5Lg3ek/LzB1eyqTcMDzG9/v7qrzPsG7K8nUHkpmJSa5Kd632OHUub1F9S0hR/TyQliMAcjxS8hYVVV5zjnnUFEUHn300SQpkx/dd999tFqtfOONN3jZZZfF1EMfYyDGpXj11VuwDXJOv7KjGwf042IwGJTn6y32hx9+uLzf3//+dz7//PNyHKqoqCCgjUECl13Ud9asWayqqmJaWhqHDRvG8vJyPvroo6yrq+Pxxx/PYcOGsba2lo2NjWxsbGRBQQH//e9/c/369Zw/f75cObfZbCQ1hdzv90uD4GmnncaJEyfy3nvvNegfifSSUCjEL7/8sl+6UjxKpJDrSa+Q60noUdGElEK+/yrkL7zwQozykZaWJpVP8WHPmDGDTzzxBM1ms0yuk52dLQM9fD6fdPloaGhgMBiUiQT6IvxEeaHE6gVQ3GslUFqrqqqkG4fYl5WVFeO6YDKZ+PHHHzMnJ4eZmZnMy8tjRkYGvV6vtFYceuihJMm1a9fSZDJx7dq17O7u5qhRo+j3+zlx4sQen0XAWOkVaWFRSGjxNyNixYjzjMICYTKZ6PF4jM9qAlEaFsDluvNsGovBwWw2s7CwUCoSNpuNRUVFDAaD8eEkEyzVi0En5lgvipFoD7PZ3Dd3ip20RifTJ2Vij+j76bbFpE+/yjMgbIrU4ZFHHkm6zoOC97T/fJKst+DJejfs2TplZWX1/by9MZPuIO4j0UYWw6qhyMqsO96f8U28Z5FYx2q1sqy8LGEslJBDRUVFSd3D4/HI8ywWS6+ydenSpbI+LpdLjgdLliyRq63C1RQAy8vLpY87EAE6yM3NZUZGBidNmkRVVVldXU1FUeh2u6WeoL+vxWKh2Wym0+nk7373O9bW1vIvf/kLi4uLJQSnw+GQE4PW1lbpsnrQQQexu7ubdrudTqeTRUVFbGhoYHl5OfPy8uj1epmdnc3777+fbW1tvPLKK1lSUkJFUWi1Wtna2sopU6YwLS2N9fX1vOWWW3jXXXdx7ty50t99ypQpXLNmDUnyrrvuYnNzMxsaGnjEEUdIK3tbWxtPPfVUOdn44x//SNLoabB8+XJWV1fzvffekxOFVatW0eVysaKigo2NjXzsscdYV1fHjo4OAngNwGQAD3E36bop2MNBRCQBaFBHALBu3Tq88cYbICmhjf785z9j9uzZUBQFpAaX9M0332Dr1q0wmUw4/PDDYbFYsGXLFrzzzjv49ttv0d3dLa8NIGHyGnGPQCAAkjCZTPjmm28knJWAjTJcS5BAsyqCltxB0RIdTJs2DcOGDYPD4YDH44HJZEJGRgaKioqgqir8fj/OOOMMHHzwwfD7/aioqMCGDRtgsVjgcDiwZcsWbNmyBcceeywKCwsBaNCI+fn5ss6nnnoq5s2bhxEjRsjqCMgrUV+z2Yzs7GxkZWVJWDMBh/X9999j/fr1US8j/LsDEehCHUKZaAuz2QxVVdHV1YVNmzYhLy8v0j7dAD4O/64JtwugJT7qAu68604UFBQAADZs2CDbZ9u2bfjqq6/w7bffavXQv69uAM7Y5gc0CMa47yYaihHGBEgCem3Hjh0xsGQ9wrvFuVVfKG5d45SRddIX121v2bJFlo25pgnJkygr4C27InXIzc3Vipj6csGBo94SucTQTr6bXUmiLQENus4AM0gA78Q5yRxnXxSlp6fvdN0A4Ouvv+69UDTF+cZ2G/XwiRoSm0WTvo9kDVhtEpKAYe2NVFVFVpaxQga51AVNLusupx8jAU2267/Vl156SSbYcTqdIInKykpUV1fD6/Vi8+bNUFUVH336EVhMYCoAm7Feog6ffvqp3OdyuWQb69vaYrHIJEnd3d0IhULy/AULFshyHr9Hbn/+uZa5TMhxATN86623IhgMAgDWr18vn/O3v/0tfvvb38Lv9wMAZs+eDbfbjRkzZiAnJwdffvklTCYTvvrqK4NcJAmPx4OMjAyYTCZMnToVDocDXV1duPHGGwEAF110ETweD0pKSrB48WIUFxfL53nttddw1FFHQVEUrF69Gi+++CIsFgvq6upw2mmnobu7G1deeSXKy8uRkZGBU045BRdeeKGU01999RVyc3OxevVqzJkzBw6HA36/H6+//jrmzZuHX//613jvvffwwAMP4PXXX0dBQQGmTZuGxx57DEcccQReffVVvP3226iursavfvUr2XbvvvsuGhsbcfPNN+O0004zJKl6+OGH8b///Q8TJ040JHKaPXs2mpubsXLlSrz11luYNm0a/vOf/8h3AeB4AL9GL6QoysAMDLtL898Z3p8s5IqisL29nS6Xy+D7lZ6eLq3KaWlpcrYt/MGEdbOurs5oTXDFWhNQFN9iMJBoDSLd70EHHcRhw4bx0ksvZXZ2Nq1WK0ltKcvlcvGMM86gy+VifX09jzzySPr9fra2thqWn1wuV9w2W7FyBUPlISqqwlB5iAceeKC8/6OPPsoRI0ZIa4KwYItEEqqq0u12x6Qu7g/r29vQhnq/76jkICLhg/AV9Pv9dDgc8h3HWKt3kTVrr7H4RrNozzjWsp3mOHENg5ITBdTGY/8uqkM/Ait7DJRL5nq6MqWlpTv9DPFQb/TbcQOR9bEThSAmI9J3SkDk68qaEBde0uFwMDMzk0AcV7e+cJy+Kuo/IOhGKohTdf+z4tzTBmP8Trhv5ubm0u12x5UzPVmMxbHXXnvNGCPii98PouNuLBaLIVZlxIgRHDZsGKuqquQ+j8fDl19+ma+99prmLiJAARpBNIXfWUb8+sVjsVKaqE6ChXtodPnovuxyuWLa7YgjjpAuM6WlpTz33HOlpf+hhx5iXl4e3W63bD9VVTljxgzZx4Rrpt1ul+Pjqaeeyrlz5zInJ4cmk0m6EjqdThYWFsoxTVyzrq6O9913n1y9HD9+PFVVpdfr5fz581lZWclgMCihH6+66irOnz+ffr+fY8aMkSvJjz76KI844ghecMEFLCwsZG1tLQsLC5mRkUGPx8Pq6mparVb6fD7abDYuWLCAc+bMkc+Rnp7OlpYWPvvss7Tb7SwsLGRFRQVPPfVUtrS08M0336TdbmdxcTEDgQBnzpzJ3Nxcvvrqq1yyZAlPOOEEjho1im63m5MmTeLGjRt5880302QyMTs7mwA2AmgHcAc0a/l7AK5ixPDzKYCfAXgDwI8AvKE7Vq7/nyzvcWU7Gd7XFPIVK1cwPTtd4mWn56Zz8eLFssOfeuqptNvtcuDKyclhWloafT4fAc3PV3xYt9xyC+12OxsaGuILOROI3PhCO5qjBwW9D5sQHMkocEOHDuXw4cPpcDjodDp51VVXsauriwsXLpRKZ2FhIV0uF7u7u3n66aezoqKCBx54IKdOndqrL5hoQ9WhGrKQimdUFIUjR46UuLJC+BUWFrKoqIhut5tlZWVxXUKSwSAWri9C6OrbvL6+Pv55jth2nTx5Ms1ms8z+lkghGOw84K4iKd4p3m9gEAcbK9BkegCa3BVuLD1NGAd64hdWUhNB/fWbBzm+/870+YRZUXt7X3EmxNHjr9Vqpd1uN2RfFa6FeXl5HD16ND0eDy0WCysqKmg2m3nMMceQ1Hyb9XFkqqqyvb2dXV1drKmpkUrp+PHjee2117KoqIgej4ejRo3i6DGjI/1OidSlrq6OJSUl9Hg89Hq9NJvNPOSQQzh//nxecskltNvtDIVCzM3N5ejRo7lixQqWlpZy3bp1HDp0KEeNGsWGhga+8MILMc8uJq7CHcXv99NsNseggPl8PumOox/33G43i4qK5HgiJpP5+fkGd8xQKGSYWIl3HwgEaDabmZWVxezsbMNEW1VVpqWlUVEUtrS0SIOXQJ4544wzeOWVV7K0tJS5ublCn/oMwPUA0qgp2SYAzwFoYEQhv4gRJfxZAEPC2z8FcCb7qOvucWU7Gd6XFPIVK1fQ4rVoM/0cRBRKndA+6KCDDAEoFouFxxxzDGtqaqgoCr1er+zQTU1NrK2t5bHHHhsrXPzQYOPGhAWMuFeBURilp6ezubmZJ5xwAu12Ow899FAGg0HOmDGDxcXF9Pv90hJvt9tZ3xBWOkfDYEFYuGghvV4vb7vtNpLkcccdx9bW1qTapa+JiswWs2bBaAMRAjEdBitUfn6+/BCFtd7tdrOmpoZutzsiwPvh+6n3B4xWRvtqkUopsineH3lvwmDvc13jKNlxrez9VMb3iMwYwIlDtMUViD+BiFZuEwVv9xkZqrd27GmVJsl2SJRrQMRECUXV4/EwPz9fonNNmTKFN954Y39UC0kGdLXF4d9wXXJycjhx4kSWl5ezpKSEubm50kL99ttv02Qy0ev18uSTT2ZtbS1nzJjBhoYGPvLII1QUhR6Phw0NDfzRj36U8D2JdxUvFkPALur7gvBrT9RmTqeTDoeDaWlpBLSVA/09i4uLCUTAEAREsr5/mM1mDhs2jA6Hg4FAgKNGjaLT6ZQGuFAoxIMOOohHHHEEQ6EQp0yZQmh5casBnAbNCv4OgLUAjmZEIQ8xopDPA3AzNMX9YwDp7KOuu8eV7WR4X1LIQ+UhzYKSA+NHc3DsR20ymQzwR4KXL18u9/l8PoZCIQN8oRQciZbzwzN7gfXd1NTEww47jMcddxwVRaHf7+fvfvc75uXlsaqqimPGjCEACQN10EEHaQLFFhGEVVVVLC8v5wEHHMDi4mI2NTXxggsuGFCFXGQIJakJzdZw24VAnAziQGPbmUwmBgIBGYSjqqoMeo0rZKPfQZTwVVU1ZhnS5/MZBpM+Z/kzIyHyymDm6OfuD6fwuFO8s5xQqeqDq0G/75HoWNR3nNDloo+cyAWit3Pi1TG1irLrWLiQAtoYK+ScPkcFAGk5F/lABIzhzsIBxs3QHXaftFgsTEtLY1tbG0ePHk2z2cyMjAzef//9XLt2rVR2N27cyAsuuIA+n0+inHk8Hg4ZMoQVFRUGZfvcc8+NkeXR/UuPqKUvG62g61lfTkxe4vVpAfs4cuRIut3uGJQlh8PBrKwsVlVVyeDUOXPmcMiQIXzvvfeoKAorKiro9/v5wAMPMBQK8c477ySATQCKAXwEIEBN6b4bwHGMKORBRhRyO4APAcwA8Hv2Q9fd48p2MrwvKeSKqmgCO/qjORuRZcGA1hlLSko4ZcoU6TNVVlbGiooKtre3s729XS4JVVRU8IgjjuDBBx/MpUuXGqHrnNCsxy5EBgoFrKquYnNzMysqKmi321lUVMTnn3+eWVlZvO6660iSd955J8vKyjh8+HCazWaeeuqpJGMhhhL5eCdL7e3tLCkpkdYDkd3soIMOYlpamlxeCoVCfPrpp+WHCRM0638wvJ1m/KDjDUQSG9wGYwIhBcRwXdnMPS/YdynvBCqEcNNJWfdTnOLeuc9Qoglg8YSFEADHjBlDi8US1x8/EAjIb/Oee+7hwQcfTEBDzBL+w+np6bKMtDwrIIbCGJ8xCNpvb2P9xEkvI8VKrf6dWa1W5ufnc/LkySRJt9u9U2MpSe3aIkO34MXa/fpLCxct1PqDWGXX9UuSMcYZYZUGIjFwIvYtOmmeiKMAwEMOOUSubIv+qigKm5ubpUttSUmJdKO1Wq3SXXfIkCGsrKxkKBSS11BVlccccwyDwSBbW1tl1tM1a9awoKCAb7zxhsRTz8/P544dO1hXV8cFCxYQmsLdCOBtaKHTWQC+RgKFPLzvVmgQDlPZD113jyvbyfC+pJBLC3n0R3M2IgrlYlBRlR6vo7cof//99ywvL+fs2bPl8egPSLErXLhoYZ/qunHjRpLkP/7xDwYCAf7hD39I+tzvv//ekOSovLycS5cu5eLFi5mVlcW1a9fKskcddZQUXoWFhZw7dy4bGhqkkj5q1Cj+9re/ZWNjI0877TT5AaIBRHV44MhAjPVJKPjCf+/QQw+NHA9CC7wqQNwg136xBQZ3oH2KoxItGQabQxBZicnfTfVJcYoRx2ocJ3hyT9dvT1qjE02aBab0nm6fvZkT4d3PmzdPKpmTJ0+WwAE2m01mwfZ6vZw7dy6HDRsmjVtvvvnmThu3yAQW8jYQtuQVcn1CnebmZu3ZhDIuVnXDz/vSSy/FxYEX+4RVXhjW9NDAHo9H4qcLeEiz2UyPx8Pjjz+eTqeTpaWlHDdunOyvfr+faWlpPOCAA1haWsqf/vSntFgsLCkpkdZzn88n27ysrIwmk4lDhw5lVlYWKysref/99/OZZ56RyrvZbObMmTNJkvn5+cL1dCMjVvEPATwD4A/oWSEfBeALACb2Q9fd48p2MrwvKeTSh7yHjwZtYKg81ON1+upz3R86//zzmZOTQ7PZzFmzZvUpNa6+fnqFnNSSAgiFfMeOHfzoo4+Ym5vLYDDIyy67jM888wzLysokrmpBQQHr6uo4ZswY1tbW0uv1ah+0sOK4YLCOm63GpcFoAYEcRFYjLNBQA/RCNQBtggRIRTQjI4N5eXkRgeMGUQWjNSsBmkKPnCh4SM+DKEW4eCcGC3kCfPAUR3HhIKjD7uB9IAPkQK3+CAVjMNavPy4wKY5wIp92fe6EeL7RwoXp7rvvZl5eHgsKClhSUsJTTz01oUKuN3AJUlWVJLl69WrOmjVLYmsXFRdp49BBII6DplfYQa/Py+nTp/Ohhx7irbfeymuvvTZmPCaNWSzlmGlFxL32MOPzRifIE2OvQHcR/V8o4qqqsqCgQJYX52ZkZMiESIqisKqqiq2trSwtLTW0Y3FxsQELXuC2V1dXS/S0G264gYDmrnLaaafR7/fT6XTK7NojRoxgfX09L7zwQubk5NDj8fD99983tC/6kRgIwAUAftzX8+T5/T1xd/K+pJCTmlJusphiAy/cIHJAZ6aTK1au2NPV3CmaM2dOjMCPpyTHE2bRZZxOp5zx6ssZsmrG4WihaLAIiVl+I6TriskRvq8eojC8bBsMBgcEQqxfVqndqOQKa0V025pMJtbX13PIkCHxz01mYqF/nl04yRi0rjQD6Nec4r20D6S4R97b40pEvxPuEvp+GAqF6PF4WFpaSlVVabfb6XK5mJ+fz7vuuotkYvfPeAY4l8vF//73v7Tb7TzmmGOYnp7O+vp6dnR0cGjT0Ei9VHBI0xCOGjWKpaWl9Pl8LCws5Lx583jmmWcyIyODlZWVdDqdHDNmDNvb2+n1ejl58mTtfSggvNBWP9NAHAqi2fjMDoeDFotFjpE2m40PPvgggQhK20UXXST95B9++GECxhWkc889lxkZGSwpKeEhhxzCsrIyOp1ODhs2jIFAgCeddBLHjh3LsWPHsqGhgTU1NRwxYgSLiopos9no9/uZn58vJxpnnHEGa2trWV9fz6OPPppbt24lSXZ0dEjj4n333cdx48axpaUlps3RR4UcwEPQAj+DfTnPcI3+nrg7eV9TyAUZ0tqHf0Plob1eGSfJRx99NCb6PUbZrkNcxUxvUVIUhQcccIAhUEYIvGSVW5fLJYNQDD6XKjRrefi/VLjTYVAwHQ6HBpGov+4gREgIY6dGOJ4PqBUR1x7dsqOwXIhl2IqKCp599tmGdzJv3jzm5eXR6UqAmZxsm6Ss6LuNEyo4gxzGbm9mGeMS5v6igOxSdsOYJ2GQ8O6cSOkD8PXjksfjkQpzPKOSfhyyWCy0Wq3Mzs6m0+lkeXk57XY7v/jiCx555JH0er30+XzMzMzk1KlTabFYGAwG2dLS0uuKsz6l+5w5c2i329nY2Ei/36/hpiMxfn08AAL9MTEOCvhAPdJJvOBLqNDGEzM0+W03tgmgxWfl5eXJc7OyspiTk8PJkyfTYrGwurqaNTU1fPPNN/nss88S0Mas8vLymImIPk5t9erVLC8vZ1dXV4/tFR3bloief/55NjQ0sL6+nkVFRczNzeULL7wQUw79sJDvLO9xZTsZ3lcV8n2ZFi9ezLS0NLrdbsMs2CBwkwicjI7I1p/fnyVXgZEq9wlreIIkKyaTiW63e+cSd+ylvNsGx3ykAshSnOLdye5deO1kJ9zhSaFiM8oZ1RQ1iawL/4ZgNODoV+U8um1LmEWiH2iyrK6uLrICaDbHKKPi/MLCQoksJgLYo41BYlsgkwkeM2aMdAFZvnw5vV4vJ06cSFVVJSDBhRdeyKVLl7K7u9ugZC5btkwirLz66qvMy8vj9OnTuWrVKoOFPBrWLxGLOh966KEJx6+amhrZPr2+Lz2uvk5eC2hhm80mA5gFgkxhYSGdTieLioqkO0t5eTmfeuopWcfLLruMqqpy5MiR/Oqrr/jiiy9KdLTCwkJmZ2cbkgVeddVVbG5uZm1tLU8++WQ5sUlWIU+WkFLIUwr5zlJ0MKU+sUA06TvwUUcdxdLS0qTvs3z5cq5evVr+X7FyBa0Oa0RIhgWnyWRiZmamzJCZDCcDM6YPDOlNKCUsa4kS5D24UcQ9v59uFyJTW6J77O3LtilOcYoHOe8mP3+93IwHSQegZwXegYg1dgQiK8qChQzWT+gLYuvgcDh4//33S79vq9XKX/3qV7LM3LlzCWgKdjLyVyjENpvNsCIiXCOF37O+DfTxN8KdRVVVBoNBDhs2jBaLhRdccAFJLZN1SUlJjEL+/fffJ4QKFIkDBYvnECu/+sRE4nxh1EpqtdkEY4ba8H4B2+h2u3nSSSfJe9vtdlqtVmZmZvKQQw7h9u3bWVtbS4vFwlmzZslJwqOPPkqSvPDCC/njH/84RjchyREjRjA3N5d+v5+LFy+W++fPny/Pnzhxokyq9NBDD/G9996T5RYvXsynn35a+tknQ9DQVZ4A8C4TKdBAUU/H+8p7XNlOhlMKefIUHUyZjEK+Y8eOpIJE9cq+fjltxcoV2sfqQMQnforxY45rzU4giEOhUIxQEawXfslarfXl6urqIsesMFrpA7ptsTSXG1XfvrhbJLD6CoEslvX0uLRCcCd9jxQTiJ0wDVqc5f646+ypoN6UW0uKk+Ee+mcoFJLyTS/XhLImy4rvwo7E34gNRhkdj03QZLoT0jouUtB7vV6qqkqHwyET4+izRAJGpbanmCHh8hGtBAsuKyuL2RfP/UUfGCqs7FardUACbhNNLPQrzdFyMjpOK9G1BbSmx+MxKPPCQm6z2VhdXS1ZZCwVLj92u525ubk0mUzSwn3//ffzxBNPNOgmgrxeL++44w6S5AMPPMARI0awrq6Oubm5vPbaa9nZ2Wk4J5G1PBmFfMeOHSRJAK+hF4W7t+N95T2ubCfDKYW8d9q0aRPHjx8f80HpAfmFICktLZUfvdlsZkFBAQOBAG02G0tKSmRyACG8QqEQa2traTKZmJ6ezsLCQqqqytLSUjY2NrKgVIuYRho0n+xSEOfDaH1OIJR2hgfEpSKR8BeDjLmHMr2xbqCKjrYXvnZCKPYYMGpFjEWrp3b0+/3JWdkHqy93P+qlT1Txk5/8pOcVjcH63CneK3jQBIwOIvSlRJyRkRG3vaLHJQBa7M5CaONGVFA9csPbfu1YQUGBPL+trU1mT3a4HQZDiMViYV1dHVVV5cUXX0yPxyOzNQqlWlEUuR0P170nnjBhQtz9++Iqp1CoAfCiiy4ioMldvbVdZPAcOXIkS0tLee655xLQFPWXX36Z6enpnDRpEoPBIMvLy2m1WqUOs2rVKra1tZE0KtSnnnpqjKup0+mM5BTRtXlLSwuffvppQ59zuVycN28e58+fT7vdztraWpkUsaKighs2bKDNZqPNZpOrD6effjoBbAawFcA6ACUAXoCWsfMNAAcwpZCnKBE98MADnDx5shQqf/3rXw2dVczkRUYw4cahKIpMOKGqKi+99FIC2gxX+M8B4NChQw0fpriW8MkDEEl45IQ2WFTF/7Cjg556Y/0sXn9uZmZmRPD1EKCkP18ovkkLXv2g58VOLfVaLBYGAgGazWaOGzeOgDYYVNdUa20nXGhMiB1sE/i47xTvBQN6ilOc4r2Ho1f7Urx3KeeJVmfFfvFexTisR5JxOBycM2cOCwsL5ZhbXl5Ok8lEi8XCyspK1tXVsbGxkYcffniPCvkZZ5zBX//61/KYuP/VV19Nh8NhiOsaP368zLjp8/nY2dlJj8fDO+64g2vXrqXf7+fw4cM5f/58Alow7yuvvMKcnByef/757OzsJACeffbZJMmHH35YBPz+C8AHAL4DcBIAOzUlvBxh/3IMsEKuIkX7BNXX1+Pll1+GqqpwuVzYtm0bAMBsNgMAfvjhBwCA3++Hx+PBjh07ZCcoKysDAHR3d+OGG24AAGzbtg2KokBRFADAmjVr5L1OOOEEVFdXIyMjA9u2bUNGboZ2gAA8ABwAugG4dBU0RTY3btyY8DmsVitMpkjhhoYGdHd3y2fZunWrPPbNN9/IY9hhvE5hYaHcttlscDqdUBQFhxxyCACgq6sLPp/PcI54VgPp6o1NADoTVj0uifdhNpvx5z//GevXr8eOHTvw8ssvAwCGDB2C999/X2u7LgCB8C+gtaOo0pa+3Tcp6t4F1xxspH+lyUi7OF0gRfse6WXMTlPmwF0qETmdzrj7hXyPR3Hl2S6mri5NeOnl9M6QxWLp0/5kKScnZ6fO7wvJMSpJ8nq9vRdSYeh3+v48ZswYQ1HRVvq+smzZMgBav9KPlTt27Ig5DwC2b98Ok8mEkpISnHvuudi4cSMURcEZZ5yBadOm4Ve/+hUyMjLw+OOP48svv4SqqrDZbPi///s/3HrrrfD7/fj888/xwQcf4N1338Wpp56KY445JuHjHX300Vi6dCmGDh2Kjz/+GJ2dnVBVFf/3f/8HANi8eTO2bNEGxXHjxqGmpgYA0NHRAZLo7u7GwoULkZOTgx9++AHfffcdAE2/CAQCGD58OHw+H77//nt5nW+//Rb19fU4/fTTsWHDBgAoAxAC4AZQCuD/FEX5F4BVAGoSVn5naKA0+13JKQt5LK1YuYIun0uzDIcDKS02i4Q0Ouecc+Qyj9/vl2loHQ4Hm5qaCJ01QwQZCsu5qqocP348CwsLZeBJe3u7YdZrs9lkKmBDauijQJwJzdrrgGb5NUNLSx/PVSDOvmhoqb6wWAHQR8a7XC4OHz58t1gYhDuK0+nkiBEjZLZQACwtLe39GhkwJDkysD51cX84Gv97Vwd2pZBTUpzivZr74p7TE9Te7mS9O2DSPIjiJZKKIbLBIMvFMyuKwkWLFhEADzjgAPlefD6fTNRjMpmYlZXF/Px8/ulPf2Jrayv9fj/tdjunT59OVVXp9/vZ1NTEMWPG0OFwsL29PUYHaW1t5bnnnsvMzEyS5COPPCIxwHsifRbQ6KREguJBQvv9fi5fvpynn346zz33XKl3CDKbzbTZbFy+fDlNJhMXLVrExYsXU1VV5uXlsa2tjTabjaFQiCRZVlbGtrY2btiwgQB4/vnns6uri8uWLRP9/h1oFvJ3AVwJ4AZoUyEzgB1MWchTBAArf7sSx550LDq2dmhW6JEA0oDObZ3o7u5GZ2cnHn/8cTlT7O7uhsVigdVqhcVikbNFMXO/7LLLAEB2iu7ubnz66aeoq6uT/5944glp/bj11ltRV1eHQCCA8vJy/O9//4vMvh8CcCs0S3InAB80K3M6NNEBAH4AhwCwAaiCNv80AVCAESNGwO/3IyMjA5WVlTCZTLj11lvR3t4On88Hs9MMtAE4EEBQO0d+ItAs0l1dXXC73Tj22GNhtVrh9XqRnp4Ou92OoqKiSENaoFmhBem3+0mqqsLr9WLr1q147bXXsHHjRnR2dsqZea+0FprHWjwyI2I97w8RRqt4H639fSYFKYtzilK0k5TQ0q3AuIK3CyisdCRF3d3dcozoiYLBYNLW+/5Y+VevXt37+dG7d8UKZD9p+/btvRfaBoMsFyvY1AyYAIBXXnlF7lNVFRMmTIDZbEZXVxe+/fZbfPfdd7j99tsBaBbntLQ0HHTQQTCZTFAUBQcddBDWrVuHHTt2YNq0abjuuut6rNJhhx2GSy65pNeqNzc345Zbbkl4/MApB+KZfzwDHA1gMbRfAOvXr8djjz2Gzs5OPPTQQ7JvvvDCCwC0/ud0OvHDDz+AJCoqKjB27FjZL9PS0tDV1YWNGzfisccew//+9z90dnbC4dAG/u+++w6qqsrzIbUKmADkA/iSZDeABdhVX95Aafa7klMWciOFykNaNHoAxFhogS4HQguK6SVgzWQyxaT8FQll9EkBRFmxLSKiFUXhjBkzeM4558ho6bVr17K4uJiAlmjAbDFr1tEZ0NKFm0DU6uqWAeIKRFLW6/Bkv/jiCx566KG02+2cMGECp06dyjvuuIOPPvqoViYA4kcgLoVmSRZ4umdqv4FAwJAt7IMPPiCg+ZFLv3EVmnVYnNsPX+qe/AL1Fg69tX+nrUcmaFi8+v9iexf5g5eUlBj3ZUDzpVcjVguRpEKyFdqqiMACTk9Q5xSnOMUGFjIiKVmxG76lXeH/HCNTdByTgC0BD3hgbXDPv3vB+oRFybI+qc/y5cvpdDr5l7/8RR632+0SilBs5+XlxYBAqKpKj8fDYDBIm80mU9yLY2KV/eSTT2ZeXh5DoRDHjRtHm83GG2+8kaeffjpXrVrFOXPmsKGhgcXFxbTb7SwtLWUgEODf/vY36Yd+4oknMj8/nwcffDDr6+vp8Xj429/+VrOIN4IYAm288yMmE7TFYqHf76fJZOLIkSNZX18vMc7Xrl1rQH6xWq0sLCzkL37xC5nRMxAI0Gq1MhgMcuPGjbTZbHS5XLTb7czKyhJj+BZoQZ1bAMyGZjF/G8DPAGziLrCQ73FlOxkebAq5Hv5v9erVnDVrFknyzTff5OOPP05SS2srllcGmmTHVKApSG0grgRxtu6/V7ftA2HSIpNFJ/R6vczJyaHJZKLP56OqqnzuueeYlZXFgw8+mGVlZRJj1WQy8c0335QpdYPBIIPBIDMyMmiz2VhTU8MhQ4ZoAYrV1bx+6fVa3cZAg0DMh6aYCYXcF65vKyKZI1UwLTNNPuNNN93Euro61tXVcdSoUVyyZIl2rl6508MQ2hAjOPQTjLgDnCXqN4qTCfxMhN8qEktEs8kcqcdgWN5N8R5g1yCow/7I/h6OxTNkDGQmSyFjrANwLYfumkIp7wtyUBz0Kz0PGhSZvZX1SZei8LoBbRwWLqQAeNNNN0kIXFVVmZmZyfz8fJrNZg4fPpwFBQXMyMggEJU0KXztoUOHcuPGjXS5XKysrGRDQwNra2vZ2NhIp9NJl8vFUaNG0WQyySBKh8NBh8Mh3SmzsrJ43nnn0WazUVVVNjU10e12s6GhgSUlJQwGg6ytrWVubi7T0tKYk5PDsWPH0mKxSDeXUCjEI488kqtWrWJ+fj69Xi9LS0tps9lYVVXFv/71r6yqqpKK9AUXXEAAPProozl9+nQ2NDSwvLxce7Zx0PSGy0FcCPndKKrCjo4ODhs2jK+//vqA61ZRelYqMVA8HiwK+ejRo7l69WoefPDBrK2tjcm0pVfAoxVy4St1++2385577iEZi5V5zTXXJFUPX5pPGywC4Y9yMYwK+eLw/ivD3KaVDZWHBqgleqcVK1cwPTcMS2QNDxwm0OVz0ew3a3UaCU2R9oIWr4UrVq5IeL1f/OIXkUEI4esdEDuQCLhGQLPeCut0Qr9COyKDWhTEYTRcpNVqTRqdxeAHWDwIBoldxGIitqfrkeIU74ucm5tLIIGlui/ISzu5gqaXqwnZioQTAz2sq36l1efzGQwniZKm9VSvgW5zfewRAKmkAtoYU99Qr5WNZ8ix6NogzuTLZrMxLS1N4rBnZGTwlVdeoaqq7OzsZFZWFi+88EIWFxdz/vz5bGtr46xZszh//nwqiiKh/hRFMcZuhffpLdqAhq8+ZcoUQ7meDEHR7ak3cCVCzknmHVgsloRjp9/vp8ViMeYVMYXb0QTNYBdua4tVQ2v5yU9+sst1GKQU8sGtkH///fe8/PLLDR06Ly9Pum6ID66xsZE2m42VlZVsbGyUEEFOp1MKo9bWVhYUFLCgoIA2m43nnXceTSYTf/azn7GxsZG1tbWsqKhgZWUlm5qaOGzYMLa0tLCpqUn7KEWAnx0RC7ngNmiKufgfVtAVVdnTTUhSU9bdfrccKNKz03tUxsmIQq5a1IgrRNQgE60YCss/AN5xxx30p/sjx/PCvyLoFLHXGjp0aJ+WD/VCe6AHicHK2dnZPa8+6Dk6sdJO4oHvE+2cwkRPcRLs9/uN7hzxMPWjg7aFpbYnS7+AWPUiRp5aLBYWFxezurqaQERRjVauAWiTA/0EISr4XP+t2mw25ubmsrCwkNOnT2coFIqdeERPIA4y/tevghqefyfb2Ww2s6ioyHAfp9Mp7xMKhVhRUdGz7OkhWF6vlCqKQqfTyYkTJxKIuOtUVVUxEAhIV4/S0lIGg0EZnCnOF1ZzIDJ5EPDEqqoyEAj0DVBgkLBhPFERgyvvcDiYnp7Or776iiT51Vdf8fDDD2dVVRVtNhtffPFFkuS9997L4cOHs7GxkZWVlXznnXdIGgNJR48eTVJLqLhy5UqDzoGUQj54FfL29nYqiiI/ApFMQKB6iH1iJimS6gAR7E673c7i4mK53+12c+7cuayurpYfWjAYpMvlIknW1NSwpqaGjY2NDAaDHD58OEnyxBNPZGZWpoayokJbAm+Dpni3QZtRzsIetZAPFC1ctJAmp86XPT839iMWPstRH/XBBx8shevTTz/N2tpa44cutrN024qGsaooCktKSvrlz5fiFKc4Pu9NmMyDgXdq0hlPQXVBU5aF9TYNmgtMUfh/WC5aLBbDva1WK+vr66mqKo877jhpZJo/f35EyUyw2piVlcWJEyfS4/Fw0aJFWu6L9DRDPYXSL5UxHXrYTj9zD+VMJpPB0uz3+w1Jh3riviYRivdu9fk99O0t3EvEdlZWVkLkMZFlNB4n4xYZHTumZ5GvZFf22WSzgwJa/MHmzZt54YUX8sc//jFJ8qijjuKyZcvY3t7Ompoarl+/nv/+9795yCGHcPv27ZoesXCh9EyIh+wSL6s5Ugr54FbIk+1g0ctGiTqbfolGHBN+ZA6Hg4qiyKAIUa6xsZFZWVkSaoiM4x5iQ4yC3ptLyGCjSZMnaUJZTDYOjrRbjPCwxhfeFotFpjDOy8tjfn6+dkyFpoQLi21YmRdZ3BL5hcPWB4GvZ1MP5yWyXqWS9uzdvKshJVO8V3DSE5BefMEtFktiuRTmnJwcPvnkkwSilKx4fdEe9T/OfZOBnk1W4RMQfAYeQB99g7tDH+o2UO832fecqNwRRxxhKONyuXoda3JycwgY3SP7mnRvsLG+fRRFkR4FYl99fT1PPfVU1tbWMj09nXV1dfR6vdy6dSvb29tZWVnJ2bNn0+Px0G63s76+no2NjTIQldQU8ocffpgvvvgiVVUlSY4cOZJer5eNjY288cYbOXbsWAJ4j0JRBv4OoJG7WNfd48p2MjwYFPLzzj8v6Q6lt4yLZSkAHDduHO12O91ut2F2XVBQIGe5Q4cO5YUXXiiFofjAFEXhkiVLSJJLlizh0qVLE9Y1WkGPdgnRB6XGmxkKOvHEE/nee+/FPSZ84Z9//nlpxd+8ebOhzNSpU3nxxRfLur7//vtsbGzkkCFD+NFHHyWsf0ZmhrbcGoBm6c8AMTXcxvGQBXKhRWWHg53EwOV0OuVkxmw202q1Rj7uaYi4vShGQWAInNH7lSvQovH1y7G2SHk9wksMJ1KwfQn2p1wZdj8PZBBfilPcH07yu8/MzIzZd9ZZZ/Ur34KiDrzrl16JilayZLmdyanQAydc1UxiktzbpCfRMybiQCBgmBjoJzn6fBkffvihvJ5BaTdDG/8yEJFPUWNJwqyoSaDwJPMMiVYCErWVfiVan94+3r30KwHR9bLb7bz88ss5Y8YM2u12mkwmLl26lJ2dnbznnnvY1tbGtWvXUlVVbtmyRRpNV61axezsbNbX10vdo7W1la+++ipJTSG/8MILuXTpUumNEK0H3X333QTwNTVlvAK7yVq+x5XtZHhPKOQrVq7Q4AWBPi+dCfcTsdwjPrCSkhKaTCbabDbDhxq9ZJORkUG3202r1cqsrCza7XYOHz6c119/PUnyiiuu4LnnntvvZ2tvb2dtbS3JnhXyzs7OXvefeuqpvPfee2PKiIBXMXno7OzkT37yE7nMlIju+c09Wlu0QRuczoRRIdcrsGmI+C2aw6yCS5cupdvtptvt5q233sqCggJmZ2dLVxSTySQHvuyCbK5YuYIfffRR5LpuEDnQkhm16e5XAGJubH8QkylVVeUSpM1m04RvSrFOcYpTvAc5kdIlFCq9wqj3UY5nde6Jzz77bLk9bdo0FhYWxtx/p63Wg2jlMJn2EW6t8d6FMN6kp6fzRz/6kXR5dbgcGuQfoKGKOUGUoH8TmL18/BHtJX4PPPBANjU1SU8CfVzD5MmTaTab6ff7abVaZaKg0047TZ5fWlrK9PR0ZmRkyLiF1tZWLlu2jGlpaZw1a5bUF6BBHv4EwH0AzmBflWugGcAtfTqnrzfZE7y7FfIVK1fQ5DJFsKrboOF897ETAWBaWlrcYD+32y33FxcXy2PTp09ndnY2KyoqCED6jhUXFzMUCrGhoYHBYJBTpkzp9/PNmTOHNptNztCFVT8YDDIQCDA3N5dOp5MOh4NLlizhiBEjZFS43W5nQ0MDW1paWFRUJCcdVVVVbG9vl5Yb8TxiRUAqwuHnt9vtDAQCLC8vp9Vqpaqqxhl3GJUFAWhCOD+8P0f7L69lgqaIm0BYwObmZg4bNowLFiyQWctIbeIwfPhwAxKLuF9GRgYLCwtptYdn6gJWTEFiJIMEA4P4yBP1hz3BvfoB7uVCe1ByCm99r2ZFUYzW5kH2jfSmZOtZKH49uUz2ei19f9bJRL3/9dSpU+V2jGvGzqw+xcm3YLPZ9rhcTZZ7c2ex2+1MS0sz7h8f/p0cbrtSEFWI2w+Fe6scE4XByIkI8k0gwf3F9VRouk7UKkJM4GycfmU2mw39IN7qSHFxca+TMYFmE12uqKhI9lVFUejz+VhaWkqLxcIRI0ZI+GYR1yd0KAHAEQgE6HK5ePHFF7OyspJms5kXXHABly5dSgB86KGHePHFF9NsNvN///sf58+fL4xs3wNYCeATAAFqSvZL7I/lGxgP4LFey/Xn4rubd7dC7vK7NEus8MW+EsQpiTuSfonM6XTS5/MxEAjIDmY2m+nxeFhbW0ubzSajzEXH/cUvfsHm5mY6nU4WFRXxmGOO4QcffMBJkyZJZbmpqYn/+Mc/BuT52tvbpfXi5z//OT0eD91uN6dNm0ZVVTl16lSSZHV1NQ866CB+88039Pl8bGlp4XnnncfrrruOoVCIoVCI8+fPZ0VFBX0+Hx9//HH5EXZ0dMgERCJpUENDA30+H6dOncqrr76afr+f2dnZvOGGGwiAo0aNMgrhckQwc3PD78IMTpo0iSS1xAZWo7CwWq0S51R85Onp6fT7/XL5LPqD71FgKrFCqjdly2KxGCxN4vqhUGiPDQopTnGK9z+2Wq1SsbZYLBJwQCRLMZRXoa1Ghv+L+JsY1llqq6qq4pZJ6PrRHwx+Xz/O6Y130wRLuJNEK7X6ZHx2u10iglVWVmqJ9URZYXAKQXPLtCFmctPV1SXjn+R7FMf9iEygFGiW98o4da2ClsTPZTz/8MMP51//+lfa7XZmZGSwq6uLiqIwMzOT27dvp8vl4sknn8yTTz6ZTz31FMvKyhgIBFhdXU2z2cw33niDn3zyCe+55x4edthh0tAIaK66emPRIYccknS7er3emBiHeAp/XDQeGCGRVVWVCDd2u13vorMj/PstgFcBbAfwIoB/AXgFQBk1ZftuAHcAeA3AhwAOYZQSjpRC3n+CFVoGTCCC8X1R1Mfg0HyHhZADNH/vOXPmcPr06TzxxBPlS3a73czJyeETTzzBadOm0eVyMRQKccaMGTJDVFpaGi0WC3Nzc1lbW8vxNMn8xgAA2VxJREFU48fz6aefZjAY5LRp0wb0+drb25mRkUGv18tnn32WBx54IGfNmsWysjL5sYmkAnV1dRwyZAgBTaB7PB5WV1fLBEJtbW2sqalhUVERzz33XJlMaM6cOUxPT6fX6+X06dNZUFAgl5mi0UvEB5CTk0O70x4RljkghumEp01z+xk5ciQ3btwoUWsAbSlKBG0KwVRVVcUXXniBdXV1LCws5BVXXGEQjkJRjsk0aQ/fcycSuOxsBP5OcQ84uClOcYr7yD1NwAeZ5bwnjl6ljMuDyCVkf2GxUi3+xw3MFJjcUfuERVms+AYCAa2cA5quop/MOMLvNxE4gVD+dVlLhUIrFGcxrimKQpfLRVVVOXLkyBhMdLGCMXHSxJg+GO2GIrajfcnFcX2fjTfRE3UEjKvB0UhB0Sz0Nj3GvtVqlSv6ADboyv8AYCOAbgDjAKwFsA5aFs8uAO8DUAEsDpdZD2AbgK8AKEgp5P0nAJqlQGB8nwfNX3kqJISgM9PJFStXyGABERzQG23cuFFuX3vttTzrrLMSll26dCkvv/zynX4eMoxaIiwbVq2zCoV8+vTpnDVrFmtra+VsMTc3l7m5uWxqauKPfvQjZmRk8I477mBraytJLUgiFApRMSlSONTX18vMYMIy7fP5OGPGDBYWFjIjI4Mej4dPPfUUW1paaDab6XA4uHz5cgLg1VdfzRUrVxiX0cIfhMVi4fLly3nooYdy+PDh/OGHH6iqKs8//3wCIKn5w2dmZtLtdrOwsJAOZ1SApRkxH/iAcJSgVBRlz8IlRrffIEoLvcd5FwWS9ZlTk6U9y3uRIj1QXFVVtde4eexrHL0K63a7DXjnFotlwCBBYwxM8TjRxMsDzV8d2jg2e/Zsrlq1ig6HgzNmzOALL7xAl8vFzz77jFVVVSTJM888k1dddRVJ8rHHHiMArl27lgsXLYy4/ArEtzA2vslk4h/+8AdOnz6dRx99tEy8J+oRDxzBZDLJCYF+YqBvN30CQH0yItHW+rYRrrICsx0ATzrpJHldaAo1oSncH4d/uwGcCOAbaBZ0EzTL+FYAOQCmhfdPAXA+NKt6C1IKef9JdaiawG5GbIdygTa3rd8Qgvfff79M/DNt2jR+8803ccsdfvjhrK+vj8HL7A9NmjzJ+BxzIp182bJlPOigg+j1ejlt2jQC4FFHHcXly5fTYrGwqKiITz/9NG02GydMmMDW1la2tLRo7ihhv23kQ5t9W5Pzl9an7BUfmhBaMX6Obqd0/4inSIvlPeED7/P5YmfRop7NPdcrKUXd0fNxr9c7+Ae92kFQh32d9bEHvfSZvnAKxzvF/eXCwsK4rnSDJd5Bb8TQu8sMdPKfgeCdlfHxztdnBE2G9QqsqqoGiN9ez0+LbOsswhqHDVcnnXQSSc3QlZGRQb/fz8bGRqqqysbGRubm5nLjxo1sbGzkWWedJRFNLBYLP/roIy1/SBsi+VBaoa1497H9olFqFEWRbdXc3Mza2lp5vdzcXAkbqW8fcY14K9c2m016AejdWBBxWbmTJKBZu3eEtzcA2BTefgyaRfwwAEdCU8I/guZ73glgPlIKef9p4aKFEf/xEYhgttpBk820W+syEGTwhRdcEPmQVVVlbm6uwYfbbDbLLGXReOkxwUAi8HIAhKVeKQ8EAvEtzQq0mbxulm82m+lyuSSKjSzboDtvFy3H9ipcBsmANyA8SAbEQc8lg6AOKU4xdg8ed785gUzWZ6E08F4mfxJZq/WuG31FsxFK5THHHEOPxyOVeDFWChcWk8lk9LNO0HbBYJAnnngiHQ4H29vbecstt1BVVZ5wwgkcMWIEhwwZwtGjR9PlcvGvf/2r9AYQ0MfRCnkgEODatWu16wuX3yuhKeRDw/d1Q/Nv169Y6gwXjzzyiEy6KLKjqqrKjIwM6W8vQBvS0tJYX19PQEMJMplMnD17tsyirqoq58yZQ0CDng4Gg9LlpqmpiRaLRb6DkpISCVcN4N/h+nwKIBOaQt4d3t4M4F/UFPI/QrOmzwDwN2gKuR3A0dAU8uOQUsh3jmKsym2gNc26VyXXERTzYVwJDU4QIEl2dHQwLy9PzigzMzN52mmnUVVVGaCYlZXFpqYmWq3WyHJSmu7jciESgKkTOvr/euGQKLuY2WyWgkXMhGPK6lOwRy396y0FBjYhNhnGAPKetoonk8QjxbuRRf/ciTiEvYb3MiVpf+JBrYxH8YCt/kT3RxGU79219U80Buitsv15xmjfanENMWmJRnHLzs6WPtw2my0uTvm48ePo8/k0P++JE3nppZdSURSee+65fPfddwmA48eP54knnsjvvvuO06dPZ3FxMadPn06Hw8Frr72WhYWF8t7p6elSyQUiyRGRD23cVSFhiQH0ayxOBiM+WlcQEx8RRKtfQddPlqLjysL/t4ePfwPgPURcVv4DTdF+CZrLypPQrOlvQVPU/6upXbgbKYV84EhgkSuqwlB5aK9UxskEFvJZWkdsbGxkKBSiz+fjzJkzJUzQvHnzCGjLX9ddd53ERze4lQiF3A9NKY+yBCfyo3Y4HLLjx1MkFUWREdgAWF5eHjk+EkQ1IrPp8Uahpaqq5i8WbXkREeZR8E+DVZHd0wp+ilOc4hQPFCdjBbZarQMzgYgyDEn3sSsxcJNHMb5EXS9ucp/eWIm9jp4TtUm09V2Uy83N7dO45nQ6DYpodNIp4WMd75oCcjF6f1L3V3TtmAGirOfyhYWF8rqZmZm87bbb5DGhZNvtdt55553yGWw2G51OJ/Py8jhs2DA5iUhLS+ORRx7JgoIC/uEPf6DT6ZSIN1lZWbzyyisJzTf8fmj+4euhBXUKN5VPAfwCwLvQ3Ff+F96/CMAmAG9CwzDfgJRCPrjp7bffZk5ODkePHt1jYp5QKMS1a9eyra2Nv/3tbzlp0iQ2Njby/vvv5/PPP8/i4mIGg0Fu3ryZzz77LF988UXD+ePGjePJJ59stPYvgBac6gZz83OpWCJBmR6Phx6Ph1VVVXJWHwqFWFVVxZKSEs6YMYNer5dWq1WLBN9NmQ0TpgNWIa0ey5YtY1lZGd1uN9PS0mKzf4WgQVfqhQB6xlndZzh6cuIKt0MqsDDFg5QTZh/cz7ikpCTx8QRKnF6mDdqYA1F3sbqaoFxcxa4vroc+aGhdgKb09UEh16NvAGBLS0v8OujkaFpamgbHG34HZ5xxBvPz86WbBABitO78mSAyQZwAg183EN/f2Ww20263690qJPt8Pon6Fi/7ZbKcnp5uCDgFIBHMRP6SPdFnYlxwdO2kN+DpM6E3NTWxpKSE6enpdDgcMuPnNddcw4ceeogLFizgm2++yccff5xtbW20WCxSf4IuOyeA5wCsARBklH4KzRI+O3p/f1hFigacjj/+ePzud7+L2f+3v/0Nzz//PH79619j3bp1eOmll+KeP23aNLzzzjvo6OiQ+9rb2wEAb731FubMmYOVK1fi6quvxpdffgmHw4Hnnnsu5noTJkxARUUF/vLUXzC0YijwGwD3AngKQAewZv0a8BgCYwFYgY07NmL79u349ttvYTab5XXWrl0Lp9OJL774Al6vFwCgqioUm6IVcPS7qZKizZs3xz/QDW3+CeC8887DRx99hB07dmDdunXYvn27sey3AO6C9gl3R3brPqoBI1UdRJ+VC0BI91+FBtREANV7pEZ7lpQ9XYEUJUNbt27d01UYFPTJJ58kPphAbOllWnd3d/xCe5pE3TcB6EhczOPxxO6M80gejweKokBRFG1sUnQfeiEAZ/heJt1JlvDv+PCvAsCqu013N0jCZrNBURS89dZbWhkTAHe4fAk0xwYAvjQfDj30UDQ3NyM7OxsA8Itf/AKrV69GdXU1cnJytILp0JKxA5r38VoAvwdQBiBN2202m7F9+3acffbZyMzMxNixYxEIBNDV1YVLL70U48ePx7hx4wxt8MMPP8Bms2Hr1q2YN28empubYbPZkJ6eDpJwuVyyrMlkQjzyer3YsWMHPv30U8P+L774AgCwbds2dHV1xT23N0p0z2RJ3Fe82+LiYgBAZ2cn/ve//2HHjh0AtPfW2dkJVVXR3t6ONWvW4LvvvsOCBQuwfv16KIqCZ555BhdccAEefvhh3H333XjiiScM9wrrXkWKoryiKMqb0N7aYSS/ja4XyeNIPrBTD6e72KDnvc1C/o9//IO///3v+fLLL/OZZ54hSb700ktctGgRd+zYwerq6h5ngvrlIZEKNhqzU8xSvV4v77zzTsP5Bx98MB966CGZBCI7O5smk0nO3GXZUmiBFmJmHsevy2w2GyCGPB4PzWYznW6nhDACoFk50nuf5fY0c9dbI3Jycvo0eza0qbA+6ANG9JaRooGfvetn7oPVDSYh70tBp/sYx7VwemFAcYnGMZasX7rXvWOB3b9LeG/xJ+8pA69+5c8SdWwA69Dnlbn6QdBuu5lFUF5PbSiwsAHNhUIkfZFjjT4xTvQ1FGjuj2LFMDNyXa/XS5vNRofDwX/+858MBAIaSo0J2tiZASIvfJ4ZvOaaa7h8+XI2NTUxFAqxs7OTS5YskcGOJBkqD2loX9HJ5gBtvJqhbVutVhYUFNDhcNBkMtHtdjMQCEgLbzAYpKqqHDp0aK9ZS4PBoAx6FKx3Jx0zZkzCMSspCMU9zHqLfXRf0WePFe9VbJeVlRms6gK55aijjhIuMF3Q/MKPgeYb3kJtojsFwD8AvAFgFQA3B0jX3ePKdjK8tynkZ5xxBr/55htu2bKFkydP5hdffGE4ftppp8V0nmnTpsnU7j6fj0cccYRUvPXLt7/85S9lx7LZbCwrK6PT6WRFRQX9fj/HjBlDRVF44YUXysCKrKwsWiwWFhYW0mK1RAaccmjZv1pgUF4zMzOlINTjfm7dupW/+c1vIp16QfgcNzRlPIkAtkTBnDvLcaGx9hblYBfzlIOmGPeN7+WceINFivcIx1VGkp1A7c9JXrIHQR0GmgcLhv5u5ETZF3fHvaPdGU0mk0T7iFfWbDbTbDbLCXIgEGBzc7N0SX3zzTdZWlqqneNBJPlcHfrk/qlvkz2agK4f7TkQ14lu/2hYRP2xyZMnSxAKu93Oww47jIFAgNnZ2TziiCP4s5/9jIBmBJ04UUtitGDBAuGyswOam8rWMM8GEATwPAAXNeX8YgBXMKWQDz5asGABAW12K2bnwudKD+V3xx13xHRQs9ksI4GT6cgWi4Xz5s2Le0xRFIk0oiiK9F+LsTqr0KwHLZAKtclp4rPPPssDDjiAQ4cOZSgUYnp6OjMyMkiSX331FUtLS7WZ/kHh6ySImBaZvKKt+2J7Vynn+x3vz4pXigcFJ70qlIpb2Cd5VynJYrwQllqhfPWm3AlggKTuE0d+2mw2Q592Op3Sd9tisTAYDNLj8dBisdDlcrG0tJSKojA/P581NTUcP348J0yYwCOPPJJzj5mrjZcAs/KykkILiRfgKhDPxPEerzEIv7NdFasVrV/o9YrDDz+caWlpciVl+vTpsv2tVqvUuWbMmMH77ruPdrudw4YNE0bQbdAydH4O4BUAzQAOgeYA+1aY/w3gV0wp5IOPnnzySQJaNLBYOqusrNQ+1JL8Hj++mIQ4UdjfYlssxwFGIZiZmSmFxpYtW1hQUEBVVenz+fjggw/Ketmddu1jVaBZthtBjIEW/KKACxct5LPPPsvx48dz6NChJCmXyWw2m0w6kJ+fHxFk8VxE9jbeTQGq/eZA/88dVEGrSbRzsvVVFIXp6ek9l9OtkiRtTYp2W0jxHuFoJa+3VNj7FSeYhO9MMN/ewoFAIKl+sDNtkWiCmZWVFbMvXl1EBmm73R4Za3t4b8lcU79fr6zLbTP2+1VhvfETAEtLS1lYWBjTVg6Hg4FAgI2NjVQUhccccwzvu+8+OhwONjU18ZRTTiGA7wE8RM0S/ho0hfxQAPdxF+m6gyj6bO+nLVu2AAA+//xzTJgwAR0dHVi7di1I4os1X2hhAVYAkXhJKIpiCLwRAQv6oESbzQZAC4poaGiQwQttbW0yUGLDhg1QFAVmsxk//elPoSgKwh0JDz74IAAgPT0dLodL84wiNHj7dwG8COAHwO1yY+7RcwEAq1evRmtrKwAt0MPn82Hbtm24+uqrYbFY0NjYGAmu8YSfa9sANGL4fn2lnQ6k3LFzp+9y+t74N5nnFYG5oh/4/f6Bq09/43OSaGdR32TKfffdd70Uimx2dnYmdV10Y/D3h/2AhFwUMrGzszNh3zAE8e0PlCBWMyaYfS8nRVFiggG///77mH4QTx7uzJggxtjounz99dcx+/R1Ef3w5ptvBgB07ujUAkA3hQu4kBSRhNVqjbtfURRDYKXcVmGQd1JGW5AU7a3fkL5/dHV1Gd7Hxx9/jM8//1z+dzg0BIqtW7fi+++/x3vvvQeS2Lp1K7Zv344tW7bgk08+ETqTE8AoRVG+AVAP4BYANwOYrihKGQAoiuJSFKUCA0W7StMfSB6sFvJonPKJkzQfpGeeeYbFxcWRwAkVxBXQfLYzEIFhSnJWHC8FLADOnTtXnqMv43A4eOutt8prCt+oyZMn87LLLouxPDm9To4ePZolJSXMyclhSUkJs7KyuG7dOpJkZmYm09LSWFBQwNdff50Wi0WuBgDQsn7aYLQs7ucz9cHAgxbyLMUpToJT/TfFcQOWo3HGU9wzJ2mV31e+t2greU8cDAZpsVjY0NDAzMxMOhwOzpkzRw+1/D4015QtAFqhZeD8CsDbAN4J82EcIF13jyvbyfBgVMhXrFxBk8ukuRIoiPwC3LFjBxcvXhxZ9lJAXAaiAlpEd66xU/j9flqtVv70pz9lUVERvV4vs7OzZadqbGyU4P0kJUbo+PHjWVtbS0VReO2119JsNjMvL49bt24lSU6bNo3p6em02Ww84YQTZHDJxo0bWVBQkPDZFi5aSJNTU/xNThPnzZ/H8vJyut1uqnbto1Vtuo93BDQhqQ82s2JglPIkhMluC2xJ+WqnOMUpTvGAczJZLlOc4mgWbmyqqkp9q6mpiUAkl4HoW6IvmUwmulwu+nw+ut1uVlRUkCQPP/xwiYpHktASATVAS+zzfxRKM/AnhBFXBpr3uLKdDA9Ghdzld2l+123Q0tK3QWaPzMrKYm1tbaTjqNCUVZGURqBYhBXYefPm0W63s6Ojg7/+9a/pdDqZn5/PnJwcKorCV155hf/5z3+oqiptNhttNhtra2tJkr/5zW8IaLPbqqoq/vWvf5V1fOihh2i1Wg37BJ111lmsrq7m6tWrDfsXLloYSSIknsul86kbD83KXxr1cfQ1RXhPynoiSLI9xbvRN36XJl1IrVqkOMUpHiychDzaWxNEpWIddm9b65MHivg7obOI/wJgw+VyUVEUHnvssczOzmZxcTHJXhXyXzCikD8GYDx3ga67x5XtZHgwKuSwIjYdfRsIE2ixWlhbW8tly5ZpyrcZmgVdKLhmaIq5SwuiJMmf//znrK2tlecJcrlcJMnu7m42NDTw5JNPJkmed955nDp1KseNG8fly5ezpaWFHR0dO/1cJqcp/nOJyYRPe0bkIXGgYby09QP8EfaquAZ124G+1cEQZCveVQVi4Ob6jDfuSHxsX1kuTHFiTg3SKU5x33i//mb2QwOKeN8XXnhh3P2KohjyouiPi/8i07gYUwWqysiRIyU8pcvl4qpVq5idnc2ioiKpg51wwgkkyQ8++IDQouJs2I0KuRK+waCm5uZmvvbaa3u6GgZSFAVYDGNwWxeAHwOKqqC7S4u6WXT6Itz+q9u18ID1APzQIOa7gEkTJuEvT/1l91a8F+rpuZAB4PTwvm8B/B8igZxmaAErW6BlPvNDyz4GaBnNEnWzeMfSAKxL4txdRKqqJp3hLj8/X2Yx65WKAbT3XMRut6eyFKYoRSlKUYr2W0o0Blssll6D861WqyG42eVyoaOjAzfddBMuvvhiFBYW4uOPP8b06dPx6quvwm63o729HVu3bsXChQvx2muvwWw246233vqQZKWiKMcBaCZ5BgAoivIYgBtIPjeAjwwAKZSV/pLqUDV0Sj19DsAGFJYWyl23/b/bsPDEhVC3qppiuRlw2V1Ycc+KQaeMA4DJaYr/XFYA+u9gIzR0FRNgc9i0eaQFQBWAA6ClA1ahhUDEUahlBDwBnAJgrK58QFdQd65ADdH+hH9FKmMkKCfKJBFALurUqzKuu59UxpMJUI9u1ygym807p4zvXGbigSfnnq5AilK0j9PeCYwx6EiPMHLppZfGLzTY5Os+RF6vF+np6fK/fgwOBCIKQWdnJxRFgdvtht1ul/sPO+ww+Q6zs7MxbNgweczlcuHPf/4z3nrrLWzfvh3ff/89/vKXv+DYY4/Ftm3bUFBQgIMPPhj19fXIyMjAv/71L7z55puApuWA5N1CGQ//P2RXKOPi4oOeB6PLSlxfazeoOlSuWLliT1ev3xT3uYTbhn7/ASBMoMvn4nHHHactHdnCZeaEz7GAGA5iJrR03wr4+uuvU1VVtre3c/jw4TSZTYQC5oZyqaoqvele41KVGlmWKiws5JgxY/q/tJee/NJgWloai0vCKcbN2rPiVEQSLkwK77NBS4wk9uvcdZJJACGfU1H2yyXKFKc4xSne064pIjM1oHOHjHa97CHZTnNz8x5vw72dE/WBUCgkUeS++OILer1eWiwW2u12FhdrY/SGDRtYXV3NYDDIH//4x1y9erVMDvXiiy/yhhtu4Lx583j00Ufz2muvZUFBAbds2cLly5ezuLiY69ev55YtW1hYWMjPP/+cJAngNe5mXXePWcgVRTlYUZQPFEX5SFGUS/ZUPfpLt/2/27Dw2IVQf69q7hz3Ay6TC7/55W8w75h5e7p6/SbxXKZVJuDHgGmVCYpZAcYA2A7gd9Ce9xUAXcC2jm14+OGHUVRUBL/br+Gt/g6a5cYRvmgegA3a5uTJk2GxWPDOO+/gqaeeQk52DmxWGzau2wi/348zTjsDTm/EtMpuyu2RI0dKTHaDZSgLPa/1jA2Xz9btMyNiZdftFzil69atQ0N9gzbr3qE9K34dbgMVWiLdLmguO3ZErCdjwtVTlEhdeyCn0wlVVTXfNPZavHdKWcxSlKIU7QT1GY+6P5bjqFsw7Dq70/kk+lIFRYGqqvJ5hVVWWmejF0p7gHh//fXXd0EN9y8SfQAAzj//fACaS2hHR4fBSk4SaWlpUBQFO3bsgKIo6OjokMdff/11jBw5Ena7HVu2bMFLL72Ev//975g2bRr+9a9/4ZJLLkEoFMKHH34IAJg0aRJ8Ph/sdjtqamrw2Wef7aYnjkO7ewYQbnQTgI8BlEBzhngbQE2i8oPRQr4/EQDNKj4LmjUY0JBWgLjlJ02epJVTIucpJiXuykFbWxtXrVrFFStX0OVzadZod/heGSAOjsyUV65cydbW1tiZdG+WZVMvx/dlTrQq0B82oWc0ncGGjpPiFO/nnNBy25/VuN2wgudwOPp2jr3n4/pgv2AwSK/Xa0Bu0Y8lfYJYVKJ+dxXvR2OXHqghEchBfn4+s7KyqKoqS0pKJLBCKBSi1+ulqqr0+/0cN26cPMfn89Hr9XLIkCE0mUzMyclhXl4ef/e737GkpIRpaWk8/PDDuW7dOk6fPp2NjY286KKLCKADwIcAxnIft5CPAPARyU9IbgdwP4AZe6guKeqFpF95PYBLAFwJYFp4fxz6y1N/wWXnXwYQUK5RkPd6HvJy8hKuHLzwwgs4fuHx6NjWoVmthwPwQbNIjIK0psybNw/PP/88VFU1Zm8jkJOTk/gBuhIfAuL4nPfx+KCmXhJZ9kQWS1SKty5oIioRbe7b9aMz8KUoRSkaWEoIhsB+XKw/5/SRRLbrpEmE3CiIyYIZvUr57bffYvPmzYZ9qqrKVcrJkycnf18hunayTXpdjehl7NqXSJ99NFEc19atW/H111+DJL7++mu43W4AwDnnnIP09HSoqoqOjg68+uqryMnJQVpaGrZv347LL78cFRUVSEtLw4IFC2AymXD11VfjyCOPxNy5c1FfX4+rrrpK3iecqfV9AOcAWLLLHjqK9pRCngfgf7r/X4T3pWgQUUdHB4YMGQJluwLcA+BOaG4bPwawEjB3meF0OlFdXY1ly5bhueeeg8vlQn19PRQocDgcGNI4BH9/9u9Ys2YNxo8fj/r6euTk5KC+vh4NDQ344IMPsPye5ejs6NSE61YALwF4HZow+ikiQq8W4LFEF7vkx+v1egFoLiZVVVVobm6G3W6Pm3ZYT4qiSGEYnSY5WlGMl0Z5n6KoptKnKh/WHAmOGeigJr0ATlGKUrQHSEFyqdX7oil4dJfvo/uLzWYzBOv1SnrFeBsMrjCKoqCzs1O6OLhcLgQCAfzwww/yeFdXFzZv3ozu7m488cQTCW8T40ojtBVPTNFeye12w+nU3DLJ3TDL2YtJGNpMJhOqqqqQlZWF2tpajBw5Elu3bpWTrltuuQVr1qyB0+lEd3c3bDYbXnrpJZhMJrjdbsycORPd3d1Yt24dVqxYgV/84hfYsGEDqqqqAABtbW14/vnn5X2POOIIsfk6gKLd9sC7yxSvZwCzAfxS938BdDiP4X2nAHgNwGuFhYW9OFWkaFfQTTfdxIKCAm7fvl0L9tQtnx0w5gAeeOCBvPDCC/njH/+Y33//PQ899FAWFxfTbrfLpcfCwkLm5ubKZSiLxcK8vDxOnz6dL774YvzlK4vu16VbFvSAmGZcJjSbzbRarXQ4HLTZbLHLjuY9vxS333Mqw+n+09ap722fYEOeB71bRjJp6+2I5FxI0qUjLS2NgDZGCHeFnvI89BYwL4IABfv9frmtKEp815hUUP0u5f4G7iqKwvz8fK5du5YnnHAC6+vrSZLp6encsmULSbK1tZWPPPIISc0NtqSkhCS5ZMkSZmVlsb29ne3t7czIyODSpUu5fv16Q7byjz76iEOHDpXXevXVVwlN/wwC+JT7uMvKagAFuv/54X2SSN5Fsplkc0ZGxm6tXIo0+u6777BmzRrk5eXhqSefQkVpBUKhEMxmM/746B/xySef4J133sHf//53eL1evPjiixg7diy2bt2K2267DQCwevVqrFmzBgBw7rnnIhAIYPXq1Xj88cfR0tICADCZo0yvndAsMvnQXCQY3t8F4KnwtgnAGGBH9w5s374dpaWlyMrKgtkSdi9xQcNNnxourwJWW89W84S0r3pWJGG86nOAVzxKDtK9Z9oZSbU/BboORFvvDMVZTHI4HIlXrFLAu4OSEq5ebezhpGD4txMRVxImKBsmry+ywgkAHo9Hrnru2LHDGIivo23btsU/ECY9DjUArF+/Xm6TjO8a00tdU9R/MpvNMasBicYWi8UCVVVRWVmJYFDrVOLcuro6fPbZZ/juu+8wZcoU/OxnPwMA/PDDD9i0aRMA4O2335bX8ng80v3F4/HIfuPz+RAIBPDCCy8AAO699160trYO1OP2n3aX5q9naJ7Cn0BLlSKCOmsTlU8Fde4eWrhooZapE6DJaWJLSwvPPvts3nvvvRw3bhyvuuoqtra20m63kyQ3btzI888/nwUFBTz++OOZlpbGxx57jAA4fPhwAuDSpUul1drj8UjLi8iYFW3J6JWTSWOfjoilbj8KihkI3tsyhvY5W2oyvKetzIIHwtrc2zVSVsEUD2buAWpwsPPeJkt3JQvruH68d7vdcttischVj4yMDGZkZLCyspJZWVlUFIV5eXlcu3YtV61axTFjxsis5kVFRayvr2d+fj49Hg+bmpo4evRoaSH/4IMPaLPZWF1dzeeff551dXXMysriBRdcwDfffJMjR45kfX09Z8yYwXXr1pHcsxbyPaKQU1PKp0GLYP0YwI96KptSyHc9xcUft4Ner5dff/01//jHP3Lq1KkcOXIk7XY7165dyx9++IFTDpoSUWBM4AEHHECXyyUVpba2tt3ywaenp0f+q4hgoofrBYAWe5Q7iwmawLdE9sllNbeuXEpp2SvY0AdSPKjY4AKR4hTvJ+x0Ovd4HfY0u1wuFhQUSKz3rKwsWiwWnnDCCQwGg1QUhRaLhYsXL6bFYuHIkSO5dOlS6W5SUVHB9vb23a4TYX/CISf5BMkKkqUkr9lT9dib6KSTTkIoFAIAbN68Gbfeeiu6urqw5P+zd91hUlTZ91RVV+fu6ZmenHqGmWFyZMhIVKKKioqCiBhQRNddF3dBF2GXNa9xDSu6IgjCinnV3+qqIGYFQUBAQIawBMkwOfX5/VFd1d0zPYE8aJ3vu19XV7169ep113u37rv33BkzMH/+/Gblv//+e3zwwQfa94qKCjz99NMhI5hnvzgbGAVlzULyffYBjpYfhcfjweWXX44tW7ZoS4E7d+5EWlqaUr8BQE8ANuCLL75AZWWltuS5ZMmSVu/phJk2fKteBw4E0Il4oQT4VPi++1Zf62uapNxthMLkErBbeQ4BVENZu7FAGVaOxe3BeQxldbQP7fibqMveJw0n29Xl1+Q60wR6AK+Osx3Hw5Ee0jWmPUG0ZzkCXdQqKyuRnJyMefPmQZZlmM1mpKam4osvvoDb7YbFYoHJZMIzzzyj8YpPmTJFO//9999HSkrKGbiLM4DT/QZwPPJrt5APGDCA3bt3Z0JCAkVRpNFopNvt5rvvvsu7776bpaWl3Lt3L0myoaGBWVlZHDVqFMeNG8eqqirW1NTwqaeeYn19Pd977z3+9re/bXYNAIplfGaATFfecAPx+uuvMz09nW+//bbi3jIeCme4BUo2zr7H9vZ8WjO0hfs+h0GxelvaKK9mKA11zOn7VK3nKj+3EbpFXZczJkajMciNp3Pnzq2fo/9XdTkO0d0xfln9FBj02pqo8/W8efOC9qvuJ7IsU5ZlbQy6+eabefvtt7eo21RWVtLr9ZIkFy5cyAsvvPAkak4nBpwBC/kZV7bbI792hZwkL7roIoqiyOuuu45du3al1WplUVERU1NT+dNPPwWV9Xq9LC4u5sqVK3ngwAEmJCSwtLSUkyZNarF+TbmeGSDjFV/ypli5ciULCwuVh3E6lAQ+40FcgbOLZaG1RDZyiHtpj2+xDYTn+NuUmZl55vtFF1100UUXXZqIzWZrZkSTZZk2m037HvgSYjQaGRERwY8//phz5szh5MmTSZJz5szhzp07uWzZMhYUFDA/P5/nnHMON23a1KKOUlZWxtzc3OPUoI4d0BVyXSEPhbKyshYzmKl/flEU6Xa7KQhC0ANhMBiYmZlJu91Oi8WivblGRkZq5ZxOp7/OMAQpouHh4czOzmZkZCS3bNlCknzllVeaB9NNANHZty34JAxB/tmnQk6rhb29YkCw5bGjBAnqcnbJKX52TqZ0yOdQF12OQYKs2S2sHKn/c4fD0WZdvXv3pizLvP7664PpeAPqTkhIoMPlCM6oLKBDG7YMBgMjIyO1/jCZTEHBmq+88ooWM2IwGBgfH8/s7GzOmTOH48aNY25urhY4SZLffvstb731VpIKTeG0adO4YMECTf9ZuXIl3333XV0h7yjya1fIX3/99dCDRsD3sLCwoElRDSaRZVkLdjOZTLz44ouDmE7U8mazOSTjybPPPkuDwcBOnTrx3nvv5ezZs2mxWCjLMnPzcv3Kd6DSmXaGB9PjkPYu2QW+iERHR/v7sGkKZ5XdpRDKakOo+tRB9xQr7LqydPZIi/zKZ5Fyrosup01aGzsDjpnNZk0pdjqdTElJ0Y6ZTCZKksTw8PBW67755pspSRIFQWBdXR0dDgenT5/O4SOGa2Ukq8RJN0/i119/zZiYGNrtdt406SY/YUIWFFdJn1L+7rvvKttWEF0Drtcam5g6b7SkuJ9kZrH2zK3qHCNJUlAAd+C2KIo0mUyMioqiJEk0mUwUBIGzZ8/m8uXL2bdvX8bGxjIxMZHnnnsuSXL58uVMTEyk2+3mlClT2q2Q19fXn7DeBV0h1xXyUJg4cSIBBC0LNbVQq398tYx63GazcejQoVq5qVOnElAUdY/Ho+1PTU2lKIoUBIFut1uzyP/www8EwC5dunDSJGWgkSSJ//jHP+hwOJiVlaXUoSqksn+wwWUg+jV5eKc2GegKQgyqbfi1Blr01YQSJzqoNEsoFELUQaStwT/oPkp9A3Goc3T/XV3aKy38V1JTU89823TR5Rcgbc0RgTR9gRKodFqtVi35HQCWlpY2P8cCfzyTKiozWFvtFEC42ijjCih7ivpK7Yumxp6mhi21TwVBYGZmJrt160ZRFBkeHs7MzEwajUbKskxRFDVmFYfDQUEQaDQa6XQ6GR0dzcjISNpsNsUQmJvLoqIidu3alUVFRXzzzTdJKm4wF1xwAQcMGMC+ffuesN6FXxPLio6WcfPkmyGaRAgmJb377NmzASipzFXYbDYA/shvj8cDQRC07y6XC4DCxhKYJEFN0mOxWIKI+Xft2gWv1wuSqKyshMPhgCAI+OmnnyAIAhwOB+Li4nDw4EE0NjZi4sSJIIma2hqFAaOzr6IeUJhJAGAxgE982758PXgA/uQlMoACBCcHMQOIaL1/qqqqIAgCJEnCoUOHWi/sQyhmmUA0NITIaNIEcXFxQQkpghhiQlVPKEymr/m+GxF8r/R9Nkn33HSfDh3af6UJysrKTm87dISG/rx2SBgMhqDvrTGltDVHBDKWJScna9s9evTQ6q2urobBYMD5558Pg8GALVu2KIXMUNi3RChz4dGmF4fCDNYWCOBwG2XU4y2MGceKUH2mJuFpdumABEyAv08lScJVV12FtLQ0eL1eHD58GNu3b4fJZILVaoUgCLj77rvx5JNPoqamBhEREVi3bh0qKysxdepU9OzZE0OHDkVGRgYuuOACAMDf/vY3LFmyBHfccQcqKysBAN999x1effVVfPLJJzgboSvkHQznDj4Xz/zzGdBCoDsAt/9YoGJdVVUVdN7Ro0chy7JWRqUdIhn0kKxfvx6AQkUUFxen7S8pKQmqT33gpk+fDpJYu3YtrrzySrz++usQRRGbNm2C2+3G1rKtQBf4qQO/gV+htvvrk4wS8vPzlS/q5NUZwKtQFNVU374aACqDYZNJTpZlWCwWREdHgyQaGxuVZZ6TgPbU87///S/oe7up3NSxq66F4wGX1tpxkgZTHe3Ar1yZOh46Nx1NoD+vpw2C6H9gH33sUS2zZ1PlG2huaGlL6W4Nffv21bYD59/t27dr9ZJKFtBnn30Woij6Fdc6KFS6XijZTkNNHR10HGqtz9o7/9psNnz33Xfab+R2u0ESUVFRcDqdIImYmBgMGzYMV1xxBXr06AG32w2TyYSysjLs2LEDX375JbZs2YLHH38c69atw9ixY9G/f3/U1NRg+/btAIDzzjsPERFtWPQ6MPSRuANhwcsL8NGyjwArlJTB30JRTi0I4mEWBEGzlqsPy/79+1FXV6dZcFVLOKC8NaqW12+//RaAokyuWLEiqIyKuro6mEwmkMTq1asRFhaGuro6REdH45VXXsGYMWNwxRVX+AebUgC7oSjiEgBVb/W98QuigITIBMiy7F/MMgI4AiALQCGUfFjnQEl5n+2rJwqADAiSAHekG9HR0XA6naiqqoLRaER4eLhSp4qm43GoAa6NQc9sNrde4ERxplOb62iOX7kydSJKyq8ap0KBOsG0DL8GiIJfbfndb3+Ho0cVc3N7VjlPBIH179+/X9vesWNHyPJ1dXV+I5oXQfkuNFgDtk/lOHSGuc+PHDmC//u//8OmTZsgyzIOHDiAhoYGxMfHo66uDoIgYMGCBUhLSwNJlJeXw+VywWg0YteuXSCJvLw8pKWlITs7G6tXr8aOHTuwatUqbN++HdnZ2QD8ngNnLU63j8zxyK/Fh9yT4ffpbpY1MwxEhHIsJSWFJpNJC3ZwOBycOXMmP/74Y82nu6CggGlpaczPz2fnzp21YBSSHDx4MN1uN81mM202G3NyckiSn376KaOiomixWBgXF6cFUFRVVXHIkCHMzMzklVdeyerqapLk4cOHFd+3cBCJICZBCWC8UvGJy8rKYn5+PmfMmBF8j2YQ5/h83QLv0YbgaHMZFCWRHo+HksHnpxcQsCKKYjBNYHTAcROIZN/2Jb7Pzr7jAojf++sxWUxMSkoiAKalpWnBrSaTiYKoB0T+4sV9hq+vS8eUk+yDe7ZwUuvSseSkBOV3kKBwNXg2kFhiyJAhdDgczMzMZFRUFI1GI9PS0rh//37GxMQwISGBdrudcXFxzMrK4rRp0zh+/Hju2bOHJPndd9+RZBCt4skA9KDOX7dCLoiCoqyaEZITHFHKn1oNflDh8Xi4b9++oO+jR4/m4sWLSfr5O202G0ly/Pjx2rFvv/2W/fr1C2rHoUOH6HA4+NRTT7XZ5kHnDWr+8mAHB503KGT5+QvmU7SIyjnn+O5JgKJAx4GyXebgwYO1l4Ghw4ZSNIr+JD3DQNwILQDG5XL52WGsUBLzSL561SCZq6CcPwj+4Es1kY9PAgNc26K00kUXXXTRRZezRozHVj4wULWptPSCoL5wSpJEs9kctF8QBMbGxvKiiy6i3W6n0+nkoEGDOG3aNKakpFCWZWZlZfGPf/wjx44dS6fTSaPRyMsuu4wHDhxgaWkp4+PjmZSUxNzcXEZHRzMjI4M5OTkcMWIESV0h1xXykwxPhofI9P3BQ2XNFEC7237C1wlUyENh1qxZnDp1arvrG3TeIL/ya2pZGVcxf8F82sJsQeeoirIgC8wvyGd2djZJKhyt8b5yESBu8PVH0yh1QFHE1QyaLjRnPkkKcY4ARsZGBg0gvzrpwJy3gb9Tm2VOEt2XThOpCwA9f8DZJjpr1UmVQCNVKAk1Tp577rkhy+Xk5HD8+PHs1asX3W43Z8yYoSnSJDl58mTOmTOn3TrH6QB0lpVfN+6ZeQ+sh6yKX9n2Jge3AzAB/3jiHyd8nRdffBGXXnppi8f/9Kc/4b777mt3fR9+8CFY4/tT1RAffvBhq+XHjhmLisMVYA0x6eZJin/bOADTAQ4m1qxZgy1btmDbtm0oP1zuD4o0BlQiQ/E3B1BcXAxZljHrz7NgbPQVqoTmrz3h2gmIi48DAl39BF99ArB/z37U1NS0+35/cTi1rpcnB2xHmXbG2LZ5KbbnYjpOGixtFzkj0F3rzy6cwGNrtVpbPBYYpxQWFhayjBYY3VLcUqj4ptBVnTSoxA5tIqBtXbp0AaDcz1//+lct9mzmzJnIzMwEAOTk5GDBggXaOPnhhx9CFEUIgoBVq1Zpdb300kswGAwQBAF2ux0PPfQQ/ve//2Ht2rXo37//Cd/fLxG6Qt6BMHbMWMx+dDbcTrfCPlIGRckoA4TXBUy6dhLGjhl7hlt5bOjVqxcAYOvWrXj55Ze1/S+++CJuueUWPPP0M0AmFJYVCUBXAAVAbW0tMjIylAFBVbQqABzybddDUbp9dcuyjJcXvAyXzQWTyYTZT83Go48+CgCY/9J89OjeA5MnT9Yi8kEoE260r76OqhScIphMpjPdBB06FFSfuUufaYaZwOunp6cf8/mB1KuiKAYHuetoNwJZU+Li4oJ+l0BlXaXXAxDE5mG3+yjFiObKN9Hc6EEo81mU77sIRUmXoBia2qlLN0UglXHgdqvBwgE8BmvWrAGgBHp/8MEHQf3QqVMnAMDmzZvxyCOPaPsnTJgASZL8Vl7fOWPHjoUkSbBYLPj+++8xcOBAPPjgg4iNjT2+m/s14HSb5I9Hfi0uK4GYv2A+PRkeCqJAT4aH8xfMP9NNOiY0TXO7ZMmSoCUq1d8LAHEulARC50FxR+mlLHXNmDGDqampNIebFVceg69cgK/62KvGMicnh9XV1Zy/YD6NJsWf3O60My8/j4Ig8M477+Tf/vY35ufnMzY21u+ecjmIWCiuDp4Qy3IG+N1qTnL2s7NCSjtAG3TR5RRK0wRrmqjB3x2gjWeFS9kvWAIT8gWK6mfdmr91ixIHJUGQ+tumghgAxddbhPbfC5VVOzExkVarNSj5nyAIFEUx6P9sMploMBjocrnYp08fLTNmsyR4ApicnEwAfO6551hUVESHw0G73c6cnByKosi5c+dy/PjxFEWROTk5HD16NAElMdljjz3GkSNHsri4mJdeeqnWbpI0m82MjY1lVlbWaddBThTQfch1hfyXgJdeeokFBQU0m800GAyMiYmhxWKhKIq02+3My8ujyWTyDwwCiEgoym/AJCgIAjMyMugKd4Uc1GRZZklJSfNjTXw/RVGkKIq0Wq28//77mZ6erhxz+q4pQVHwTU3OdwUMmG0p5Gq7O0g0uy666PILEd2X/djlGF+mvv/+ewJKFujbbrtN25+fn6+RBvzud78jAI4bN44AGBUVRavVypEjR/KGG26gIAhcuXIl4+LilOsngegPZZ4Ja6MNdhAjoGSuDiAciIiIaOarbTAYaDKZtIzV6nG320232619d7lclCRJU9IlSaIoijSbzUFKeVRUFHv06EEAvOiii/iXv/xFY2VLSEig1WrlM888w/Hjx1OSJGZnZ/Ozzz7T6vz888/pdrtpNBrZq1cvGgwGCoLA/fv302w202Kx6Aq5rpDrOBNYt24dzz//fG7cuJFZWVna4GAwGILe6tV9zQYms39bHVAAMDY2loIgaOlzmw5SzQJMWhiQdeoxXXTRRZezRzpakLWqzIaaSzp16qQpygDat7ohwP/SdQwvEoIg0Gw2B/WPui2KIgsLC2m1WrVjBoOBffr00YI1IyIiaLFYCCjW9LFjxzIsLIyyLFOSJKalpfGzzz7T0ttnZ2czJyeHRqORQ4YM4YcffsjIyEjNOh8dHc0vv/ySHo+HRqNRY0FxuVy6Qq4r5DrOBP7+978zLi6O2dnZ2gARERFBu91+zAOfyWRqtk8URcbGxgYNMs3O1S1KuuiiSztEtQSe6Xb8IqSJMtnSmB+kyPYIUY/zJLXnJEkzFw8o85qqzDbdf7Kvf9555xFQjFL//ve/Kcsyi4qKmJ+fz7y8PEZHR2tl0tPTmZmZyaSkJNrtdpaXl7O4uJj5+fkURZFdu3bl0aNH2dDQwClTprC0tJT5+fn8xz/+oc3hF110EfPz84OolH+NgK6Q6wr52YhAf3erw0qz3UwIoEE20Gg0Mjc3V0vgE2rAUgdok8mkLbFJkqS9yUuSxLy8PK38woULgwZLQRCCJwMBLVomgqwXbUwo7ZW2BmGNJ10XXXTpUHLKlPGO4n/egdrc7r4O8KEOlFCKcUdo76ledQ00TKnXcjgcQe3r27cvAcXFZtCgQTSbzRRFkWPHjqXZbObw4cMpiiLz8vJ4xRVX8Nlnn+WsWbNIkjU1NezSpQu3bNlyhjWJjgWcAYVcUK7bsVFaWsrly5ef6WboCIEFLy/AxN9NRFVmFbAKQA0UVhQvlMjxM0UdJkAZqtoLg+8cte0hYLPZ/FH2avlQ6ZB1tI1j/X1Od306dPxaEQPg57aLCYKAtvQHq9UKl8uFXbt2nZy2/QLgcDhQXl7e/IDKtNKEvnXo0KH4z3/+A1mWkZKSgk2bNsFkMiEmJgbl5eU4dOgQHA4HampqEBUVhcbGRiQnJ2Pjxo0wGAyora3FkiVL0L9/f1RUKBzCl156KVavXq0xyBw5cgTPPvssBg8efGpv/iyCIAgrSJaezmvqtIc6Tgh3zbxLUcbXQFFkrQD6AIg88bqjoqLaLtQS2qucqU9AAxTlupUXiLr6Oj9FlFpex/HhZCvPujJ++hGKc1lHh0QQBV5baEkZb6ItkNT4pwPrl2UZEyZMQFhYGKqqqpor4/aAbQmn7X9ksYTmtm2pb84///xjKt9eNFXGtfq8CFLGVfrAnTt3AgDCw8MRHa3w9MbExMDhcMBms0EQBNTW1oIkKioqEB0dDUEQUF1drVEHN70mSfz973/HqlWrsGrVKpSVlenKeAeArpDrOCFs/2k7sAHASCgK6kUABgIYC8ANYNSx1RcXF6dtHz169CS1shW0MraKogir1apxdtfX1QO16sHWz20Xfs1P36/53n8p+KW+BB0DRb/NZjt17Wgv2jEOnZSV8BDGCqfT6fd/9aG+vh7fffcdqqsDCOYFKLkeDFD4twOtwcRpUcqD2hOAlvrm3XffPaby7YWqVKsQBCEkH77X60VMTIzGDd61a1dcffXVAIDt27fj4MGDWtKdmJgYNDQ0IC8vDwcOHMDq1avRuXNnhIeHQ5IkXHPNNaiursbevXvR0NCAIUOG4JlnnkF9vWJV2rhxYxDHuo4zA31a1HFCSE5LBvYDSAZQ5/sEFMW1GkB8++uSJAl79uzRvnfu3PmktbMlCF7/TGCQDdrAGBsbC1mWUV9fj4iICIiiCMkg+RMISdAyfQbBDODcdl68LXeeX7IFUs+CqKMFBCa7Oeloz4xX23YRFR1CiQnUD49xRm+WSCipSYEoAK3kK1IzV1osFgwcOBAmkwmCIOCnn36C3W5HfLxvAoiFMh+oCXK8Ae0WcUwvQceCLl26hLRoq5b91hBYpsX/ZMsJPjWodaifBw4cCF2V1YqioiKEhYUhPDwcALBv3z7k5eVBlmUIgoD4+HhYLBYYjUaUl5dj3759KC8vx7hx45CRkYHq6moMHToU1113HX7/+99DFEXcd999mqHrggsuQE1NDa6//nrk5OSgpKQEeXl5uPHGG9HQcDakbP5lQ1fIdZwQ7pl5DwSTAGwH4ILyCfgV8k2hzwvMcqZCTbOroqysTNtOTEw8OQ02AAjQ8wOXWwUIMBqNEAQBkiQhMTERbrcbVVVVihXIS6AKiqIsoln2M1ESlVWCb9QdyseIESO06xgMvjRu5uBzEWp19JdqgdRxXDjRpfKQ6IAzQGNjY9uFjhfH+iJ4Ct8N2sqW25LltFUEtFeURDgcDgB+S76qMKto1tc7mtR3BMDhgO9NFFCv16t9pqSkIDExEaKoXLeyshK7d+9WCu5DaANDGJRxrqbtWwNaqCMEVN/olStXKsFyTZRiq9WK3r17NztPy7jpw+WXXw5RFNHY2Kj8FoHX90Dxtw/VxoBMmyQhSZLmXnLLLbf4iwoCvF6v1o8bNmxAeXk56uvrIYoiCgoKkJeXh/T0dKxduxbXXHMNrrvuOs09paGhAYIgYMmSJRgzZgwqKiqwb98+PPnkk7jmmmsAAFlZWfjqq68wZcoUHD16FDfeeCOWLVuG1atXY82aNVi7di2WLFmivVydbBw+fBhPP/30Kan7F4fTHUV6PKKzrHRsTLp5EgWHQJwDhbJqPIj044soDwsL07YD2UmOhzYxSNSofQlKhk7ffqvVyujoaLpcLpLk0qVLGR4eztTUVCYnJ3PChAkkSZvNxvj4+ODo/0gE03aZQUSgVe5ZLTL+bGRh0OWMis6h3wHlFGbRPNm/t5rHITCzZOB4GyQqdawEJatkwDE1c6XZbGZKSopGAZiSksKIiAhKksTk5GT269ePoigyIyNDqS9UX5nRnCHrWO6rnRS3al82pcltMVPriV7fEnr/kCFDmv2uapI8dW5wOByUZZmRkZGUZZmiKDIyMpIGg4GyLDM6OpqJiYmUJIlut5sej4crVqzggAEDmJ+fz4EDB3Lbtm1cuXIlBw0axFtvvZUlJSWUZZlPPfUUc3Nzm2XODoVhw4bx0KFDIY/NmDGDDz30EO+5554Wz3/kkUf4zDPPNMvafbYAOu2hrpCfrZi/YL6S9lcdZNUBp2voicHtdgd9b23Qi46ODhrEJElSsqEBwVywsc2vo52nDpASgrJpxsfHMyoqipGRkayrq2NCQgK7devGvn378pprrqHD4aDZbNYmolDcs0FiDNjW+dB10UWXE5CQCvOxZAMOkbUYaGXMdbW/bkmSGBERodUXmKQmPDycksGX9l08czzvanvUT9WwE/hS4nK52jxfzfZ8rNcPfAGy2+1afeqcEvibmEwmba4TRTFonjSZTDQYDExJSWFOTg4vvfRSLWumLMuUZZkmk4k5OTns0aMHw8PDmZycTKvVyvDwcGZmZlKWZRoMBprNZhYXFzMsLCyIg3zJkiXs27cvL7jgAqakpPCPf/wj58+fz65duzIvL4+bN28mSe7du5dZWVlMTEykKIr87LPPSCpK+oQJE9ivXz+mpqYyNTWVQ4YM4ejRo2k2m1lYWMgpU6a0qkfU19efWkXlGABdIdcV8rMVM2bMoCvSpVjHZ4KY4BuUzgsYoAzNJ4img6NWRmg+KIaaRMLDw5mUlNR6nYBfUZZBhEMbICMjIxkREUG3282cnBzabDb+/PPPrKurY2pqKrt06cLvvvuOoihSkiR26dKl5cnMgaC0xyGvr4suuuhyOkRCs/FWVfBUK3dycrL/uAjFcCH5ywuCQEmSaDQam42pLpcrKBOzIAiUZZkFBQVKmcEB9banvQKImIDv1mO7X1EUGRMTQwB86KGHtPbGxcUxLCyMgiAwKipKu3+73d58nmhB2lsuUGRZptVq1c4NVMIDpT1856Io0uPxaFk01f02m037jURR1L6rPOSt1SkIArOysgiAF154YdAxlfs8IyODPXr00K5pMBhoMBgYExPD6OjooPaZzWYajUZaLBZOnTpVu4YoinS5XJw1a5bWLkEQeNlll/Hhhx9mcnIyXS4X+/Tpw759+3LlypXs3r078/PzedFFF/HgwYMkyX79+vHbb78lSe7bt48ej4ckuXbtWnbt2pWFhYXMz8/nxo0bT4pOA10h1xXysxWlpaXKwGuDopSP8w2wjoBBIDl4sFInhZycHGV/XxDdoCjNEojO0AYyQRA0S0agtSI1NZW/+c1vGBcXR4fDETSodOrUqcVBL9B6LggCMzIyuGjRIv7xj3+k0Wik0Wik1Wql0WgMslKFsnDooosuupzV0oqrSODYKUkSTSYTTSYTnU6nlqxHlmWWlJTw8OHDdLp8ydeiQSS0XneQSAg2XLSyEqC2KSIiggaDQRvLi4uLCSgrn6qyGXh/qrVeTSnfdF44GUncWqojcD5qT18HSufOnbX50uFwaNdQXW4EQaDBYAh5bZfLxdzc3GbXUftMkiSWlJRobZdlmSNGjNCS9KWkpDAtLY0A6PF4tHMDXwweeOABOp1OWq1WyrLMXr16EQC7devGxMRE5uTk0O120+128+jRo1y8eDEB8LHHHuOcOXNos9k095f8/HwuXbqUJDl9+nTedtttJFtWyG+55RbOnz+fJFlbW8uqqqqTotNAV8h1hfxsQFlZGcPCwrS007Is+60qUvsGrRYnhYiWB6lAhTvQMqMONuo+deC65JJL+MQTT1AQBO3NXZIkbelOHYxUy4q63dQaot1jG4Nmi/felqVH9yfXRZdfppymZ7uZNVRdqWvFOt2i1bdJm1NSUuh0Onnrrbdq++x2OwsLCzUXCNXqOWTIEKWMAcSfQNhPw70GiGooMZvNwUYTi+++WvDtblp3qDFeEATN0t70nMA2ZWZman3rdru1bdl07JlGA+e0tvzdi4qKgrJ6qmIwGELuD5S//e1vQeXHjBnTal9ERUVRFEXNxz2wbZIkaRmx4+Pjef311we9HAX+NkuWLOGcOXPYu3dv3nbbbTx8+DCTkpI0XWPz5s0sLi4m2bJCvmDBAubk5PD+++8/adZx8swo5B0wxl5HR8eHH36II0eO4P3330dDQ4PCZaoSBnSHQg3Yz/ddgD8y3bctiiKMRiNcLhckSfInACKUiP4iX3mFIAButxsAUFVVFUSFKIoiRFHEoUOHlOp9EfQq3+y7776Lu+++GyTh9XphNBrR2NiIhoYGpKcrXF4OhwOSJMFgMKCxsRFer9fPPOAjRCGp8bWq30OiJXKIqhb2axW2cVyHDh1nJ0I826eCLUdl6dCgjkWtjC0tjmNNdm/duhV1dXX44osvAAAJCQmoqKjA999/j4aGBtTV1cHr9YIk/vvf/yonNQD4KxTO8ZOMpveqMtFIkqSN07W1tfCqlDpm+OlqVRpeg29/ExYdte6mfWMwGEASR44cCdmewDb9+OOP2hxy4MABbbu+9tgzyantqKysbJOWcNWqVaitbc7Z2dDQEHJ/IB566KGg8kuWLAnqi8D/rCRJOHLkCLxeL+rr6+H1eoOOe71eLRHR7t278cEHH2Dz5s3a8by8PHz44YfweDzIy8sDoLD/NL2/rVu34u2339a+GwwGrU01NX5anjFjxuDtt9+GxWLB8OHD8fHHH7d6rx0ZukKu45jx+uuvQxAE/P73v0dJSYmyUx3wSgDI8FNlqe/FAERBRGFBIZKTk+H1eiGKIgwGA6ZNm6ZkUROgDJBlAMIBlCuDbU1NDYxGI+Li4pCdnQ1BEBAbGwuv14u6ujotE1zTwdRoNGo8wSkpKRoXq8Fg0CjHVGVfzYYWBJ2WVYcOHaHgbmF/O/IEtagIN8EJKe6qvtJKFc2U+FYgiiJWrVoFIHisVGn7jqfO44XKCa5SyQZSL6p9SxJ1NXX+kw5CmYd+8n1vgNJHLRhRTCYTDAYDSktLERcXp9ELdu7cGdOmTUNYWBj69OkDo9EIp9OJxMREGAwGFBcXa3PKiaAtSsyTjZ9/Dk7PqtFV+hD4n21sbAwyUAHBPO2BCjxJnHvuuUE0xykpKbjuuutQWVmJkpISvPHGG/jmm2+wcOFC7N69G+Hh4fj000+xdetWPPPMM+jXr5923ooVKwAAr776qlbfli1b0KlTJ/zmN7/ByJEjsXr16hPpijOL022SPx7RXVbOLAadN4gw+VRrExjhjqDFYtGOC5JAeHzLo7dBoQOUfUuEcaAcpizVeTwejhgxgnPmzGFSUhInT56sLUPFxMRozCkAKJpEWiwWzp8/n6mpqSwpKWFKSgpHjhxJs9nMnj17KmUL0Cxg0mg0Bi2zybJMl8sVFM0eyJZyPAE7upwForsC6dIR5BT9D9sTDNiaBLrhAS1TALbmKnFMPtctBbw3kcDxuLV7HDBggNY2l8vF+Pj44Htq2u+egP3JIC6EMncEuLHIRpndunXj2rVrKcsyV69ezZdffpl9+/ZlRkYGBw8ezNdff50HDhxgQ0MDY2Nj6XA4mJ2dzQsvvJAzZ84kySAWMQCKO6O57Xv/pUhQsLBPmroaiaLIpKQkpqamMjExkU6nk3a7nQ8++CC7d+9Oi8VCg8HAvLw8PvLII1y/fj3z8/NZVFTEu+66S3NZue+++5iTk8PCwkIOGTKEBw4cOCl6D3Qfcl0h72gYdN4gxQ9wPIjpvk+fArx+/XqSZGpqqj+i/1Yo1Fm+wdDqtPLiSy4mAEZGRvLhhx9mVlYWBw4cyMmTJ7Nbt25ctmwZY2Ji2K9fP0ZHR7Nbt25cuHAh+/Tpo/iaqQOrABYWFioPtsF3nagQA68uuuiiy0mQjIyME69HQDDDVOKJ1ddu7uyTyJEuCMIJvwCcbDlmQ4rUZDsGzWgee/bsycmTJ2tc4F26dGF2djaTkpIYGxtLURQZFRXFiIgILly4kK+88gojIiJoNptpsVgYERFBT4anfe2JCv5NU1JSznifBrZH7V+j0Rjkgy6KirFM5Z9XfwtRFJVcHVDmerWsGhzar18/WiwWfvjhh+zXrx8HDBhAj8fDSy+9lJdeeim/+OIL9ujRg2lpaSTZLq70UwnoPuQ6Oho+WvYRMApAKhR3klQAFyjHiouLYbFYsHPnTsTHxitZ8P4O4Ai0ZaaEmATkZOdAEATU1dVh9uzZqK6uxrp167Bw4cIgfztAScE8ZswY/PWvf8XXX3+t+Iq5ACQAsAHfr/9eKdgAJRPoPmi+3h0apyDJ4lmDwDiCptBHIB0dGJs2tZBq+FjRAH+W0P/5Plv577fmrtLQ0ACDwaC5CTTL5qk+b8foPeJyuYKuHROjpKHs2rUrDAYD8vPzj63CdqDFlPQBUFPYq1mO1XOa+i7Lsqz1hbrfbAlIiRzontII4GcomZX7+6+zZs0avPzyyxAEAXFxcXj66adx6NAh7N69G3V1dRAEAQkJCRg0aBCuvfZajB49Grfccguqq6tBEAcPHsS2LdtC3GiIG9vn32xoaMDWHVsDbrrNbjmpCMxYDShuKWp/19XVaf7dqotQdXU1BEHQ3HNMJhO8Xi92794NSZJw4MAByLIMh8MBo9EIh8OBzz77DDU1NZg4cSJWrFihucVIkoSLLroIsizDZDI1c5/5NUGfDnW0jloAyU325Sgf1dXVqK6uRm1tLXb+b6f/Tc9LnD/ifPzpT3/Cxo0b8de//hVWqxVHjhzBhg0bsHXrVuzevRsHDhzA4cOH0b9/f+zZswdLly7F1q1bcdttt2Ht2rVolBoBJ4BbAdwAYAqAKwGv7FWUcPXf6whu3sSJE9t/f9a2i7QKc9tFACi2hdOEY063faqh2lZC4dS7nOrQcWZxHP99xUAXAr5Hu7GxUTNkNPPbVp83L4JSuLcFNWBRVXhVxejbb79FfX39KfPNDRyvVCUwEGpAvnqf6n2r842K6OhobVvd31DvDwTSrqNeTgDgAfCt76sgoLq6GqNHj4bdbsf27dtx0003ITMzE16vF++//z7i4uJw4MABfPbZZ3j55ZdhtVoxa9Ys5Obmoqba57h/FYAwBBuKWvBVD3rxUsvIPjmNaPp/IxUiBJVwgSTsdjsA5f8hyzIkSUJNTQ1EUdSIFKKiokBSI0mora1FdXU1jh49qv1+Bw4cQEFBAdavX69dL9BnvsX//q8AHWzm1tHhYAKwvcm+7fCzqoTAxRdfjHnz5uG222477svOmzcP3hovcB78/9I1AN6D8pJgApAPoDeAAf7zikuKsX///vZfqC0GlKZoariqCVnqtMPh9L+VqAPfqWBzaDeajiz6SKPjTCDwf9fSStoxKK1nHD7dW5bbqbGFCDIN9cJusVg0ZTgsLOx4W9cmml67sbExSAFriUkkOjpaG9cClfZAC7vFYkFsbGzQtdRz0tPT/S8u6vsLAawHoMT9Izw8XCMbsFqtcLlc2L17N2JiYhAZGYnS0lJIkgSz2Yzzzz8fkiRhwIABcLlcuO+++5S5IQzKKrILof9XAbefmZkJg2zwkxloNwX/73YKxk11XkhPT9eUbO3SkgRBEDQGs4qKCoiiCJvNBlmWQRKNjY0YMmQIJElCRUUFoqOjNaIEt9sNSZJgt9shyzI8Hg9cLhfmzp2rEDcAuOSSS/DRRx/h+++/b7WdDodDY2v5tUCfJk8Rtm7dCqu1ufnV4XDg888/B4BmD0NbGD16NK6//noAwL333gsA+Mc//oF58+adYGtbxqC+g4DXoDCfNPo+X/PtbwFvvPEGVq9ejcjISG1fQUHBMV336quvhmAUgpXxjwAMBxAF4FIApVAG1ICg9pUrV+L1118/pmsdEwJf3gVo7RNEZZBTJ4tmk15Lk35bOrPKXiO3Xrb8aPOBq1VLw0l+8psp/02JBnRLuI4TQLP/V1PlOuD5stvtfqUt8H/XEmtSXQv7OzDq6trZ6EPNd4ViQqmurtaYMw4cOBB88CSOFS6XC2FhYUG/JwMGVYfDoc0bge4nWVlZABSFsX///tr5PXr00M5955130LVrV+2Y1+vVqA+Tk5ORnJyM/v37Iz4+XjsHAmCz2WAwGGCxWEASVqsV+/Yp/iQk8fnnnwe9AKnuMz169MCKFSv8LxFE8Bgd2M3qeOgALFYLjEYjZFlGfV29ct7tAWUFKAq9G0AGgPFQDFAGBP3vJUlCly5dkJycDIPB0Oy+Bg4cCJfLBZPJBJvNpinZmZmZuOqqq7BlyxbU19fDYrGgV69euPHGGzFixAi89dZbyMjIgCzLGDZsGK644gocPXoUH374IRwOB0RRxMaNGzFo0CBUV1fjvvvuw8SJE1FTU4N169YhPj4e8fHxWLNmDaZOnYqxY8di7NixWLt2LQYPHowVK1agpKQEb731FrZu3QqzWVlmLi0txdKlS7VbKCgogCRJKCwsxKOPPopfBU630/rxyNkY1FlWVsbc3Fzt+7Bhw3jo0CF6PB7u27ePJGm1WoPOeeONN/jDDz80q6u+vr7ZPpvNdpJb3DKCWFYkcNB5g07LdePj45UI+PG+AJjxIGb62nGXb/siEG40T4BhCA7ikSSJ4eHhzQNYhCbnNDmuZnRryswSSmRZ5r333stRo0b5k1Ko9V8ChYGmaTuNLdfXmhhkQ5upkc+otJJlTxdd2hTTGb6+LscmaibMEIGfagIZIEQgpg1akKuaXEg99sILL2jbWVlZQeOdul1UVES73U4AHDFiBDMzM4Pb5FS2o6OjKUmSFqAoSRLT0tI4YMAArU0mk4miKLJ79+7Mzs4moCSjKyoqYkJCAmtqapiUlMTevXszOzubQ4YMYVxcXFAKeshQ5qVYXxuSfO3oB8ICGmwGFhQWMCkpiZs2bVICQF3wz2Nqn4yBMucVQiFVMEMhVfDd28nCsmXLWFBQwPz8fJ5zzjnctGkTy8vLSZJer5eTJk3iI488EnROoA7TFD/88ANTU1N5++23n7Q2nilAZ1n55Sjkn376KQE/LZQkSTz//PMJKJRARUVFWlSymopWFEWazWYOGTKEkiQFDV7p6ek0m8202+30eDxBA5jD4eDUqVMZFxdHg8FAg8FAp9PJ3/zmNySVl4MBAwYwPz+fAwcO5LZt2477vo73RcBms/HTTz/Vsnuqg/e9996rfXc6nTSZTJRlOYgCURMJwRlBVQW6JZaVJvvVDJ4tShsZNa1Wa4s0X126dGm9bgOITk3a1EZq6NYU7nYxLbSSpe+knhPQZq2OEC8aCQkJx133GROdwUeXX5CcdIrXwDEZYGpqqjaPAeBzzz1Hl8vF7t27B2dBvgTKOOEEcQu0sfyaa66hIAiMiopiXV0dzz33XGZlZXHOnDkcOnQo3W43N23axGeffZYGg4GPPfYYy8rKaLPZ+N5771GySkQ2FIrDmSB+47uOoBhMYmNjCSjbNpuN1157Lbt3787Y2FhGRETQarUyLCyMHo+HCQkJ/Oqrr9ijRw+aTCaWlJTwp59+4rhx43jBBRcwIyODRqORiYmJJEmL02ewEUHEQzEmBbDrCILAuLg4fvvtt/zoo49YWlrKpKQk5bgVRH5AvwoB/SuDGAXFIGUGJat03PN3e/DII4+wsLCQ2dnZHDNmDCsrK0/p9ToqoCvkvxyF/JVXXiGg8KOq1oHU1FTtgevUqZP2kPbo0YMulytooOvbty+zs7M1RV2tIy8vT6PistlstNls2iCXlZWlpaWNiIhgdHQ0ly1bxvPPP5+33347i4uLmZiYyMjISJJkRUUFJ0yYwK5du7KoqIhvvvkmSXLOnDm8+OKLOWTIEKanp/OOO+4gSf7xj3+kKIosLCzkmDFjSJIvvfQSu3btysLCQk6cOJENDQ0h+8Nms/Gll14iAF588cVsbGykyWSix+PRrNA9evRgz5496Xa7ed555zE7O1ur2+bwKdPnQBlw1b6ahNBKpAPEaN92kvJZXFxMSZIYFhbW7glHVTItFov2khRqUgu07Khpg9uUNpS9kpKSVo93aAu5Lrro0uGkRYU8cAyVjr3enj17BqVEt9vtlGWZ999/P4uLi7UxXuOnvtV3zTugKM6RSj1RUVEcP348LRYLu3btyltuuUVTyDdu3KjNcZGRkbRaraypqWFZWRlTU1P9FJU9oCjkg6GMsSoFr9XKiRMnEgDDwsKYkpLChQsXctSoUVq69k8//ZRRUVEElPH1nXfeYUNDA4cOHcqwsDDabDZmZGSwrKyMnTt3Zm5uLteuXcvS0lI6nA7l+uq4blT6UhAE2u125ubm8oYbbmBDQwPHjx/P3Nxc5uXlcexVY+mOdfsNTIGGGgeIi6Eo43al7sFDBp9q9UUHdYX8F6mQp6SkaMp3oAL19NNPa98DlTlV+Y6JiWFYWBhlWabBYNDKqG/qADhw4EDeddddmivFK6+8olmg7XY7r7vuOh49epTh4eFMTEzkli1bWFdXx/DwcJLktGnT+NJLL5EkDx06xIyMDFZUVHDOnDlMTU3l4cOHWV1dzeTkZG7fvp1ksIV83bp1PP/881lXV0eSnDRpEufOnRuyP2w2G2+77TYCyqpBTEyMtnSoDuKiKNLpdFIQBIaHh2sDvCiKdIb5lFwTgpezBTRXyAXfAD/eN7m4wLCIMG1loT3W5XZz/f4S5DgmYF100SWEdPBnqV0W8lLfZ6DBwAxt3BUEQUt8k52dzVtvvZXXXHMNJUnihRdeSAD85ptv+PjjjzM6OpojR44Mmgskq+R3P1RlpHK91lwdKisr6fV6SZILFy5kUVER161bR6PRSJK87rrrKJpFoguI3vBzjEeAokVZjVbvPykpiYBi9BgxYgSTkpJ4wQUXUJIkdunSRZt7ZFnmI488wv79+9NoNDIvL49Op5Ovv/46k5OTKQgCb7nlFo4cOZIA+LeH/8bE1EQCYERUBPv268uwsDAWFBRw4MCBtFqtXL58Obt06cLCwkIWFhYyLy+PR48e5ZIlS9ivXz+OGjWKgiAoLxgBCfkggZmZmVy8ePFxaCQ6jhXQecjPTix4eQEiEyIVLk+TAMEg4OprrgYAjBkzRisXGExTUVEBr9cLh8OB7OxsGAwGZGZmonPnzujRowcOHz6Mo0ePwmw2IycnBy6XC4IgBKWsHT16NA4ePKhwdfuuFRMTo6Wkf/PNN+FwONDQ0IA+ffogNTUVgD9Y5oMPPsD999+PoqIi9O/fHzU1Ndi+XaFUGTRoEMLCwrTrb9vWnFv1o48+wooVK9C1a1cUFRXho48+wpYtW7Q+MVlNEMwKv2llZSVee/01iKKIsrIyzJo1CwaDAaIoora2FkajEdnZ2dp9jhs3DrIsw2KxwGKxID9P4cAV60WFZUXFn9CcupBQIufnQglEPQxUV1Rr99ZSJH9goFFLZU4VjMbgqM+goNC4U3zxFii5dOjQcYz4JTxLy32fDNhXA23cJYlDhw4BAA4fPowZM2Zg6dKlIIkPP/wQ0dHRSElJwW9+8xsUFxdrQaIzZ87E3/72N9ww/gbgFQCPwU8U8CEwadIkPPzwwy02a8WKFSgqKkJBQQGefvppGI1GbNq0SZsTn3/+edx47Y3AOiiSDiUg9bAyb1gsFm1cVefi8vJybNmyBSSxbds2NDY24siRI3C73SgtLUX37t3xxhtv4JNPPkFDQwO2bNkCSZLw6KOP4siRIyCJ999/H2+99RZEUcSMu2dg3659EAQBL734Ev5wxx9w5MgRFBYWoqamBtXV1bjmmmuwatUqpKenw2KxQJIk9OjRA//+97+xcuVKPPbYY7BarSgvL0dKXAoGDhyIvTv2gg0MCmJdsWIF+vXrhy5dumDIkCHN0t3rOAtxut8Ajkc6soV8/oL5NLgMwZksw/zWhejoaM6ePZsAtOATAPzzn/+sZT8rKChQ3qgjIpiVlcWYmBgC0LJ/lZSUsEePHgQUK3Lnzp0JgKNGjeJNN91Eg0EJ8LPb7fzmm2+YkZFBj8dDQRA4d+5cdu3alT179iSpuKNcdNFFJMmSkhJu2LCh2T3NmTOHkydP1r6PGDGCS5YsIRlsIX/iiSc4derUkH0CGcEZPg3Q/IrffvttkgzqD1XUVQPVat40SCg2NrbF7HmtZZLraFnmTorogW8dS9rrb34SMyjqosvxiBrgqI3B7YgdUWOWnE5ns9TogwYNosViodlspsOhuG4IgkCj0cgBAwawsLBQW72VZbmZ22BCQgK7du2qzSFlZWXMy8sjSf75z39maWmp5vLh9Xq5ePFizWUTUFxQWhrjVXfP4+mjltwCVRecUH105ZVXElCyVYZakZAkid26daPRaGRkZCTj4+NZVVXFlJQU5ufns7i4mACYm5vLxMRERkVFsbi4mCQ5fvx4mkwmXnLJJZRlmSUlJdyzZw8XLVrECRMmnIgqo6MJoLusnH0KuTve3Xz5bTyUgA7fA6gOFIEP51133UVJkpoNFkajMcgHWVXaO3XqpNXjcrmYlpYWdNxqtWrpaiVJYkREBDt16sTbbruNK1asoMlkYmZmJgcOHMjvv/+epOKyMnnyZG0Z8LvvviPZukLucrk0F5UffviB6enp/Pnnn0mSBw4c4NatW5XIcROC+0UGMTR4YFfvUYtQDzG4qdsRERFtDqBhYWGMjo4+oUH4rJKoYz/npAd16fLLFj2o9dj763j7LNB3uDXXl4jQ+0PGxoQoazAYmJWV1e5YGpWhSo2fUcdug8HACRMmaMagwHgbt9vN5cuXa+N24Hg8bNgwzp07l4BisGpsbOSIESOYnp7OLVu2kCTvv/9+zpo1S5tXVFx11VWaQadfv36asctsM/v7rSS4/eqcoLa56f01nStkWdYUbtU4BihzUJ8+fQiA06ZN087r1KmTFiwqiiKNRqNGttC9e3caDAbOnTuXv/nNb7Q+GjNmDFevXs3u3bsTUEgbbDYbIyMjNddOm83GJ554gj/++KNmXBs/fjwB8O9//zvtdjtNJhNjY2OZl5fH884771SpOb9KQFfIzz6FHIBiAZ4ZIAH0RBAUhdNisbBXr15csmQJJ0+erD30nTt35vjx41lWVqa9UdtsNi5btozfffcdzWYzXS4XR44cyYMHD9Jms7Ffv3785ptvOGXKFObm5jI9PZ1ZWVksKChgfHw84+PjWVRUxD59+mgD3HvvvceioiIWFBTw3HPPJUlWVVVx4sSJzMvLY05ODkeMGEGydYX8D3/4A7OysrSgzkWLFrGwsJD5+fksKSnhl19+SUEUQvfLrUqfkIoFRGXdIMmHHnqIBoOBVquVDoeDl19+OdPT02kwGBgREUG73a4xyKj++TfddBMTExP9dIbtnQhPgEnkVyXmDtAGXXTRpcNKayuPLcXhWCyWIBIDQRBoNptpMBgYFxfH/Px8JiUlMSkpiXv37uWrr77Kbt26MS8vj/Hx8bzvvvtIKgr5tdddq9RjhsJUIqAZe1WgYedYDRJNjWPqPeXn5x9T3zQN9FfjpSwWCx0OB1NSUjTj2rXXKvfUuXNnPv744+zZsycBJeB12LBhBMC//OUvjImJ4dVXX81zzjmHHo+HRUVFdLvdXLZsGUeNGnXqlZ9fOKAr5GehQm5EaAu5E5qC2BpCUQyd7bRDIS3kM6FQNwVEoKsUgqRfIbfZbIyPj+fo0aM115zLLruM5513HouLi2kwGLhkyRIKgsCbbrqJNpuNguQb+DqByEOwYh7mG6Dbo6zrirouuuiiy3FLoPKrrtg2FaPRyH79+hFQVnuXLFmiMZgYDAYOGzaMJpOJQ4YM4eeff05ZlpmXl8fBgwcr7p0t0d06QMSAiEaLY3l6errWhqb71P2B7i9aPglAo+RV7019qTAYDEGWdLUPTCZTkGuLGggLKFZ1m81GWZaD2hIdHa0p4EajkWazmTExMRRFkWPHjtVWNZ577jk6HA6KosjBgwczLi6OgwYNYm5uLl0ul8aMRpLvv/8+e/ToweLiYl566aUaz7iO1gFdIT/7FHJ3rDvYV3o8FCXQAsIGuuPdZ7qJpx0hfcjH+77H+b5f6h8kv/jiC86aNYuiKDIyMpIZGRns1KmT5qYTHh7Ovn370mw202Qy8dtvv6Uoipw4caIy+BmgSDaIHN9gHBdisG5pIrEEDO6tKeW6wq6LLroAWtwPgJCxMEFiabs+dWUUgLbi90tyLwvMq6BS+arHVCVXlmUKgkBJkmgwGDhjxgx27dqVkZGRrKqq4oCBA5RzCqHMIYnw5zvwsWnBBL/C3g/KXBzQjkDXGvWz6YtDoCX80ksv1fYH8qi7XK6geKfExMSQ932sbF2iKIbOwQGE7LNQoh7r2bMnO3fuTEmSGBMTw6KiIl5xxRWMjIxk37596XA46Ha72alTpyAFXocC6Ar52aeQz18wn6JF9AfYGaFk2rKAkk3i/AXzz3QTzwjmL5hPo8UYTNsUh2C3Ht/AecUVV9BsNlMQBKanpzMuLo5XXHGFFqyq+i5aLBZ6PJ4ghVwbiExQfKp9SSAQ7tuvflfL5YcexJoOfLIsNw/oCUWxqEvICUMXXX4R0sFpDM+UyLLcolLY1I1FVbID9wXmWEhKSqLBYOCTTz6pZdoMDw/nqlWr6HA4GB0d7bcim6Eo2jNBXA7FyBN4/aaGFwnKvOBq+V5ae/FRAzsD70kQBE6cOJHZ2dk0GAyapTrwvKbntCTnn3++5nMOKHEAqjuMOr+pdQ8aNEgLmL366qu1PlSP9+nTh6IoMi8vT3MBuuSSS7TspB6PR4sB+POf/8zU1FRedNFFfOGFF4KojXUogE57ePZh7JixmPf8PLjD3YARQB2AesAd5sbc2XMxdszYM93EM4KxY8aitqoWrCE8GR6FMuv6gAISgEHK5s6dOzF69GjcfffdSEhIwOWXX44ffvgBmZmZWLJkCYqKinDdddfhL3/5C7Zu3YrS0lJkZ2dj2rRpSt0WKAIA5wGYAeBC3/coAA7f9SwA1gMIh0KVaAFgBiAEtElUKLHqG+oVaqy+vmMClCHUz1ypowkCaT116PhF4HhoDA2tH5YkqX31tLPYmUB9fb1GNxhIFwtAsfQ1+d7YGNyRq1at0s7dvXs3rFYrZs6cCVEUIQgCnE4nampqkJaWBpfL5R9bbgBQD+ApAEuhjN8qBAAVvm0rgLt9nxUAPFB+FwlADIBcKPM1FIrb5557DgCQkJAAg8GAoqIiPPjgg4iLUzhnJ0yYgNzcXHTq1AlGoxEfffQRBEGAzWZDdU21QmHscmhNiY+PhyzLQfesfjcajUhJSYEoitixYwf69++v9WFjYyPKysoAAEuWLIEgCNp5l1xyCSorKyGKIj788EOsX78eADSa4x9//BEAYDAY4PV6ER0djQ0bNqB379644447cPvtt6O4uBhJSUlITk7GoEGD0KNHD+zcubNFamMdpxe6Qn4SMHbMWOzfvR+s9b3p1BL7d+//1SrjTbH9p+2ACcD2JgdiAZiAZcuW4cUXX8TMmTMBAFdddRXy8vJgMBgwceJEjBo1Cs8++yymTJminbp27VqkpKTgnpn3KAMWAQwH0A0Kr+3rAGwAfvZxlzcCIkSgAUAlIDQIQDWUQfpqANMBXAVlALfBr+Rv8V3Q49tvC2j/8T49J2OitZ+EOnTo0HHy0UYKg6bKacsFT7wpJwMtvUConN4k0bt3b21/UwUdUJTEls5tbGxEeXk59u/fj4aGBpDErl27MGXKFBw8eBCbNm1SclZYROBLAHsATIIybu8HIACiSQToq1cCEAbF+OKFosBvADASQITv3MugzBcAsrOzcc89yjySmJgIkti4cSMmT56MPn36wOv1Yu7cuTh48CB27dqF3NxcXHfddVi/fj2OHDmi5awoLy/X7m/Pnj1B/WGz2RAbGwtAeQHYuXMnvF4vjhw5gm+//VZTuquqqmCxKNalkSNHIiEhATabMunMmjVLs6Tecsst6Ny5M0RRxAMPPACSqKhQ3kZURV0QBNjtdqxduxYHDx4EEPzfM5lMkCQJDQ0N2qeOMwtdIddxypGclqwMhK9BUZbVZBCvAYP6Dgp5zssvv4xVq1Zhw4YNmDZtWot1jx0zFnOemQO3xQ0sBDALEF8RgQrAE+/B/AXz0VjVqAz8vk/WEt5ar2JdHwUgFcogngrluwglEYYI4H++C20DUAUl4ZAKtenN55/WcZwTbdCkVtlyuQ6HNiyGOnT8omAOvdtuV96imyYBkyQJdrvdr8jaEHpmbuFFXpRamMaPdVxqAS2tfDkcDgiCAIPBEGRdjYiI0JRKFU2VvcDvUVFRCAsLAwBs3rwZgGKB/+abbxATEwOSqKmpgc1gA1YC2AvgHgBP+iowALf/5naIoojsrGzYzDZgN4C1UMbxBigrtP8H4BCUY0sAKAZllJeXQxAElJSU4KuvvsKLL74IURRRXFyM3bt34+abb0ZtbS0GDx4Mg8GAgwcP4s233lRWAkQohpsoKJZ3H1JTUzFx4kQIggBRFPHJJ59oKwdqQjybzYauXbuivLwcJCEIAiIjI9G5c2cYDAbMnz8fe/bsQU1NDQwGA7p06QKDwQBBELBx40btnJiYGDgcDvTs2RM2mw0OhwOiKOLQoUMoLCxEYWEhXnjhBTz88MNYs2ZNsxUMHR0HukKu45Tjnpn3wNpoVVxHFgGYpXzmpOTgww8+DCq7dOlSlJaWHlP9TVcoGqsaQS+xdePWoFWKa665BhMnTsS8efMA+Cz3yU0qS4ayxHk3gNuVXbLRZ4F3QVHYh0F5cuKhTJIm37kXACiA332mJSQCKG7fvQVap4ImtaZjqgkdF7rhRUcotKC4nvWoCb1btWA2dWVobGxEdXW1X1GqRGjXuBZe5L2NLbiKtUfvCtAArFar9rIQ2EZ1X1Mrt+pOYjKZUFNTA0FU3gAOHjqIzMzM0JfzuaSo2wCwd+9eHD58GJIkaX0gCALq6uqwZ88eTbmvqalR+qACyqevn8OsYXj+uedBEgaDAZ06dVIObIC/z4wAOkN5SVkF4FMAu5RDq1atwvbt20ES69evx7/+9S8UFhZi8eLFMBgM2LBhA1atWoWdO3eie/fumDNnDpavXq6MuQYo43kCgHFQsoMC+Omnn7Bw4UKMHTsWjY2NKC0txc8//4yYmBgUFxcjPT0dNTU1+Oijj2Cz2SDLMlwuF958801s2rQJZrMZBw4cQENDA2w2G6Kjo/HNN9/AbDbD6/Vi3rx5WLt2LYxGIyZOnIjKykosW7YM5eXluOKKK5CWlgaPx4MPPvgAy5cvh8vlQkREBObNmwen09naP0LHmcTpdlo/HunIQZ062of5C+bTk+GhIAr0ZHjOSLDr+PHjuXjxYu27J8MTmrIyyr8tmkWGR4QrgbtqUFESlOBOF4h4+AO/zgGR7ttWA0mbUi4KvvIqE8Cxcn3/2hO16EF2upxEaRqsd6bb01FEzRERGGwYeDwpKYmXX355cDCjse16T2cfm0wmjeWkuLjYP+YGjCFGo5EkGR0dzdGjR/Pvf/87o6KiKMuyxoeemJjI/fv3Mzk5mbIs84EHHvBfJw1EKhQ2l2nQcpDoOPuBMxDUKfAsWL4oLS3l8uXLz3QzdHRAVFZWonfv3toSXu/evdGnTx/MmjULERER8Hq9SE9Px9tvv40//vGPqKqqQrdu3fDkk0+ivKJc8a0ToVjvDQAOQgn4qQawWbFQNzY2+i1JbgBHoATvGqEMyyKUJVEVagBo0+1QMEC3IOvQ8QuHzWZDZWXH8TNTgxO3b9+OqqoqSJIEWZYhyzLKy8thMBjQ0NAAo9GIuro6yLKsBXGeTRAEASSV4Mvq6iD3GzV41Gw24+eff4bZbIbBYIDVasWiRYswYMAAkMRVV12FFStWICMjA1u2bEF4eDg+//JzZVx3+SqLBnAAwDBAfFVEY1UHCQDQcdwQBGEFyWNbrj9B6C4rOs5qPP/88zh48CCOHDmC6upqeDweREdHw+v14oUXXsDw4cORkJCgRdEHwh3hRv9+/WE2m4HD0AKEsAYQtiu+f41sVPZdA2WJMh7AAF8FasCQqlCboTxRgQp4U2XchGDfTl0Z1/FrhRXBz8JJ8nk+5Rh+7KecsDJuQ8j+aRpwGcjKERI+IhA1gLK6ulo7VFNTowUmqvWqbnKqe0lTBAVwdoDfryXGl3POOaeZL7woirBYLKioqIDFYsFll12GqqoqHDhwQOuHnTt34vDhw/B6vSAJu92OW265RWFVUw0xF0JxZawE8Dpw44QbT/l96vhlQlfIdZzVOHDgAHbt2oWEhASkp6fjs88+w9KlSyEIAu688058/PHHMJvN2Lp1a7Nz3W43pk+fjqeeeArXX389kpKScGj/IciyjMxOmcoA7oWiVC+C4rP4PTTfQ80/MdBPEWg2MamTmyRJygDO5mVUJCc3dWrXoeMXiiq0/vLaUfHesZ/SbqrDllCJ4FU3H+Lj4+F2u5XdPmtwQ0MDZFkOvqYTymxfoZx/9OhRbN68WVNYmzK/qMprfn4+AARZxyVJ0hRfkykgeKUD0DS2tOKvUgiqFnFAuedDhw6huroaNTU1ePfddxEeHo7OnTtj7NixKCgoQI8ePSDLMhYvXowtW7bgm2++wd13342XXnoJE6+fqPyH5wG4F0A9MGn8JDz91NOn7X51/LKgK+Q6zmpERUXhlltuwSOPPIKEhASMHTsWn3zyCcxmM9asWYMbbrgBDQ0NISmd1AAjURS1iUUQBAiCgDlz5igR7QZBeUrSoXyKUKL0Af8EJCkBT0K54Ffg4T+mTnZBk55aJjq4Tdu3N+WG7IAIfJnoAJOwDh0dHe2mOgxAZGRk850CNCs3AOzbt08LFg20DtfX12vXFCUROArAC8RE+6hAukFhvgpEgDagKuS7du8K+q7ei6r41tQERLB2oNU+q9Ua9F0NDLXb7UG/RefOnVFQUACLxYIePXpg+vTpGDx4MGJiYrB69WqUlpbitttuQ2FhIdatW4fi4mK8/PLLGDZsGJ79x7OK76/X5wNcT10Z13FC0BVyHWcVbp58Mww2hfrJYDPgw48+xL///W8MHjwYd9xxB7766it4vV4IgoCKigq8+uqrrdb3r3/9CwCwe/duhIWFISwsDBaLBc899xxEUURCbIKiZG+EYgn3QnFtAbQJKCsjCwBAkcqEGaiwtjQPq4rs3mPtgQ6AwBcO3VVSh44WEdIy3ga7jNPphMFgwP79+5sfJICjCguK3W6HLMuastypUyekpKRAlmUYjUaNTcMoG+F2u+FyufDzzz8rcSvrlXqCEODRoSqt+/buU3YYcXJwCjWOwBeSQFccQFkRoI/fPDChUXx8PGprayGKIjZt2oR//vOf2LRpE7Zt24YuXbrg6NGmnaRDx6mDrpDrOGtw8+Sb8cy8Z9B4WSMwHWi8rBFvv/829u7dC4/Hg8svvxxbtmzB+eefj+rqagwZMgRdu3ZttU6z2Yy7774bS5cuxWWXX4bIuEgcOXIEL8x5AXV1ddi5c6dSMMAQpPpoxsXFwWazobCwULHAq/OBmrRHRnMO7qZlOoDfpY4zi5b8c3Wc/WhsbMSjjz7q3yFAiTsJgN1uR1FREZKTkyFJEr766quQdcmyrAUd9urVC7GxsaioqIDNZkNycjI2b96M0tJSZGZmQpIkzJo1CwUFBSgoKIDZbMbho4eVitIB/KaFBvvcz9VkNMj3tTkKJ2esIk5oVU1NZtMWrFZrkIIeeI7Kn04SK1aswJEjR1BRUYE9e/Zg+/btaGhoQEJCAubOnYsdO3bglVdeAQCsW7cOa9asOf7G69DRFk43rcvxiE57qIMkJasUkqZQskrHVV+/fv347bffklRoGWWnTFh8FFbjoVAWjldorCRJ4m9+8xtOmDCB8fHxTEhI4COPPEJJkrhv3z6FAisaRDgUKkTZVw/850Ol3BJ99YYFUHC1QOdndViD99nPPCWaLrro0oKEeD61Zx9o9pwLgkBZlmm1WhkREUFBEOjxeNq8TmCdMTExtNlsBECHw0Gj0agdM5vNfqpBMaAOC5QxCgimUvXRsAbRGQogrE3akOZvP0QQRb46m14HoMPlULab1gEoY6Z6DYN/v0pXqN6rIAiUJIlGozG4P0NI4P3Lstz8N2jS/4HXiYqK4quvvkqTycTCwkIKgqD1aVhYGI1GI81mM9PT09mtWzdu2LDhmOcdm812TPufeeYZzp07lyQ5Z84c7ty5M+h4WVkZc3NzSZLffvstb731VpLNaX6PBcOGDeOhQ4dIkkuWLOGIESNIkm+99Rbvu+++46rzbAPOAO2hbprRcdagsaoxZCKfk0ExddfMu1Av1SvLyRdBydoJ36dZsXQtXrwYb731Fg4ePIiIiAgkJSWhW7duSExMVMpGAciGQotY7xMJgNGXVEMAMBZKIomFStu1ZeLAWwgwAFWVVwXvHw+/L/vZDH1lQMcvERX+TaPRCIfDEew/3mSoIon6+npUVVXh4MGDIIlt27Y1S56jQrX0MiB4saqqCrW1Cu9q//79IYoiRFFEz549tXTrADTrNwAgEsr4pK7gCQCcgNCoXPeyyy9T9scByAHQI7ARAI4oCdNsdpu/vqyAa+QAEIFevXrhnF7nKOeMhJI0LR4KXWA8FJpZEUrSnjT/JQxGA7KysrSEOSNHjsS8efNQW1uLJ598UusXi8WixQKpGU8LCwsBKFbytLQ0CIKApKQkFBcXa9lSw8PD8de//hWAkqI+NjZWC/q88847kZWVpbVl0aJFeOSRR3DeeefBbDbjD3/4AzZt2oSvv/66xQRIJxM33XQTrr76agDA/fffj9/+9rctli0tLcUTTzwRtC9U/FRTDB8+HIcPH9a+v/fee3C5XEFlUlJS0KtXL0ydOrXdbddxjDgRbR7AZQB+gKJWlDY5Ng3AZigJaocE7B/q27cZwNT2XEe3kOsgT76FPBCCKChWGgFKcofAa0wHBVFo9fz5C+bTGGFU2jcdfgu4GZx08yR/20dBSTwkBFioAKJLgPWss++4qYn16AafNLFAnXXJcmwdoA266HKM0t6kNoFJdAItsxaLpXldqvXZ1HqddrudoijS6XRSlmWGh4dr9YwdO5Y333wzAfDNN9/UzrFarXS73cp3A4LHkxDXkAxSaOt8KoiYgHNN/nsQRZFms5nuGHezvurWrRsvu+wydu3a1W9xt4LoBv9KYSuJzkaMGEGbzcY5c+bwoosu4vDhw0mSI0eOZGxsLAHFAt69e3dt22w2MzIykikpKbTb7SwpKaHT6aTRaGR0dDR/+9vfMiwsjKIocvHixYyIiOCll15Ko9FIk8nEXr168dNPP9XGdZPJxPT0dGZnZ9NkMtHlcjE8PJydOnXif//732OaYx588EGOHj066B5jYmKC/iNWq5UOh4N2u11rs3osOztb2xZFkd26dQtaSRAEgWlpaXS5XDSZTFo9ycnJNBqNNBqNtFgs7NatG//+978zIyODgiAwMTGRnTt3ZnJyMnv27Mnc3FxKksTIyEgWFxezU6dO2v81ISGBkydPZllZGZ1OJ3NycmixWGixWLhgwQKSZGNjIydNmsTMzEyee+65HDZs2HFb6s8kcAYs5CeqkGcDyASwFAEKOZT34++hsC6nAvgJyjuy5NvuBCVM5HsAOW1dR1fIdZDkpJsnKUvCgUqvXVF4TxSeDI/ibhKOkEq/J8PTZh3zF8ynO943MRlBd6xby0gKgLgYSnbP8SBu9W2ryqk6MbtA9AAhg6JJVLLfdWkycbUykanLtc2Uh1g0V+R1CS1n2wuOLh1WAt0nAEVBV90o2iNBriM+EQQhSFGLjIyky+UioGSnBMCIiAgaDAZardbm9TZxHbFYLHQ6nYyIiKDJZGKXLl2UYyJCPgtGs5EXXXQRs7OzGRMTQ7vdzt69e/PKK6/kQw891GxcHDVqFM1mc8h7gRFEMohzm993amoqTSYThw0b1moftebCovaLIAg0m8202Wza2HjTTTc162Oj0cj77ruPDocjaAwNCwvjwIEDaTabg9yAVN2kc+fOTElJYVFRESdNmsT8/HxWVVWxsLCQhYWFFEWRycnJ2nlOp1PbliSJI0aMIKC46qhuOYIg+F+ofOUcDgdFUaTFYglqd2RkpObWI8sy8/LyKAgCjUYjs7KymJKSwvDwcC5dupRGo5GyLDMlJYUmk4mRkZG02+202WwsLS3l+eefTwD0eDycNm0aBUHg4MGDmZyczOjoaF5++eUsKyujIAg899xz2djYyCFDhjA6OpokuXjxYg4bNoyNjY3cvXs3XS6XrpCfDoVcq6S5Qj4NwLSA7+8D6OmT91sq15LoCrkOFZq1GYpl/GQo42QLPuQ+pd8YYdQU6+OFZJWClf3boFjKxysD6jP/eIbhMeH+wdcgsaioiO5Yt9+yZQBxCTQ/TwCKYh6gaIuiyIiICIqiGGQ9CbLGt2ahDjv2SS+k6EqtLroESVxc3HGfG6gc9u7dm8nJyQTA+Ph4AoqvOOC3wl955ZXMysqi1Wqlx+PRFL5ACQ8Pp8fjYWJiImVZ5t13383/+7//U5RWo0CMAxEH4mYo45UVFCRlpTAuLo41NTUkqfkah8K//vUvOp1OHj16lEePHmVYWBhHjx5NQRBosVuUsegc+Mcn31gWGxvLrKwsZZ8AZcyL9ZVx+u8hJSUl5EuHqqwmJSVp+3r16uV/2QBYUFBAALzsssu0F6f8/HxGREQwPDycCxYs0MbT/v37s1+/fhQEgd27d+d3331Hi8XCiIgILl26lAkJCfzd737HhQsXMjIykklJSUH9YLPZ+Oijj2ovTGazmdHR0drvZrfbtXapyrYoirzwwgu17waDQRuHrVZr0Iud2v7w8HA6HA7Nmn755ZfznXfe0cqp5xuNRnbq1Ik2m01b0bnwwgsZHR3Niy66iLIsc+vWrYyNjWV0dDRjY2MZGRnJQYMG8ZxzzmFZWRkdDgfnz1fmxfvvv59Go5Ekedttt/GFF17Q7v3iiy/WFfIzrJA/CeCqgO//BHCpT54P2D8OwJNt1a8r5DpOB+YvmK8owBIUyw1Ad7z7hJVx0mfdB/zuMBN8k4+r+eSrWkCcTqd/v+Brk6N9k3dIa5QqxoBtGe2ynLd3uT6ovcd7ri666BIkLT1DoihqFvOUlJSg8haLRQtKVBU/l8vFzMxMAmBJSQnNZjNNJhNjY2NZWVlJkuzVq5cSJKq68LmhKcGJiYkkySFDhnDUqFF86aWXWF5e3uK4d8kll9But2uWYpfLxQEDBhAACwsLFaux2PzeLrnkEpL07xsMIsW3PQEhXXwCFVRVSXc4HCH7K7A/7Xa7pihbLBbN9aWwsJCpqakUBIHFxcWMjIzUFOOEhAQaDAZGRkZyypQpvOmmm5ibm8tDhw4RUF4oJt08iYI54N5EZQVDFEVKkhQ0RqvtlGWZFouFNpuNBoOBiYmJmjVdFEXabDa63W7269ePpaWl2vkZGRkElBcxp9PJhIQEAuC4ceN400030WKxMCkpiQsWLCCgvNQ999xzHD58OPv06UNJknjxxRfzmWee4Xnnnaf9V8xmMwcMGMDu3bvT6XTSZrOxT58+LCsrC7J8P/TQQ5RlmaSukJ+ItBkaJgjCh4IgrA0hI9s690QgCMJEQRCWC4KwfN++fafyUjp0AADGjhmL/bv3gw0Ea5UHZP/O/Rg7ZuwJ1/30U0/D7DQDgXl/vACSlcQd4eHhKC4uhslkgs1mg9lsRk1NDUaPHq2UNUIJwPo9/AGR46HQJ8ZBcQ7zcQULgoD09HQ4HA4lCMwOJfU2ABQEbEdDCbKSAVwM/3VCBFwq49MxIKD4MZ97puE80w3QoSM0AjNkiqKIUaNGAVCCG3fv3o3c3FwUFxfjwQcfRF1dHVwuF2JjY3HJJZfA4/EgLS0NtbW1kCQJ77//Pqqrq/Gf//wHXbp00ZLpFBcX48knn8Tll1+ujAUHAKlBwqSbJ2HHjh0AgHfffReTJ0/Gd999h65du7YYOEgShYWFWLVqFVatWoUJEyagf//+EAQBq1atwrZt2/D4Y49r9LQGgwGiKOKNN95AWlpAlGfgM+mBEhgPhbZWzW4sCALMZjPCwsJgt9shCAKsVqsSUA8gLCwM5557LhITE1FYWIioqCiEhYXh5ZdfRnp6OqxWK95++23Y7XZ06dIFq1atQnJyMmw2G1JSUtC3b1+kpaXh3HPPxaZNmyBJEmJilERL8fHxqKqqwr59++ByueByufDMvGfA0QRKfe32ArW1tRpv/Lhx47RbUrnRRVGEx+OB1WoFqfCmx8bGgiRkWUZVVZV2P4Hjanl5OQBgzZo1CA8PD/oNZFlGY2MjamtrNRrf9evX45577sHKlSuRmpqKiIgILF26FOvXr0dFRQUEQUBUVBRycnLw1VdfITw8HGFhYaiqqgqdsCoAvXv3xmuvvQav14uff/4ZS5cubbW8Dj/aVMhJnksyL4S81cppOwEkBXxP9O1raX+o684mWUqyNCoqqu070aGjg+P5Z56H8d9GoAyKMi4A0mYJIy8cifDwcKxbtw51dXWoqKjA4cOHYTAY8PHHH0M2ygojQjWAp+FXdpcBsELJHFoHheVFBBISEkASNTU1yqDtS5cNQFHCD/m2BSjsCHEA/htw/DYoa1cIKKeipQQhTZX4szmDp54LRMcpgKpInQhUJayurg6iKOKTTz4BSRw+fBi1tbXYuHEjVq5ciaeeegqNjY1oaGjAO++8gw0bNuDw4cN44IEH8Je//KVdmUNfmvcS0lLT8MUXX6ChsgGPP/Y4fvjhB3i9XuzYsQMDBgzAAw88oPF4qwhM3vbmu2/iyy+/xPbt21FRUYG3335bS8SzcOFCAMCCBQuwcuVKAAqLTKACreEggvnbA7KVqgqoJEmoqalBbW2tlgjozTffRHx8PABFMd2xYweOHDmCjIwMDB48GFVVVZBlGQcPHkR1dTV++OEH9O/fH+vXr8fmzZtRWVkJi8WCfv36YeDAgbDZbFi2bBluvvlmyLKMo0ePYujQofj8888xZMgQ/PGPf8Thw4fx49YfgVFQEjBtgGL4mAblZQIKa9e8efM01hw142lYWBiGDh0Ki8WCxsZGGAwGjcXL6XTCZDLh559/xieffIJNmzZpfbB3r5JhbseOHejcubPWd16vF2PHjoXX68XevXtxzz33AABGjRqFI0eOYPfu3VizZg0kSUJERASef/55/PDDD5AkCYcPH8bKlSshCAI++OAD7NixAzk5OUhISGj1fzNq1CgkJiYiJycHV111FUpKSoJ/Sx0t42SY2dHcZSUXwUGdW6BM0Qbfdir8QZ25bdWvu6zo+KVg0s2TKFoCgojMRvbp04cul4tdunRhXFwck5KSaDKZmJCQwMcff5zzF8z3L3tOD1imDQMxVlkKRbSyvCxKInv06MH//Oc/TE5O5l133cX5C+ZrS7RWu8/fUgAxEH62A9+SKgZCcaMJDBp14cR8wlthUtDlNIvu239a5WS4a6nBeup31b/YYrHwwgsvZOfOnTUfYqPRSLvdzrfeeosRERG88cYbWVpaSpfLRYPBwNmzZ/Onn36iKIrMzMzklClTgnimSXLy5MmcM2cOSXLlypUURZEFBQXMycnh7NmzWVdXx969ezMvL4+5ublBvNQhA+8lvxuexWLh1KlTmZ3jZwxpyb1NdSMBoMT2RPm2s6DF0QiCQKvVqrmiNA3QvOeeezQXDLXf0tPT2alTJ77zzju02+3s0qULo6OjabPZaDKZ+M4772juPGazmQ6Hg3//+99ZW1vLCRMmaO4m6enp/Pjjj0mSt99+O0tKSigIAv/whz8obZjqa+d4EOdBcf25AcRVyj1brVY6nU66XC4uWrQoiId8xowZQQGy6rE5c+Zw5MiR7N+/P9PT0zlz5kzabLYgLnKS9Hq9nDJlCnNzc5mXl8dFixaRJF988UXm5uayqKiIffr04ZYtW0iSHo+H+/btY01NDYcOHcqsrCyOHDmS/fr145IlS4LKtAeqG9P+/fvZqVMn7t69u13ndSTgbPMhh7LQ/T8AtQB+RnDA5l1QGFV+BDAsYP9wKInIfwJwV3uuoyvkOs52BCniZhCZ8G8LyuQxevRoGgwGCoJAk8lEo9HIv/71r6yrqws9sQcGQ/k+JUniU089xZEjR7J79+7s2rUr165dS5vNRovFwmnTpvnPN8HvJ9q0bjXJhzOgbHsUCAnNg0ZbUshb8l3XFXhdzmLp27dvyP0mkykoUU1gIF9Tsdlt2nZERETQMZfLxcjISN52223s3bu3RnfXs2dPXnLJJXz88ce5ZcsWpqWlaeNPYJKYQOVN3V9fX9/i2NVSwpqm9ZItUNOOUQLw161bR7PZzK7duwYr7V3999apU6ege9X8pA3wx76I0Ma7yspK7tq1i127dmV+fj7z8vL44osvkiT37dvHyy+/XAvs7Nev34kM4ccErR+AkDS6AE7atVr7fc4U+vXrx8LCQmZnZ2svd2cbcLYp5KdLdIVcR0dCU+tFWwhpNfIp4ujkn3xkWWZkZCRNJhMFQaDL5aIkSbRarZoFrKioqE3GE0mSOGjQoJDHVP7i45b2KsuhyhlO4Lq6/CrllAcEn4KXv8BnLJBZRWVEac89q0GYALRgPiA48+T8+fM1SkCr1crk5GQaDAbGx8fTZDJRkiR+/fXXrKioYHp6OtPS0lhUVMRnn32Wubm5nDBhAm02G3Nzc9m3b1+Wl5fzmmuuYV5eHvPz8/nqq6+SVBS+O++8kwUFBezevTv37NlDknz77bc1zu8BAwZwz549Spv7Qsnc6YGyuuZjR1Ep+yCAyIVfQR0Khc2lSR9ER0fz3XffVco7QZwPZQxRx1BT+5Ta1NRUiqKoscKcDmhjvmohn4mTnjtDx6kFOmJQpw4dv2Zs3boVeXl5eOKJJ5CdnY2xY/0BnosXL0Z2djYGDBgAALjyyitRUFCARx99VCtz7733YvaLsxV/QheAf0Bx2BoNJYLjiP9ahw8fhsfjwRdffAGv14tDhw6hoaEBv//97/HQQw9BkiT86U9/gt1uh9PpxF133YXevXtDEARER0fjjTfeQExMDCRJwuTJk2E2m5WKA/y+Dx06BCA4A2DTbIAtQoQyXYaC6kNu932yyXcRwVkBY6Fn69TRJpR5UYEgCFqw3olgzJgxARc4gYpCxEnExsZqAXZAsN/4zz//HFRWlmVIkgSTyQRZloP219TUQBAEyLKM0aNH4/bbb4fRaITRaITdbkdKSgoyMzMxYsQIAEq2zu3bt6OhoQG9evWC0WhEY2MjevbsiaSkJGzfvh33338/fvzxR9x000344Ycf8OKLL6Kqqgpdu3ZFeXk5XC4X5s6di//973/YvXs3pk6diq5du6KyshKLFi2C0WjE1q1bMXDgQGRmZuLWW2/V2lRRUYEHH3wQklVSxrT9UOJQbgBwCBAtIlavXg1B9P12lwR0RDKAyuZ9eeDAAVx55ZWwWqxKXMd7vj7fDuA1YFDfQe36mbZs2YLGxkaYTKZ2lT8ZePqppzHp6kmK3/trUOKGGn2frwETr5l42tqi4yzC6X4DOB7RLeQ6zhTU5d3IyEimpqayd+/evOKKK/jQQw/xnHPOYbdu3VhcXMxu3boxOTmZhw8fZnJyMhsbG0lSs2zhThBjfBaeaAS7rPisQiaTiX/+85+5fPly9u3blyUlJRw8eDB37dqlua0UFBRQkpSMesOHD2dycjLtdjt79OjBuLg4zYdSpxrUpcPLqU5U1Z4VmUB+/jb86zXrrrpP8J9XWlrKLl26cPHixaypqWGvXr1oMpm05DwGg4EkNeu30+mk2Wym1Wrl1VdfHWQRj46OZlJSEm+88UampKRovtFGo5Hjxo2jzWbTygTS+l1zzTUEFOo+NdOj6jet7g+8n5ycHAKKK0d+fr527S+//JIANLcWAFqCmYKCAhoMBv7444+a9d9oNDIlJYVDhgxRLMMyiGL4LdkCOO7qcSwrK1OureZf6A+iHxSrcby/XWryHvW+IyIigrKfAkpiHZWmce7cuUE+yl999RU/+ugjkuQ999zDP//5z5w9e/bpnDaCcKpyZ+g4tYDusqIr5Do6FsrKyoJcRNRMaoGTQ2xsrMZ7q/p+NytnCZjIW1BEZFlmVFSUNjmH4ucFmivbKudw4L6ggCg9Q+fpF90PvmNJwDMQ9Ky4A8q0opCrWQ8jIyObHevcubOWZMVut3Pw4MFcsmQJMzMz6XA4NMVVjQ8BwKlTpwa5sKhjjMfjocViCXJRU1/qrVYrb775ZgJgamoq7Xa7lvxLkiS63W4mJCRodQWNAfCnXjeZTLTZbBpv+cUXX8zIyEiazWZGRETwkUceIaDwc6s+3bIss1OnTrTb7ZRlmX379mV8fDwTExN57733sqioSPPRLi0tpSD7slBaJYZHhLOsrExRyAUosS8uEFeC6AMt8FNt53XXXae9aBiNRsbExNDhcFCSJD7xxBNMTEzknj172NDQwKeffprPPvts0Jjd0NDAW265hc8//zyvuuoqXnXVVdy0adMZmD10nM2ArpDrCrmOjgXVqmO1Wrl9+3ZN2VYnD6PRyLi4OI0JYeDAgezVqxdFUeSTTz4ZPHmr1rh4BCnoCYkJNJvN7Nu3L++9997g1NrqOQHWPjWFcmDdarrlM6746KJLBxdt1Qponpk28OW1lZcq1eqcmJjIrKwsLVV9dnY2p0+fzpSUFCYlJXHEiBFB6c8BJUvjl19+qVmv1bpEUWRaWhoB/0u30Wik1WplQkICx44d26wdgS8Xbreb8fHxWn0PPPBAUEComt3TYDDQZrPxkksuIQCmp6czMjKSkZGRlCSJy5cvJwD27NmTt99+OwGwurqaJDl8+HB2796dZrOZkiTRaDTS4/EwLCxMU8ibxtjk5uayrKyM+/fvV9oWBmWFUB3TAgLGw8LCSFJLea/2l7qt+udfeeWVvPjii5mdnc3c3FweOHCAaWlpzMjIoCiKWiKi2tpalpWVccCAAczPz+fAgQO5bds2rW2///3vmZ6ezszMTL733nunbV7R0fEBXSHXFXIdHQtlZWWapUn9DLRQS5LE2NhYLYObIAiaQq2WNxqNwVZtAzhm7BjKskyn06lNcEOHDmV0dLS/nCmEwvBLl2O1LOuW6F+XnEbaRvWZVRX4hQsXNnvpNRqNmhINKJZkQRCYnJxMk8nUzNXihCTwvx5i1SvoRR7gxIkTg14+1LaomSJHjhxJQBnDAhXgqKgorbxqxXe73Xz//fc5ZcoURkZGMiYmhhdeeCEFQWBqaiqnTJnSpkJOkiUlJf422nz3dAp+07CwMM2aDyhj8dy5cxkTE9Os7DnnnMPY2FhKkkRJkijLMtPT07lkyRKazWampqYyJiaGTqeTiYmJfPzxx0mSDz/8MHNzc5mbm8tHH330NM9MOk41oCvkukKu48wi0N+vKdXfjz/+SLPZrE3UTqeTTqeTDz74ILt27crw8HDGxsbyyJEjFASBF110EQ0Gg+aD3rlzZwqCwLvuuoszZsyg3W6ny+XiwIEDaTababFYeOutt/qvqfphOtuYgOTjm7ROmdibKA/dO0CbdOmYcqzK2Cl8AbPZbPz000+177GxsSwpKdEU1MB05yq1oSRJjImJ0SzZJpNJc1tr7VqBL+gxMTFBCnGLEkgnakZIpTwwbXrT66gW8pKSEsqyTI/HQ0EQ6HQ6tVU/lc8bABMSErT06ZGRkczKymJGRgZzc3NpNpsZHR3NrKwsvvLKK1y1alW7xlfNjxxQfMmtJ+e3C+QfDxR1dcJqtXLhwoXa/uHDh9NqtVIQBH799ddMSkqiIAh8++23aTAYmJOTQ6/Xy969e3PkyJGcMWMG7777bhYXF5Mkly9fzry8PFZUVLC8vJw5OTn87rvvTuXUpOM0AzrLig4dZw43T74Zz8x7Bo1dGhVGlCsAjPAfLygo8Ge/hJIhrqGhAc899xxWrFiByspKpKam4uGHHwapZIkzGAzwer0oKSlBv379AACPPfYYvv32W0RGRsLr9WLcuHGora1FTU0NLBYLzFYfO8paAE8BqPE1oKXsl/Ut7D9VT3dbZAUVUKY9lWDi61PUDh1nPwITRlrbUZ4B2yee+DIIlZWVOOecc7Tve/bswXfffYd9+/YBULIrqs/+smXLIIqilh68rq4OVVVVqK2tRW1tLerq6lq/Dfpv5OjRozh8+LD2PYhFJvCZD9yug5LtNwAxMTEYOHCgVpfRaER4eDhyc3MRFxeHI0eOQJIkrFq1CvX19bBYLLj55psxdOhQ2O12WCwW5OXlaaxL119/PdxuN3r37g2HwwGTyYRLL70Ua9euxbp161BcXAyj0YgZM2bgrbdaS9zdAqKgZAYGTui3FARBS0evQpIkCIKAgwcPAlBYaK699lqtb//73/+iuroaJPHPf/4TP//8M0RRxMiRI9HY2Ij9+/cDANLT0/Htt98CUNhv1IyYn332GS6++GLYbDbY7XZccskl+PTTT4//JnTowKmbsnXoOOug0RNuADASCj1hCrTJora2FrIsw+FQ8jb/73//Q2RkJGRZxiWXXIKGhgasWLEC9913HwwGA6xWKwRBwPr167Fjxw6UlZVBFEU0NDRgw4YNOHToEMrLy/HYY49hyJAhEAQBDz30EBrqGpQL1kNJGa3O7W1nuw5+olUmtXayxKkTsSAIsFqtzfZrqA3eb7FYlB1u33EXADOAvr7vub7PnOD2HVc6ZZ0q8ZeLqmMs33BKWtEqAhVpr9cb9P1YYDT6uUjr6+uDnj1BELSU6ugEf5p4dRyQ0GzmNplMqKqqwrfffotbb70VoigiLi4OxcXF2LlzJyoqKlBXV4ePP/4YjY2NMBqN6N69O5588klYLBYkJibCYDBg48aNaGxsxAUXXIAZM2Zg2bJlGD16NJxOJ26//Xbce++9AIDU1FT85z//wffff49169bh7rvvDnmfC15egMi4SAgGAYJJQGpqqnJAALA54J5CdGNTektBELSxNxAk/RSvUMYlq9UKkli0aJG2Pzs7GwaDAYIgIC8vT/vtSkpKACgvXMuWLcP555+Pqqoq/Pe//8Unn3wCURRx9OhRLF++HEOGDAl5nzp0nAycZBuDDh1nLxqrGhVO3P1QPgHFivMnKMrwLLRp+TpRnDv4XHz05UfKi0EyFM7dhQiejBt9n174ucFVA1EPAF9AmfDU+Uyd7EQ0s6rJsoz6esXE7vV6IQgCXC4XysvLIQgCSEIQBM0aGAj1uzZxdgPwf1As+lYAS5vcnB3KS4JPoVetdhoEhJyYg3B8+o+Oswyh/m+nAup//NRdACH/s4HjSENDg/YMCYKAwsJCrFy5Ujm4CcosbYDyXAkALACyASiGWww6bxA2b9wMSZLw4IMPorq6Gunp6aitrUWfPn2QkZGBHTt24KuvvkJWVhbq6+vRt2/foOvv3bsXO3bsQFhYGOx2O95++21s27YNiYmJuOGGG1BbW4vvvvsOV199dbtvfcHLCzBh0gTUV9crbR4FwAngeQDVvnvZ5SscwthgNBpRW1urfScZNC6pEAQBNTU12vfs7Gz88MMPAID169dr/6X169drqxxlZWVa+YcffhgNDcrbXefOnbFx40bU1NTg0UcfxdixY3HgwAGsX78e33zzDdasWQMAOOecc3DNNddg6tSpIIk33ngDL730Urv7RoeOUNAt5Dp0+CBZJUUBjoTyGYjtaNtV4yTgo2U+ZTwVitKdCmBgQAFVR2mEYjmLCNgnAFgFZfIWABRAecIFBCvoASgpKUFZWRkMBgOys7NhNBpRXl4Or9eLxYsXQ5IkxMXFYdq0aYiKioLRaEReXh4AwG63IyIiwm/h+z8oSYi8AA77PmUoyoMIRYEwAIgBYAE2lG0IbsxxGMx1/DLRqjJ+Emct9YWzPWgpgZbRaITH4wl9UgtVG41GrT5ZljUFMywsDDt27IDJZIIoikoiHS+UcQBQ7r0CmjJuNBmx7+d92LVrF6KiojBmzBhMnDgRgwY1T5oza9YsdO/eHb1790ZWVhYAZXXrzTffxLBhw5qtWC1duhSFhYUoLi7Gv/71L9x2220tdU1I3DXzLtQ31ivjmDqmiVCUcwnai3lLCFTGA9H0BSrwuyRJmjIOADNnztT+Sw0NDRg5ciQAaG49cXFx2Lp1K7Kzs2G1WpGSkoLy8nI0NDRgzZo1+POf/4whQ4bgo48+QmRkJKZPnw5AGTevueYadOvWDd27d8f111+P4uLidvaMDh0t4HQ7rR+P6EGdOk4HtHTHKk/uePiTW9jBQecNOuVtAHzXnBkgtwYEKo0HPRke3nnXnZRNPlYFI0KfN90XBKd+H+8LBjsHtEZbOX/B/JBt2LlzJzMyMrTkRu3F/AXz6Y73UbzJvmurQXiyr19nghjlC+gKDNQTfH2u7jMGbOuiS3ulPcmAjlOaspgEitPpDH1MhJ/iVFTSx8uyzISEBAqCwLS0NBYWFjImJobdunUjSXbp0oW5ubksLCxkcnIyJYNECMpz39Iz6/F4uG/fvpDHvv/+e8bFxZEklyxZwhEjRmjHbDZbyHNWrlzJd999t8Xv7YEgCv7nOAZELIjfgLgNyvOvjk3jQaT4ys30fQY8/4Ig8JZbbtESBiUnJ/Occ87hqFGj6HQ6KQgCMzMzOXPmTHbt2pX5+fnMyMhgRkYG8/PzmZ+fT6PRyKuuuooxMTGMjY1lbm4uu3XrxsWLF4fsl8mTJ3POnDnHdL86flmAHtSpQ8eZg5ruWFohKRbeRQBmKZ+Deg7Chx98eOobYUJz6/wuKJY2EbD+nxX3zLwH9/z1HtTV1CkPci3hjneHtuq74E/Z/CqAGsCzx4PZj87G2DFjm11+3rx56N69O+65554WLYItYeyYsdi/c7/SpjqCXp+QEBoFZeXhYQD/ATAcwHQogbMSlKlXXXVWreytQbU8hh1TE3WczWjPCtUp9CtXXbtC4ejRoy2f6HuMoiOjUV1djfr6euzbtw9ZWVnYvHkzVq1ahUWLFiEqKgoAQBKvvfYaVq1ahW3btqGhvgH0Els3bg35zLaGhx56CP/85z9x8OBB/O53v8Ptt98OAPj4448xdqxS11133YXCwkL06NEDP//8MwAlaPXWW29F165d0bVrVyxevBjvvfce7r77blx77bXo378/OnXqhCeeeEK71rx581BQUIDCwkKMGzcOyWnJiqudBOAo/NZ9QHFZEQE8A+DfUMpFAvjAd7wRkI0yRFHE888/j3fffReZmZmIiIjArl27cOedd+LHH3/EZZddprmNPP300ygvL8fq1asxbtw47Nu3D5s2bcLGjRvxj3/8A7NmzUJ9fT3i4+NRU1ODQ4cOYfXq1cfUnzp0nFKc7jeA4xHdQq7j14JB5w1SrPRNrPMQQXesu0UL2fwF82mMMAadJzpFCkZ/xrwzmbLZHetWONXDffc0E8TtIO7yfVepG0UQcQimtzOHsDymdQBrrC6nV9pj/TYde73Dhg3TtptmwQ2STAT9L+12OwE0SxYWilvbarWyT58+dDqdFEWRU6dOpdvtZl1dHUly3rx5HDp0KEly2rRpnDx5Mr1eL0meEJ3el19+ydjY2KD7c7lcjIqK0qgZnU4nw8PDabfbGRsbq2XwFEWRNpuNaWlpQansZVlmUVERCwoKKMsyt23bxrVr1zI5OZklJSXMz8/n8OHD+ezsZ/19IUCxirvhz8xZAGWsGgOF0jEbRCqIcSCsCmXhrl27+P777/OGG26g1+vl119/zYiICL7//vv87rvv2LdvX+1es7OzuX379qDyjY2NHDFiBD/55BOWlZVREAR++eWXbfabbiHXAd1CrkPHrxsffvAhBvUcFGSdNzYaMf+l+di/e3+LFrKxY8bihb+/AM8XHgj3CPB84cG8Z+bBW6swQTRUNuDpp54+rfcSBBHARVBWHtSA2b0AnoNiMa8HYpNikZWZBWOV0c8s0Ql+q6croL6fTnWDdZxxNKX5bA/LkCVgW11FCZjl7HY7AIWyVGUHWrZsmcbSYTKZIMuyxnaitUMAsA2AzV9vZWUlBEHAsGHD0KtXL6WoJCmqr9pWg+IjrgYVGo1GmM1mzJw5E3a7HU6nE7GxsRpVKgBMnz4d9fX1KCgoQG5urua3fDzo0qWLdh9msxkXX3wxysvLQVKzkA8dOhQ33ngjRFEESdjtdq0fUlNTsWfPHgiCgOLiYtx5551wuVyor6/H999/j6ioKNxxxx34+OOPUV1djUceeQSrV69Gly5dsO6HdUq/NkKJJTngk4W+/lkDZYx7FUAigK1QVvJeAaQGCcOGDcO3336LDz74AB988AGKi4sxceJEhIeHY8eOHSguLsbevXuxa9cufP/99wgPD0dSUlJQ+ZKSEmzYsAGbNm0CAHg8HvTo0eO4+1OHjlMJnWVFh44OhuN1jRk7ZuwxL2mfLhzcc1BRxNWA2VQA6b7tzwCIwMUXXIz3/vse6o7U+ZUpGxQKRS+CFTIbgMrT1nwdZwJNFXC2UC6QyeRwiPI+9ydBEDQ2jerqajQ2Kheora1FYmIi/ve//6G+vh6CIMBsNqO6utp/PqG4VAVQiS5evBiLFy/GAw88AI/HA1mWcffdd2P69Ol48cUXceONN2LDhg1ITU3FwYMH8cADD6Cmpgb//Oc/YTKZsHXr1pC3Y7FY8Oyzz7bZPe2BLMuIi4vD3r174XA4UFBQgKVLl6KiokK7/yVLluDzzz9HRUUFKioq0NDQgNraWrhcLixduhSjRo3C+vXrkZubiwMHDmi86fn5+Thw4AB27dqF6upq1NTUaLkWxo8fj8suuww2qw2DBw/Gss+X4cjBI0jOSMY9M+/B2DFjsWbNGtxz7z3416J/KYp4HQADIEDA3Dlz8Z//+4/yM5KYNm0abrzxxmb3d9lll+HVV1/Fnj17MHr06FbLb926FTabrV399uSTTx5nj+vQcfzQLeQ6dOg45UhOS1aU774A3oIyATcC+A4AgUGDBuGfi/6Jbb22AbdC8SMHgPVQFK5GKH6nKlpSxqNOUoONbRfR4YfGm3060ZTWs7WiPqq8mpoaCIKA2tpa1NfXQ5IkWCwW3H777SAJg8EAt9sNl8vlP1mGEqvQC4BHvbSA66+/Hq+99hrOO+88/Otf/4LRaMSKFStQUFCg+SwPGDAAFosFCQkJ+PTTT/GnP/3ppHZBUyx4eQHs4XYIJsXCL5gEfL/mezQ2NsJisSA/Px9Hjx7VqA0BYNKkSbjppptgtVphsViwf/9+5ObmwmazoUuXLqivr9deZEgiJiYGMTEx+Pzzz5Geno6XXnoJffv2RXV1NQ4cOADAz2IyePBgVFVW4a4/3gVvoxdvvvImxo4Zi59++gn5+flYtHARunbtiih3FCAAZqMZc5+fi8HnDcayZcvQrVs3DBkyBC+88AIqKhQH9J07d2oJekaPHo1Fixbh1VdfxWWXXQYArZbXoaND43T7yByP6D7kOnSc3Zi/YD6t0VbFf/RiKL7kTX10BSi+5HKIY7p0OGnV37otyWr9uMVi4RdffMHU1FQ6nU6aTKbgMoExBmnQfLY//fRTOhwO9u7dm//6178oiiL/8Y9/cNWqVczJyWFYWBjvvvtuOhwOhoeH83//+x8TExP56aefUhAEjhkzhllZWZRlmWabWWEGEUBcCuJKKL7ONvDFuS+yT58+nDx58pl+tDTMXzCfkk1SYjUCY1B8jCU2m41vvvkmZVmm0+mkxWIhAN51110cOHAgTSYTzWYzN2/ezNmzZzM+Pp4Wi4Vms5kxMTEsKirin/70JzqdTmZkZJAkc3Jy+P7775MkExMTmZqayoKCAhYWFvK3v/0t9+3bx9zcXMbGxjI7O5s33ngjSfKWW25hbm4u8/PzecUVV7CmpoZer5dTpkxhbm4u8/LyuGjRIu3eHnvsMebl5TEvL489evTg5s2btWN5eXns379/UF+EKl9WVsbc3NxT/TPo+IUAZ8CH/Iwr2+0RXSHXoePsx/wF8+nJ8FAQBXoyPLz/gfspSRJFUVQUq9uhBLA6z7yyqUv7pd2KuQP+wEw16DGQGtMnkiRREAQtUDIuLo7p6en+MkYEv7T56hQtIrds2aIFKxqNRhqNRo4YMYIPPfQQR48eTUEQNMXTYDDw4MGDfPvttzWF32azURRFms1mvjDnBYomUQk0Fn3XEqEFRz/++OPHpJDX19ef8DP0zDPPcO7cuSGPeTI8wUHTqozy9Y8k0mQy0WKxsLCwkMOHD6ckSYyPj+ctt9zCIUOG0GQy0el00mg0UpZlRkZGMisri6+//jpdLhfNZjOTkpJYUlLCgoIC5uTkcPbs2SQVasTu3bszPz+fI0eO5MGDB0mSM2bM4EMPPXTC965Dx+kEzoBCLijX7dgoLS3l8uXLz3QzdOjQcRIxc+ZM/OUvfwEAUKBCa6dm8CPal7lTxy8aU6ZMwSOPPAIgRLKgJnEEoihClmUUFxdjxYoVmj94WFgY0tLSkJqaildffRVutxsVFRUQRRG5ubnIzMzE66+/jurqauTk5GD79u2oqanBb3/7W7z2+mvYWrYVJBEVFwXWEy+++CKcTicuv/xyHDp0CP/85z9xxRVXYOrUqVi6dCmqqqpQUFCAhQsXYunSpZg+fTrCw8OxYcMGbNy48ZT1lSgpQZn4E4IDYhsB/FVx2/E2nvrspzp0/BIgCMIKkqWn9Zq6Qq5Dh47TgQUvL8CNk29EZVWlFsClMagkA9gJJYDOAKAeioJeC0Ux7wKgpSFA9pVvCvV8FRKC2C9OJWe1jvYjMBX6iaSyV1OkA4pPu1pXp06dUFtbi+3bFaL+yMhIREZGYvPmzVra+sBrqt9dLhcKCwvx6aefgiTCw8MhCAJ69uyJJUuWoKqqCiTx8MMPo6amBvfeey/CwsLQqVMnfPbZZygpKcHDDz+MQYMGwel0IikpCTk5ORg6dCiWL1+OJ598ErfffjsWL14Mt9uN7t274+mnn4YkSbDb7bjtttvwzjvvwGKx4K233kJMTIzGzjJlyhT89NNPmDx5Mvbt2wer1YotO7Zg19FdSvZeN5T8BRUACgGsBjyRHky6bhLmz58PURQxbNgw3H///c3qee6557RMnjp0/FpxJhTyM+6O0h7RXVZ06Di7EdK/9dwAt4VQfuOBrgyOALcBIJj3uY/fnSDos6lEgogGEeH77ADuHh1NNPehX5A0dakRRZGSJIU8Huir7nK5tOOCINBmszExMVE7bjAYmvWXIAi0Wq3atuoCYzAYOGbMGL7xxht0uVxMSEjgokWLOGTIEEZGRnLv3r2cNGkS58yZw/DwcALg22+/zRkzZrBfv36cNWsWyWD3j4EDB3Ljxo0kya+++orpGT63HhGEB8SfQIyE5hY0+ZbJFEWRlZWV3LlzJy+44AIuWbKEXbp04caNGzlnzhyOGjWKAwYMaPb8tpTRU4eOXyqgu6yEhm4h16Hj7EZK5xRs27VNYS85DIX+sAsUDnJA4R2vhmIhV1fVW7BiS5IEks1dGHScepzGlQWbzYbKyuPntjSbzaipqWmznCRJGgWgLMtaRk6TyYTa2lrIssJ16PV6kZmZiXXr1iE+Ph579uwJOtfr9SIsLAwOhwO7du3yT7KCAFlWsk6aTCaNB9xsNqOhoQH19fWIiIiAzWbDgQMH0NDQgLq6Onz66af48MMPce+99yIpKQlJSUn45ptvkJWVhWXLliE8PBx2u127R4PBAEEQUFFR4bf4+9y+IiMjcdlll+Gf//wnvv76a+zatQvDhw/HnXfeiYceeggxMTEQRRFVVVWw2WzIz8/HO++8o/WR3W7XWEsaGhpgMOiMyTp+2dBdVlqArpDr0HF2QxAFwArgUijuKdsBvA6gHMAwAN2h0CFGAfgcinJ+CYD3fWUC4HA4IAgC6uvrERcXh23btmlKUUsIdGfQ0TZcLpdGXRfky+9As9/jbIaaNKetebCl/4+qkKuKfFP3l/DwcJSUlODjjz/WlHNRFOF2uyFJEn7++Wet3rb+ozabDbW1tWhoaAh6cWjalsA222w2lJeXa4mOWrqW2i6SMBqNqKurgyRJSElJgcPhwHfffQeLxYLi4mIcOHAA27ZtQ21tLSRJQlpaGv79738jIyOj1T7UoeNswplQyHUech06dJxySBZJUcZTofhyrwagJszbAeBJKIlXjFD8wb0A8gD81lfGrXz84Q9/wNGjR3HkyBHMmzcPAFBaWoorrrgCgGLVtDvt2nVVhcvpdAY3SIDiex6Q6EVFi9a/lkZLcwv7Q0Fou0hHgKaMA35lHDg+ZfwMUJS3F6oVW4WavbMpWlKUVQV44MCBWn2BqKqqwtKlS7X9JNHY2Ii9e/di9+7dQfW29cJYWVmp8YE3VcYD2xJYX3l5uXZdLfNowLWMRqN2PC4uDl6vFz169ECPHj1QX1+PLVu2ICsrC5Ikoba2FlOnToUkSTjnnHMwd+5cnHfeeXjkkUeQmJjYatt16NDRNnSFXIcOHaccjdWNimVcxUgAWfAHbN4MYDQU5S0V/gQ/232f/QHZKOOBBx4AoASITrhpArZs2YKvV36NRa8tgslkQl1dHWqra9GzZ0+43W4t2YmmwKT46k7yXVuEouyHQ1OW77rrLmVDQLAC3Q/+/YGI9H22NpqqOn6gvnaWKOfHikDFD2FQXnpO472KYhvTmhn+3z4QJvizcx4DJEnCRx99FPJYbW1tq9b3NtvaQtmgPg7xHVBcdgJhtVqblVGVeLPZrNXx9ddf46uvvgKgKOpvvPEGGhsbIYoiPvnkE+zduxdLlizBtddeiy+//BKrVq1q8UVGhw4d7YeukOvQoeOUorKyUlEm7gUwC8DbAOYAeAqKgvoSgAcA3AdgHZQsnp0BPAtggVKH8IaAiPAIDBo0CD179sS4CeNQUVuhKPCyUk9tQ63mm7tmzRocOnQItbW16NatmzLSGQAcBFAERdFXFcXzARxV6giPDMfefXuVegUAYwCU+Mou8d1QU/3qqO+zNQNnKL/rM+ktGNF2kaYKXWtQlTlBEBAVFaX5XUOCsupxrPcq45iyrqrXt9vtsNuVFRJRFJGdnR2srBoC2tI3oAIRCiOPr9kJCQkaU0tbaGxs1CzXodCa5ftY3KgCyzZV8pt+F0URpaXBq+3JyclB32VZ1laOYmJi8Je//AWSJGHUqFHo1q0bZFlG//79tXTzdrsdn3/+Ofr27YsNGzYgIiICtbW1ePrpp/Hxxx+36x7U3+aaa67Bq6++2q6yu3btwqWXXgoAePHFF3HLLbcElZs5cyb+9re/tbteHTo6KnSFXIcOHacU//nPf9CzZ09Y3VbgCgDbAFQB8AITrp0Ad7gbCfEJikK0GYrLyucA9kBTZPv164cePXqgX79+WL5iORhHYBKAG6EofInQFOLy8nK4XC5kZmZCEASsWrUKleWVihW8HMB/oSjbtVDcZN4DkKbs+/vjf8fsF2cr/u7jAGQAuBCK9d7nCWOxWBCfEA+T1aTsUBXyQNjQ4SzgQcrlwbbLtBoQ2cSrRz1PFEV07tzZr6AehNI/rcUAClAoKgNRD+W3aYJAK3FcXByys7MhyzIsFgtEUYQkSZp/drdu3fDTTz8hKirKf52LoLwkEMB3CF65sEGjz9y3bx+AEK5OHQSqT3hLFvawsDB89tlnQft++OEHAMr/V5ZldOrUSVPkd+zYgT//+c8AgJUrV8LpdKKhoQHffPMNhg0bBkB58aisrMQHH3yAyspKDBs2DL1794bH48Hq1au167z44ovYtWtXyJeUxsZGnH/++a3eW9Pz4uPjdSVbx68Dp5vW5XhEpz3UoePsxY8//siwsDA6HA5KBoVuTjbKFASBr732GseNG0eS/Otf/8qLLrqI119/PVNSUuh2u1lXV0dRFFlVVUW3282ffvpJoXa7CMQ1IJJ9tG4uP/XcjTfeyHPPPZezZ8/m/v37SfqyGI5HcAbD8SCifJ8ytPTbgK/O6U3KTwcFUWh2f2VlZco5UQF0e+mtUPEJrRw7DbR/kiQxLCwsZNmhQ4cec/2yLLevrHRs9TYVURSDaAaHDx/Oc889lwaDgYIg0Gg0MisrixMnTmReXh5zcnI4YsQIkuSAgQMUak2VcjMRhBmECUSur23hIOIRmoIzoC9VOSm/TyvXam+ftNROm80W8phK+Wg0Gmk2m7V6VJpHtT+DqEVbEUmSmJmZSbPZrPWL2Wym0+lkYmIirVYrLRYLu3TpotXrcDhoNpvZuXNn5uXlcdasWezTpw8HDRpEWZZ5/fXXUxAEnnfeeVy/fj1zc3O5efNm5uXlMSoqir179+YzzzzDHj168I477tCoIMePH8/FixeTJJcvX86+ffuypKSEgwcP5q5du0gqGVazs7OZn5/P0aNHn9rBT8dZC5wB2sMzrmy3R3SFXIeOsweTbp5EyapM+pJV4oUjL2S3bt34/PPPs2/fvvR4PCwsLKTZbOYbb7yhKeQ7d+5kYmIi+/XrxzvuuKNFhVwQBaIfFCX8Jp9S0xuEUbkeqfAyT58+nR6Ph/v37+f8BfNpjDD6FbLxIOw+hUJWFIqamhqSVNoeKgX5eNCT4eH8BfPpyfBQEAV6Mjwce9VYhft5vE8E+FPEnwI5Hq5wQRCCuLcDxWg0npCCGXiuy+ViXFyc8l309zEAIidgu7WXks7wv2AFcMq/9NJLBECLxcJ9+/Zx7ty5dLvd9Hq9rK+vp9Vq5dtvv93u/6XWBvWlbCaIbr7/UwuKsqqkZmRknLLft+lvIctyM+VYLaO+pMTExNBgMFCWZTocDu0lyfj/7d15dFT1+fjx92fuZGYyWUlCFpYkEAGzkUBRWcqXWKXiUlApKCICUsDoV5QiVqVa1HKq54daa0kBrVIVV7TqQWytBdSvKC0KyiJLkKCALGExgcBkmef3x525SdhBICzP65wcZu69c++dez8kz3zm+Twfj6fRPb7nnnskNTVVLMsSv98v0dHRUlBQIPHx8eLz+SQ6OlrmzJkjV155paxbt07atm17wPnFxMQ4H+iSk5OlS5cuAkhaWppTgz05OVk8Ho9ER0dLenq6WJbltNsePXpIXFyc/PznPxeXyyUej0csyxKv1yuvvvqqAJKSkuKc94ABAyQyMlLcbre0a9dOUlNT5bzzzhPLsiQvL09atWol48ePFxGRHj16yIUXXii9e/cWn88nt912m4iIvPLKKzJ8+HAREUlLS3P+r+/cufOk/A5UZz4NyDUgV+qMVnxrsR2ENQx8vUh6erqIiJSUlIjL5XIC8k2bNkl6erqUl5dLdXW1JCQkiN/vlxUrVsgvfvELef755wWQW265Ra6++mp5+OGHJa1Fmrj8Lrt389eh4NdtB2/FtxZLaWmpcz5dunSRxYsXi4g9OVFii0Q7sPAgiamJ8uLMFw/+Hrw0nsRoKOKOd0vxrcXiT/bXL+9p74v4UEA3PBRsZp+cYO10+jmmIL5hAB7LMX9LEOGJkCeeeEKaN28uUVFRsm3bNhk3bpz4fD5p166dtGnTRqKiouSZZ5456raKsdtmo29DfAjtQstjQ/8e5rwO1ovcaJnhwG8GGgb7B5nEKi4uTvx+v8TGxoplWZKVlSWWZUlkZKS0bt1atm3bJr1795bmzZtLSUmJ3HTTTfLggw9KWlqatGzZUhITEyU7O1s6dOggb7/9toiITJ06VWJjY0VE5LnnnnMC1cTERNm7d+9Br8+6deskMTHReU9paWni8/nE4/E4AX9iYqJkZmaKMUa8Xq9ERkYK2D3k4dd6vV657rrrpLi4+IDAHZD8/HznNR9//LEAUlxcLFFRUZKYmCher1fOP/98sSyr0XHatWsnIiJDhgyRvLw8EbED8uTkZFmwYIHzwSQnJ0fy8vKkd+/eIiJy2WWXSf/+/eWFF16QysrKo24v6tzSFAG55pArpU6Y6TOmQ3/qyxu2AXrAt99+i8/n47777qOwsNDZPi0tjYkTJ9KtWzd69OhBYWEhMTExZGdn89RTT/Hcc89hjOHdd99l+fLlzJ07l//p+T/876/+F6/bC49j55kHoUO7DpRMKWH8+PHk5+eTl5dH9+7dKSgoAGDwDYMp31hu//ILCOXflzP4hsEHvIeSKSUUjyjG7DXwMvAwRM+OZsaUGcz51xyqLq+y39cKYClQTX35xLBjGJDoOMm/jS0rVH8wAvv8j5PXayd823+zDr7OEc7RbphPvxc7FAMSEhKIjY09bLURr9eLP9JPXV1do/2LCDfeeCOXX345F1xwAa+99hojRow46veRcV6GXXXHQ301n33AWuoH4f4UuBD73uxXdcflcjllA8HONzfGkJaWVr9RMhAJxOKU7qQG+x5k0/g+hNpQp06dCAQCVFRUEB0dTWlpKb169aKmpoaVK1eSlJSEx+PhnnvuYerUqbz88su8+uqrTJkyhaysLK6//npWrFhBZGQkLVu25JFHHmHcuHG0bdv2gGvw85//nKeeesp5PvC6gbij7AmGsrKzqKiwB0hERUXh9XpJTk4mJibGqbk+dOhQ7rvvPkQEv99PIBAA7Pu6efNm5/GcOXPIzMwE7FKLCQkJuN1uXC4XX331FVFRUdTU1LB7926MMdx11132PcrIAOwBrV6vl5iYGHr37k2rVq1o0aIFeXl5/OMf/2DLli3Oe8jPzyc6Oprc3FwuueQSpk2bxtKlS3n//fcBePfdd7ntttv44osvuOCCCw47IFepU0kDcqXUCVNXtV95Q4Ae9j/79u1j586dfP755yxZssQpMTd8+HBWr17Nf/7zH7Kzs5k0aRJg/zGeO3cuwWCQb7/9ltWrVzN37lxeeeUVnnzySfbt3lffu1AnrFy5EoA333yTpUuXsmzZMp588smjqpSxv5IpJQQDQSRg77+yvJLBNwzm27Xf1r+/j7DLNzYHrggtywB+R33pxoOV17NoPMgxHNMZDqzZ7eWYB4ce7P0mJCRw00032U9qsAfWHqNw0BwIBBoN4gwf0xjDI488wjXXXFP/olCsE+Fu8ImlQQntuLg4IiMjDwjuw/sND9Q8WPB/2WWXsWTJEubMmcN///tf8vLy2Lp161G/n0kTJ+Hf6beD8jewq/v4qJ8t9g6gJ3bN/G7Y9zEZaAG0sYPEmpoaevTogWVZDBkyhPj4eHswrAntK4D9AWRP6FoY+z1TC3wNfBM6mVbgifHQrVs3tm3bxtixY2nZsqUT3A8ZMoTRo0c3Kl3YuXNnvvzyS/785z9TXV3NI488QmZmpr1/7OojAwYM4PXXX6e4uNhZ3tCf/vQnFi1aRMeOHUlISOD1N1+nbkAd3A/BfkGnXKiIEB8fz6ZNm5zXitg11adNmwZAUVERWVlZgH1tHnvsMcCu1pOcnMzUqVMB+55u2bIFEXu23Xnz5h1yRtZwO7Msi7i4OGpqasjIyCAnJ4elS5fy6quvcsMNNzSqy+52u+nQoQPbtm2joqLCmQ11+fLlBINBvvvuOy6++GIeffRRfvjhB2cGUqWa3Knukj+eH01ZUerMYPmtg+Zeh3O7D6dz587Ss2dPJ7/zdNRocGg41aE/B6bpxGOnszTHTqcxCENCaQlRh0jnCKfegJ0G40LIt9cbl5FLL73UycM9VE64x+ORjIyMA1JLDponfrA89yOkaOz/Ex4ACEhSUpJERUU1SkcIr+vbt69z7s2aNXPSD1JSUiQmJkZcLpcYYyQiIkI8Ho+kpKSIZVliWZZ07dpVYmJiZPLkyZKbmysZGRmybds2ERH54x//KM2aNZOkpCTp2rVro3SloxEeD4DByS/HS+MxBCDcHrqX4TZ9v738UIpvLbbv82CECQgd7NSVzMzMA44dHotwsPSpoxFOu6ipqZGrrrpK3nzzzePaj/GZA//vtmicShNOxwnfS8uypFmzZs6y8PJwvnfD14afh9NvwutuvPFGZ9nHH38slmXJQw89JFFRUdK+fXuJiYmR3NxcGTZsmFiWJYmJiRITEyO9e/eW7OxsufDCCyU+Pl5E7JSVyy67TEREFi9eLAkJCdK2bVvJycmR6dOnS3V1tfTo0UPy8vIkNzdX/vCHPxzXtVJnPzSHXANypc5kB80hj7Zzu88GL858sT6HfP/BgL5QcN08FIxHIcQ0yBO+MBRot6wPwjv/pLO9rncoiA8HxO1DAXMz+3mrVq2kc+fOTv7sCy+84ARCh/vx+XyHroKyf/7yIfK6r776arEsS1q0aCEtWrSQ3NxcZ7BgUlKSxMfHS0REhOTl5Ul2drZkZ2fL5MmTJS4uTvr16+cErldddZVERkZKfn6+WJYlycnJIiJyzTXXSHR0tMybN08GDRokKSkpsm7dOhER6dOnj+Tm5spdd9110PtRV1cnBQUFsnr16hN2j52gPC50f30c14fM7j26N7qmqampsmbNmhN2nmHjxo2TgoIC6dChg9x+++0SDAaPaz/AQSsLhc/f7/dLXFyc9O3bV3r16iXjxo2Ttm3bSs+ePeXiiy+W+Ph46dy5s/Tt21ciIyOloKBAysrKpKCgQDwejyQmJorf75c+ffqIiMjDDz8sMTEx4vV6JT09Xb788ktZt26dnH/++dKvXz/Jz8+XpKQkmTx5soiI3HHHHZKSkiIiIgsWLJB27dpJYWGhTJgwQTIyMkSkcX68iMiVV14p8+bNO/6Lq85ZGpBrQK7UGW//ahZnSzAe1mhwaMMPHz0bBNTNEK4JrQv3iIcG91luS/r/sr+0bt1aCgoK7KAtFqEoFAS5Efz2toktEiU1LVUGDBggq1atkhYtWgggnTp1kuXLlzu95A17zhsG4C6Xq1GP9VH97NdzHi6H16tXL8nLyxNjjHO8rKwsadmypSQkJIjX65W0tDS5//77JT8/XyIiIuT222+XqKgoERHp399+zyL2YMJOnTpJfn6+dOnSRYqKimTevHmybds2GThwoOTn50t2draMHj36kPdh+fLl0qZNG/n1r399Qu9vRrsMu0c7IvQBav/7fJZ9yAzDc/APHuFrcDz27dsnNTU1ImIH0QUFBSfwjJU6eZoiIDf2cU9vXbp0kUWLFjX1aSillGPmSzO5Y9wdbN+xHaohsUUi27dshxtpPFjvM+Af0L1Hd3Zs30FOTg4vvPACq1evZsyYMaxYsYLtO7fDRcClwHSgAoqHF1MypYQZM2awaNEiunXvxu3jbmfn5p0HnIvL5arvZWngpZdeYuTIkQfP0XVTP3jRjT3w0GBP2rQXsGDsHWPp3KkzQ4YM4Xe/+x39+/enb9++BAIBkpOTAXu6+UGDBjFx4sQfcTVPHzNfmsmosaPswbvpwLfges3OaQ/uDWL5LUYNG0XJlJKmPdETzHiMPSHW1Tjvm7eAWrDEonbPsQ9+XLNmDQMHDiQYDOLxeCgpKeGCCy44kaet1ElhjPlcRLocecsT6FR/AjieH+0hV0qdCTjY1/5Dj9zDeLBvFRrmN7siXfU9te5Qb3tr7AluGvRmG2OcHvHY2FjZsGGD08MNCFc22H7/cnz7PW/VupX885//FBERn88n/fr1k0AgIElJSU7t+Orqalm2bNlJv66n2onK7z6TOOU+m4XSl5ohRNrf+pxt3wYodSRoD/nBaQ+5UupMEJMUw+6rdjfuIV8Krrdc1NXWHfV+nF7aDlV2acV+2NVJ3gMadpDHAJX1T40xtGnThvXr11NXV0dkZCQ+n4+Kigrq6vY7vh+7Nzz0OhHB6/WSkJBAVVUVFRUVuN1uzj//fJYvX87ChQvp0qULb7zxBjfeeKNTRm/kyJFcc801Tok6dea69bZbmfrsVGSf2NV/BIpHFp913wYodSRN0UOuZQ+VUuoEmfqnqbjfdtsl9OqAdeD+t5vnn3/+mPYzYeIEO2ViJXYw3gZoD/wMiAL+B0ihvnxi6M9GRkYGGzduxLIsLMuiqqqKTz/91Knf3bAkoqkzWG6L2NhYUlJS6NixI7169SI2NpaYmBgiIyO55ppr+Oqrr4iMjKRLF/sg/fv3Z8aMGURERGBZFi+88ALff//9cV4xdTopmVJCcG/Q7rELCFItGowrdYpoD7lSSp1AM1+ayYSJE/h27bekZ6UzaeKkg05AdDguy4VMEPg98Fvq65NPwa553rAHfh3wDmQkZVC2uuyg+ysuLub1118nNTWVjIwMZsyYQfPmxzN7kTpXFBUVMXnyZOeDmFLnEu0hV0qpM9zgGwZTtrqMYF2QstVlxxyMA6RnpduD6pKon0USoJwDJ15KB3baE90cyl/+8hfKy8tZtmwZ7777rgbjZ4Fhw4Yxa9YsADIzMykvLz/mfVxxxRXMnj2bBQsWHHS/lZWVlJRoD7lSp4IG5EopdZqZNHES/vf8cD7wNvUpMPE0DtCxnye2SDyuwF+dOcrKysjOzmbkyJHk5uby/vvvEwgEePrpp/n+++8pKiqif//+VFXZAwOGDRtGcXExXbt2pW3btsyfP5+bb76Z7Oxshg0bRjAYZM6cObz88ssMGjSIzp07M2DAgEazXlZWVjJp0iTuvPNOZ9nTTz/N2LFjj+s9HDCOQSnl0JQVpZQ6DYVTX9aXrseKtKjbW0dUXBR7avdAf5zSdP73/Ex/YroG5Ge5xx57jLvuuguoH4QbExNDSkoKpaWleL1eAoEAERERtG/fnhUrVjgDdX0+Hz/88AOxsbH4fD62bt2Ky+XC5XJRV1eHx+OhtrbWCZhbtWpFRUUFgUCAQCCAMYZmzZrRvn17Pv/8c1wuF6mpqQwdOpQHH3yQxx9/nMcff5wdO3bQrFkzrrrqKu6++26uuuoqSktLiYuLIxAI8NZbb3HJJZc05WVU6qhoyopSSimgPvVFgkLtnlokKOzeuZsXp71IxoIMzCRDxoIMDcabQMO0jqMVHR192PUPPPAAH3zwAVdccQW7du1qtG758uX8+c9/pmXLlgC8+eab+Hw+9uzZQ0VFBQA1NTU0a9YMt9tNeXk50dHRdOvWjdraWubNm0dCQgIVFRUMHz6cTp06AXZg7/P5qK2tJRgM0qFDB1q2bMmGDRto27YtF110EZZlMWLECKZMmcIvf/lL4uLi+O6773C73SxatIinn36aqVOnkp+fz8aNG4mPj2fHjh289dZbrFmzhtraWkpKSrj88st18K9Sh+E+8iZKKaVOF4NvGKwB+Ak0ceJEoqOjnd7nY1VbW4vb/eP/lD700EMAXHrppQesmzt3LldccQWvvvoqLpeLQYMGsW/fPowxbN26FbDnFAkH11VVVezZs4eFCxcSDAYZMGAAu3fvBmDAgAG8+OKLtGnThm+++YasrCzWrl2L3+93es8BduzYwc9+9jM++ugjPB4P06dPZ/ny5ezYsYP09HQCgQC1tbXExsaSlZXFl19+ycUXX0x5eTkffvghLVu2pE2bNqxbt47+/fuzbt06ysrKfvR1UupspT3kSimlzkoPP/ww8+fPZ8GCBZSVlfHSSy8d1evKysrIy8tzno8aNQpjDG63m7Zt2/LOO+8wfvx4YmJiiI2NpXv37nTt2pW8vDyaN29Ox44dyc/PJy8vj0WLFjF//nz27t1LZmYmYAfFcXFxxMTEkJCQwN13382wYcMYOHAgCQkJjBkzhqjoKIzLYIxhzJ1jKCkpYefOnbjdbrZv3050dDSRkZFOKUufz0dubi5+v59gMMj111/vDN6NjY0lMzMTYwxr1qyhvLyc9HR7dPDGjRsREbp168Ynn3xCr169sCwLj8dDr169MMbQoUMHPv74Y7Zv386jjz7K3r17SU9PZ/78+XTt2hURYejQoSxZsoRRo0bx29/+ljvvvNNJlwmX4aytPfbZPpU6V2hArpRS6rSXl5fnTHQ0cOBAAO666y4iIyOJjIzE4/Hw4YcfEh8fz+DBg+nRowcPPvggl112Gb/4xS+44IILuOmmm8jOzqZ///7ceOONjBkzhr/+9a/84Q9/YNasWRQVFbFo0SKmTZvG2rVr8Xg8XHfddTzzzDMAeDweNmzYQGVlJdu3b2fHjh3U1tby6aef8v3337NmzRq6devGz372M+bNm8fatWvZf5zWe++9x+LFi0lISGDDhg0sW7aMV199lT179gAQDAZ55plnqKIKbgLGA91w1lVXV/PSSy+xZ88eqqqq8Hg8AAQCAbKysqiurnZyw9u1awdAaWkpAF6vl1tuuYXo6Gi2bduGz+fj2muvRUT44IMPuOiii9i5cyfBYJC4uDh++OEHRIQxY8aQnp5OVFQUr7/+OnPmzGH9+vVs3ryZ3Nxc1q1bx2uvvca6dev4+9//TseOHdmwYcNJbhFKnV10UKdSSqnTyq5du3jqqaeYOXMmLpeLr7/+2lnncrkIBoPOc7fbTXJyMps2bSIhIYEdO3bgcrmIj49n586dBwTElmU5gxddLrtPKryN3+/H5XJRWVnJ0YiNjXVyuMPCAy7DcnJyyMjI4L333jvgPZSVlZGbm0tVVRUul4u4uLj6c07DnoU1CsgAVgO7Gh8jMjKSBx54gHvvvRdjDHl5eSQmJrJgwQIsy8Lr9bJr1y5SU1OJiYlh48aNzvsMTxb1xhtv0KdPH6qqqhAR6urqEBGCwSCPPPII9957Lz6fD4D09HQqKyvZu3cve/fuJTs7m1mzZvH222/zxBNPUF5eTrNmzUhLS+O3v/0t999/P2VlZezevZvJkyeze/duJk6ceFTXVqmm1BSDOjUgV0opdVopKyvj4osvpqys7IAAd39JSUlYlsWWLVsaBds/+clPWLx4McFgELfb7QSb+wsHrlVVVURERFBbW3vY4zU8RkP7f1CIiIigpqYGy7KcADcyMpKOHTuycOFCkpKSyM3N5cMPP3TSOgKBQP15RgKZwF7AD9QAa+x9H+/f7U2bNlFUVMTKlSudDyNHsmvXLi688EIKCgp4/fXXj+u4Sp1ptMqKUkqpc17v3r2dAYBHCj7Ly8vZsmULQKOA+IsvvnAeNyzpB5CWluY8rqurc2p3x8XFHfF4h6ql3fDYgFPPOyIiwllXU1PDkiVLALuXeuHChRhjqK2txRhDMBi0A2ULiAH6AVuAjtiBuQcwhz29Q3r++ee56KKLmDRp0lEH4wDx8fGsXr1ag3GlTjINyJVSSjW5mS/NJCktCWMMpZtK7VQNaBSA+v3+A17ncrmcMn4RERHOIEcRwbIsAFJTU4mKinJes3PnTudxRESE8/iA2S6PEPwe7Hz2V11d7TxOTU11Av6NGzcSDAYxxuByuWjVqhUxMTH4/X78Xj9sBaYCvbGD8u+BIBQXFx/xmAdz00038d133zFgwIDjer1S6uTSgFwppVSTmvnSTIaOGsr2H7aDF7i8wcoGQXG4JzssnCayePFiAAoLCwGcMoThnulwOkpYuNoJ0GhmSuo7zm1HyAwJB9vGGOeDQLi2dzhY/81vfoPP58MYQ0pKihMQp6amUlJSgsvl4pJLLmHt2rXU1NQwevRotm/fjtfnhSrgHeBDoA6KRxVTMkWnslfqbKQBuVJKqSZ1x/g7qPPUgQ8IAC2oD4at+u32T7UIBoNERETQpYud6hmuJhIurxcOkisrK0lISHBed8i0lGNIzQ6nmoT3F96niLBv3z7nw8Nrr71GQkICxhg+//xzZs6cCcC2bdvo1KkTgwYN4t///jc1NTVUV1ezdetWfD4fb/39LTq06UCnTp0Y879juPLKKzUYV+ospgG5UkqpJrV903a7gsgu7KB8a4OVDVK2ExMTAbtyidvtxuVyUVhYyMcff4zL5XIqlBhjiIuL4+qrrwagR48exMXFYYwhNTXVqRpygK37PTc0+kAA0LNnT8BOdUlOTsYYQ//+/Zk/fz4Al19+Ofn5+URERDh1uLOysujVqxc5OTkMHz6cqKgoHnjgAbZt28aSJUtIS0sjISGBTz/9lOeeew6APn36sHLlSr744guefPJJZs+efUzXVCl1ZtEqK0oppZqU8Rm7kghAMrARiAY2Y1caKcOZrMbr9bJo0SKysrK49NJL+frrr0lKSmLz5s1UVFSQkpLCrl27CAaDTJkyhaqqKn79619TV1dHSkoKCxcudILyQCBgp8i4sQdNPhA6hznAV8A+sPwWo4aNYsXyFUyePNnpjT+Y6OhoZ0bMQ6mqqiI/P58vvviCuLg4br/9dt577z3mzJlD+/btj/8iKqVOGC17eAgakCul1NnLGGMHxi7sSiKpwDqgOvS8+uhL/R1Lmb6klkls377dfuIDrgbSgW+BN+DFaS8y+IbBx/OWDuqDDz5gxIgRjB07ljvvvPOE7VcpdWJpQH4IGpArpdTZy/JbBNODsBY7j9vCDsa9QCZkBDMoW112wo8bHkxaVx3KizH1xy0eoQMolTpXaR1ypZRS55zRw0fbaSpZQAT2JDjNgQvBv9PPpImTTspxB98wmL9N/5udmx4KxhNbJPLisy9qMK6UOqU0IFdKKXVQ3bt3PyXHKZlSQvFNxVjfWbAPO02lHDI2ZzD9iemHTBuZMWMGmzZtcp7/6le/YsWKFcd07ME3DKb8+3IkYFdKKd9YfkLTVJRS6mhoQK6UUuoAZWVlVFRUHPX2M2bMYODAgcyaNQuAoqIi9k81HD9+PFFRUQwZMsQpGRj2pyf/RO0ee9p62SdIUChbXcaE+yYcOGFPg2M2DMifeeYZcnJyDnuODbdXSqnTxY8KyI0x/88Ys9IY85Ux5u/GmPgG6+41xpQaY1YZYy5rsLxPaFmpMeaeH3N8pZRSJ8/XX38NwPz58ykqKuKXv/wl559/PoMHD3YGWd5zzz3k5OQwYsQIPvnkEwBatmzJ+vXrnefR0dEAlJSU4Ha7Wb9+PVlZWVx77bX07NmTvn37kpOTQ11dHePHj+eCCy6gY8eOTJs2zTmXRx99lPz8fAoKCrjnnnuYNWsWixYtYvDgwRQWFrJ3716KiopYuHAhAC+//DL5+fnk5eXxm9/8BrAD8vPOO48JEyZQUFBA165d2bJly6m5mEopdRg/alCnMebnwFwRqTXGPAogIr8xxuQALwMXYk/x8AEQrue0Gnsy4A3Af4FBInLY7xh1UKdSSp1a//d//0fPnj1JSEhgx44djdaFp3uvq7MHQ1qW5Tx2uVxERUWxZ88eZ6ZMgG7duvHpp58C9iyVMTExrFmzBoD8/Hzuuusu7rjjDnbv3o2IkJSURF1dHS6XC4/Hw6ZNmxg3bhzTp0+nVatW9OjRg5UrV/LYY49RVFTE6NGjmTZtGn/84x9ZuXIlTz31FG3atOGnP/0p33zzDV26dKGkpIQ9e/aQmZnJihUrGD16NB999BGJiYkkJSUxY8YM0tL2n65TKXWuOeMGdYrI+yIS/t7xM6BV6HE/4BURCYjIOqAUOzi/ECgVkW9EpBp4JbStUkqp08jYsWMBqKiocGa8DPP7/YiIM3NmVFSUs87tdjvBuMfjwbLsmXXCPdfhbcrKygA7gL/hhhsYNmwY+/btQ0SIiopi69atBAIBtm7dyubNmwkGg7z44ov06NGDZcuWYVkWW7faM/ns2bOHiy66iC5dulBYWEhhYSHXXXcdK1euJBAIkJubS21tLV26dCEiIoJvvvkGt9vNZ599xk9/+lM+//xzbr75ZiZMmHDSrqdSSh3Oicwhvxl4L/S4JfBdg3UbQssOtVwppVQTmvnSTDLbZ+KyXGS2z8Tvt2fqCQaDB9QAr6qqIhgMOj3glZWVzrrq6mpneU1NjdNz3rC3vKKigpqaGmf5vffe60w5X1dXR0VFBSJC3759McYQHR2NZVls2bKFDz/8kMLCQt5//33Wr1/P9ddfD0C7du2c/S9dupQ5c+aQn5/P3LlzG+WNu91ujDGsWrWK7777jjlz5lBYWMjvf/97NmzYcMKup1JKHYsjBuTGmA+MMcsO8tOvwTYTgFpg5ok6MWPMKGPMImPMom3btp2o3SqllNrPzJdmMuy2Yazvvh6ZIKzPXs9HH30EgHDotMaYmBgAWrRo4SzLzc3F7XYDEB8f7/SuN5yuvrq62nlsjCEpKQmwe95dLheZmZkUFBQwe/ZsLMsiOjqaYDBIcnIyHTt2ZO7cuVx++eW0bt2ap59+Gp/Px/DhwwEIBAL89a9/xefzMW/ePEaMGMHixYvp1atXo3MXEVq3bk3fvn1ZsmQJS5cu5f333z/ua6iUUj+G+0gbiMilh1tvjBkGXAVcIvXdKBuB1g02axVaxmGW73/c6cB0sHPIj3SeSimljs8tY26htl8ttAkt2IzdXRME8QlUNd4+/Kt+3759ADTsNFmzZg1+v5+Kigp27tzpBOThbcHupQ7nmbvdboqKinjrrbec/YbTWcIa9pD/8MMPTJ48mQ8//JAbb7yRW265herqasrLy4mJiaG6uhrLspg0aRK9evVi7dq1/OQnP6Ffv34888wzzjE6dOhARUWFc+41NTWsXr2a3NzcE3BFlVLq2PzYKit9gLuBviLS8Ff2O8D1xhivMaYN0A74D/YgznbGmDbGGA9wfWhbpZRSTWT39t32lPFhtdizZYI9Uc8hhNNOGpYwDC8LExGMMY160Wtra/F4PM72s2bNwu/3Y1mWEzB7PB68Xi+RkZEAeL1ezjvvPABmz55NaWkpOTk5rFq1isjISIwxvPvuu1x88cWMHDmShx56iPj4eAYNGkTv3r0BGDZsGOnp6RQWFlJXV8c//vEPKisrKSgooLCwkAULFhzrpVNKqRPix1ZZKcWe3Hh7aNFnInJLaN0E7LzyWuBOEXkvtPwK4I/Yv+6fFZEjTsGmVVaUUurkuPW2W/nLM3+BQdT3kH+P/f2kADFAKEW8a9euLFq0yOn1jo6O5rzzzmPv3r2sWLECl8tFv379WLhwIbt27XJ6wY0xBAIB4uPj2bp1K6WlpVx99dV8/fXXtG/fnl27dlFVVUUgECAnJ4ft27fz0EMPMXToUK699lpWrVpFWVkZ+fn5AHz66aeMGTOG5s2bc//99zN//nzGjh3L4sWLT+3FU0qdlZqiysqPCshPFQ3IlVLq5HBHuamrqQM/cDV2T/m32IVrq7G/RxWI9EfSIrUFxhi2bt1KixYt6NChAx06dKB58+bMnj2bkSNH8tRTT1FRUUFZWRkbN25k1apVdO/eHWMMzZo1o7KykkAgwK233srzzz9PXV0d2dnZfPLJJ9x7771Mnz6dYDBIs2bNiImJ4dlnn+WWW26hrKyM2bNnM3nyZGbPns2OHTu4+eab+eabb/D7/UyfPp2OHTs23YVUSp01NCA/BA3IlVLq5DDGQBdgKXZQvguIB/ZgB+RxYNVa/G3633RKeaXUOeGMq0OulFLqzGb5LcgFCoC92GkqewEDeCHKRGkwrpRSJ5kG5EopdQ4bNWwUvAFkA+OBodgjg4DiEcXs3rlbg3GllDrJjlj2UCml1NmrZEoJAFOfnYrsE/AAAsUji511SimlTi7NIVdKKaWUUipEc8iVUkoppZQ6x2hArpRSSimlVBPSgFwppZRSSqkmpAG5UkoppZRSTUgDcqWUUkoppZqQBuRKKaWUUko1IQ3IlVJKKaWUakIakCullFJKKdWENCBXSimllFKqCWlArpRSSimlVBMyItLU53BExphtwPqmPo8mlASUN/VJqNOCtgUF2g5UPW0LKkzbwomTISLNT+UBz4iA/FxnjFkkIl2a+jxU09O2oEDbgaqnbUGFaVs4s2nKilJKKaWUUk1IA3KllFJKKaWakAbkZ4bpTX0C6rShbUGBtgNVT9uCCtO2cAbTHHKllFJKKaWakPaQK6WUUkop1YQ0ID/NGGP+nzFmpTHmK2PM340x8Q3W3WuMKTXGrDLGXNZgeZ/QslJjzD1NcuLqhDLGDDDGLDfGBI0xXfZbp+3gHKb3+dxijHnWGLPVGLOswbIEY8y/jDFrQv82Cy03xpg/hdrGV8aYzk135upEMsa0NsbMM8asCP1tuCO0XNvCWUID8tPPv4A8EekIrAbuBTDG5ADXA7lAH6DEGGMZYyxgCnA5kAMMCm2rzmzLgGuBjxou1HZwbtP7fE6agf1/vaF7gH+LSDvg36HnYLeLdqGfUcBfTtE5qpOvFhgnIjlAV+C20P99bQtnCQ3ITzMi8r6I1Iaefga0Cj3uB7wiIgERWQeUAheGfkpF5BsRqQZeCW2rzmAi8rWIrDrIKm0H5za9z+cYEfkI2LHf4n7A30KP/wZc3WD582L7DIg3xqSdkhNVJ5WIfC8iX4QeVwJfAy3RtnDW0ID89HYz8F7ocUvguwbrNoSWHWq5OjtpOzi36X1WACki8n3o8WYgJfRY28c5wBiTCXQCFqJt4azhbuoTOBcZYz4AUg+yaoKIvB3aZgL2V1QzT+W5qVPnaNqBUkodjoiIMUbLpZ0jjDHRwBvAnSJSYYxx1mlbOLNpQN4EROTSw603xgwDrgIukfq6lBuB1g02axVaxmGWq9PYkdrBIWg7OLcd7v6rc8cWY0yaiHwfSkPYGlqu7eMsZoyJwA7GZ4rIm6HF2hbOEpqycpoxxvQB7gb6ikhVg1XvANcbY7zGmDbYAzX+A/wXaGeMaWOM8WAP+HvnVJ+3OmW0HZzb9D4rsO/50NDjocDbDZbfFKqw0RX4oUE6gzqDGbsr/K/A1yLyeINV2hbOEtpDfvr5M+AF/hX6KuozEblFRJYbY14DVmCnstwmInUAxpj/Bf4JWMCzIrK8aU5dnSjGmGuAp4DmwLvGmCUicpm2g3ObiNTqfT63GGNeBoqAJGPMBuB3wCPAa8aYEcB6YGBo8znAFdiDvauA4af8hNXJ0gMYAiw1xiwJLbsPbQtnDZ2pUymllFJKqSakKStKKaWUUko1IQ3IlVJKKaWUakIakCullFJKKdWENCBXSimllFKqCWlArpRSSimlVBPSgFwppZRSSqkmpAG5UkoppZRSTUgDcqWUUkoppZrQ/wcVABgOwPDXfwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize embeddings\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "#\n",
    "words = ft_model.wv.index_to_key\n",
    "wvs = ft_model.wv[words]\n",
    "\n",
    "# instantiate tSNE\n",
    "tsne = TSNE(n_components=2, random_state=42, n_iter=5000, perplexity=5)\n",
    "\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "# train tSNE\n",
    "T = tsne.fit_transform(wvs)\n",
    "\n",
    "# get labels\n",
    "labels = words\n",
    "\n",
    "# generate the plot\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.scatter(T[:, 0], T[:, 1], c='green', edgecolors='k')\n",
    "for label, x, y in zip(labels, T[:, 0], T[:, 1]):\n",
    "    plt.annotate(label, xy=(x+1, y+1), xytext=(0, 0), textcoords='offset points')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "1d3c0a17",
   "metadata": {},
   "outputs": [],
   "source": [
    "def average_word_vectors(words, model, vocabulary, num_features):\n",
    "    \"\"\"Calculate the average word vectors.\"\"\"\n",
    "    \n",
    "    feature_vector = np.zeros((num_features,),dtype=\"float64\")\n",
    "    nwords = 0.\n",
    "    \n",
    "    for word in words:\n",
    "        if word in vocabulary: \n",
    "            nwords = nwords + 1.\n",
    "            feature_vector = np.add(feature_vector, model.wv[word])\n",
    "    \n",
    "    if nwords:\n",
    "        feature_vector = np.divide(feature_vector, nwords)\n",
    "        \n",
    "    return feature_vector\n",
    "\n",
    "\n",
    "def averaged_word_vectorizer(corpus, model, num_features):\n",
    "    vocabulary = set(model.wv.index_to_key)\n",
    "    features = [average_word_vectors(tokenized_sentence, model, vocabulary, num_features)\n",
    "                    for tokenized_sentence in corpus]\n",
    "    return np.array(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "dc1a116c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.107033</td>\n",
       "      <td>-0.257712</td>\n",
       "      <td>0.934672</td>\n",
       "      <td>0.559356</td>\n",
       "      <td>0.686412</td>\n",
       "      <td>0.447946</td>\n",
       "      <td>-0.869933</td>\n",
       "      <td>1.894799</td>\n",
       "      <td>1.459755</td>\n",
       "      <td>-1.916570</td>\n",
       "      <td>0.397112</td>\n",
       "      <td>0.515644</td>\n",
       "      <td>-0.852542</td>\n",
       "      <td>1.297885</td>\n",
       "      <td>0.549002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.254809</td>\n",
       "      <td>-0.601824</td>\n",
       "      <td>1.379148</td>\n",
       "      <td>-0.504156</td>\n",
       "      <td>1.079953</td>\n",
       "      <td>-0.067444</td>\n",
       "      <td>-0.208640</td>\n",
       "      <td>2.384728</td>\n",
       "      <td>1.941919</td>\n",
       "      <td>-1.668585</td>\n",
       "      <td>0.818902</td>\n",
       "      <td>0.808836</td>\n",
       "      <td>-0.674634</td>\n",
       "      <td>1.334193</td>\n",
       "      <td>0.660699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.200457</td>\n",
       "      <td>-0.251950</td>\n",
       "      <td>0.974955</td>\n",
       "      <td>-0.058585</td>\n",
       "      <td>0.529732</td>\n",
       "      <td>0.056667</td>\n",
       "      <td>-1.031909</td>\n",
       "      <td>2.050088</td>\n",
       "      <td>0.897333</td>\n",
       "      <td>-1.652671</td>\n",
       "      <td>1.246199</td>\n",
       "      <td>1.045728</td>\n",
       "      <td>-1.006129</td>\n",
       "      <td>1.082732</td>\n",
       "      <td>0.167535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.276970</td>\n",
       "      <td>-0.927038</td>\n",
       "      <td>1.301236</td>\n",
       "      <td>1.378255</td>\n",
       "      <td>2.153215</td>\n",
       "      <td>0.830994</td>\n",
       "      <td>-1.514208</td>\n",
       "      <td>2.617846</td>\n",
       "      <td>2.296189</td>\n",
       "      <td>-2.033484</td>\n",
       "      <td>0.183094</td>\n",
       "      <td>1.497957</td>\n",
       "      <td>-0.442312</td>\n",
       "      <td>2.048704</td>\n",
       "      <td>1.078270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.030012</td>\n",
       "      <td>-1.079105</td>\n",
       "      <td>1.141255</td>\n",
       "      <td>0.954837</td>\n",
       "      <td>1.089585</td>\n",
       "      <td>-0.739677</td>\n",
       "      <td>-0.837649</td>\n",
       "      <td>2.022358</td>\n",
       "      <td>0.571746</td>\n",
       "      <td>0.305362</td>\n",
       "      <td>-0.568516</td>\n",
       "      <td>1.962668</td>\n",
       "      <td>0.417151</td>\n",
       "      <td>0.705438</td>\n",
       "      <td>1.112067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3850</th>\n",
       "      <td>0.710247</td>\n",
       "      <td>-0.181846</td>\n",
       "      <td>1.662149</td>\n",
       "      <td>-0.244348</td>\n",
       "      <td>0.338349</td>\n",
       "      <td>0.084848</td>\n",
       "      <td>-0.439830</td>\n",
       "      <td>2.344689</td>\n",
       "      <td>2.584766</td>\n",
       "      <td>-2.150858</td>\n",
       "      <td>0.741672</td>\n",
       "      <td>0.486941</td>\n",
       "      <td>0.221942</td>\n",
       "      <td>0.534831</td>\n",
       "      <td>0.672667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3851</th>\n",
       "      <td>0.554523</td>\n",
       "      <td>0.388400</td>\n",
       "      <td>1.710742</td>\n",
       "      <td>-0.329672</td>\n",
       "      <td>1.001149</td>\n",
       "      <td>1.012695</td>\n",
       "      <td>-0.067734</td>\n",
       "      <td>2.623292</td>\n",
       "      <td>2.291972</td>\n",
       "      <td>-1.810588</td>\n",
       "      <td>1.093883</td>\n",
       "      <td>-0.345498</td>\n",
       "      <td>-0.213911</td>\n",
       "      <td>0.827246</td>\n",
       "      <td>0.766652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3852</th>\n",
       "      <td>0.560451</td>\n",
       "      <td>0.008423</td>\n",
       "      <td>1.499729</td>\n",
       "      <td>-0.494843</td>\n",
       "      <td>1.720102</td>\n",
       "      <td>0.195239</td>\n",
       "      <td>-0.689705</td>\n",
       "      <td>2.700927</td>\n",
       "      <td>1.748089</td>\n",
       "      <td>-0.899192</td>\n",
       "      <td>1.334562</td>\n",
       "      <td>1.449301</td>\n",
       "      <td>-1.035777</td>\n",
       "      <td>0.849274</td>\n",
       "      <td>1.448733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3853</th>\n",
       "      <td>0.319070</td>\n",
       "      <td>-0.642746</td>\n",
       "      <td>0.823955</td>\n",
       "      <td>-0.326663</td>\n",
       "      <td>1.859310</td>\n",
       "      <td>1.282834</td>\n",
       "      <td>-0.756218</td>\n",
       "      <td>1.870903</td>\n",
       "      <td>1.425502</td>\n",
       "      <td>-0.000835</td>\n",
       "      <td>1.503368</td>\n",
       "      <td>1.952871</td>\n",
       "      <td>0.070900</td>\n",
       "      <td>1.094950</td>\n",
       "      <td>0.973991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3854</th>\n",
       "      <td>0.540972</td>\n",
       "      <td>-0.072429</td>\n",
       "      <td>1.342036</td>\n",
       "      <td>-0.539194</td>\n",
       "      <td>1.560155</td>\n",
       "      <td>0.096151</td>\n",
       "      <td>-0.438405</td>\n",
       "      <td>2.930018</td>\n",
       "      <td>1.393340</td>\n",
       "      <td>-1.632607</td>\n",
       "      <td>0.824118</td>\n",
       "      <td>0.903267</td>\n",
       "      <td>0.078116</td>\n",
       "      <td>1.690626</td>\n",
       "      <td>1.279346</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3855 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            0         1         2         3         4         5         6   \\\n",
       "0     1.107033 -0.257712  0.934672  0.559356  0.686412  0.447946 -0.869933   \n",
       "1     1.254809 -0.601824  1.379148 -0.504156  1.079953 -0.067444 -0.208640   \n",
       "2     1.200457 -0.251950  0.974955 -0.058585  0.529732  0.056667 -1.031909   \n",
       "3     0.276970 -0.927038  1.301236  1.378255  2.153215  0.830994 -1.514208   \n",
       "4    -0.030012 -1.079105  1.141255  0.954837  1.089585 -0.739677 -0.837649   \n",
       "...        ...       ...       ...       ...       ...       ...       ...   \n",
       "3850  0.710247 -0.181846  1.662149 -0.244348  0.338349  0.084848 -0.439830   \n",
       "3851  0.554523  0.388400  1.710742 -0.329672  1.001149  1.012695 -0.067734   \n",
       "3852  0.560451  0.008423  1.499729 -0.494843  1.720102  0.195239 -0.689705   \n",
       "3853  0.319070 -0.642746  0.823955 -0.326663  1.859310  1.282834 -0.756218   \n",
       "3854  0.540972 -0.072429  1.342036 -0.539194  1.560155  0.096151 -0.438405   \n",
       "\n",
       "            7         8         9         10        11        12        13  \\\n",
       "0     1.894799  1.459755 -1.916570  0.397112  0.515644 -0.852542  1.297885   \n",
       "1     2.384728  1.941919 -1.668585  0.818902  0.808836 -0.674634  1.334193   \n",
       "2     2.050088  0.897333 -1.652671  1.246199  1.045728 -1.006129  1.082732   \n",
       "3     2.617846  2.296189 -2.033484  0.183094  1.497957 -0.442312  2.048704   \n",
       "4     2.022358  0.571746  0.305362 -0.568516  1.962668  0.417151  0.705438   \n",
       "...        ...       ...       ...       ...       ...       ...       ...   \n",
       "3850  2.344689  2.584766 -2.150858  0.741672  0.486941  0.221942  0.534831   \n",
       "3851  2.623292  2.291972 -1.810588  1.093883 -0.345498 -0.213911  0.827246   \n",
       "3852  2.700927  1.748089 -0.899192  1.334562  1.449301 -1.035777  0.849274   \n",
       "3853  1.870903  1.425502 -0.000835  1.503368  1.952871  0.070900  1.094950   \n",
       "3854  2.930018  1.393340 -1.632607  0.824118  0.903267  0.078116  1.690626   \n",
       "\n",
       "            14  \n",
       "0     0.549002  \n",
       "1     0.660699  \n",
       "2     0.167535  \n",
       "3     1.078270  \n",
       "4     1.112067  \n",
       "...        ...  \n",
       "3850  0.672667  \n",
       "3851  0.766652  \n",
       "3852  1.448733  \n",
       "3853  0.973991  \n",
       "3854  1.279346  \n",
       "\n",
       "[3855 rows x 15 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get document level embeddings\n",
    "ft_doc_features = averaged_word_vectorizer(corpus=tokenized_corpus, model=ft_model,\n",
    "                                             num_features=feature_size)\n",
    "\n",
    "# createa a pandas DataFrame\n",
    "pd.DataFrame(ft_doc_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a49892d",
   "metadata": {},
   "source": [
    "<a name=\"AffinityPropagation\"></a>\n",
    "## 2.4 Document Clustering with Affinity Propagation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59511040",
   "metadata": {},
   "source": [
    "Now that we have our **features for each document**, we can cluster these documents using the **Affinity Propagation** algorithm, which is a clustering algorithm based on the concept of “**message passing**” between data points and does not need the number of clusters as an explicit input which is often required by partition-based clustering algorithms.\n",
    "\n",
    "*More info on Affinity Propagation algorithm [here](https://www.geeksforgeeks.org/affinity-propagation-in-ml-to-find-the-number-of-clusters/)*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "b1755b09",
   "metadata": {
    "scrolled": true
   },
   "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>Sentences</th>\n",
       "      <th>Cluster Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Project Gutenberg eBook of The Odyssey, by Homer\\n\\nThis eBook is for the use of anyone anywhere in the United States and\\nmost other parts of the world at no cost and with almost no restricti...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>You may copy it, give it away or re-use it under the terms\\nof the Project Gutenberg License included with this eBook or online at\\nwww.gutenberg.org.</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>If you are not located in the United States, you\\nwill have to check the laws of the country where you are located before\\nusing this eBook.</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Title: The Odyssey\\n\\nAuthor: Homer\\n\\nTranslator: Samuel Butler\\n\\nRelease Date: April, 1999 [eBook #1727]\\n[Most recently updated: January 4, 2022]\\n\\nLanguage: English\\n\\n\\nProduced by: Jim Tin...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BOOK II.</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3850</th>\n",
       "      <td>General Information About Project Gutenberg-tm electronic works\\n\\nProfessor Michael S. Hart was the originator of the Project\\nGutenberg-tm concept of a library of electronic works that could be\\...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3851</th>\n",
       "      <td>For forty years, he produced and\\ndistributed Project Gutenberg-tm eBooks with only a loose network of\\nvolunteer support.</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3852</th>\n",
       "      <td>Project Gutenberg-tm eBooks are often created from several printed\\neditions, all of which are confirmed as not protected by copyright in\\nthe U.S. unless a copyright notice is included.</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3853</th>\n",
       "      <td>Thus, we do not\\nnecessarily keep eBooks in compliance with any particular paper\\nedition.</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3854</th>\n",
       "      <td>Most people start at our website which has the main PG search\\nfacility: www.gutenberg.org\\n\\nThis website includes information about Project Gutenberg-tm,\\nincluding how to make donations to the ...</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3855 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                                                                                                                                                    Sentences  \\\n",
       "0     The Project Gutenberg eBook of The Odyssey, by Homer\\n\\nThis eBook is for the use of anyone anywhere in the United States and\\nmost other parts of the world at no cost and with almost no restricti...   \n",
       "1                                                      You may copy it, give it away or re-use it under the terms\\nof the Project Gutenberg License included with this eBook or online at\\nwww.gutenberg.org.   \n",
       "2                                                                If you are not located in the United States, you\\nwill have to check the laws of the country where you are located before\\nusing this eBook.   \n",
       "3     Title: The Odyssey\\n\\nAuthor: Homer\\n\\nTranslator: Samuel Butler\\n\\nRelease Date: April, 1999 [eBook #1727]\\n[Most recently updated: January 4, 2022]\\n\\nLanguage: English\\n\\n\\nProduced by: Jim Tin...   \n",
       "4                                                                                                                                                                                                    BOOK II.   \n",
       "...                                                                                                                                                                                                       ...   \n",
       "3850  General Information About Project Gutenberg-tm electronic works\\n\\nProfessor Michael S. Hart was the originator of the Project\\nGutenberg-tm concept of a library of electronic works that could be\\...   \n",
       "3851                                                                               For forty years, he produced and\\ndistributed Project Gutenberg-tm eBooks with only a loose network of\\nvolunteer support.   \n",
       "3852               Project Gutenberg-tm eBooks are often created from several printed\\neditions, all of which are confirmed as not protected by copyright in\\nthe U.S. unless a copyright notice is included.   \n",
       "3853                                                                                                               Thus, we do not\\nnecessarily keep eBooks in compliance with any particular paper\\nedition.   \n",
       "3854  Most people start at our website which has the main PG search\\nfacility: www.gutenberg.org\\n\\nThis website includes information about Project Gutenberg-tm,\\nincluding how to make donations to the ...   \n",
       "\n",
       "      Cluster Label  \n",
       "0                24  \n",
       "1                24  \n",
       "2                24  \n",
       "3                24  \n",
       "4                 0  \n",
       "...             ...  \n",
       "3850             24  \n",
       "3851             24  \n",
       "3852             24  \n",
       "3853             19  \n",
       "3854             24  \n",
       "\n",
       "[3855 rows x 2 columns]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import AffinityPropagation\n",
    "\n",
    "# instantiate AffinityPropagation\n",
    "ap = AffinityPropagation(random_state=42, preference=-50, damping=0.9, max_iter=5000)\n",
    "\n",
    "# train AP\n",
    "ap.fit(ft_doc_features)\n",
    "\n",
    "# get labels\n",
    "cluster_labels = ap.labels_\n",
    "\n",
    "# create pandas DataFrame\n",
    "cluster_labels = pd.DataFrame(cluster_labels, \n",
    "                              columns=['Cluster Label'])\n",
    "\n",
    "# concatenate the 2 DataFrames\n",
    "pd.concat([corpus_df, cluster_labels], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "705a4e69",
   "metadata": {},
   "source": [
    "__PCA Visualization__\n",
    "\n",
    "We can also **visualize how each document in positioned in each cluster** by using **Principal Component Analysis** (PCA) to **reduce the feature dimensions to 2-D**.\n",
    "\n",
    "*More info on PCA [here](https://www.youtube.com/watch?v=HMOI_lkzW08)*."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "c56c6a23",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "\n",
    "# instantiate PCA\n",
    "pca = PCA(n_components=2, random_state=42)\n",
    "\n",
    "# train PCA\n",
    "pcs = pca.fit_transform(ft_doc_features)\n",
    "\n",
    "# get labels\n",
    "labels = ap.labels_\n",
    "\n",
    "# generate visualization\n",
    "categories = list(cluster_labels['Cluster Label'])\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "for i in range(len(labels)):\n",
    "    label = labels[i]\n",
    "    color = 'orange' if label == 0 else 'blue' if label == 1 else 'green'\n",
    "    annotation_label = categories[i]\n",
    "    x, y = pcs[i]\n",
    "    plt.scatter(x, y, c=color, edgecolors='k')\n",
    "    plt.annotate(annotation_label, xy=(x+1e-2, y+1e-2), xytext=(0, 0), \n",
    "                 textcoords='offset points')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76daa077",
   "metadata": {},
   "source": [
    "<a name=\"corex\"></a>\n",
    "# 2.5 CorEx"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bb0a1be",
   "metadata": {},
   "source": [
    "_More info on [CorEx](https://github.com/gregversteeg/corex_topic/blob/master/corextopic/example/corex_topic_example.ipynb)._"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33fa5b4a",
   "metadata": {},
   "source": [
    "The topic model assumes input is in the form of a __doc-word matrix__, where rows are documents and columns are binary counts. \n",
    "\n",
    "We'll vectorize the data, take the top 20,000 words, and convert it to a sparse matrix to save on memory usage. \n",
    "\n",
    "Note, we use __binary count vectors as input to the `CorEx` topic model__."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "9e5fa560",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3855, 5725)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import scipy.sparse as ss\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "\n",
    "# transform data into a sparse matrix\n",
    "vectorizer = CountVectorizer(stop_words='english', max_features=20000, binary=True)\n",
    "doc_word = vectorizer.fit_transform(norm_corpus)\n",
    "doc_word = ss.csr_matrix(doc_word)\n",
    "\n",
    "doc_word.shape # n_docs x m_words"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f44432d",
   "metadata": {},
   "source": [
    "Our doc-word matrix is 3,855 sentences by 5,725 words. \n",
    "\n",
    "Let's get the words that __label the columns__. We'll need these for outputting __readable topics__ and later for __anchoring__."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "d104e475",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['abandon', 'abate', 'abet', 'abhorred', 'abide', 'abides', 'able', 'aboard', 'abode', 'abominable']\n"
     ]
    }
   ],
   "source": [
    "# Get words that label the columns (needed to extract readable topics and make anchoring easier)\n",
    "words = list(np.asarray(vectorizer.get_feature_names()))\n",
    "print(words[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c6eabcd",
   "metadata": {},
   "source": [
    "The __main parameters__ of the CorEx topic model are:\n",
    "\n",
    "* `n_hidden`: number of topics (\"hidden\" as in \"hidden latent topics\")\n",
    "* `words`: words that label the columns of the doc-word matrix (optional)\n",
    "* `docs`: document labels that label the rows of the doc-word matrix (optional)\n",
    "* `max_iter`: number of iterations to run through the update equations (optional, defaults to 200)\n",
    "* `verbose`: if verbose=1, then CorEx will print the topic TCs with each iteration\n",
    "* `seed`: random number seed to use for model initialization (optional)\n",
    "\n",
    "\n",
    "We'll train a topic model with 25 topics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "50942d97",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import corextopic.corextopic as ct\n",
    "import corextopic.vis_topic as vt\n",
    "\n",
    "# Train the CorEx topic model with 4 topics\n",
    "topic_model = ct.Corex(n_hidden=4, words=words, max_iter=1000, verbose=0, seed=1)\n",
    "\n",
    "# fit model to data\n",
    "topic_model.fit(doc_word, words=words);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ccc28d6",
   "metadata": {},
   "source": [
    "The CorEx topic model provides functionality for easily __accessing the topics__. Let's take a look one of the topics."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "ef7d1f2d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('say', 0.09926673241572775, 1.0),\n",
       " ('father', 0.05857472767832096, 1.0),\n",
       " ('answer', 0.04717486517435672, 1.0),\n",
       " ('son', 0.04395668562172449, 1.0),\n",
       " ('ulysses', 0.03964742480368083, 1.0),\n",
       " ('come', 0.029782346232644363, 1.0),\n",
       " ('shall', 0.029745672809366908, 1.0),\n",
       " ('man', 0.029329201780131542, 1.0),\n",
       " ('home', 0.025664982403053024, 1.0),\n",
       " ('house', 0.02381322786584103, 1.0)]"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print a single topic from CorEx topic model\n",
    "topic_model.get_topics(topic=0, n_words=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9a946b75",
   "metadata": {},
   "source": [
    "The topic words are those with the __highest mutual information with the topic__, rather than those with highest probability within the topic as in LDA. \n",
    "\n",
    "The mutual information with the topic is the number reported in each tuple. CorEx also returns the \"sign\" of each word, which is either `1` or `-1`. If the sign is `-1`, then that means the __absence of a word is informative__ in that topic, rather than its presence.\n",
    "\n",
    "If the column labels have not been specified through words, then the code will return the column indices for the top words in each topic.\n",
    "\n",
    "We can also retrieve all of the topics at once if we would like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "6b3ffa38",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0: say, father, answer, son, ulysses, come, shall, man, home, house\n",
      "1: project, work, odyssey, gutenberg, copy, electronic, line, term, foundation, license\n",
      "2: ship, sea, wind, water, sail, land, shore, wave, rock, day\n",
      "3: wine, drink, hand, make, head, cloak, round, eat, meat, shoulder\n"
     ]
    }
   ],
   "source": [
    "# Print all topics from the CorEx topic model\n",
    "topics = topic_model.get_topics()\n",
    "for n,topic in enumerate(topics):\n",
    "    topic_words,_,_ = zip(*topic)\n",
    "    print('{}: '.format(n) + ', '.join(topic_words))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb00c150",
   "metadata": {},
   "source": [
    "We can also get the column indices instead of the column labels if necessary."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "d99bd878",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(4306, 0.09926673241572775, 1.0),\n",
       " (1836, 0.05857472767832096, 1.0),\n",
       " (214, 0.04717486517435672, 1.0),\n",
       " (4611, 0.04395668562172449, 1.0),\n",
       " (5218, 0.03964742480368083, 1.0),\n",
       " (920, 0.029782346232644363, 1.0),\n",
       " (4410, 0.029745672809366908, 1.0),\n",
       " (3033, 0.029329201780131542, 1.0),\n",
       " (2422, 0.025664982403053024, 1.0),\n",
       " (2459, 0.02381322786584103, 1.0)]"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_model.get_topics(topic=0, n_words=10, print_words=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e629da41",
   "metadata": {},
   "source": [
    "If we need to directly access the topic assignments for each word, they can be accessed through `clusters`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "119d39b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 2 1 ... 1 1 3]\n",
      "(5725,)\n"
     ]
    }
   ],
   "source": [
    "print(topic_model.clusters)\n",
    "print(topic_model.clusters.shape) # m_words"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7bc69511",
   "metadata": {},
   "source": [
    "As with the topic words, the __most probable documents per topic can also be easily accessed__. \n",
    "\n",
    "Documents are sorted according to log probabilities which is why the highest probability documents have a score of `0` ($e^0=1$) and other documents have negative scores (for example, $e^{-0.5} \\approx 0.6$)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "db0ab3fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NOTE: 'docs' not provided to CorEx. Returning top docs as lists of row indices\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[(2969, 0.0),\n",
       " (1459, 0.0),\n",
       " (1088, 0.0),\n",
       " (560, 0.0),\n",
       " (2534, 0.0),\n",
       " (1090, 0.0),\n",
       " (191, 0.0),\n",
       " (3137, 0.0),\n",
       " (2118, 0.0),\n",
       " (364, 0.0)]"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print a single topic from CorEx topic model\n",
    "topic_model.get_top_docs(topic=0, n_docs=10, sort_by='log_prob')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "37b18695",
   "metadata": {},
   "source": [
    "`CorEx` is a __discriminative model__, whereas `LDA` is a __generative model__. \n",
    "\n",
    "This means that while `LDA` outputs a __probability distribution over each document__, `CorEx` instead estimates the __probability a document belongs to a topic given that document's words__. \n",
    "\n",
    "As a result, the probabilities across topics for a given document do not have to add up to 1. The estimated probabilities of topics for each document can be accessed through `log_p_y_given_x` or `p_y_given_x`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "id": "494bf39c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3855, 4)\n"
     ]
    }
   ],
   "source": [
    "print(topic_model.p_y_given_x.shape) # n_docs x k_topics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8fe43335",
   "metadata": {},
   "source": [
    "We can also use a `softmax` to make a binary determination of which documents belong to each topic. \n",
    "\n",
    "These softmax labels can be accessed through `labels`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "c9f63a03",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3855, 4)\n"
     ]
    }
   ],
   "source": [
    "print(topic_model.labels.shape) # n_docs x k_topics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52561258",
   "metadata": {},
   "source": [
    "Since CorEx does not prescribe a probability distribution of topics over each document, this means that a document could possibly belong to no topics (all 0's across topics in `labels`) or all topics (all 1's across topics in `labels`)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45d9445f",
   "metadata": {},
   "source": [
    "__Overall TC__\n",
    "\n",
    "Total correlation is the measure which `CorEx` maximize when constructing the topic model. It can be accessed through `tc` and is reported in nats."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "id": "c212e433",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4.100564503284473"
      ]
     },
     "execution_count": 183,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_model.tc"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5bb82d3",
   "metadata": {},
   "source": [
    "__Model selection__\n",
    "\n",
    "`CorEx` starts its algorithm with a random initialization, and so different runs can result in different topic models. \n",
    "\n",
    "One way of finding a better topic model is to restart the `CorEx` algorithm several times and take the run that has the highest TC value (i.e. the run that produces topics that are most informative about the documents)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50a8fbc7",
   "metadata": {},
   "source": [
    "__Topic TC__\n",
    "\n",
    "\n",
    "The overall total correlation is the sum of the total correlation per each topic. These can be accessed through `tcs`. \n",
    "\n",
    "For an __unsupervised CorEx topic model__, the topics are always sorted from high to low according to their TC. \n",
    "\n",
    "For an __anchored CorEx topic model__, the topics are not sorted, and are outputted such that the anchored topics come first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "id": "48d4877d",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4,)"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_model.tcs.shape # k_topics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "id": "efce013a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.14 , 1.032, 0.984, 0.945])"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_model.tcs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "962b3654",
   "metadata": {},
   "source": [
    "__Selecting the Number of Topics__\n",
    "\n",
    "\n",
    "One way to choose the number of topics is to observe the distribution of TCs for each topic to see how much each additional topic contributes to the overall TC. \n",
    "\n",
    "We should keep adding topics until additional topics do not significantly contribute to the overall TC. This is similar to choosing a __cutoff eigenvalue__ when doing topic modeling via `LSA`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "id": "9f58f487",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,5))\n",
    "plt.bar(range(topic_model.tcs.shape[0]), topic_model.tcs, color='#4e79a7', width=0.5)\n",
    "plt.xlabel('Topic', fontsize=16)\n",
    "plt.ylabel('Total Correlation (nats)', fontsize=16);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2489f458",
   "metadata": {},
   "source": [
    "After initializing `CorEx` with __25 topics__ from which only the first 4 were the __most informative__, we __re-initialized the model with just those 4__. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "522ef268",
   "metadata": {},
   "source": [
    "__Pointwise Document TC__\n",
    "\n",
    "\n",
    "We can decompose total correlation further. The topic correlation is the average of the pointwise total correlations for each individual document. The pointwise total correlations can be accessed through `log_z`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "id": "e1972c53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.333, 12.089,  1.1  ,  1.298],\n",
       "       [ 1.539,  8.207,  1.1  ,  1.298],\n",
       "       [ 1.516,  4.404,  1.098,  1.297],\n",
       "       ...,\n",
       "       [ 2.121, 10.033,  1.107,  1.661],\n",
       "       [ 2.121,  4.474,  1.1  ,  1.298],\n",
       "       [-0.227, 26.037,  1.099,  0.884]])"
      ]
     },
     "execution_count": 189,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "topic_model.log_z"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22ec8080",
   "metadata": {},
   "source": [
    "__Anchoring for Semi-Supervised Topic Modeling__\n",
    "\n",
    "\n",
    "Anchored CorEx is an extension of CorEx that allows the \"anchoring\" of words to topics. When anchoring a word to a topic, CorEx is trying to __maximize the mutual information between that word and the anchored topic__. \n",
    "\n",
    "So, anchoring provides a way to __guide the topic model towards specific subsets of words that the user would like to explore__.\n",
    "\n",
    "__Anchoring a single set of words to a single topic:__ this can help promote a topic that did not naturally emerge when running an unsupervised instance of the CorEx topic model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "id": "4f69ac5c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Anchor one word to the first topic\n",
    "anchor_words = ['ulysses']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "id": "947b9249",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Anchor the word 'ulysses' to the first topic\n",
    "anchored_topic_model = ct.Corex(n_hidden=4, seed=2)\n",
    "anchored_topic_model.fit(doc_word, words=words, anchors=anchor_words, anchor_strength=5 );"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "id": "0d9cdaa4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0: ulysses, suitor, laertes, eumaeus, swineherd, noble, stockman, alcinous, forget, hit\n"
     ]
    }
   ],
   "source": [
    "topic_words,_,_ = zip(*anchored_topic_model.get_topics(topic=0))\n",
    "print('0: ' + ', '.join(topic_words))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a992d78",
   "metadata": {},
   "source": [
    "__Anchoring single sets of words to multiple topics:__ This can help find different aspects of a topic that may be discussed in several different contexts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "id": "4bc1200d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Anchor 'ulysses' and 'odysseus' to first topic, 'penelope' and 'wife' to second topic, so on...\n",
    "anchor_words = [['ulysses', 'odysseus'], ['penelope', 'wife'], ['telemachus', 'son'], ['poseidon', 'god']]\n",
    "\n",
    "anchored_topic_model = ct.Corex(n_hidden=4, seed=2)\n",
    "anchored_topic_model.fit(doc_word, words=words, anchors=anchor_words, anchor_strength=5);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "id": "95c2113e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0: ulysses, odysseus, return, eumaeus, noble, swineherd, long, minerva, stockman, answered\n",
      "1: penelope, wife, say, house, servant, hand, make, good, woman, shall\n",
      "2: son, telemachus, father, answer, atreus, pisistratus, laertes, nestor, tell, reply\n",
      "3: god, sea, wind, ship, wave, jove, sail, raise, rock, heaven\n"
     ]
    }
   ],
   "source": [
    "for n in range(len(anchor_words)):\n",
    "    topic_words,_,_ = zip(*anchored_topic_model.get_topics(topic=n))\n",
    "    print('{}: '.format(n) + ', '.join(topic_words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "id": "fcafa91d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# instantiate tSNE\n",
    "tsne = TSNE(n_components=2, random_state=42, n_iter=5000, perplexity=3)\n",
    "\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "# fit tsne to GloVe vectors\n",
    "T = tsne.fit_transform(anchored_topic_model.log_z)\n",
    "\n",
    "# create labels\n",
    "labels = anchor_words\n",
    "\n",
    "# create visualization\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.scatter(T[:, 0], T[:, 1], c='red', edgecolors='r')\n",
    "for label, x, y in zip(labels, T[:, 0], T[:, 1]):\n",
    "    plt.annotate(label, xy=(x+1, y+1), xytext=(0, 0), textcoords='offset points')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a1572b64",
   "metadata": {},
   "source": [
    "__Anchoring different sets of words to multiple topics:__ This can help enforce topic separability if there appear to be chimera topics.\n",
    "\n",
    "We can continue to develop even more involved anchoring strategies. Here we anchor \"ulysses\" by itself, as well as in two other topics each with \"man\", \"travel\", and 'ithaca' to find different aspects around the word \"ulysses\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "id": "07208cfd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Anchor with single words and groups of words\n",
    "anchor_words = ['ulysses', ['ulysses', 'man'], ['ulysses', 'travel'], ['ulysses', 'ithaca']]\n",
    "\n",
    "anchored_topic_model = ct.Corex(n_hidden=4, seed=2)\n",
    "anchored_topic_model.fit(doc_word, words=words, anchors=anchor_words, anchor_strength=3);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "id": "b6ad51b8",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0: ulysses, laertes, minerva, stockman, mind, throw, shoulder, hit, sang, skill\n",
      "1: ulysses, man, house, good, son, make, hand, drink, shall, father\n",
      "2: ulysses, travel, bow, noble, phaeacians, alcinous, king, string, trial, forget\n",
      "3: ulysses, ithaca, say, answer, home, suitor, tell, come, telemachus, speak\n"
     ]
    }
   ],
   "source": [
    "for n in range(len(anchor_words)):\n",
    "    topic_words,_,_ = zip(*anchored_topic_model.get_topics(topic=n))\n",
    "    print('{}: '.format(n) + ', '.join(topic_words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "id": "adb939b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Print topics in text file\n"
     ]
    }
   ],
   "source": [
    "from corextopic import vis_topic as vt\n",
    "vt.vis_rep(topic_model, column_label=words, prefix='topic-model-example')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "9fcf69ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# instantiate tSNE\n",
    "tsne = TSNE(n_components=2, random_state=42, n_iter=5000, perplexity=3)\n",
    "\n",
    "np.set_printoptions(suppress=True)\n",
    "\n",
    "# fit tsne to GloVe vectors\n",
    "T = tsne.fit_transform(anchored_topic_model.log_z)\n",
    "\n",
    "# create labels\n",
    "labels = anchor_words\n",
    "\n",
    "# create visualization\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.scatter(T[:, 0], T[:, 1], c='red', edgecolors='r')\n",
    "for label, x, y in zip(labels, T[:, 0], T[:, 1]):\n",
    "    plt.annotate(label, xy=(x+1, y+1), xytext=(0, 0), textcoords='offset points')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9848ea0c",
   "metadata": {},
   "source": [
    "<a name=\"conc\"></a>\n",
    "# 6. Conclusions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "95a9a0b0",
   "metadata": {},
   "source": [
    "Deep Learning models are entirely new to me, thus, I should probably have tried less models!\n",
    "\n",
    "The __anchored CorEx model__ seemed to perform far better than __LDA__ as the following output makes perfect sense:\n",
    "\n",
    ">0: ulysses, odysseus, return, eumaeus, noble, swineherd, long, minerva, stockman, answered <br>\n",
    "1: penelope, wife, say, house, servant, hand, make, good, woman, shall <br>\n",
    "2: son, telemachus, father, answer, atreus, pisistratus, laertes, nestor, tell, reply <br>\n",
    "3: god, sea, wind, ship, wave, jove, sail, raise, rock, heaven\n",
    "\n",
    "I did not manage to find a proper way to visualize the topics and the clustering of the topics/words, and got a bit lost on this project while searching for a way to do that. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}