{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5886e925",
   "metadata": {
    "id": "5886e925"
   },
   "source": [
    "# Lecture 10 - More Text Normalization and NLP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b788b5bb",
   "metadata": {
    "executionInfo": {
     "elapsed": 1224,
     "status": "ok",
     "timestamp": 1675189729193,
     "user": {
      "displayName": "Scott Wehrwein",
      "userId": "11327482518794216604"
     },
     "user_tz": 480
    },
    "id": "b788b5bb"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import spacy"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f95f47be",
   "metadata": {
    "id": "f95f47be"
   },
   "source": [
    "# Announcements:\n",
    "* Lab 4 pre-lab is available\n",
    "* Data Ethics 2 moved to Monday\n",
    "  * Before class: Read an article and answers a few questions\n",
    "  * In class: hands-on data analysis activity, reproducing some of the analysis/results from the article\n",
    "* This is a good time to be thinking about project topics\n",
    "\n",
    "# Goals:\n",
    "* Know the meaning and purpose of some basic text normalization operations (from natual language processing):\n",
    "  * Sentence tokenization\n",
    "  * Lowercasing, contractions, punctuation, canonicalization\n",
    "  * Stemming\n",
    "  * Lemmatization\n",
    "  * Stopword removal\n",
    "* Know how to use a Counter to count frequencies\n",
    "* Understand the notion of a word vector and calculating similarity among words\n",
    "\n",
    "* Get some hands-on practice using the above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7d6a5da8-f4b3-45ad-af86-481bfabc0c8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load(\"en_core_web_sm\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7fd4f078-5b64-4ebb-b660-6ef88f592514",
   "metadata": {},
   "source": [
    "### Counting Frequencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2deaa412-b509-4df3-bb28-e8c2fd86e079",
   "metadata": {},
   "outputs": [],
   "source": [
    "quote = \"You can please some of the people all of the time, you can please all of the people some of the time, but you can't please all of the people all of the time\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "555dc79a-9c3e-4f58-ae30-fdbed70a5b6f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'You can please some of the people all of the time, you can please all of the people some of the time, but you can not please all of the people all of the time'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quote = quote.replace(\"can't\", \"can not\")\n",
    "quote"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "82d29af2-6ff4-49a6-bd74-71ae2d588440",
   "metadata": {},
   "outputs": [],
   "source": [
    "q = nlp(quote)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e9e16e65-ef42-4488-b5d1-01fc205ad004",
   "metadata": {},
   "outputs": [],
   "source": [
    "import collections\n",
    "\n",
    "counts = collections.Counter([str(x) for x in q])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a88ddf94-20e3-4ee7-af78-174e70be73d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('of', 6),\n",
       " ('the', 6),\n",
       " ('all', 4),\n",
       " ('can', 3),\n",
       " ('please', 3),\n",
       " ('people', 3),\n",
       " ('time', 3),\n",
       " ('some', 2),\n",
       " (',', 2),\n",
       " ('you', 2),\n",
       " ('You', 1),\n",
       " ('but', 1),\n",
       " ('not', 1)]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts.most_common()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c54ddb03-6a48-4c33-909a-6740a2650b61",
   "metadata": {},
   "source": [
    "### Word Vectors and Similarity"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6445c39-7c08-4731-b58a-ce6dca67f464",
   "metadata": {},
   "source": [
    "Language models often have a way to map a word or token to a **vector** that, somehow, encodes the word's meaning in a bunch of numbers. These are often called **embedding vectors**, and we'll be seeing more about them later in the course.\n",
    "\n",
    "But for now, let's just take a look at one thing this enables: similarity calculations."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f093d3e-2fef-4584-8630-f99405a12e66",
   "metadata": {},
   "source": [
    "The small language model we've been using (`en_core_web_sm`) does not have word vectors built into it, so we'll reach for a larger one that does. You can still compute similarities with the `_sm` model, but they won't be as intelligent."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1500ed15-ef24-4648-95fe-7eef36288bff",
   "metadata": {},
   "source": [
    "The small model is now in the default JupyterHub environment, but the large one is not, so for the moment I have to do this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "732abebe-c414-43bc-b791-1284cecff11b",
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load(\"en_core_web_lg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ccc203b5-e9bf-4627-9591-a37e7fd781db",
   "metadata": {},
   "outputs": [],
   "source": [
    "words = \"cat dog tiger king queen castle waffle pancake\" \n",
    "ans = nlp(words)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b580734f-238a-4660-b487-0ab9f61fb6b0",
   "metadata": {},
   "source": [
    "Here's the word vector for \"cat\":"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "542f11c6-e264-427c-bf6d-1bf9686a0e06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.15067  , -0.024468 , -0.23368  , -0.23378  , -0.18382  ,\n",
       "        0.32711  , -0.22084  , -0.28777  ,  0.12759  ,  1.1656   ,\n",
       "       -0.64163  , -0.098455 , -0.62397  ,  0.010431 , -0.25653  ,\n",
       "        0.31799  ,  0.037779 ,  1.1904   , -0.17714  , -0.2595   ,\n",
       "       -0.31461  ,  0.038825 , -0.15713  , -0.13484  ,  0.36936  ,\n",
       "       -0.30562  , -0.40619  , -0.38965  ,  0.3686   ,  0.013963 ,\n",
       "       -0.6895   ,  0.004066 , -0.1367   ,  0.32564  ,  0.24688  ,\n",
       "       -0.14011  ,  0.53889  , -0.80441  , -0.1777   , -0.12922  ,\n",
       "        0.16303  ,  0.14917  , -0.068429 , -0.33922  ,  0.18495  ,\n",
       "       -0.082544 , -0.46892  ,  0.39581  , -0.13742  , -0.35132  ,\n",
       "        0.22223  , -0.144    , -0.048287 ,  0.3379   , -0.31916  ,\n",
       "        0.20526  ,  0.098624 , -0.23877  ,  0.045338 ,  0.43941  ,\n",
       "        0.030385 , -0.013821 , -0.093273 , -0.18178  ,  0.19438  ,\n",
       "       -0.3782   ,  0.70144  ,  0.16236  ,  0.0059111,  0.024898 ,\n",
       "       -0.13613  , -0.11425  , -0.31598  , -0.14209  ,  0.028194 ,\n",
       "        0.5419   , -0.42413  , -0.599    ,  0.24976  , -0.27003  ,\n",
       "        0.14964  ,  0.29287  , -0.31281  ,  0.16543  , -0.21045  ,\n",
       "       -0.4408   ,  1.2174   ,  0.51236  ,  0.56209  ,  0.14131  ,\n",
       "        0.092514 ,  0.71396  , -0.021051 , -0.33704  , -0.20275  ,\n",
       "       -0.36181  ,  0.22055  , -0.25665  ,  0.28425  , -0.16968  ,\n",
       "        0.058029 ,  0.61182  ,  0.31576  , -0.079185 ,  0.35538  ,\n",
       "       -0.51236  ,  0.4235   , -0.30033  , -0.22376  ,  0.15223  ,\n",
       "       -0.048292 ,  0.23532  ,  0.46507  , -0.67579  , -0.32905  ,\n",
       "        0.08446  , -0.22123  , -0.045333 ,  0.34463  , -0.1455   ,\n",
       "       -0.18047  , -0.17887  ,  0.96879  , -1.0028   , -0.47343  ,\n",
       "        0.28542  ,  0.56382  , -0.33211  , -0.38275  , -0.2749   ,\n",
       "       -0.22955  , -0.24265  , -0.37689  ,  0.24822  ,  0.36941  ,\n",
       "        0.14651  , -0.37864  ,  0.31134  , -0.28449  ,  0.36948  ,\n",
       "       -2.8174   , -0.38319  , -0.022373 ,  0.56376  ,  0.40131  ,\n",
       "       -0.42131  , -0.11311  , -0.17317  ,  0.1411   , -0.13194  ,\n",
       "        0.18494  ,  0.097692 , -0.097341 , -0.23987  ,  0.16631  ,\n",
       "       -0.28556  ,  0.0038654,  0.53292  , -0.32367  , -0.38744  ,\n",
       "        0.27011  , -0.34181  , -0.27702  , -0.67279  , -0.10771  ,\n",
       "       -0.062189 , -0.24783  , -0.070884 , -0.20898  ,  0.062404 ,\n",
       "        0.022372 ,  0.13408  ,  0.1305   , -0.19546  , -0.46849  ,\n",
       "        0.77731  , -0.043978 ,  0.3827   , -0.23376  ,  1.0457   ,\n",
       "       -0.14371  , -0.3565   , -0.080713 , -0.31047  , -0.57822  ,\n",
       "       -0.28067  , -0.069678 ,  0.068929 , -0.16227  , -0.63934  ,\n",
       "       -0.62149  ,  0.11222  , -0.16969  , -0.54637  ,  0.49661  ,\n",
       "        0.46565  ,  0.088294 , -0.48496  ,  0.69263  , -0.068977 ,\n",
       "       -0.53709  ,  0.20802  , -0.42987  , -0.11921  ,  0.1174   ,\n",
       "       -0.18443  ,  0.43797  , -0.1236   ,  0.3607   , -0.19608  ,\n",
       "       -0.35366  ,  0.18808  , -0.5061   ,  0.14455  , -0.024368 ,\n",
       "       -0.10772  , -0.0115   ,  0.58634  , -0.054461 ,  0.0076487,\n",
       "       -0.056297 ,  0.27193  ,  0.23096  , -0.29296  , -0.24325  ,\n",
       "        0.10317  , -0.10014  ,  0.7089   ,  0.17402  , -0.0037509,\n",
       "       -0.46304  ,  0.11806  , -0.16457  , -0.38609  ,  0.14524  ,\n",
       "        0.098122 , -0.12352  , -0.1047   ,  0.39047  , -0.3063   ,\n",
       "       -0.65375  , -0.0044248, -0.033876 ,  0.037114 , -0.27472  ,\n",
       "        0.0053147,  0.30737  ,  0.12528  , -0.19527  , -0.16461  ,\n",
       "        0.087518 , -0.051107 , -0.16323  ,  0.521    ,  0.10822  ,\n",
       "       -0.060379 , -0.71735  , -0.064327 ,  0.37043  , -0.41054  ,\n",
       "       -0.2728   , -0.30217  ,  0.015771 , -0.43056  ,  0.35647  ,\n",
       "        0.17188  , -0.54598  , -0.21541  , -0.044889 , -0.10597  ,\n",
       "       -0.54391  ,  0.53908  ,  0.070938 ,  0.097839 ,  0.097908 ,\n",
       "        0.17805  ,  0.18995  ,  0.49962  , -0.18529  ,  0.051234 ,\n",
       "        0.019574 ,  0.24805  ,  0.3144   , -0.29304  ,  0.54235  ,\n",
       "        0.46672  ,  0.26017  , -0.44705  ,  0.28287  , -0.033345 ,\n",
       "       -0.33181  , -0.10902  , -0.023324 ,  0.2106   , -0.29633  ,\n",
       "        0.81506  ,  0.038524 ,  0.46004  ,  0.17187  , -0.29804  ],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans[0].vector"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3bda1c0e-950f-4d65-8b1e-09df58b3a2a1",
   "metadata": {},
   "source": [
    "That doesn't mean much to me, but if I use the `similarity` method, it will calculate how **similar** two words' vectors are.\n",
    "\n",
    "Here's cat vs dog:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "8849431e-b972-4faf-b8fd-0403ece5d0d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8016854524612427"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans[0].similarity(ans[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b1350b5-82ff-4bc8-835a-846782c833cb",
   "metadata": {},
   "source": [
    "And here's cat vs waffle:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "3d6a6215-276f-4f8e-ae20-4e2c1a2d76a9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.22979263961315155"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans[1].similarity(ans[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2101be4-ce98-463d-b765-080b310cd9f4",
   "metadata": {},
   "source": [
    "Let's visualize the similaries among all eight of the words in our list:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "3f4148d2-d314-4265-a48b-f467c9ec437b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjAAAAHPCAYAAABN3eA+AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAws5JREFUeJzs3QVYFFsbB/A/KSDSKXZfO7C7G2yvSpjYgd3Y3frZ3Xnt7i5QWkAQQUAaFJXe/Z5zkIWFxYteFtjl/fnMIzNzdnbPzszuO+85Z1ZBKBQKQQghhBAiQxQL+gUQQgghhPwuCmAIIYQQInMogCGEEEKIzKEAhhBCCCEyhwIYQgghhMgcCmAIIYQQInMogCGEEEKIzKEAhhBCCCEyhwIYQgghhMgcCmAIIYQQInMogCGEEELIH3v06BF69uyJkiVLQkFBARcuXPjXxzx8+BANGjSAmpoaKlSogJ07d/7281IAQwghhJA/9v37d9SpUwfbtm3LVXl/f39069YNLVu2xNu3bzF37lxMmjQJ586d+63nVaAfcySEEEJIXmAZmPPnz6NXr145lpk1axYuXbqEd+/eiZaNGTMGLi4ueP78ea6fizIwhBBCCBFJTEzE169fxSa2LK+wIKVTp05iyzp37gxHR0ckJyfnejvKefaKipjkyA+QB8lnNkIeeCwLgqxboZQCeaClqApZ550UBXng/z0U8kC3WAnIOq/w1zLzvbRy22EsXrxYbJmDgwMWLVqUJ9sPDQ2FsbGx2DI2n5KSgsjISJiamuZqOxTAEEIIIfJAkJonm5kzZw6mTp0qtqxYsWLI66amzIRCocTlv0IBDCGEEELEgpW8DlgyMzEx4VmYzMLDw6GsrAx9ff1cb4cCGEIIIUQeCAWQBU2bNsXly5fFlt26dQvm5uZQUVHJ9XaoEy8hhBAiDwSCvJl+07dv3+Ds7Myn9GHS7O/AwEBRk5SNjY3YiKOAgADeTMVGIu3fvx/79u3D9OnTf+t5KQNDCCGEkD/GRg+1bdtWNJ/ef8bW1hYHDx7E58+fRcEMU758eVy7dg329vb43//+x2+At2XLFvTt2/e3npcCGEIIIUQOCAuoCalNmzaiTriSsCAmq9atW+PNmzf/6XkpgCGEEELkgUA2+sDkFeoDQwghhBCZQxkYQgghRB4Ii1YGhgIYQgghRB4I8uZGdrKCmpAIIYQQInMoA0MIIYTIAyE1IRFCCCFE1ggogCGEEEKIjBEWsQwM9YEhhBBCiMyhPjCEEEKIPBAUrQwMBTCEEEKIPBBSAFOkLFq0CBcuXBD9imZBc3R2w4HjZ+Hp5YuIqGhsXrkA7Vs1Q2F22iUQh5w+IvJ7EirqF8f01tVQ30w3x/LXvD7joKM/PsX+gKaqMpqVM4B9yyrQUVfNt9dsaNsFpmN6QcVIF/E+nxDosA/fXr2TWLZE0xqodnZZtuVurSYgwS8423I9ixaouGMaYm68hO+IVZCWLtbd0Gt0H+ga6eLT+0DsW7wH7155SizbpEtTdLbuivLVK0BFVQWffAJxcuNxOD96K1auxwgLdLHqCgMzQ8RFf8Wza89wdPUhJCcmS60e7aw6o9toS2gb6SLE5xOOLTkAn9eS90Vl82oYONsaphXNoKquisjgSDw4fgs3910RlTGrXBq9p/6NcrUqwLCUEY4t2Y9b+69C2vraWmLI2L+hb6QPfx9/bFy4DS6v3CSWbdO1JfrYWqJyjUpQVVXBB++P2Lv+IF4+fC0qYzm4O7r274wKVcvzeW83H+xYuQeezl5Sq8PQkYMwftIIGBkbwtvLFwtmr8DL504Sy7Iyi5fPQu26NVChYlns3XkEC+aszHHbvfp2w679G3D9yh0MHTIB0jRoWD+MGGcFQ2MD+Hp/wIr5G+D0UvJnvKGRPmYtnoIadf5C2QqlcWTPKaxcsCFbORu7QRg0tC9MzYwRE/0FNy/fxYbl/0NSYpJU60JyRn1gCpn4+ARUrVQBc6eOgyy46R2KtQ+9MaJRBZwY0gT1SupiwoU3+Pw1XmL5t8ExWHDTDb1qmOGsdTOs6V4HHmFfsOSO5C9eadCzaI4yi4YjZMtZeHSehrhXnqhydAFUSxr88nGuLcfjbd1hoinB/3O2Mqpmhii90BZxLzykWAOgec8WGO4wEme3nca0bpPh+coDCw4tgkFJQ4nlqzeuAZfHzlhmuxjTu0+B23NXzN2/AOVrVBCVadWrNaxn2eLUppOY2G4cts3YihY9W8Bqlq3U6tGoRzMMWTgMl7edw8Ju0+H9+h2mHZwHvRz2RWJ8Iu4cvo4VAxZgTofJuLz1LPpOG4Q2gzqKyrDAJiIwDGdWH0VseAzyQweLtpiyeAIObjkK204j4fzSDRuPrYGxmZHE8nWb1MGrR46YajULQ7vYwenZW6w7tAJValYSlanfrC5uX7iL8f3tMcpiPEKDw7D5xDoYmvz6OP1Tln26YunKOdi0bic6tOyNl88cceLsbpiVMpVYvlgxVURFRvPyHu6/DqpKlS4Jh6Uz8fxpRoAmLV0tO2LO0qnYuekAere3guMLZ+w+uZkHHpKoFlNFdFQsdm7aDy+P9xLL9OjbBdPmj8f/1u1B9xYDMN9+Kbr16oip88aj0N3ITpAHk4yQiwBGIBBg9erVqFSpEooVK4YyZcpg+fLlfN2sWbNQpUoVaGhooEKFCliwYAGSk5NFv5C5ePFiuLi4QEFBgU+SfjUzP7Vs2hCT7GzRsU1zyIKjbz7yYKRPzVKooKeJGW2qwURTDWdcgySWdwv9gpJa6hhcryzMtDVQz0wXfWuVhmfYl3x7zcajLBB58i4iT9xBgm8QPjnsR1JIFIxsuvzycSmRsUiJyJiytTcrKqLCNnsErzuJxMAwqdbBYmQv3D11G3dO3kKQbxD2L96LqJBIdLHuKrE8W39h5z/wdX2Pzx8/49iaI/z/hh0aicpUrV8NXk7v8PjiQ0QEhcPl8Vs8vvgIlWpnfKnmtS4je+LR6Xt4eOouPvsF4/iSA4j+HIX2Vp0llg/08MeLS08Q/P4TIoMi8OzCI7g9ckaVhn+Jyvi7+uHUysN4efkpkpOklznKbJBdf1w+cQ2Xjl/FR99AbHLYhvCQcPSxsZRYnq0/uv0k3rl445N/MHau2otP/kFo0TEj2+owYTnOHbqI9x6+CPANxMrp66CoqADzFvWlUocx44fi+JFzOHb4LN77fODZlODgUAwdMUhi+U+BwZg/ewXOnLyIr1++5bhdRUVFbN+zFmtXbkXAR8mfC3lp6JjBOHf8Is4eu4gP7z/ybAoL/gYN7SexfPCnz1gxfz0unr6Gb18l16OeeS28eeWKK//c5OWfPniJq+dvoWbdjOOu0DQhCfNgkhFyEcDMmTOHBzAsOPH09MTx48dhbJwWbZcoUYIHJWz55s2bsWfPHmzcuJGvGzhwIKZNm4YaNWrg8+fPfGLLSO4kpwrwLjwOTcvqiy1vUlYfLp9jJT6mtqkOwr4l4LF/BP/59ajvibjzPgwtykvOHOQ1BRVlFK9dEV8eiqeTvz50RnHzar98bI2bG1DnzT5UPbUYJZrVzLa+pP0ApER94cGRNCmrKKNirUrZmn+cH79FtQa5+0Blwbp6cXXExcaJlr177YmKNSuicp3KfN64jDEatDWH0z1HSIOSijLK1awI98fi+8L9sQsqNaiaq22UqVGel/V+Kd2M17/tj6q1q4o1/zBsvpZ5jVzvDw1NDXzNtD+yUlMvBiVl5V+W+VMqKiq8KejBvadiyx/eewrzRvX+07anzRrPMzUsOJI2FRVl1KhTjQcYmbH5eg1r//F2WfMT226tetX5fKmyZmjVvhke3hZ/v0j+kvlOvHFxcTww2bZtG2xt01LdFStWRIsWLfjf8+fPF5UtV64cD1hOnTqFmTNnQl1dHZqamlBWVoaJiUmOz5GYmMinzBQTE3m2pyiLiU9CqlAIPQ3x90FfQxVRP8Tfr3R1S+pgeZdamH3NFUmpAqQIhGhdwRCz2vw6eMgrynoloKCsxLMpmSVHxkLLSEfiY5LDY+A/Yzt+uPpBoZgKDPq25kGMV78F+PYyrelL07waDAe1h0fHqVKvQwk9LSgpKyE2Sx1iI2KhYyi5DllZ2vWCmkYxPLvyRLTsyeXH0NLXxvJzq/kXKvtivn74Gv7ZfhbSUEK3BK/Hlwjx7NuXiFhoG/y6Hhuf7/75Piji/KbTPINTUHT0tKGsrIToSPHmquiIGOgb6eVqG4PHDIC6uhruXrqfY5lx8+wQERqJ148l90n5L/T0dfnnYER4lNjyiIgoGBn/eZNVw8b1MNi6L9q36IX8oKunw+sRFREttjwqIgoGRuIXWr/j2oXb/D06dnkvPzdYoHT8wFns2XoIhYpAdrIneUHmA5h3797x4KJ9+/YS1589exabNm2Cr68vvn37hpSUFGhpaf3Wc6xcuZI3NWU2f8YkLJw5+T+9dnmhkGVeKGFZOr+ob1jzwBt2jSugaVkDRH5PxKYnPlh+7x0Wdczd1WpeELIXmZmCgoSFaRL8QviU7ruTN+8vYzLGEr4vPaFYXA0Vtk7Bxxk7kBKT91fHOcryetkHaw5VENPCohUG2g/GypHL8CUqI3io0aQm+k0YgN3zd8LnrTdMy5lixCI7xIRH48yWU5AWIT9istTjXx6zvP98qBVXQ8V6VTBglhXCA0J501JBYhnFP9kfHXu1w8hpQzFz2HzEREnOXFqN+xsdLdtjfL8p0u00mq0O2euVW8U1i/Omo2mTFiA6WnK9pCXba+b74s/qwTRqVh+j7YdjyazVcH3jjjLlS2PusmmICIvEjg37UGgIKYCRKSyLkpMXL17g77//5sFH586doa2tjZMnT2L9+vW/3UQ1dar4lbViXPbRJ0WNrroqlBQUsmVbon8kZcvKpDvw2p9nYWzN00ZWVDEsAXUVJQw/8xrjm1WCYXHpZrVSouMgTEmFSpZMhYq+NpKzZAJ+5dsbH+j3ac3/LlbOBMXKGKPywbkZBRTTQjjzgLN8tFJiQGheVYGPDkpNSYWOofhIL20DbXzJkpWR1Pl3wtpJWDt2FVyfuIitGzzdCg//uc/71TCB3gFQ01DD2FUTcHbr6f/0BSCxHjFxP+shvi+0DLTx9V/qERkUzv8P8g7k9e41eUCBBTCx0V+QkpIKfUPxbIuugQ6is2QCJHX+nbd+JubaLcoxszJ4zEDYTrTCxIHT4PvuA6QhOiqGX9yxUTuZGRjoZ8vK5Fa58qVRpmwpHDm1Q6w/DBMc5Y5m5l0R4P8JeSkmOpbXI2u2Rd9AL1tW5ndMmj0Gl85c4/1qGJ93flDXUMeSdXOxc+P+PD83SBHpA1O5cmUexNy9mz2F/PTpU5QtWxbz5s2Dubk5LxsQECBWRlVVFampv+51zZqKWNYm81TUm48YFSVF/GVUAi8CxT/g2HwdU8lNAPEpqdkOOkV2mSfpqkkKhMkp+O7qB+1WdcSWa7Wqg++OuR+eqlGzPG9aYhJ8g+HebjI8Ok0VTbG3XiPumTv/OykkMk/rkJKcAj83X9RpKd43oU7LurwT7q8yLxPXT8GGiesk9msppl4MgixXcKmpAp5OY9mEvJaanIKP7n6o0UJ8X9RoURu+Tt653xBr7iqmgoLC9oe3qzcatTIXW87m3Rw9fpl5mb9xNhaOX4Znd19ILDNk7EAMn2KNKUNmwsv1N96T38QGNrg6e6B1W/FbNrRq2wyOr8T7WuWWr88HtG7SE+1b9BZNN6/dw9PHL/nfIUF5F9SnS05OgYeLF5q1biy2vFnrRnj72vWPt8ua99hgkcwEqak8QyWNc+OPCQR5M8kImW9CUlNT4yONWJ8WFow0b94cERER8PDw4KOSAgMDedalYcOGuHr1Ks6fPy/2eNYvxt/fn98HplSpUrzTb0EGJz9+xCMwKKO5IjgkDF4+ftDWKgFTE8lDMguSVf1ymH/TDdWNtVHbVBv/uAUhNC4B/WqX4uu3PHmP8O8JWNa5Fp9vXd4QS+964rTLJzQrp8+bkNgw7JrGWjDSVMuX1xy25xLKb56M7y5++ObkDUOrjlA1M0D4kZt8fanZVlAx1YP/5C183nhkDyR+Cuf3i2GdgA36tIZe92bwHbmarxcmJiPeO1DsOVK/fuf/Z12eVy7tvYDJG6fCz/U9vN94oePgLnwI9c2j1/l6q1k20DPRxxb7jaLgZfJGe+xbtAc+b71EWY+khCT8iPvB/3595xUf3eTv/gE+zj68CWnw9CF4fftVtg/vvHJj72WM3jCJjxzyfeONtoM7Qr+kAe4dS8sC9Z85BLrGetg9bSufb2/dhY+2YiOWmCoNq6HrKAvcOZRW7/TOwWaV044/1o9H11gfZaqXQ8L3BN7UJA0ndp+Bw5a5eOfqDXdHD1ha9YSxmTHOH77E14+dM4oPf14yeaUoeHHYPBcbF26Fu5Mn9H5mbxITEvE97ruo2chuxnA4jF+Gz59CRWXiv8cj/ofk2xT8Fzv/dxDbdq2Gy1t3OL5yhvXQAShVyhSH9p/k6+c5TIWJqREmjpktekyNWml914pravAsB5tnI798vP2QmJgEr3fiw5K/fElrYs26PC8d3Hkcq/+3GO7OnnB2dMMA694wLWWCk4fSOhGzoc9GpoaYPWGR6DHValbh/2sUV4eegS6fZ/Xw8/Hny+/fesxHN71z84bLGw+ULV+KZ2Xu3XwstXPjTwiFsjMEOi/IfADDsNFHrOPWwoULERISAlNTU4wZMwYjRoyAvb09JkyYwPvJdO/enZdlN69L17dvX/zzzz9o27YtYmNjceDAAQwdOrTA6uLu9R7DJ84Sza/Zupv/b9m1A5bPn4bCpnNVE3xJSMLuF36I/JGISvqa2GpZjw+VZliAEvo1QVTeooYZvien4pRLIDY+9oZmMRU0Kq2HyS3SRr7kh+hLT6GkW4KPGuI3svMOhI/1MiQFR/D1Ksa6UM10PxUWtJReMBSqJnoQJCTxQMbHeim+3HuDgvL08hOU0NHCgMl/Q9dID4E+AfweLxE/68CWGWaqQ+chXfiX+ejlY/mU7t6Zu9g6bRP/m/VzYVmwwTOsePDzNeorHO+8wtG1R6RWj1dXnkFTpwQsJ/fnTWLBPoHYMGwFon7Wg93cTs8so1lDQVGBBzWGpY1481M4u9/LmmO4/zPg4XU31sXSaxnNxOwmeWx698Idq/52kEo97ly6D21dLYywt+Uddz94+/N7vLDhuwxr0jDJdB+S3lYWfH/MWGnPp3RXT93AUvu0mx/2te3F71Gycu8SsediN7xjU167+M913gl26szxMDYx5EHG4P6jEfQpRHTjOrNSJcUec+/JBdHfdevVRN8BPREYEIyGtSX3ScwP1y/e5h2rx08byZvE3nv5YfSgKaKMD1tW0kx80MaFe8dEf9esWx09+3ZBcGAI2punDYPfsSGtmWjynLH8vWH3jWFBzaYV21GoCAtPMJUfFITUePdHkiOl0xad35LPpF2hyzqPZdK/v4S0rVBKgTzQUsy/OypLi3fSn/X7KGz8v0sn45TfdIuVgKzzCpf+TfwSnDPuSP1fqNXtAVkgFxkYQgghpMgTFK0MDAUwhBBCiDwQFq0ARuZHIRFCCCGk6KEMDCGEECIPBDQKiRBCCCGyRkhNSIQQQgghhRo1IRFCCCHyQFC0MjAUwBBCCCHyQFi0AhgahUQIIYQQmUMZGEIIIUQeCIpWBoYCGEIIIUQeCCiAIYQQQoiMERaxX6OmPjCEEEIIkTnUhEQIIYTIAwE1IRFCCCFE1giLVgBDTUiEEEIIkTnUhEQIIYTIA0HRysBQAPOHks9shDxQ6W8PefBm+ULIuq+CcMiDqNQfkHUpcjKao3xxE8gDJQVqLMgVYdEKYOioIIQQQojMoQwMIYQQIg8ERSsDQwEMIYQQIg+ERSuAoSYkQgghhMgcysAQQggh8kBQtDIwFMAQQggh8kBAAQwhhBBCZI2waAUw1AeGEEIIITKHmpAIIYQQeSAoWhkYCmAIIYQQeSAsWgEMNSERQgghROZQBoYQQgiRB4KilYGhAIYQQgiRB8KiFcBQExIhhBBCZA5lYAghhBB5IChaGRgKYAghhBB5IChaAYxcNCG1adMGU6ZMKeiXQQghhJB8QhmYAnTaJRCHnD4i8nsSKuoXx/TW1VDfTDfH8te8PuOgoz8+xf6ApqoympUzgH3LKtBRV0Vh4+jshgPHz8LTyxcRUdHYvHIB2rdqhsKiuk0H1BnTDRpGOojxCcazRUcR+sr7Xx9nbF4ZFmfnI9o7COc6z5NYpqJFE3TYPgH+Nxxxa+QmSIuFTU8MGNMf+kZ6+OgTgO2LdsDtlbvEsi26NoeFdU9UrFEBKqoqCPAJwKENR+D40ElUpnP/jpi5cUa2x3ap2B3JiclSq0cvWwsMGjMAekb6+OjzEVsdtsP1lZvEsq26toCljQUq16jI6+HvE4AD6w/h9UNHsTJWEwfDrJwZlFWUEOQfjFO7zuDWuTuQpn62vWA9bhAMjPTxwecj1i/cAueXrhLLtu3WCv1seqFKzcq8Hh+8/bF7/QG8ePBKVKZClXIYM3MEqtWuipKlTfn2Tuw5I9U69LW1xJCxf0PfSB/+Pv7YuHAbXHLYF226tkQfW0tUrlEJqrwOH7F3/UG8fPhaVMZycHd07d8ZFaqW5/Pebj7YsXIPPJ29pFqP3rYWGDxm4M96fMQWh//lWI/WXVuit01PVPpZD1Z+3/pDeJXpmGJlbLIcUyd2ncHNc7dRqAiFKErkIgMji256h2LtQ2+MaFQBJ4Y0Qb2Suphw4Q0+f42XWP5tcAwW3HRDrxpmOGvdDGu614FH2BcsueOJwig+PgFVK1XA3KnjUNhU7NkYzRZZ4e3WSzjXZT4PXLodmQHNkvq/fJxqCXW03TQGwU88ciyjaaaPJgsG4/ML6X5At+nZGuMWjcHxrccxustYuL1yw8ojy2FU0lBi+dqNa8HpsRPm2szH2G7j4fzMBcsOLEGlGhXFyn37+h396g0Um6QZvLSzaIOJi8bh8JbjGNl5NA9c1hxdCaOSRhLL12lSG46PnDDTei5GdR2Lt8+csergMv4lmu5rbByObDmGcRYTMazDKFw/dROzN8xEw9bmUqtHR4t2mLZkEvZvPoIhnUbg7UsXbDm2FsZmkutRr0kdvHzkiMlDZsC680g4PnuLjYdWoWrNyqIyaupqCAr4jG3LdyEyLArS1sGiLaYsnoCDW47CttNIOL90w8Zja3KsQ90mdfDqkSOmWs3C0C52cHr2FusOrUCVmhn7on6zurh94S7G97fHKIvxCA0Ow+YT62BoYiC1erS3aIPJi8bj8JZjGNbZjh9T646ugnEOx1TdJrXx6pETplvPwfCuY/DmmTPWHFye5Zj6ikNbjmG0xQTYdhiFq6duYO6GmWgkxWPqj5uQBHkwyQiZC2C+f/8OGxsbaGpqwtTUFOvXrxdbHxMTw9fr6upCQ0MDXbt2xfv378XK7NmzB6VLl+bre/fujQ0bNkBHRydf63H0zUcejPSpWQoV9DQxo001mGiq4YxrkMTybqFfUFJLHYPrlYWZtgbqmemib63S8Az7gsKoZdOGmGRni45tmqOwqWXXFV4nH8DrxAPE+obw7Mu3kChUt2n/y8e1XDUcvheeI+yNr8T1CooKaLd1HBzXn8PXwHBIUz+7vrh+8gaunbiBQN9P2L5oJ8JDItDTpqfE8mz9qR1n4O3ig2D/EOxbfQDB/sFo2rGJeEGhEDERMWKTNA0Y1Q9XT17H1RPXEOAbyLMvESHh6JVDPdj6EztOwcvFm18F71m1j//frGNTURnn5y54fOMp315IwGec3fcPPrz7gNqNakqtHkNGD8TFE1dx8fgVfHwfgA0LtyIsJBz9bHtLLM/WH95+HJ4uXvjkH4TtK3cj0D8ILTtmnC9s3Zal23Hr4l0kJSVB2gbZ9cflE9dw6fhVfPQNxCaHbQgPCUcfG0uJ5dn6o9tP4p2LNz75B2Pnqr28Li06ZmRaHSYsx7lDF/Hew5fvj5XT10FRUQHmLepLrR4DR/XHlZPXeV3Yc252+B+vR28bC4nl2frjmY6pXT+PqRaZjqm3z13w6MYTvr3ggBCc2fcP/N59QJ1GtVCoCCiAKdRmzJiB+/fv4/z587h16xYePHgAJ6eMNPjQoUPh6OiIS5cu4fnz5xAKhejWrRuSk9OuIp8+fYoxY8Zg8uTJcHZ2RseOHbF8+fJ8rUNyqgDvwuPQtKz4FX+Tsvpw+Rwr8TG1TXUQ9i0Bj/0jeJ2ivifizvswtCgv+YqbSKaoogTDWuUR9Ei8qYXNs+ahnFQd0ApaZY3htPGfHMs0sO+NhOg4eJ98KNW3X1lFGVVqVYbjozdiy50eOaGGefVcbUNBQQHqmho8W5GZenF1HH9xBCdfH8Pyg9kzNHlej9pVxJp/mNcPnVDTvEau66GhqY642K85lqnfoh5KVywFlxeSmxDyoh7ValfBi4cZzT/Mi4evUdu8Zq7rUZzvj5zrIU2sDlVrVxVr/mHYfK3f2hfZj6nM1NSLQUlZ+Zdl/ns9qog1/zBs/neOKXVN9V++xgYt6qFMxVJwfiG5iZDkD5nqA/Pt2zfs27cPhw8f5oEHc+jQIZQqVYr/zTItLHBhQUqzZmlXAceOHePZlgsXLqB///7YunUrz8pMnz6dr69SpQqePXuGK1eu5Pi8iYmJfMosNTkVxVSU/qgeMfFJSBUKoadRTGy5voYqon6IP0+6uiV1sLxLLcy+5oqkVAFSBEK0rmCIWW2q/dFrKKrU9EpAUVkJ8RHimSs2r2EoOQunVd4YjeYMxKU+SyFMlZxeZcFP1b/b4FynuZA2bT0tKCkrZcuOsHk9w5z7UGXWf3Q/qGuo4eHlR6JlgX6fsGbqOnx454/iJTTQZ0RvbL6wEXadxvCsTd7XQxvKrB6R4vWIjoyBnpFerrYxcHR/qGmo495l8aCxeIniOOd0ivdpSE0VYOPczXB8nHGhk5d0eD2UEZ1lf7B5A8Pc1cNqzN+8yej2pXsoCGl1UOLvfdY6sD5WuTF4zACoq6vh7qX7OZYZN88OEaGReC3VfZG9HuwYy209Bo0ewM+Nu5cfZDumLjidFh1T6+duklo9/phQdpp/ilwA4+fnx1OpTZtmpPb09PRQtWpV/ve7d+/4B0njxo1F6/X19fl6to7x9vbmzUaZNWrU6JcBzMqVK7F48WKxZXO7NcS8HhnP8ycUsswLJSxL5xf1DWseeMOucQU0LWuAyO+J2PTEB8vvvcOijrm7siC/6OymwBYJJTYLtd86njcLffEPlfgWqhRXQ7stY/Fo5l4kxHwrwDoo5KoPX1vLNrCZao2Fwx0QG5WR8Xv3xotP6dxfe2Dnje3oNawX/rdwO/KxGhL3RVbtLdti2DQbzB2+UKwezI9vPzCikx3PKDVoUR/jHcYiJPAzb16SlqyvmdeDn9W/1rlXe9hNH4ZpQ+cgJks98lv2OuTumOrYqx1GThuKmcPm51gHq3F/o6Nle4zvNwVJidJtEst2/PB6/HtFOli2w/BpNpg9fIHEY2pop1HQ+HlMTXQYx48p1rxUaAgogCm0/u0AzGk9W85OxKx/53a7c+bMwdSpU8WWpR6cjT+lq64KJQWFbNmW6B9J2bIy6Q689udZGFvztN78VQxLQF1FCcPPvMb4ZpVgWFzy44g41sQjSEmFupF4tkXdQBvxkdn7E6loqsOobgUY1CyLFstsRUGNgqIiRn08hKuDVyMx9hu0yhihy4FposexMgwrc6r1DHwNyLs+MV+ivyI1JRW6Wa4odQ10smUzJHX+nb5uKpaMXoY3T97+siw7L7xdvFGqvBmk4Uv0F6SkpGbLGunq6/5r3xvW+XfW+ulYOHoJnB6LN6Wlv/bgj2lZI18PP5StVAZWEwZJJYCJ5fVIyXaFr2ugi6h/qQfr/Ltgw2zMGrUQrwrwaj6tDqnQN8x+TEVHRP9r599562dirt2iHDMSbESQ7UQrTBw4Db7vPiDf66HP6hHzr51/56yfjvmjF8PxX46p9x5+KFepDKwnDC5cAUwRI1OdeCtVqgQVFRW8ePFCrNOuj48P/7t69er8g+Tly5ei9VFRUXz9X3/9xeerVauGV6/E26pZn5lfKVasGLS0tMSmP20+YlSUFPGXUQm8CBQfWcDm65hKbsaIT0nNtrMUMwVlJHcEyamIcPNHqZbifRPYfJijeGdvJikuHqfbz8bZzvNEk+eRe4jxDeF/h7/1Q6zf52xlPt56g5Bn7/jfrINwXkpJToGP23s0aCneEZLNezh6/jLzMnPjdKyYsAov74mfAzmpWL0ioqQ0AobXw9UH5q0aiC1n8+6OHr/MvMzZMBNLxq/Ai7sZ5/qvsIsWNlxZWvXwcvVB41YNxZazeVdHycPa0zMvDpvmYt64JXh69zkKEquDt6s3GrUSH1XD5t1+sS9Y5mX+xtlYOH4Znt3N+FzObMjYgRg+xRpThsyEl+u/36rgv9fDBw2zHFMN/+WYYpmXeRtmYdH45Xiey2MKUjym/phQmDeTjJCpJiQ28mjEiBG8Iy9rGjI2Nsa8efOgqJj21V65cmVYWlpi1KhR2LVrF0qUKIHZs2fDzMyML2cmTpyIVq1a8ZFHPXv2xL1793D9+vVsWRlps6pfDvNvuqG6sTZqm2rjH7cghMYloF/ttP48W568R/j3BCzrnNbLvXV5Qyy964nTLp/QrJw+b0Jiw7BrGmvBSFMNhc2PH/EIDMroNxEcEgYvHz9oa5WAqYnk4Yz5xW33dbTdPBYRrh8Q5uSLv4a05cOfPY/c5esbzR6A4ia6uD9lV9qoHG/xkWHxUV+RmpgstjxrmaSvPyQuzytnd5/D7M0zeQDg6eSJ7kO6w8jMCJePpDWFjpg9HAYm+lg9Za0oeJm9aSb+57ADnm/eQfdn1iMpIRHf49Jeq7W9Fd69ecdHJ7HOmL2H9+KdeLfM3wZpOb3nLOZtns1HR3k4eaKnVVo9Lh65zNfbzR4BA1MDrJi8WhS8sPLsvh6ebzxF2ZvEhCR8j/vO/x4yYVDaaKuAEKioKKNJ+8bo3K8j1s/ZLLV6HNt1Cku2zsc7Fy+4Onmgj5UFTMyMcO7wBb5+/NzRMDIxgMOk5aLgZfGW+Vi3YDPcnTxEGYMEvj++izqksnvBMOzCzdDEEFVqVMKP7/EI+hic53U4sfsMHLbMxTtXb/5lb2nVE8Zmxjh/+BJfP3bOKD78ecnklaLgxWHzXGxcuBXuTmxfpNUhMVMdWLOR3YzhcBi/DJ8/hYrKxH+PR/wPybeM+K9O7TmDBZvn8FFF7HVZWvVIq8fPY2rM7JH8mFo2eZUoeFmweTYfVeWRwzFlPWEQvH4eU2y/NG3fGF37dcK6OdK7z9MfEVATUqG2du1a3pnXwsKCByjTpk3Dly8Zqf8DBw7wEUY9evTg/WVYsHLt2jX+AcA0b94cO3fu5H1a5s+fj86dO8Pe3h7btknvQ1qSzlVN8CUhCbtf+CHyRyIq6Wtiq2U9PlSaYQFK6NcEUXmLGmb4npyKUy6B2PjYG5rFVNCotB4mt8h55ExBcvd6j+ETZ4nm12zdzf+37NoBy+dnNLUUBL/LL1FMtwQaTOnNb2THbkp33WYtvgWnZRrYMk0z6d2nIi88uPwQWrpasJ4yhHd4/egdgDk28xEenNZUxZozWCCQrodVd/7BO3nFRD6lu3n6Fu+4y2hqaWLq6ik8uGFBja+7L+z7ToO3s/Sumu9desDrYWtvzV+zv/dHzLKeg7D0ehjri92/w8KqB6/H1BWT+ZTu+umbWGm/hv/NOmBOXTmJf+GzL1PWOXnZpJX8uaSFdb7V1tXCyKlD+Y3s/Lz9MdlqJkKDwvh6tszEzFhUvo+1Ja/H7FXT+JTu8qnrWDxlBf/b0NgAx+8cEK2zGTeIT+x+K6P7TsrzOty5dJ/XYYS9Ld8X7OZ67B4v7N4tkurQ28qC12HGSns+pWP3SFlqnxYc9LXtBdViqli5d4nYc7Eb3rFJGu7+PKaG2dv8rMdHfo+XsJ/10DfWEzumWIDD6jF9xRQ+pbt2+gaW/zymWEfxaSsnw+jnMRXg9wlLJq3gz0UKjoKQ2h94xsbLywuPHz/O9Rv3Y0fGl4AsU+mf8cEjy/bXXQhZdxrSvXdMfkkWpkLW/RBI/74r+UFZ4c+bugsTJQWZ6u0g0dNg6Y8wi9+XNrr2v1IfkXZRU9jJVBNSXlm3bh0fhl28eHHefMSGYm/fLr1RFoQQQojUCakJSe6xTrxr1qxBXFwcKlSogC1btmDkyJEF/bIIIYQQkkuyn5f7A6dPn0Z4eDji4+Ph4eHB78xLCCGEyDKhQJgn059grRjly5eHmpoaGjRo8K9dMthNZuvUqcN/0of9LNCwYcP4qOHfUSQDGEIIIUTuCArmt5BOnTqFKVOm8FHBb9++RcuWLfkd7wMDAyWWf/LkCf/NQjaqmCURzpw5g9evX/92SwgFMIQQQoi89IER5sH0m9htSVgwwgIQds+1TZs28Z/w2bFjh8Ty7F5u5cqVw6RJk3jWpkWLFhg9evS/3pMtKwpgCCGEECLCfvvv69evYlPW3wNMx25Xwn5QuVOnTmLL2Tz7nUFJ2G8VBgUF8VucsIHQYWFhOHv2LLp3747fQQEMIYQQIg8EwjyZ2O//aWtri01smSSRkZFITU3lN5bNjM2HhobmGMCwPjADBw6EqqoqTExMoKOjw39s+XdQAEMIIYTIA0He9IFhv//HbhCbeWLLfkXSbwzmdId7T09P3ny0cOFCnr25ceMG/P39f3tATZG8DwwhhBBCcv79PzblhoGBAZSUlLJlW9hI36xZmXQsm8Puis9+FoipXbs2vy8b6/y7bNkyPiopNygDQwghhMgDQf6PQmJNQGzY9O3bt8WWs3nWVCTJjx8/RL9hmI4FQczv/DgAZWAIIYQQeSAsmF+Snjp1KqytrWFubo6mTZti9+7dfAh1epMQa34KDg7G4cOH+Tz7IWX2Ez5slBL7PcLPnz/zYdiNGjVCyZIlc/28FMAQQggh5I+xzrjsJnRLlizhwUjNmjX5CKOyZcvy9WxZ5nvCDB06lN8Jn/2IMvtBZtaBt127dli9Ou1X53OLfszxD9GPORYu9GOOhQf9mGPhQT/mWLR+zPHHhlF5sh2NqXsgCygDQwghhMgDQcE0IRUU6sRLCCGEEJlDGRhCCCFEHgh//2cAZBkFMIQQQog8EBStJiQKYP6Qx7IgyIM3yxdCHgx3XgJZF95gAeRBoEISZJ1jQgjkgXPUB8iD0iUMCvolyAThH/yStCyjPjCEEEIIkTmUgSGEEELkgYCakAghhBAia4TUhEQIIYQQUqhRExIhhBAiDwTUhEQIIYQQWSOgJiRCCCGEkEKNmpAIIYQQeSCgJiRCCCGEyBoahUQIIYQQUrhRExIhhBAiDwTUhEQIIYQQGSMsYqOQKANDCCGEyANB0crA0I85EkIIIUTmUAaGEEIIkQeCopWBoQCGEEIIkQfCotUHRiaakB48eAAFBQXExsYW9EshhBBCSCFQKAOYNm3aYMqUKaL5Zs2a4fPnz9DW1oYsM7TtgtrPd6KB3ylUv74Omo3+yrFsiaY10DD4fLZJraKZxPJ6Fi34+kr7ZkuxBkB1mw4Y9GwDRvjuR59rS2HSqGquHmdsXhmjPh5C35vLcyxT0aIJRgcdRae9Gfu+IDk6u2H8TAe0tRiCms274u6jZyhMGlh3wIQnGzHH+wBGXlmG0g1z3helzatg6DkHTHPeidneBzD27lo0HtFFrIyishJaTuqN8Y828G3aXV+Biq1rS70ebaw6YeXj/2G79zHMv7walRtWy7FsJfNqmHV2KTa+3Y//eR3Dkrub0GFEd7EyLf9uj5mnl2CTywE+2R9dgHJ1KiG/9R/aG1dencGLj/dw7OY+1GtcJ8ey7bq1xo5Tm3DP4woev7+FQ1d2oWmbRlJ/jWNG2+K993N8++qHly+uo0XzXz9nq5ZNeDlW3sfrGexGWWcrM2niSHi4P0LcF1/4+73G+rWLUKxYMdF6X58XSEkKzjZt2ZzzZ8Pvsho+AI/eXINX8CtcunsCDZvUy7GsobEBNu1aibsvL8Iv4i0WLJ8hsVwJrRJYsmYOXnrc4du9/fw82nRogULXhCTIg0lGyEQTkqqqKkxMTKT+PMnJyVBRUZHKtvUsmqPMouEImLsb3157wdC6E6ocXQD3NpOQFBKZ4+NcW45HatwP0XxK1NdsZVTNDFF6oS3iXnhAmir2bIxmi6zwZN5BhL72QXWrduh2ZAZOt52FbyFROT5OtYQ62m4ag+AnHlA3lByEaprpo8mCwfj8wguFRXx8AqpWqoBe3TrBft4yFCbVezRB54XWuLbgAIIcfVB/cDsMPjQTOzrMxFcJ+yI5PhGvD91C+LtAJMUnokzDqui2YjiSfiTi7Yn7vEzb6f1Rs3dzXJ29F5G+ITx46b/bHgf7LEKoR4BU6mHeoxkGLhyGYwv2wNfRG62HdMSkg/Pg0NEe0RLOi8T4BNw/fANB7wKQGJ/IAxrrFXZI/JGIxyfu8DJVm9TAq0tP4PfGB8mJSeg82hL2R+bDoeNUxIZFIz90smyPGUsmY+Xs9XB+7Yq+1r2w7fg69G1lhdDgsGzl6zepixePXmHryp349uUbLP7ujs2H18C62yh4u7+Xymvs398CG9YvwoSJc/Hs+WuMGmmNK5ePoladNvj0KSRb+XLlSuPypSPYu+84bIdORLOmDbFt6wpEREbh/PlrvMygQb2xYvkcjLSbhufPHVGlcgXs27uRr5s2YxH/v0mzblBSUhJtt2aNarh54yTOnbuSJ/Xq3qszFiyfiYUzlsPxlTMG2/bDgVPb0alZb4QEh0r8fomOisH/NuzB8DHZAzJGRUUZR/7ZiaiIaIwbNh2hIWEwNTPB92/fUZgIZSj4kMsMzNChQ/Hw4UNs3ryZNxux6eDBg9makPbs2YPSpUtDQ0MDvXv3xoYNG6CjoyO2rcuXL6NBgwZQU1NDhQoVsHjxYqSkpIjWs23u3LkTlpaWKF68OJYtk96XlPEoC0SevIvIE3eQ4BuETw77kRQSBSMb8avgrFIiY5ESkTFl+7VRRUVU2GaP4HUnkRiY/YMxL9Wy6wqvkw/gdeIBYn1D8GzRUR64VLdp/8vHtVw1HL4XniPsja/E9QqKCmi3dRwc15/D18BwFBYtmzbEJDtbdGzTHIVNk5Fd8fbUAziffMCDjVtLjuLr5yiYW3WQWJ4FIB6XniPifTC+BEXC7fxTfHjkhjKNMrIdtfq0wNP/XYLvfRfEfoqA09G7+PDQFU1GdZNaPTqO7IEnp+/hyal7CPULxqklBxHzORKtrTpJLP/J4yNeXXqKkPdBiAqKwMsLj+HxyAWVG2ZkM/dO2YIHR2/hk+dHhPqF4PDsXfxc/6t5TeQXq9EDceHEFZw/fhn+7wOwbuFmhAaHo79tb4nl2fpD/zsOT2cvBPoHYdvKXfz/1p2kd4VvP3kU9h84if0HTsDLyxfTpjvgU1AIxoy2kVh+tJ01Aj8F83KsPHvcgYOnMM1+jKhMk8YN8OyZI06evICAgCDcvvMIp05dRIMGGZm8yMhohIVFiKZu3TrA19cfDx89z5N6jRxnjdPHzuPU0fPw8/HH0nlr8TkkFEOGD5BYPvhTCJbMXYN/Tl1B3Nc4iWX6D+kNHR1tjLa2h9MrZwQHfYbjy7d45+GTJ6+ZyEkAwwKXpk2bYtSoUbzZiE0sUMns6dOnGDNmDCZPngxnZ2d07NgRy5eLpx9v3rwJKysrTJo0CZ6enti1axcPhLKWc3Bw4AGMm5sbhg8fLpU6Kagoo3jtivjy0Fls+deHzihunnO6nKlxcwPqvNmHqqcWo0Sz7B/AJe0HICXqCw+OpElRRQmGtcoj6JG72HI2z5qHclJ1QCtolTWG08Z/cizTwL43EqLj4H3yYZ6+ZnnF9oVprfL48NhNbLnfIzeUapDzvsjMpEZZlKpfGQEv34mWKakqIyUxSaxcckISSpvnrpnwdympKKNszQrwfOwittzjsSsqNsjdc5auUY6X9XmZc/ZRVV2VP9f32G/ID8oqyvirdlU8f/BKbPmLh69Qp2HugigWcGkUV8eX2OwZ17zAMs3169fG7Tvi59zt2w/RtIm5xMew4IStz+zW7Qc8OFFWTkvmP332CvXr10JD87p8vnz5MujStR2uXb+b4+sYMrgPDh46lUf1UkbNOn/h8X3xYIjNN2iYcxPev+nQpTXeOrryJqTX7+7hxpNzGGc/AoqKhewrVEBNSAWK9XNhKT2WWUlvNvLyEm9W2Lp1K7p27Yrp06fz+SpVquDZs2e4ciUjBckCldmzZ8PW1pbPswzM0qVLMXPmTB60pBs8eLDUApd0ynoloKCsxLMpmSVHxkLLSDxrJFoXHgP/Gdvxw9UPCsVUYNC3NQ9ivPotwLeXnryMpnk1GA5qD4+OUyFtanoleB+J+IgvYsvZvIah5DpolTdGozkDcanPUghTJfeOZ8FP1b/b4FynuVJ53fJIQzdtX3yPFN8XbF4zhya6dJNfbIXGz335aNM5nsFJxzIyTUZ2Q+BLL0QHhKN88xqo2qkBFKT0Ia2pWwJKykr4yjKLmcRFxELbQPIxlW7N853Q1NPij7+06TTP4OSk76whiA2NhudT8YBPWnT1dPgXenSEeHNVVEQM9A31c7UN67GDoK6hjluXpHNhYmCgx19jeJh4M114eCSMTYwkPoYtZ+vFyodF8iCEbS80NBynT1+CoYE+Hj44z4Mwtm7HzkNYs/Z/ErdpadkFOjpaOHT4dJ7US1dfl9crMly8GTUyIor3dflTZcqVQqnSJXHh7DUM+3s8ylUoy4MZJSVlbF23C4WGoGiNQpKJPjBZeXt782ajzBo1aiQWwDg5OeH169diGZfU1FQkJCTgx48fPEBizM0lX21klpiYyKfMkoSpUFXIaMfNDWHW5kkFBQkL0yT4hfAp3Xcnb6iWNIDJGEv4vvSEYnE1VNg6BR9n7EBKjOS0p1Rkfb28CkKJzULtt47nzUJf/LO3OzMqxdXQbstYPJq5Fwkx+XN1LE+yvu/sCyOHw0nkUP8lUNVQg1m9Smg3eyCiP4bxpiXm5qLD6LFqJMbeW8f3c0xAGJzPPELd/q2kWQ1ke8msHtmXilnTfyGKsXOgXmX0mTUEEQGhvGkpq86jLdDIogXW/u2AlMRkFOz+kXyuZNWlVweMmT4c9razEZPloid/jiHhb9Up8/LWrZpizuxJvF/Nq9dvUbFiOWxcv4QHN8tXbMq2veFD/8aNm/fx+XNYgdbr3ygqKPKmr7n2SyAQCODu8g7GJoawm2BbuAKYIkYmAxh2ILIDMuuyzNhBxvq89OnTJ9vjWZ+YdKzvy79ZuXIl31ZmIzWrwk4r51FEmaVEx0GYkgqVLJkKFX1tJGfJaPzKtzc+0O/Tmv9drJwJipUxRuWDmTIXimnviXnAWbi1moDEAMmBw59gTTyClFSoZ8kYqRtoIz5LJoBR0VSHUd0KMKhZFi2W2YqCGnY1z0YjXR28Gomx36BVxghdDkwTPY6VYViZU61n4GtA4ekTU1j8iEnbF5pZjicNfa1sWZmsWN8WJtz7E4obaqP1lD6iAOZHdBxO222EUjEVaOhoIi4sBu1n/y16TF77FhOH1JRUaGepRwkDbXz9l3pEBqUdF8HegdAy0EbPyQOyBTCdRvVEt/F9sGHIEgR7BSK/xETH8r52+kbi2RY9A11ER0b/a+ffhRvmYKbdfLx87Ci118i+jNlrZF/CmRka6iM8TPL+DgsNh7FxlvJGBnzwQ1RUDJ9fvGgGjh07x/vHMO7uXiheXAM7t6/BipWbxT6ny5QxQ/v2LdFvwMg8q1dMVAyvV9Zsi76BXraszO9g70lySgr/Xknn6/MBRiaGvNkqOTmjb2WBEhStTryFMoBhTUgsW5KTatWq4dUr8fZlR0fxk71+/fo8U1Op0n8fPjlnzhxMnSreTONWzSrXjxcmp+C7qx+0W9VB7I2XouVabP6meD1+RaNmed60xCT4BsO93WSx9WYzB0NJUx2BC/f9cmTTnxAkpyLCzR+lWtbExxsZ7zWfv+WUrXxSXDxOtxcf0l3DpgNKNq+O26O3IC4wgv/wWNYyDWf0g6qmOp46HPnlyKaijO2Lz27+qNCyJrxvZuyLCi1rwUfCvsgJuwZQUs0+6i41MZkHL6yZqVrXhvC8knHM5qXU5BQEuH/AXy1q422m86B6i9pwvv069xtSUIByMfGPsk52Fug+oS822y5DgNsH5KeU5BS8c/VGk9YNcf/6I9FyNv/gxpNfZl4cNs7FnLEOeHInbzq05oQFHW/euKJD+1a4ePGGaHmHDq1w+fJNiY958dIJ3bt3FFvWsUNrODm5igZHsGYvQZabqbHPcnasZc2CDLUdyJukrl3Lu2YyFkiw7EiLNk1w62pGsyKbv309o7n0d7HRTJZ9u4rVoXzFsjyoKzTBC0MBTMErV64cXr58iY8fP0JTU1Ms6mUmTpyIVq1a8ZFHPXv2xL1793D9+nWxrMzChQvRo0cP3gG4f//+vLOVq6sr76z7u6ON2D0MMt/HgPnd5qOwPZdQfvNkfHfxwzcnbxhadYSqmQHCj6R9WJSabQUVUz34T97C541H9kDip3DE+3zinYAN+rSGXvdm8B25mq8XJiYj3lv8qjL1a9qQvqzL84rb7utou3ksIlw/IMzJF38NacuHP3seSfsAajR7AIqb6OL+lF1pTRDeQWKPj4/6yr8cMy/PWibp6w+JywvCjx/xCAzKaMYLDgmDl48ftLVKwDSHfgL55cXe6+i1cSxCXP0R/OY96g1qB+2S+nA6lrYv2s0ciBImurg4dSefN7fpiC/BkYj62SzJ7hnTZFR3PrQ6Xcm6FaFlostHLJUw0UNr+z48Y/ZsV94Mb5Xk9t4rGLFhIgJc/fiw51aDO0CvpAEeHkt7Xb1nDoausR72T9vG59tYd+bDq9mIJaZSw2roNMoC9w9dF2s2spz6N/ZO3ozIoAho/czwJH5PQOKPBOSHo7tOYdnWBfB08YKrozv6WFnCxMwYZw+f5+snzh0DI1MDLJi4TBS8LNm6AGsXbIKbkwf0DfXSXnNCIr7FSWeo7sbNe3DowGY4Obnw4GTUCCuUKW2GXbuP8PXLl81GyZKmGDY87UKJLR83dhjWrXHA3v3HeKfe4cP+xhDr8aJtXr16G1Mm2+GtsztevXqLShXLYbHDDFy+clvsc5x9VtvaDMSRo2d+ebH6J/ZuP4INO5bD7a0n3ji6YJBNX5Q0M8XxA2f4+hkLJsHE1AjTxs0XPeavmmmdxjU0NaCnr8vnWZDn650W/B7bfxq2owbBYeUsHNpzAuUqlMF4+5E4uOc4ChPhf2gmk0WFMgPDOueyzrfVq1dHfHw8Dhw4ILa+efPmfPgza9aZP38+OnfuDHt7e2zblvYhx7BlrE/MkiVLsGbNGt6ZjGVuRo7Mu3Tl74i+9BRKuiX4qCEVI10eZPhYL0NScFq6VsVYF6olM9KzLGgpvWAoVE30IEhI4oGMj/VSfLn3BgXF7/JLFNMtgQZTekPDSAfR3kG4brMW34LTMiVsmabZn3eUK2zcvd5j+MRZovk1W3fz/y27dsDy+RnNXgXB88oLqOtqotWk3tA00kGETxBODF3LgxSGLdMqqS8+VH3WQOiUNoQgRYCYwDDcW30STscyrlKVi6mgzfQB0C1tyO8P43vfGRem7EDiz6BSGhyvPIOmjiZ6TO4HbUNdhPh8wpZhKxD9sx46RrrQy3RMKSoqoM/MwTAobYTUFAEiAkPxz5pjeHTstqgMC3JUiqlg7M60Tv7pWGffy5vSvsSk7dbFu9DW1YLd1GEwMNKHr9cHTBwyHZ+D0vp6GBjr84AmXV8bS94UMXfVdD6JXvOpa3CYnHc3eMvszJlL0NfTxfx59jA1NYK7hzd6WlgjMDAtODQxMUaZ0iVF5T9+/MTXr1u3CGPH2iIkJAxT7BeK7gHDLF+R1ky0ZNFMmJmZICIiGleu3saChWkXXuk6tG+JsmVL8WHYee3qhZvQ1dPGpBl2MDQ2hM87Xwz/ezwf+swYGRugpJn4fcWuPczoRFy7bg306t8dQYHBaFkv7RYCn0PCYNN3DL/J3fVHZxD6ORwHdh/Dzs3i300kfykI5SRkY8Ou2Wilx48f58vzvTaTfD8HWfNG4d/7AMmC4c5LIOtWNVgAeRCoID4UWxY5JmS/kZsscov+CHlQuoTsXxj5R4nfLkAavo6SfP+k36W1JyMzW5gVygxMbqxbt47f/4V1wmXNR4cOHcL27dsL+mURQgghBUMgF/kI+Q9gWCde1jQUFxfH7/GyZcuWAmseIoQQQkj+ktkA5vTpvLnxESGEECIPhJSBIYQQQojMERStJqRC9kMOhBBCCCFy3IRECCGEkEyK1k8hUQBDCCGEyAMhNSERQgghhBRu1IRECCGEyANB0erESwEMIYQQIg8EKFIogCGEEELkgLCIZWBoGDUhhBBCZA5lYAghhBB5IECRQgEMIYQQIgeE1IRECCGEEFK4UQaGEEIIkQcCFCkUwBBCCCFyQFjEAhgahUQIIYQQmUMZmD+0QikF8uCrIBzyILzBAsi62U5LIQ8Slk2GrDt5tjzkwW4DJciDuJT4gn4JskGAIoUCGEIIIUQOCItYAENNSIQQQgiROZSBIYQQQuSBAEUKBTCEEEKIHBBSAEMIIYQQWSMsYgEM9YEhhBBCiMyhJiRCCCFEDgiLWAaGAhhCCCFEHggVUJRQExIhhBBCZA5lYAghhBA5IKQmJEIIIYTIGqGAmpAIIYQQQgo1akIihBBC5ICQmpAIIYQQImuENAqJEEIIIaRwK/TDqNu0aYMpU6ZIXDd06FD06tUr318TIYQQUhibkIR5MMkKme4Ds3nzZgiFQsiKLtbd0Gt0H+ga6eLT+0DsW7wH7155SizbpEtTdLbuivLVK0BFVQWffAJxcuNxOD96K1auxwgLdLHqCgMzQ8RFf8Wza89wdPUhJCcmS6UOFjY9MWBMf+gb6eGjTwC2L9oBt1fuEsu26NocFtY9UbFGWh0CfAJwaMMROD50EpXp3L8jZm6cke2xXSp2l1odmAbWHdB0dHeUMNRBxPtg3Fx8BJ9ee0ssW9q8CtrPGQT9iqZQUS+GL0GReHP8Ll7uuyEqo6ishObjLFC7X0toGesi6sNn3F11En4PXVHQHJ3dcOD4WXh6+SIiKhqbVy5A+1bNUFioNO8G1XZ9oKClC0FoIBLP70HqB8nnhdrgKVBp1D7b8tTPgfixejz/W7l2U6h26A9FQ1O2YyCIDEHS/QtIcbwv1Xr8ZdMBdcZ0g7qRDmJ8gvFi0VGEvpJ8TGVmbF4ZPc7OR4x3EP7pPE9imQoWTdB++wR8vOGI2yM3QVr62faC1dhBMDDSwwefj9iwcCucX0k+htt2bYW+tpaoUqMyP78/ePtjz/oDePHwdcbrrlIOo2eMQLXaVVCytCnf3om9ZyBtg4b1w4hxVjA0NoCv9wesmL8BTi+dJZY1NNLHrMVTUKPOXyhboTSO7DmFlQs2ZCtnYzcIg4b2hamZMWKiv+Dm5bvYsPx/SEpMQmEhpFFIskNbWxs6OjqQBc17tsBwh5E4u+00pnWbDM9XHlhwaBEMShpKLF+9cQ24PHbGMtvFmN59Ctyeu2Lu/gUoX6OCqEyrXq1hPcsWpzadxMR247Btxla06NkCVrNspVKHNj1bY9yiMTi+9ThGdxkLt1duWHlkOYxyqEPtxrXg9NgJc23mY2y38XB+5oJlB5agUo2KYuW+ff2OfvUGik3SDF6q92iCzgut8WTbRezpPg+Br7ww+NBMaJXUl1g+OT4Rrw/dwuH+S7Gj/Qw82XYBbab3R71BbUVl2k7vj/pD2uGmwyHs6DATTsfuov9ue5jUKIuCFh+fgKqVKmDu1HEobJTrtUCx3iORdPs0fqybjNQPHlAfvQgKOpKPqYR/duPbAuuMyWEohN+/IsXliaiM8Edc2vY2zcD3NROR/PIO1AZNhlK1elKrR4WejdF0kRXebr2E813m88Cly5EZKJ7DMZVOpYQ62mwag5AnHjmW0TTTR+MFg/H5hRekqaNFO0xdPBEHthyGVaeRcH7pis3H1sDYzEhi+XpN6uDlI0dMsZoJmy6j4PTsLTYcWoUqNSuLyqipqyE4MATbVuxCZFgU8kNXy46Ys3Qqdm46gN7treD4whm7T27mgYckqsVUER0Vi52b9sPL473EMj36dsG0+ePxv3V70L3FAMy3X4puvTpi6ry0oLmwEArzZpIVhb4JKasbN27wwOXw4cPZmpBYc9OkSZMwc+ZM6OnpwcTEBIsWLRJ7vJeXF1q0aAE1NTVUr14dd+7cgYKCAi5cuCDV120xshfunrqNOydvIcg3CPsX70VUSCS6WHeVWJ6tv7DzH/i6vsfnj59xbM0R/n/DDo1EZarWrwYvp3d4fPEhIoLC4fL4LR5ffIRKtStJpQ797Pri+skbuHbiBgJ9P2H7op0ID4lAT5ueEsuz9ad2nIG3iw+C/UOwb/UBBPsHo2nHJuIFhULERMSITdLUZGRXvD31AM4nHyDSNwS3lhzF189RMLfqILF8qEcAPC4955kaln1xO/8UHx65oUyjaqIytfq0wNP/XYLvfRfEfoqA09G7+PDQFU1GdUNBa9m0ISbZ2aJjm+YobFTb9ELyy9tIfnELgrAgJJ7fC0FsJFRaSD4vkPADwrhY0aRUpjKgrsmDlHSpvu5IcXvBtyeMCkXyo8sQhHyEUvnqUqtHLbuu8D75AN4nHiDWN4RnX76FRKG6TfZsUWYtVw2H74XnCHvjK3G9gqIC2m4dhzfrzyEuMBzSNNhuAC6euIqLx6/io28ANjhsRVhIBPrZSG6mZ+uPbD8BTxcvfPIPwvZVe/j/rTpmZPfYui1Ld+D2xXtISsqfTMXQMYNx7vhFnD12ER/ef+TZlNDgMAwa2k9i+eBPn7Fi/npcPH0N375+k1imnnktvHnliiv/3OTlnz54iavnb6Fm3b+kXBvZsX37dpQvX55/tzZo0ACPHz/+ZfnExETMmzcPZcuWRbFixVCxYkXs379ffgOYkydPYsCAATx4sbGxkVjm0KFDKF68OF6+fIk1a9ZgyZIluH37Nl8nEAh4wKOhocHX7969m7+B0qasooyKtSpla/5xfvwW1Rrk7gRgQZZ6cXXExcaJlr177YmKNSuicp20Kx7jMsZo0NYcTvccpVKHKrUqw/HRG7HlTo+cUMO8eu7roKmBr5nqwLB6HX9xBCdfH8Pyg9kzNHlJUUUJprXK48NjN7Hlfo/cUKpBxpXjr7CsSqn6lRHw8p1omZKqMlKypJKTE5JQ2rxqHr1yOaSkDMVSlZDqJX5esHmlcrk7L1Qad0SqjwuEMRE5P03l2lA0MkOqX85Zjv96TBnUKo/gR+JNqWyeNQ/lpMqAVtAqa4w3G//JsUw9+95IiI6D98mHkCZ2frNmnpeZmn8YNl/bvGauz28NTQ18yXJ+5ycVFWXUqFONBxiZsfl6DWv/8XZZ8xPbbq16aZ91pcqaoVX7Znh4+ykKE6FAIU+m33Xq1CneV5V9n759+xYtW7ZE165dERgYmONj2Hf53bt3sW/fPnh7e+PEiROoVi3jolCu+sCw6G7u3Lm4ePEi2rbNSN1nVbt2bTg4OPC/K1eujG3btvE3qWPHjrh16xb8/Pzw4MEDnp1hli9fztdJUwk9LSgpKyE2MlZseWxELHQMc9cEZmnXC2oaxfDsSkaq/Mnlx9DS18byc6v5hwf7ELp++Br+2X42z+ug/bMOWbMjbF7PUDdX2+g/uh/UNdTw8PIj0bJAv09YM3UdPrzzR/ESGugzojc2X9gIu05jeNYmr2noluD9Vb5HfhFbzuY1DbV/+djJL7ZCQy/t8Y82neMZnHQsI9NkZDcEvvRCdEA4yjevgaqdGkBBUaauEfKVQnEtKCgpQRAnfl6wzIqi1r+fF6zPjNJfDZBwZF32lWoa0Fx8kH0zsysXJJzdgVQfyX0g/iu1n8fEjwjxYyo+4gvUczi/tcobo+GcgbjSZymEqZJ7TbLgp+rfbfBPp7mQNh09bSgrKyM6Uvz8joqI5v3dcmPImIG8yejOpXsoKLp6Orwe7HVnFhURBQOjXzfn/cq1C7ehp6+LY5f38s9aFigdP3AWe7YeQmEiLKA+MBs2bMCIESMwcuRIPr9p0ybcvHkTO3bswMqVKyW2pDx8+BAfPnzgrSVMuXLlfvt5ZSKAOXfuHMLCwvDkyRM0apTRhJJTAJOZqakpwsPTUq8syitdurQoeGH+bXvpqS42ZZYqTIWSgtLvVSRL4yI7EXLT3tjCohUG2g/GypHL8CUq40OyRpOa6DdhAHbP3wmft94wLWeKEYvsEBMejTNbTkEqsr7gXNahrWUb2Ey1xsLhDoiNyvjCevfGi0/p3F97YOeN7eg1rBf+t3A7pEX4B/viUP8lUNVQg1m9Smg3eyCiP4bxpiXm5qLD6LFqJMbeW5fWJBYQBuczj1C3fyup1UF+/NkxxTrzCuO/8+aibBLj8X3tZCgUU4NS5TpQ6zUC8VGhvHlJarKdGxKW/WwWard1PG8W+uIfKnFTKsXV0HbLWDyeuReJMZKbNfLvvPj3ndGpV3vYTRuG6cPmIibT+V1Qsr3mXNYjJ42a1cdo++FYMms1XN+4o0z50pi7bBoiwiKxY8M+yJtECd95rJmHTVmxpkEnJyfMnj1bbHmnTp3w7Nkzidu/dOkSzM3NeSvJkSNHeKuJhYUFli5dCnV1dfkKYOrWrYs3b97gwIEDaNiwIT+pcqKioiI2z8qypiOGHcC/emxOWAS5ePFisWVVtSrjL+3cNQ+w0UGpKanQyZKp0DbQxpcsWRlJnX8nrJ2EtWNXwfWJi9i6wdOt8PCf+7xfDRPoHQA1DTWMXTUBZ7eeztMRWl9+1kE3y9WYroEOYrJctUnq/Dt93VQsGb0Mb56INxdkxV6zt4s3SpU3gzT8iImDICUVmlmujDX0tbJlZbJifVuYcO9PKG6ojdZT+ogCmB/RcThttxFKxVSgoaOJuLAYtJ/9t+gxJDvW+VaYmgrFErrInINQ0NTmWZjcNB/xkUWpKRI2LoQw8jMPjQTB/lA0Ls1HJsVLIYBhTTzsmNIwEj+m1A20ES/hmFLRVIdh3QrQr1kWzZbZioIalq0b8fEQrg9ejYTYbyhRxgidD0wTPY6VYViZ061nIC4g7/rExEZ/QUpKCvQNxc9vPQNdRP9LnzTW+XfB+lmYbbcQrx5njDAsCDHRsbweWbMt+gZ62bIyv2PS7DG4dOYa71fD+Lzzg7qGOpasm4udG/cXmtGwwjx6GZK+81jLRtY+pUxkZCRSU1NhbCzeSZrNh4ZKDtBZ5oUlJFh/mfPnz/NtjBs3DtHR0b/VD0Ym8tusc8/9+/d589HEiRP/eDusfY21ybFsTrrXr8XbfCWZM2cOvnz5IjZV0cp9R9mU5BT4ufmiTkvxURB1WtblnXB/lXmZuH4KNkxcJ7FfSzH1YhBkGbSfytLRCmmBW15idfBxe48GLeuLLWfzHo6Sh7ymZ15mbpyOFRNW4eW9V7l6rorVKyJKSiMWBMmp+Ozmjwotxdv1K7SshSAnySMQJGFvr5KqeLDMpCYm8+CFNSlU69oQ3rcK9gO9UEtNgSDIF0pVxc8Lpap1kfox5/OCl6lUE4qGJXnn31xhpwNrTpLSMRXp5g+zLMcUmw9zzH5MJcXF42z72XzIdPr07sg93vmX/R3+1g9f/D5nKxNw6w1Cnr3jf38Picrz89vL1QeNW5mLLW/Uyhyuju6/zLws3DgH88cvwdO7EjJh+Sw5OQUeLl5o1rqx2PJmrRvh7es/v6WBurqa6EI4nSA1lX8O5PVnbWHoAzNHwnceW/YrWd+HXyUM2HvJ1h07doy3gnTr1o03Qx08eBDx8fHylYFhqlSpwoMYNtKItXGyNrbfxfq6sGDI1taWp67i4uJEnXh/dRBKSp39bvPRpb0XMHnjVPi5vof3Gy90HNyFD6G+efQ6X281ywZ6JvrYYr9RFLxM3miPfYv2wOetl6ivTFJCEn7E/eB/v77zio9u8nf/AB9nH96ENHj6ELy+/SrbyZYXzu4+h9mbZ8LH1QeeTp7oPqQ7jMyMcPnIFb5+xOzhMDDRx+opa0XBy+xNM/E/hx3wfPMOuj8zUEkJifj+sw7W9lZ49+YdH53EOgD2Ht6Ld+LdMn8bpOXF3uvotXEsQlz9EfzmPeoNagftkvp86DPTbuZAlDDRxcWpO/m8uU1HfAmORJRfWp+c0g2rosmo7nxodbqSdStCy0SXj1gqYaKH1vZ9+BX1s11p701B+vEjHoFBGf2JgkPC4OXjB22tEjA1kTxENr8kPbgAtSFTkfrpPQQfvaDStAsUdQ2R/DTtvFDtYQNFbX0kHEs7L9KpNO6E1I9e/L4xWal26IfUQF8Ioj5DQUkFStUbQKVhOySe2SG1erjtvo42m8ciwvUDwp18UW1IWz78+d2RtGOq4ewBKG6iiwdTdqU1MXoHiT0+PuorD34zL89aJunrD4nL88rx3aexeMs8eLp6w83RA72tesLEzAjnDqdlHcbPsYOhiQEWTV4hCl4Wb56H9Qu3wN3JU5S9SeDn93f+N+uXx+4Fk54dNzQ1QJUalfDjezyCPgZLpR4Hdx7H6v8thruzJ5wd3TDAujdMS5ng5KFzfD0b+mxkaojZEzKyCdVqVuH/axRX51knNp+clAw/H3++/P6tx3x00zs3b7i88UDZ8qV4VubezcdS+awtaMVyaC6SxMDAAEpKStmyLazrRtasTOauHWZmZnxEcbq//vqLBz1BQUG8/6pcBTBM1apVce/ePR7EsDfsd7HHsOHSrKMRa4qqUKEC1q5di549e/JUljQ9vfwEJXS0MGDy37wZJtAngN/jJSI4rYmBLTPMdD+VzkO68JN/9PKxfEp378xdbJ2WFryxfi5shw+eYcWDn69RX+F45xWOrj0ilTo8uPwQWrpasJ4yBHrsRnbeAZhjMx/hwWmpbNbZjwU06XpYded1mLxiIp/S3Tx9i3fcZTS1NDF19RQe3LCgxtfdF/Z9p8Hb+d9vAPanPK+8gLquJlpN6g1NIx1E+AThxNC1PEjhr8lIR+yeMLzPwqyB0CltCEGKADGBYbi3+iScjmV0VlQupoI20wdAt7Qhkn4kwve+My5M2YHEn186Bcnd6z2GT5wlml+zdTf/37JrByyfn9FEURBS3j5BooYWinX+GwpaehB8DkD8rsWiUUWKWnpQ0M1yTxg1DSjXaYbEf9LqkY2qGtT6j4WCtj6QnARBeBASjq7nzyUtHy6/RDHdEqg/pTdvSor2DsINm7X4FpyWKWHLipsZoDC7feketHW1MNLeljfB+Hn7Y4rVLD4EmWHLTDLdS6WPlQU/v2etnMqndFdOXcdi+7SOm+xGcsduZzQJWI8dxCd2z5gx/SZLpR7XL97mnZLHTxvJn/+9lx9GD5qCkKC0L1i2rKRZRj9I5sK9Y6K/a9atjp59u/D717Q3t+TLdmxIayaaPGcsjE0M+X1jWFCzaYX0+unJym8hqaqq8mHTbLRv7969RcvZvKVl2vuXVfPmzXHmzBl8+/YNmpqafJmPjw8UFRVRqlSpXD+3grCwNN4VkKdPn/L7wvj6+vLsTG71LiP53iey5qtAvKOWrGqjJPnGZ7JkttNSyIOEZdL5YspPJ8/+ekSarNgtyHkYqyyJS8l9s0Jh5RX+790V/ivf6p3zZDuVPG/+9jBqa2tr7Ny5E02bNuW3KNmzZw88PDz4fV5Y81NwcDC/BQrDAheWcWnSpAnva8P6wLDEQuvWrfnjckumMjB5gXUYYhEfS1GxoGXy5Mk8Gvyd4IUQQgghaQYOHIioqCh+37XPnz+jZs2auHbtGg9eGLYs8z1h2Hcwy9CwPq1sNJK+vj6/L8yyZcvwO4pcAMP6vbA79X769Im33XXo0AHr168v6JdFCCGE/CeCAmhCSsdGEbFJEtY5V9KgmvSbzP6pIhfAsDv45nQXX0IIIURWCQswgCkIRS6AIYQQQuSRkH6NmhBCCCGkcKMMDCGEECIHhEVsTDEFMIQQQogcEFITEiGEEEJI4UYZGEIIIUQOCGgUEiGEEEJkjbCIBTAy8WvUhBBCCCGZURMSIYQQIgeENAqJEEIIIbJGQE1IhBBCCCGFGzUhEUIIIXJAWMQyMBTAEEIIIXJASH1gCCGEECJrBEUsA0PDqAkhhBAic6gJ6Q9pKapCHkSl/oA8CFRIgqxLWDYZ8kBt/mbIuuDzCyAP1IUqkAfKKkoF/RJkgrCIZWAogCGEEELkgKCIBTDUhEQIIYQQmUMZGEIIIUQOCFG0UABDCCGEyAEBNSERQgghhBRulIEhhBBC5ICwiGVgKIAhhBBC5IAARQuNQiKEEEKIzKEMDCGEECIHhKAmJEIIIYTIGEERG0dNGRhCCCFEDgiKWAaG+sAQQgghROZQBoYQQgiRA8IiloGhAIYQQgiRAwIULdSERAghhBCZQxkYQgghRA4IqQmJSEs7q87oNtoS2ka6CPH5hGNLDsDn9TuJZSubV8PA2dYwrWgGVXVVRAZH4sHxW7i574qojFnl0ug99W+Uq1UBhqWMcGzJftzaf1WqO7CXrQUGjRkAPSN9fPT5iK0O2+H6yk1i2VZdW8DSxgKVa1SEiqoK/H0CcGD9Ibx+6ChWxmriYJiVM4OyihKC/INxatcZ3Dp3R6r1aGPVCZ35vtBBiE8QTi05gPevvSSWrWReDX1nD4EJ3xfFEBUcgUfHb+POvoz3uuXf7dG0T2uUrFqazwe4fcD5tSfw0cVXanVQad4Nqu36QEFLF4LQQCSe34PUD54Sy6oNngKVRu2zLU/9HIgfq8fzv5VrN4Vqh/5QNDQFFJUhiAxB0v0LSHG8j8LA0dkNB46fhaeXLyKiorF55QK0b9UMhUVD6w5oNro7ShjqIPx9MG4sPoLA194Sy5Yxr4IOcwbBoKIpVNSL4UtQJByP38WLfTfEyjUZ3gXmVu2hbWaAH9Fx8Lz2CnfXnEJKYrJMnN+ZtbNoi0U75uPxjaeYN2IhpKmPrSWGjBkIfSN9+Pt8xCaHbXDJoR6tu7ZEH16PSlBVVcEHn4/Yt/4QXj58LVbGduIQlPr5OfXJPxgndp3GjXO3UZgIULRQBiafNOrRDEMWDsPhBXvg4+iFtkM6YdrBeZjTcQqiQyKzlU+MT8Sdw9fx6V0AEuMTUMX8LwxdMRqJPxLx4ETaScMCm4jAMLy+9gyDFwyTeh3aWbTBxEXjsGHuFri/doeFdQ+sOboSNm2GIzwkPFv5Ok1qw/GRE/as2odvX7+h68AuWHVwGcb0mID3Hmlf7F9j43BkyzEE+n5CcnIymnVoitkbZiImMjbHD8L/yrxHMwxcOAzHFuyBr6M3Wg/piEkH58Gho30O+yIB9w/fQBDfF4k8oLFeYcf3xeMTaYFW1SY18OrSE/i98UFyYhIPjuyPzIdDx6mIDYvO8zoo12uBYr1HIvHsTqT6e0KlWReoj16E7yvHQxgbka18wj+7kXj5YMYCRSUUn7kFKS5PRIuEP+KQdPs0BOFBEKakQLlGQ6gNmoz4b7FI9XqLghYfn4CqlSqgV7dOsJ+3DIVJjR5N0GWhNa4uOIBARx+YD24Hq0Mz8b8OM/ElJCpb+aT4RLw6dAth7wKRHJ+IMg2roseK4Uj+kQinE2kBY61ezdBh1kBcnLkHn5x8oF/eFL3Wj+brbi49KhPndzpjMyOMWzgaLi9cIW3tLdpiyqLxWDt3E1xfu6O3dU9sOLoag9sMRZiEetRrUhuvHjlh56q9iPv6DT0GdsXag8sxssc4+Ig+p77i0Jaj+OgbiJTkFDTv0BTzNszin1OZAx0ix31gvn//DhsbG2hqasLU1BTr169HmzZtMGXKFL5eQUEBFy5cEHuMjo4ODh7M+OANDg7GwIEDoaurC319fVhaWuLjx49ijzlw4AD++usvqKmpoVq1ati+fbtoHSvLnueff/5B27ZtoaGhgTp16uD58+dSrXuXkT3x6PQ9PDx1F5/9gnF8yQFEf45Ce6vOEssHevjjxaUnCH7/CZFBEXh24RHcHjmjSsO/RGX8Xf1wauVhvLz8FMlJ0rkiy2zAqH64evI6rp64hgDfQH51FhESjl42PSWWZ+tP7DgFLxdvnllhH3Ts/2Ydm4rKOD934VdkbHshAZ9xdt8/+PDuA2o3qim1enQc2QNPTt/Dk1P3EOoXjFNLDiLmcyRaW3WSWP6Tx0e8uvQUIe+DEBUUgZcXHsPjkQsqZ9oXe6dswYOjt/DJ8yNC/UJwePYufpz91Vw69VBt0wvJL28j+cUtCMKCkHh+LwSxkVBp0VXyAxJ+QBgXK5qUylQG1DWR/DIj05Xq644Utxd8e8KoUCQ/ugxByEcola+OwqBl04aYZGeLjm2ao7BpOrIr3px6gDcnHyDSNwQ3lhzFl89RMLfqILF8qEcA3C89R8T7YMQGRcL1/FP4PXJDmUbVRGVK16+MQCcfuF18xsv4PXaD26XnKFm7vMyc34yioiIWbJuLA+sOISTwM6Rt0Kj+uHzyGi7/rMcmh//xAIxlWSRh64/tOIl3P+vBAhmWYWnRMSO79/a5Cx7eeMK3FxwQgtP7zsHvnR/qSPFz6k8zMII8mGRFvgYwM2bMwP3793H+/HncunULDx48gJOTU64f/+PHDx50sADo0aNHePLkCf+7S5cuSEpK4mX27NmDefPmYfny5Xj37h1WrFiBBQsW4NChQ2LbYmWmT58OZ2dnVKlSBYMGDUJKSgqkQUlFGeVqVoT7Y2ex5e6PXVCpQdVcbaNMjfK8rPdLDxQEZRVlVKldJVtW5PVDJ9Q0r5GrbbAvdA1NdcTFfs2xTP0W9VC6Yim4vJCc7s2LfVG2ZgV4PnYRW+7x2BUVc7kvStcox8v6/GJfsOwYe67vsd+Q55SUoViqUrasCJtXKpcRVP2KSuOOSPVxgTAmIuenqVwbikZmSPUrmGNOViipKKFkrfI8wMiMBSSlG1TO1TZMapTlAUvAy4wmZdb8VLJmeZjVqcDndUsbonLbOnh/T/xzpLCf37b21oiN+sKDI2lj9ahauwpeZanHy4eOqGVe87fqwbIuOTFvUR9lKpbG23zIKP1uHxhhHkyyIt+akL59+4Z9+/bh8OHD6NixI1/GgopSpUrlehsnT57k0fzevXv5QZaebWFZGhYMderUCUuXLuWZnT59+vD15cuXh6enJ3bt2gVbW1vRtljw0r17d/734sWLUaNGDfj6+vKMTV4roVsCSspK+BLxRWz5l4hYaBvo/PKxG5/vRgk9LSgpK+L8ptM8g1MQtPW0oayshJjIGLHl0ZEx0DPSy9U2Bo7uDzUNddy7/FBsefESxXHO6RRvf05NFWDj3M1wfJz7wPZ3aP7cF18jYsWWx+ViX6x5vhOafF8o4dKm0zyDk5O+s4YgNjQank/zPhBTKK4FBSUlCOLE68AyK4paOv/+eC1dKP3VAAlH1mVfqaYBzcUH2TcBIBAg4ewOpPrk/RemPNHQLQFFZSV8jxQ/v9m8pqH2Lx879cVWaOilPf7BpnM8g5PO/fILaOhrYfhZB7DvFBYQvz5yG092XJaZ85sFP90HdcWIjnbIDzo/68Fed2asXnpGurnaxuDRA6CuoYa7lzP2Rfrn1CWnM6LPqXVzN+G1lD6nSCELYPz8/HiWpGnTjPSinp4eqlbN3VUvw7I1LMgoUaKE2PKEhAS+/YiICHz69AkjRozAqFGjROtZZkVbW/yDpHbt2qK/WXMWEx4eLjGASUxM5FNmqcJUKCko4XcIIf5DFSwI+7efrljefz7UiquhYr0qGDDLCuEBobxpqaAIs7xgFkcKsy6UoL1lWwybZoO5wxciNkr8i/fHtx8Y0ckO6sXV0aBFfYx3GMtTzax5SVqyvWK+L35djzX9F6JYcTVUqFcZfWYNQURAKG9ayqrzaAs0smiBtX87SK2zZZrsOyMXu4J35hXGf+fNRdkkxuP72slQKKYGpcp1oNZrBOKjQnnzEvmXvSHx5Pj1Y/b3XwJVDTWUqlcJHWYPRPTHMN60xJRr8hdajbfk/WqC3vpBr5wxujpYI25SLB5tEW9qL4znNzufF2ydg7UzNuBLTM7ZjPzZF5JO+uw6WrbDiGm2mDV8PmIkfE7ZdhrJ68UyMJMcxiE4MIQ3LxUWAtlJnshWAJObk4B/oWcpxzp2phMIBGjQoAGOHTuW7bGGhoY8kElvRmrcuLHYeiUl8WBDRUVF7HnTty/JypUreZYms9ra1VBXJ3d9A+Ji4pCakgodQ/GrYy0DbXyNFD9JsooMSut0FuQdCG0DbfSaPKBAApgv0V+QkpIKPUPxqxhdfV3ERIhf7UjqHDhr/XQsHL0ETo/fZFvP9nnwxxD+t6+HH8pWKgOrCYOkEsB8+7kvtLPsixJ8X4hfQee0L4K9A/m+6zl5QLYAptOonug2vg82DFmCYK9ASIPw+1cIU1OhWEJXrL1aQVObZ2Fy03zERxalSmgyFQohjPzMP+sFwf5QNC7NRybFUwCTox8xcRCkpEIzyzFVXF8L3/7lmIr9lNaEF+79iWdr2kzpIwpg2k7rB5fzT0RZGVZGVaMYeq4cgcdbL+bqM7Ugz2+zciVhWsYUKw9mdLhWVEz7rL0XcAtWrWx5v7e8FPuzHvqGetnqEf0v9WCdf+eun4F5oxfjdQ6fU0E/P6fee/ihXKWysJkwpHAFMChaEUy+9YGpVKkSDxpevMi46ouJiYGPj49YEPL5c8YB/f79e97vJV39+vX5MiMjI769zBPLsBgbG8PMzAwfPnzItp41Jf2pOXPm4MuXL2JTLe3cZ45Sk1Pw0d0PNVrUEVteo0Vt+DpJHmYpkYIClItlBF75ifW893H1gXmrBmLL2by7o8cvr8zmbJiJJeNX4MXdl7l6LhZQsmGZ0sD2RYD7B/zVIiMDx1RvURt+v70vxOP/TnYW6D6xHzbbLufDqKUmNQWCIF8oVa0ntlipal2kfpQ8LF9UplJNKBqW5J1/c4V9HrLmJJLz7khORYibPyq2FO9jUbFlLXxyep/7d04BUM503LPh1cIsPy8sSBWkpUUUCv/5HegbCNt2I3h2NX16eus53j5z5n+Hh+Tc/+q/1MPb1QcNW5mLLW/UqgHcHN1/mXlZsGEWHMYvw7O7EjKTOXxOseakwkSYR5OsyLcMDOtsy5p2WEdeNnqIBRusIy3r05KuXbt22LZtG5o0acKzIbNmzRLLlAwZMgRr167lI4+WLFnC+88EBgbyEUVsu2x+0aJFmDRpErS0tNC1a1fe9OPo6MiDpalTp/7Ray9WrBifMvvd5qMbey9j9IZJfOSQ7xtvtB3cEfolDXDvWNoXSf+ZQ6BrrIfd07by+fbWXRAVEslHLDFVGlZD11EWuHMooyMcaxM3q1xK1HlN11gfZaqXQ8L3BN7UlNdO7zmLeZtnw9vFBx5Onuhp1R1GZka4eCStTd5u9ggYmBpgxeTVaXWwbMvLb3H4HzzfeIqu7hITkvA97jv/e8iEQXx7rGe/iooymrRvjM79OmL9nM2Qltt7r2DEhokIcPXjw55bDe4AvZIGePhzX/SeOZjvi/3TtvH5Ntad+fBqNmKJqdSwGjqNssD9TPuCNRtZTv0beydv5qPGtH5ejSd+T0Dij7TMYF5KenABakOmIvXTewg+ekGlaRco6hoi+Wnaa1LtYQNFbX0kHNso9jiVxp2Q+tGL3zcmK9UO/ZAa6AtB1GcoKKlAqXoDqDRsh8QzO1AY/PgRj8CgtCtgJjgkDF4+ftDWKgFTE6MCfW3P915Hn41jEeLqj09v3qPBoHbQLqkPx2NpfdbazxwILRNdnJ+6k883tOmIL8GRiPRLqw8bRt1sVHc+tDqdz503aDqyG0I9PiLI2Q96ZY3Rblo/eN9+ky2wKYznd1JiMvy9xUeIsuHWTNbleenEnjNw2DyHj45yc/JAL6seMDYzxvmf9Rg7eyQMTQ2xZPJKUfCycPMcbHTYBvccPqdsJgzmo5TSP6eatm+Crv06Yc0c8fOLyPF9YFjwwTrzWlhY8H4s06ZN49mMdKzz7bBhw9CqVSuULFkSmzdvFhulxIY8s9FHLLBhnXTj4uJ4xqV9+/Y8YGFGjhzJy7HnmjlzJooXL45atWqJhmoXlFdXnkFTpwQsJ/eHjqEugn0CsWHYCn5TNIbd3E7PzEBUXkFRgQc1hqWNeJNHeGAYzqw5hvs/v2QZXWNdLL22XjTPbpLHpncv3LHqb4c8r8O9Sw+gpavFRxXoG+nxD6FZ1nMQFpzWtKJvrA/jkhlfJBZWPXhgNXXFZD6lu376Jlbar+F/s85yU1dOgqGJIRITEhHo9wnLJq3kzyUtjnxfaKLH5H7QNky7qeCWYSsQHZx2DxidLPuCpb37zBwMA74vBIgIDMU/a47h0bGMm1ixIEelmArG7pwu9lyss+/lTWfyvA4pb58gUUMLxTr/DQUtPQg+ByB+12LRqCJFLT0o6BqKP0hNA8p1miHxn92SN6qqBrX+Y6GgrQ8kJ/H7wSQcXc+fqzBw93qP4RNniebXbE2rh2XXDlg+f1oBvjLA48oLaOhqovWk3tA00kG4TxCODV3LgxSmhJEOD2gyn9/sHi86pQ0hSBEgJjAMd1afhNOxjI7hj7Ze4H1S2k3vjxImevgR9RXed9/i3trTUqmDNM7vgnD30n1o62phuL0Nr8cH74+YZj0bocFhEuvRy6onr8eMFVP4lO7q6RtYZp8WrKlpqGHGyikw+vk5FeAXiEWTVvDnKkwEKFoUhHnZkPoH2H1g6tati02bNkGW2JbrC3ngn/zvfSZkQVWV3I2UKMw29Mr7TE1BUJsvvexZflneYAHkwb2UvM/EFoRkOfhqfh4s/WDnrOmQPNlOv8/Z+5kWRvRjjoQQQgiROfRTAoQQQogcEKJoKfAAht2AjhBCCCH/jaCIvYHUhEQIIYQQmVPgGRhCCCGE/HeConUfOwpgCCGEEHlAd+IlhBBCCCnkqAmJEEIIkQNCFC0UwBBCCCFyQEB9YAghhBAiawQoWmgYNSGEEEJkDjUhEUIIIXJAiKKFAhhCCCFEDgiKWB8YakIihBBCiMyhDAwhhBAiBwQoWiiAIYQQQuSAAEULNSERQgghROZQBoYQQgiRA8Ii1omXApg/5J0UBXmQIkyFPHBMCIGsO3m2PORB8PkFkHXznJZCHmwp3xnywEBNu6BfgkwQoGihAIYQQgiRAwIULdQHhhBCCCEyhzIwhBBCiBwQomihAIYQQgiRA4Ii1omXmpAIIYQQInMoA0MIIYTIAQGKFgpgCCGEEDkgQNFCTUiEEEII+U+2b9+O8uXLQ01NDQ0aNMDjx49z9binT59CWVkZdevW/e3npACGEEIIkZNRSMI8mH7XqVOnMGXKFMybNw9v375Fy5Yt0bVrVwQGBv7ycV++fIGNjQ3at2//R/WlAIYQQgiRk1FIgjyYfteGDRswYsQIjBw5En/99Rc2bdqE0qVLY8eOHb983OjRozF48GA0bdr0j+pLAQwhhBBCRBITE/H161exiS2TJCkpCU5OTujUqZPYcjb/7Nkz5OTAgQPw8/ODg4MD/hQFMIQQQoicdOIV5MG0cuVKaGtri01smSSRkZFITU2FsbGx2HI2HxoaKvEx79+/x+zZs3Hs2DHe/+VP0SgkQgghRA4I82g7c+bMwdSpU8WWFStW7JePUVAQb3sSCoXZljEs2GHNRosXL0aVKlX+0+ukAIYQQgiRA4I8CmFYsPJvAUs6AwMDKCkpZcu2hIeHZ8vKMHFxcXB0dOSdfSdMmJD2ugUCHvCwbMytW7fQrl27XD03NSERQggh5I+oqqryYdO3b98WW87mmzVrlq28lpYW3Nzc4OzsLJrGjBmDqlWr8r8bN26c6+emDAwhhBAiBwQF9Lysucna2hrm5uZ8RNHu3bv5EGoWmKQ3SQUHB+Pw4cNQVFREzZo1xR5vZGTE7x+Tdfm/KRIBTLly5fgYdTYRQggh8khYQM87cOBAREVFYcmSJfj8+TMPRK5du4ayZcvy9WzZv90T5k/IVRPSwYMHoaOjg8Kqr60l/nlxAg8/3MLBG7tQp1GtHMu26doSW06uw3W3C7jrfRV7Lv0PjVs3FCtjObg7dp7fgluel/m09dR6VK9bTap16GfbCxdfnsJT/zs4cnMv6jaunWPZtt1a4X8nN+C2+2U88LmB/Zd3oEmbRmJlKlQphzV7l+LSq9Nw/PwYg0b1R0HoP7Q3rrw6gxcf7+HYzX2o17hOjmXbdWuNHac24Z7HFTx+fwuHruxC0yz1yg9/2XTA3882YJjvfvS6thQmjarm6nHG5pUx4uMh9Lm5PMcyFSyaYFTQUXTcK/2gv6F1B0x+shHzvQ/A7soylGmYcz3KmFfB8HMOmOm8E/O8D2DC3bVoMqJLtnJNhnfBhHtreRn751vQeYEVlIupoKA5Orth/EwHtLUYgprNu+Luo5yHmRaE4SMH463bPYREuOPeo/No0sw8x7LGxobYvW8DXr65icgv3lixal62Mj0sOuHuw3/g/8kJn0Jd8PDpJQz421LKtQAGD+uHu44X4fbpKf65cwTmTXK+y6uhsT7W71yGG8/PwSvsFeYuE++8ms529CBexjXwCR46X8GcpVOhWkxVirWQLePGjcPHjx/5cGs2rLpVq1Zi380PHjzI8bGLFi3izUdFOoApzDpYtMWUxRNwcMtR2HYaCeeXbth4bA2MzYwklq/bpA5ePXLEVKtZGNrFDk7P3mLdoRWoUrOSqEz9ZnVx+8JdjO9vj1EW4xEaHIbNJ9bB0MRAKnXoaNEO05ZMwv7NRzCk0wi8femCLcfW5liHek3q4OUjR0weMgPWnUfC8dlbbDy0ClVrVhaVUVNXQ1DAZ2xbvguRYVEoCJ0s22PGksnYt+kwBnUchrcvXbHt+DqYmGXvgMbUb1IXLx69woQh0zGk03C8fvoGmw+vEauXtFXo2RhNF1nh7dZLON9lPkJfeaPLkRkoXlL/l49TKaGONpvGIOSJR45lNM300XjBYHx+4QVpq9GjCbostMbjbRexs/s8BL7ygtWhmdDOoR5J8Yl4degWDvRfiv+1n4FH2y6g3fT+aDCorahMrV7N0GHWQDzcfJ6XuTRzD2r2bIL2MweioMXHJ6BqpQqYO3UcCpvefbphxep52LBuB9q0sMSLZ444fW4vzEqZSizPvryjIqOxYe0OuLtJPlZiomP59jp3GICWTXvi+NFz2LZjFdq1byG1enTr1RFzl03Dzk370avdEDi+eIs9J7fANIfzmfXhiI6Kwc6N++Hl8V5imZ59u2D6/AnYtnY3ujbvj7lTlvLnmTY/rROqvA2jlhUFEsCwHserV69GpUqVeE/nMmXKYPnytKvBWbNm8aFVGhoaqFChAhYsWIDk5GTRY11cXNC2bVuUKFGCdwZinYdYj2YW3Q0bNozfmpgN3WITi+okYWXs7Ox4uxvbBuvxzLYrTYPs+uPyiWu4dPwqPvoGYpPDNoSHhKOPjeSrEbb+6PaTeOfijU/+wdi5ai8++QehRceMTlEOE5bj3KGLeO/hiwDfQKycvg6Kigowb1FfKnUYMnogLp64iovHr+Dj+wBsWLgVYSHh6GfbW2J5tv7w9uPwdPHir337yt0I9A9Cy47NRWXYui1Lt+PWxbv8hkgFwWr0QFw4cQXnj1+G//sArFu4GaHB4eifQ73Y+kP/Ow5PZy9en20rd/H/W3eS3odyVrXsusL75AN4n3iAWN8QvFh0FN9ColDd5te35G65ajh8LzxH2BtfiesVFBXQdus4vFl/DnGB4ZC2piO74s2pB3hz8gEifUNwY8lRfPkcBXOrDhLLh3oEwP3Sc0S8D0ZsUCRczz+F3yM3lGmUkXksXb8yAp184HbxGS/j99gNbpeeo2Tt8ihoLZs2xCQ7W3Rsk3EOFBbjJgzH0cNnceTQGfh4+2Hu7OUICQ7lWRlJPgUGY86sZTh14gK+fo2TWObpk1e4evk2395H/0Ds2nEIHu7eaNI058zOfzVszBCcPXYRZ45ehN/7j1gxfwO/uGNZGUmCP33G8nnrceH0VcR9/SaxTL2GtfHmlQuu/HOTl3/64CWu/nMTter+hcJEUEB34i1SAQzr0MMCGBaceHp64vjx46LhViwwYekmtnzz5s3Ys2cPNm7cKHrskCFDUKpUKbx+/ZqnqdjNcFRUVHhvZ3b7YhaQsPY2Nk2fPj3bc7OhWt27d+dDvlgbHdtG/fr1+W8xREdHS6W+yirKqFq7Kl4+fC22nM3XMq+Rq22wgExDUwNfYyV/UDBq6sWgpKz8yzL/pQ7ValfBi4evxJa/ePgatc1r5roOxXkdvqKwYPX6q3ZVPH+QtV6vUKdh7uulUVwdX/KpXooqSjCoVR7Bj9zFlrN51jyUkyoDWkGrrDHebPwnxzL17HsjIToO3icfQtqUVJRQslZ5HmBkxgKS0g1yl80yqVGWBywBL9+JlgW+9kbJmuVhVqcCn9ctbYjKbevg/b3fT1EXFewztE69Grh/74nY8vt3n6BR47y7IGrVuikqVS6PZ0/FPwvzioqKMmrUqYanD16ILX/y4AUPQv6U0wtn1KjzF2rXS/u8Ll3WDK07NMeD2+LvF8lf+d6Jl40BZ4HJtm3bYGtry5dVrFgRLVqkXb3Onz9frPPttGnT+A9FzZw5ky9jHYFmzJiBatXSrrgqV874oGN3C2RfJiYmJjk+//379/kQLjZGPX2c+7p163DhwgWcPXuWZ2ayYm16WW+jLBAKoKiQu/hPR08byspKiI6MEVseHREDfSO9XG1j8JgBUFdXw91L93MsM26eHSJCI/H6sRPyWlodlPlrzozNGxjmrg5WY/7mTUa3L91DYaGrp/OzXuLBaxTbN4a/bo5JZz12ENQ11HHr0l3kBzW9ElBUVsKPiC9iy+MjvkDdUHIfMK3yxmg4ZyCu9FkKYarkJDELfqr+3Qb/dJqL/KChm1aP75Hi9WDzmobav3zs1BdbofHzfXiw6RzP4KRzv/wCGvpaGH7WAVBggZIyXh+5jSc7LkutLrJOX1+XnwcR4ZFiy8MjImFk/N+apEtoacLD+wmKFVNFaqoAM6YuwoP7TyHN8zky2/kcDQOjP6/H1Qu3oGegi+NX9vLvGBYoHdt/Bru3HII83gdGVuR7APPu3TseDOT065MsiGCZFF9fX3z79g0pKSk8q5J5uBb7wagjR46gQ4cO6N+/Pw+AcotlXNh29fXFv5zi4+P57zJIwm6hzO4amJmZZlmUKlEOv4NlfzJjJ0KWRRJ17NUOI6cNxcxh8xETFSuxjNW4v9HRsj3G95uCpETpNcVkrwPr+f7vlejcqz3spg/DtKFzcqxDQZJYr1zsnC69OmDM9OGwt52NmMh8rlfW18dSvxJeM2sWard1PG8W+uIv+dbeKsXV0HbLWDyeuReJMZLT6NKS7X1OO6h+aX//JVDVUEOpepXQYfZARH8M401LTLkmf6HVeEtcXXAAQW/9oFfOGF0drBE3KRaPtlyQYk1kX7bzAOwz6r99KX6L+47WzS1QvHhxtG7TFMtWzOHNSax5KX+PqT+vR6NmDTDGfhgWz1oFFyd3lC1fGvOWT0dEWCS2b9iHwkKIoiXfAxh1dfUc17148QJ///03DxY6d+7MMyonT57E+vXrRWVYvxZ2G+KrV6/i+vXr/IegWJnevSX3V5DU/8bU1FRij+icRjBJuq1yh6o9kFux0V+QkpIK/SyZCl0DnWxX/pI6/85bPxNz7RblmFkZPGYgbCdaYeLAafB99wHSkFaHlGwZI10DXZ6t+LfOvws2zMasUQvxSgrZof+CdTJMq5d4QMuutqIjo/+18+/CDXMw024+Xj52RH5hTTyClFRoGIkfr+oG2ojPks1gVDTVYVi3AvRrlkWzZbaioEZBUZGPRro+eDUSYr+hRBkjdD4wTfQ4VoZhZU63noG4gLztE/MjJq0emlmyRsX1tfBNQj0yi/0Uwf8P9/7EszVtpvQRBTBtp/WDy/knoqwMK6OqUQw9V47A460X//MXsjyKiorh54GRsaHYckNDfUSE/7fO9ez99v+QNoTW3e0dqlStCPtpY6QSwKSfz4ZZzmd9A11ERvx5PabMGYOLp6/xfjWMzzs/nnVdun4edmzcT8dUUQlgWJMPC2Lu3r3LMymZPX36lI8bnzcvYzheQEBAtm2wTr5ssre3x6BBg/ivWrIAhvUmZ7+z8Cusvwvr/8LSjKyJ6k9vq5zb5iMmJTkF3q7eaNTKHA9vZLSZsvlHN5/+MvMyb/0sLBy/FM/uirfpphsydiCGTbbG5MEz4eXqDWlhdfBy9UHjVg3x4Ppj0XI2//Dmk19mXhZsmIN54xbj6d20L5jChNXrnas3mrRuiPvXH4mWs/kHmfaVpMyLw8a5mDPWAU/u5G+9BMmpiHTzh1nLmvh4IyNwYvMBt7IHiElx8TjbfrbYsuo2HVCyeXXcGb0FcYEREAoE2cqYz+jHg5/nDkfwPSTvR4ilJqcixM0fFVvWhNfNjHpUbFkLXhLqkSMFQFk1Y4i0inoxCAXiQYqANZuxq3Cepcqb1y9P2EAJl7ceaNO2Oe90m65Nu+a4dvVOnj4XyzxLa/hxcnIKPFy80Kx1Y9y+lnGR2rx1Y9y98ef9uljTtyDLMcWaw/ghxTPpheOgEqBoyfcAht1tj400Yn1aWMDRvHlzREREwMPDg49KYn1cWEalYcOGPMty/vx5sWYe1v+lX79+KF++PIKCgnhn3r59+/L1LCBhzUMsOKpTpw4fycSmzFizE7tTYK9evXhHYnb74pCQEN6hly1jdxKUhhO7z8Bhy1z+Zenu6AFLq54wNjPG+cOX+Pqxc0bx4c9LJq8UBS8Om+di48KtcHfyhN7P7E1iQiK+x30XNRvZzRgOh/HL8PlTqKhM/Pd4xP+Iz/M6HNt1Cku2zsc7Fy+4Onmgj5UFTMyMcO5wWlp+/NzRMDIxgMOk5aLgZfGW+Vi3YDPcnTxEGaiETHVgnWjZvWDSOxIamhiiSo1K+PE9HkEfg5Efju46hWVbF/ARUa6O7uhjZcmHUJ89nHbsTZw7BkamBlgwcZkoeFmydQHWLtgEt0z1YvuGpcvzg9vu62izeSwiXD8g3MkX1Ya05cOf3x1J64fTcPYAFDfRxYMpu3jqPMY7SOzx8VFfkZqYLLY8a5mkrz8kLs9Lz/deR5+NYxHi6o9Pb96jwaB2fAi147G0erChz1omujg/dWdavWw64ktwJCL9Qvg8u2dMs1Hd+dDqdD533qDpyG4I9fiIIGc/6JU1Rrtp/eB9+022wCa//fgRj8CgtNfOBIeEwcvHD9paJWBqIvl2BPll+7b92LFnLZzfuuP1q7ewHTqQD6E+sO8EX79g0TSYmhpj3Oi0/ohMzVppo3CKF9eAvoEen09OSoa3d9ootynTRsP5jTv8/QOhqqqCjp1aY+CgXphu7yC1ehzYeQxr/rcE7i7v4PzaFQNs+sC0lAlOHDzH10+bPx7GJkaYOSHjNfxVM+1HBVlnfD19XT6flJQMPx9/vvz+zccYNnYw3rl5w+WNO8qUL82zMvduPuJZ/cJCUMSi8wK5Ey8bfcQyIAsXLuTBA2vSYbccHjFiBM+qsB94Yv1k2GghVjZ9ODT7wSh2tz8bGxuEhYXxH5Hq06ePqH8KG4nEtpN+V0DWvJR1KDWLllmwwrI8w4cP58ET6/TLbroj6Yen8sqdS/ehrauFEfa2vBnmg7c/v8cLG97HGBjpi913pLeVBf9yn7HSnk/prp66gaX2q/jffW178SuZlXuXiD3X3vUH+ZTXWOdbVoeRU4fy1+vn7Y/JVjMRGiS5Dn2sLXkdZq+axqd0l09dx+IpK/jfhsYGOH7ngGidzbhBfGL3vRnddxLyAxvCzeplN3UYr4Ov1wdMHDIdn9PrZSxer742lrwT39xV0/mU7tKpa3CYnPPN4fLSh8svUUy3BOpP6c2bkqK9g3DDZi2+BadlStiy4mbSuR9QXvK48gIauppoPak3NI10EO4ThGND1/IghSlhpCN2TxjWrMXu8aJT2hCCFAFiAsNwZ/VJOB3L6Bj+aOsF3t2B3R+mhIkefkR9hffdt7i39jQKmrvXewyfOEs0v2brbv6/ZdcOWD4/4xwpCOf/ucY7wc6YlfYF/87TBwP7jULQp7SAiy0rVbqk2GMePUu7AGPq1a+F/gMtEBgQhLo10+7Lwy4g125YhJJmJkiIT8D79x8wZuR0/lzScu3CbejoamP8tJG8A7KPlx9GDZqMkKBQ0WcOC2gyu3j/uOjvWnWrw6JfVwQFhqBdAwu+jPVzYVmWKXPHwtjEENFRsbh/6xE2LN+OwkSIokVBWFhyXzKmSck2kAcpwl83ucmKVGHhuQr6U+OUC/4+JXkhWEn2P1LmOS2FPDAu3xnywEDt16PSZIFPhPT7yc0sNyhPtrPmY1rWrbArEr+FRAghhMg7AYoWCmAIIYQQOSAoYo1I9FtIhBBCCJE5lIEhhBBC5IAQRQsFMIQQQogcEKBooSYkQgghhMgcysAQQgghckBYxBqRKIAhhBBC5IAARQs1IRFCCCFE5lAGhhBCCJEDAmpCIoQQQoisEaJooQwMIYQQIgcERSyEoT4whBBCCJE5lIEhhBBC5IAARQsFMIQQQogcEFITEiGEEEJI4UYZGEIIIUQOCFC0UADzh/y/h0IelC9uAnngHPWhoF/Cf7bbQAnyQF2oAlm3pXxnyIMw/5uQByGd7Ar6JcgEITUhEUIIIYQUbpSBIYQQQuSAAEULBTCEEEKIHBAI6UZ2hBBCCCGFGmVgCCGEEDkgRNFCAQwhhBAiBwRFLIShAIYQQgiRA8IiFsDQjzkSQgghROZQBoYQQgiRAwIULRTAEEIIIXJAQE1IhBBCCCGFG2VgCCGEEDkgLGIZGApgCCGEEDkgQNFCo5AIIYQQInMoA0MIIYTIAWER+y0kCmAIIYQQOSAoYn1gFGU92rSzs4Oenh4UFBTg7OwscVmbNm0wZcqUgn65hBBCCMkjMh3A3LhxAwcPHsSVK1fw+fNn1KxZU+KywmLoyEF47XoHAWEuuPXwHBo3bZBjWSNjQ+zYuw5PHa/jc4wnlq6c88tt9+rbDWFfvHDw2DZIU19bS/zz4gQefriFgzd2oU6jWjmWbdO1JbacXIfrbhdw1/sq9lz6Hxq3bihWxnJwd+w8vwW3PC/zaeup9ahet1qevuYxo23x3vs5vn31w8sX19GieaNflm/Vsgkvx8r7eD2D3SjrbGUmTRwJD/dHiPviC3+/11i/dhGKFSsmWu/r8wIpScHZpi2bl+dZvfrZ9sKFF6fw5MNtHL6xB3Ub1c6xbNuurbDt5HrccruE+97Xse/SdjTJsi8qVCmH1XuW4uLLU3gd8giDRvZHfuhla4FTz4/itt917Lm+A7V/cUy16toC60+swSXXc7judQnbL21Fw9bmOZZvZ9EWj4LvYvm+JZC24SMH463bPYREuOPeo/No0izn12VsbIjd+zbg5ZubiPzijRWr5mUr08OiE+4+/Af+n5zwKdQFD59ewoC/LVEYODq7YfxMB7S1GIKazbvi7qNnKEw0+1vA7PIRlHl+DSbHtqNYvX/5HlBRgc74YTC7egxlXlxDyYuHUdyyS8bqCmVhsNYBZleOouybOygxuA8KaydeQR5MskKmAxg/Pz+YmpqiWbNmMDExgbKyssRlhYFln648CNm0bic6tOyNl88cceLsbpiVMpVYvlgxVURFRvPyHu5ev9x2qdIl4bB0Jp4/fQ1p6mDRFlMWT8DBLUdh22kknF+6YeOxNTA2M5JYvm6TOnj1yBFTrWZhaBc7OD17i3WHVqBKzUqiMvWb1cXtC3cxvr89RlmMR2hwGDafWAdDE4M8ec39+1tgw/pFWLlqC8wbdcaTJ69w5fJRlC5dUmL5cuVK4/KlI7wcK79q9VZs2rgEvXt3E5UZNKg3Viyfg6XLNqBm7TawGz0N/fv3xIplGUFmk2bdYFa6rmjq3OVvvvzcuSt5Uq+OFu0wdfFEHNhyGFZ8X7hi8y/2Rb0mdfDykSOmWM2ETZdRfF9sOLQKVWpWFpVRU1dDcGAItq3YhciwKOSHdhZtMHHROBzechwjO4+G6ys3rDm6EkYlJdejTpPacHzkhJnWczGq61i8feaMVQeXoXKNjGMqHXsvxi0cDZcXrlKvR+8+3bBi9TxsWLcDbVpY4sUzR5w+tzfH81v15/m9Ye0OuLtJPr9jomP59jp3GICWTXvi+NFz2LZjFdq1b4GCFh+fgKqVKmDu1HEobDQ6tYHe9LH4su84QgaPQeJbNxhtXQklE8nHFGO4egHUGtVH1OL1CO49DJFzlyPFP1C0XkFNDSnBnxGzZS9SIvLn3PjTYdTCPPgnK/I1gLl8+TJ0dHQgEKTFeKx5hzXzzJgxQ1Rm9OjRGDRoEKKiovj/pUqVgoaGBmrVqoUTJ06Iyg0dOhQTJ05EYGAg30a5cuUkLpMkKSkJM2fOhJmZGYoXL47GjRvjwYMHUq37mPFDcfzIORw7fBbvfT5gwZyVCA4OxdARgySW/xQYjPmzV+DMyYv4+uVbjttVVFTE9j1rsXblVgR8DJJiDYBBdv1x+cQ1XDp+FR99A7HJYRvCQ8LRx0byVSFbf3T7Sbxz8cYn/2DsXLUXn/yD0KJjM1EZhwnLce7QRbz38EWAbyBWTl8HRUUFmLeonyev2X7yKOw/cBL7D5yAl5cvpk13wKegEIwZbSOx/Gg7awR+CublWHn2uAMHT2Ga/RhRmSaNG+DZM0ecPHkBAQFBuH3nEU6duogGDTIyIJGR0QgLixBN3bp1gK+vPx4+ep4n9RpsNwAXT1zFRb4vArDBYSvCQiLQz6aXxPJs/ZHtJ+Dp4sX3wfZVe/j/rTLtC7Zuy9IduH3xHj9H8sOAUf1w9eR1XD1xje//rQ7bERESjl42PSWWZ+tP7DgFLxdvBPkHY8+qffz/Zh2bZjsvFmybiwPrDiEk8LPU6zFuwnAcPXwWRw6dgY+3H+bOXo6Q4FCelcnp/J4zaxlOnbiAr1/jJJZ5+uQVrl6+zbf30T8Qu3Ycgoe7N5o0zTmzk19aNm2ISXa26NimOQobrSF98e3CDXy7cJ0HITHrdiA1LBwl+kk+ptSaNYRag9oInzgXCa/eIPVzGJI8vJHo6ikqk+TpjdhNu/Hj1gMgORmFuQ+MIA8mWZGvAUyrVq0QFxeHt2/f8vmHDx/CwMCA/5+OBRKtW7dGQkICGjRowJuC3N3deb8Wa2trvHz5kpfbvHkzlixZwgMc1lT0+vVricskGTZsGJ4+fYqTJ0/C1dUV/fv3R5cuXfD+/Xup1FtFRQW169bAg3tPxZY/vPcU5o3q/adtT5s1nl/JseBImpRVlFG1dlW8fCj+nrL5WuY1crUNFlRqaGrga6zkD2xGTb0YlJSVf1nmd973+vVr4/adjOOLuX37IZo2kfwlwIITtj6zW7cf8OAkPZv39Nkr1K9fCw3N6/L58uXLoEvXdrh2/W6Or2PI4D44eOgU8mpfVKtdReK+qG1e87f2xZc8eJ//Sz2q1K6C1w8dxZa/fuiEmr91TKkjLvar2HJbe2vERn3hwZG0sf1bp14N3L/3RGz5/btP0Khx3gTiTKvWTVGpcnk8k3KmVaYpK0P1ryqIfyF+TMU/d0KxOtUlPkSjVVMkevpAy3YgzG6cRMnzB6EzxQ4KxVTz6UWTP5Wv7Sva2tqoW7cuD1JYcML+t7e3x+LFi3lg8/37d/j4+PBOtyw7Mn36dNFjWWaF9W85c+YMz5iwbZUoUQJKSkq8qSidpGWZsSYmlskJCgpCyZJpzQjsedi2Dxw4gBUrVmR7TGJiIp8yEwoFUFDIXfynp6/Lv/wiwsVTjxERUTAy/vOmkoaN62GwdV+0byH5qjsv6ehpQ1lZCdGRMWLLoyNioG+kl6ttDB4zAOrqarh76X6OZcbNs0NEaCReP3b6z6/ZwECPv+/hYZFiy8PDI2GcQzqZLWfrxcqHRfIvKba90NBwnD59CYYG+nj44Dz/AmXrduw8hDVr/ydxm5aWXaCjo4VDh08j7/aFcrZ9ERURnet9MWTMQN5kdOfSPRQU7Z/HVEzWYyoyBnq5rMfA0f2hpqGOe5czgk4W/HQf1BUjOtohP+iLzu8sx01E5H86v5kSWprw8H7Cm5RTUwWYMXURHtwXvxAiGZR0tKGgrARBlPgxlRodAyV9yceUcilTqNWtCWFSEiKmOUBRRxv6cyZBSVsLUYvXydTbKyxiw6jzvQ8MC05Y4MLe6MePH8PS0pJ3tH3y5Anu378PY2NjVKtWDampqVi+fDlq164NfX19aGpq4tatW7x56L948+YNf+4qVarwbaZPLAvEghtJVq5cyQOmzNP3xOjff/IsB5eCwp8fcMU1i/Omo2mTFiA6Ohb5JevrZV/gualCx17tMHLaUMwfuwQxUZJfr9W4v9HRsj3mjFyApMQkKb9m4W+UF1/eulVTzJk9CRMmzkXDxl3Qt/8IdO/WAfPmSh7pNnzo37hx8z4+fw7775X5D/VK16lXe9hNG4Z5YxfluC/yU9aXnNvzor1lWwybZoNFY5ci9mc91IurY8HWOVg7YwO+xIhnZaQt2/5A7vbHr3yL+47WzS3QvnVfLF+yActWzEHzFr/uhE4k3FI/7aCS/Nb8PG8i563kTUcJT18hesNOFO/ZSeayMIIi1olXuSACmH379sHFxYW3U1evXp03GbEAIiYmhv/NrF+/Hhs3bsSmTZt4/xfWV4UNhf6vbfOs/w3L0Dg5OfH/M2OBjCRz5szB1KlTxZZVKpX7dujoqBikpKTAMMvVmIGBfrasTG6VK18aZcqWwpFTO0TL2PvJBEe5o5l5VwT4f0JeiY3+gpSUVOgbil/F6BroIDoi+l87/85bPxNz7RblmFkZPGYgbCdaYeLAafB99yFPXjPrh8Led2MTQ7Hlhob6CA+LkPiYsNBwPkJErLyRAZKTkxH186pu8aIZOHbsHO8fw7i7e6F4cQ3s3L4GK1ZuFvvSKlPGDO3bt0S/ASORt/siJdu+0DPQ5Rmxf+v8u2D9LMy2W4hXeZDl+i++/Dym9Ax1xZbr6usi5l/qwTr/zlo/HQtHL4HT4zei5WblSsK0jClWHlwmWsb6VDH3Am7BqpUtQgLytk8MOy7Y/mAjB7MeZ396fqdjx5L/h7SLNne3d6hStSLsp43h/WNIdqmxXyBMSc2WbVHS1eFZGElSI6ORGhEJ4bfvomXJ/oFQUFSEkpEhUj4F01tdSOV7Bia9HwwLTFiwwq4a2f8sK5Pe/4VJz85YWVmhTp06qFChQp70UalXrx7P7oSHh6NSpUpiU07NTmx4rJaWltiU2+Yjhn35uTp7oHXbjA6TTKu2zeD4Kq0/0O/y9fmA1k16on2L3qLp5rV7ePr4Jf87JCgUeSklOQXert5o1Eo8cGPzbo4ev8y8zN84GwvHL8Ozuy8klhkydiCGT7HGlCEz4eXqnWevmb3vb964okP7VmLLO3RohedZ2sjTvXjpxNeL1aFDazg5ufIvKUZdQx0Cofh1Cjum2EUeO54zG2o7kDdJXbsmuX/Mn+4LL1cfNJawL1wd3X+ZeVm4cQ7mj1+Cpznsi/zE6uHj6gPzVuK3E2Dz7r84pljmZc6GmVgyfgVe3E3rE5cu0DcQtu1GYEQnO9H09NZzPlqJ/R0eIjlw/a/HmctbD7RpK96htU275nj1MiO4ygvs+GIjmEgOUlKQ9M4H6o3Fjym1Jg2Q6JLRKTezRBcPKBnoQ0FdTbRMpUwpCFNTkRqe98eLNAmL2CikfM/ApPeDOXr0KO90mx7UsI607IOAZWgYFlCcO3cOz549g66uLjZs2IDQ0FD89ddf/+n5WdPRkCFDYGNjw7M8LKCJjIzEvXv3eKanW7eM4bJ5aef/DmLbrtVweesOx1fOsB46AKVKmeLQ/pN8/TyHqTAxNcLEMbNFj6lRK+1+KMU1NaBvoMfnk5OS+aiExMQkeL0TD+i+fEnrkJl1eV45sfsMHLbMxTtXb/4FY2nVE8Zmxjh/+BJfP3bOKD78ecnklaLgxWHzXGxcuBXuTp7Q+5kxSExIxPe476JmI7sZw+Ewfhk+fwoVlYn/Ho/4H/H/+TVv3LwHhw5shpOTCw9ORo2wQpnSZti1+whfv3zZbJQsaYphwyfzebZ83NhhWLfGAXv3H+OdeocP+xtDrMeLtnn16m1MmWyHt87uePXqLSpVLIfFDjNw+cpt0Qi79C8bW5uBOHL0DA9w8tLx3aexeMs8eLp68wCyt1VPmJgZ4dzhi3z9+Dl2fF8smrxCFLws3jwP6xdu4fsiPXuTkGlfsE617F4wDOvXY2hqgCo1KuHH93gEfZTOVejpPWcxb/NseLv4wMPJEz2tusPIzAgXj1zm6+1mj4CBqQFWTF4tCl5Y+S0O/4PnG09R9iYxIYnXIykxGf7eH8We49vXtFF8WZfnpe3b9mPHnrVwfuuO16/ewnboQD6E+sC+tCzdgkXTYGpqjHGjZ4oeU7NW2mcZy96x85vNs/Pb29uXL58ybTSc37jD3z8Qqqoq6NipNQYO6oXp9g4oaD9+xCMwKEQ0HxwSBi8fP2hrlYDpL4Yr54evx87BYOksJL7z4SOJSvTpDmUTI8SdSzumdCaMgJKRAaIWph1T36/fhfbIIdBfNANfdh6Coq42dKfY4dvFmxCmN2UrK/+/vfsAj6rY4gD+jyGQgCSkkIQuYEOK9Cog0uTxKIKC9F6UIgGkd2nSpPjoCCId6b1KU4r0XqQIBAyEDtJC7vv+EzbsJhtIICF3ds/v+5Zs7i7ZezM3u+fOnDmjasEobknU/3d7NyuM+/cRduHZ7yGxhWsUfMSHRCmSUqpUKZWLYglWGKBwKOnSpUuRAUrPnj1x9uxZlC9fXk2j5iykqlWr4tatW6/8+kzW7d+/Pzp06IDg4GCVY1OkSJEEC15oycJV8PZJhfadWqkhDQYZtb9ogYtPT352P6dLb1ubZOO2xZH3c+fJgeo1KuH838EokKs0EsP6pb/By9sTTYIaqGTRMyfOqhovrN1Cfv6+CEwXEPn8z+pWVh+K3w4KUjeLFXNX47ugwep+9QZV1RXloMm2hcYmD5+mbq9q/vyl8PXxRo/uQUiTxh+Hj5xApcr1cP58xAdyYGAAMlrVhDl37oJ6fNiwPvjqqwa4dCkE7YJ6YdGilZHPGTAwYpioXx9OxQ/E1avXsXzFOvR8+oZoUaZ0cWTKlF5Nw45v65ZuVG3RNKiB+r2fPnEW7Z7TFtWetkXnQe3VzWL53FXoGxQRcHKIc+a6nyIfq/dVLXVjzZiWn0cEePFt49JN8PT2VLOGeE4xyOhcrytCgq+ox30DfBFgVROmct3/quNoP/AbdbNYNW8NBgUNQWJZtHCl+vv+tjP/vv1x7OhJ1Py8WeTfN7exXpO1LX9EBP6UJ29OfFGzMs7/fRG5c5RS2/i+N3REH6RNF4gH9x/g1KkzaNm0o3qtxHb4+Ck0btM58vshYyaqr1UqlMGAHh0Scc+gpjpf9/JEqmZ14erng0enz+FK2254cjninOI2BjQWxv0HCPm6M3w6tUbgjLEIv3Ub/67bjJtjp0Y+xzW1L9LOmRD5vVf9Gur2YPcBhDRP3ON1Zi6Gs6Utx5MAr/itFptYMqewP2ymm92hCdPr9Drl8csKR+Dh4gbdHb79apMFzCLk7Bo4gkvlXs+MsoTECr4JrXT6cvHyczZcXAsdmKNMrRBCCCFeSbiTDSFpvZSAEEIIIZyT9MAIIYQQDsBwsh4YCWCEEEIIBxDuZCmtEsAIIYQQDsCAc5EcGCGEEEJoR3pghBBCCAcQ7mR9MBLACCGEEA4g3MkCGBlCEkIIIYR2pAdGCCGEcACGzEISQgghhG7CZQhJCCGEEMLcZAhJCCGEcACGk/XASAAjhBBCOADDyXJgZBaSEEIIIV7J2LFjkTlzZri7uyNfvnzYunVrjM9duHAhypYti9SpU8PT0xNFihTBmjVr4vyaEsAIIYQQDpLEGx4Pt7iaO3cu2rVrh+7du2Pfvn0oXrw4KlSogPPnz9t9/pYtW1QAs3LlSuzZswelSpVCpUqV1P+NCxfD2fqc4kmA1/twBJlTBMIR7A49Bd3l8csKR+Dh4gbdHb5t/41XNyFn435Va0aXyjWH7jLtXZ/gr5EnsFi8/Jx9//wep+cXKlQIefPmxbhx4yK3ZcuWDVWrVsWgQYNi9TOyZ8+OmjVrolevXrF+XcmBeUneyVLCEbi6OEYnXIaUftDdnbD7cARJ3FyhOz93LzgCR/jgp7RrJyb2LjjVNOqHDx+qm7VkyZKpW1SPHj1SvShdunSx2V6uXDn88ccfsXq98PBw3LlzBz4+PnHaT8f49BJCCCFEvGCviZeXl80tpp6U0NBQPHnyBAEBATbb+f0///wTq9cbPnw47t27hxo1asRpP6UHRgghhHAARjz1wHTt2hXt27e32Wav98Wai4uL7b4YRrRt9syePRt9+vTBkiVL4O/vH6f9lABGCCGEcADh8ZTSGtNwkT1+fn5wdXWN1tty5cqVaL0y9pJ/mzRpgvnz56NMmTJx3k8ZQhJCCCHES0maNKmaNr1u3Tqb7fy+aNGiz+15adiwIWbNmoWKFSu+1GtLD4wQQgjhAIxEqsTL4aZ69eohf/78qqbLxIkT1RTqli1bRg5JBQcHY/r06ZHBS/369TFq1CgULlw4svfGw8ND5dvElgQwQgghhAMIT6SqKJz+fO3aNfTr1w+XL19Gjhw5VI2XTJkyqce5zbomzIQJExAWFoZWrVqpm0WDBg0wbdq0WL+u1IF5Se/7F4Aj8HVzjOnglx5cg+6SuSaFI/B2exO6u/boDhzBugyp4AgcYRq1m1+WBH+NbP4F4+XnHLuyCzqQHhghhBDCARiymKMQQgghdBPuZIX1ZRaSEEIIIbQjQ0hCCCGEAzBkCEkIIYQQugl3siEk6YERQgghHIDhZD0wkgMjhBBCCO1ID4wQQgjhAAwjHM5EAhghhBDCAYTLEJLeNm3apJbwvnnzZmLvihBCCCESiOTAvEa1Gn2O9X8uxoHz27Bg3XTkK5Q7xuem9vfFsHHfYdUfv+LoPzvR9bv2dp9Xv3kt9Zz9f2/Fb/uWo0u/ICRNlnAl6T9rUBnzt8/ExtOrMWXVeHxYMGeMzy1ZoThGzh6C5QcXYu3xZZiwdAwKlswf7TlTVo7D6qNLsf7UCkxbOxHlq5dFQqvbuAa27F2J48G7sHTDbBQonCfG56YO8MPICYOwYecSnL66Dz0HfGv3eSk9U6LfkK7YeWS9+rnrti/Cx2U+SrBjcITziao1qIIF22dh0+k1mLpqwgvPqVGzh2LlwUVYf3w5Ji79EYVKFoj2nJ9Wjsfao8uw8dRK/Lx2Ej59DedU7UafY8PuJTh04XcsXP8L8hd+TnsE+GL4+P5YvX0BjofsQrf+9tujQYta6jkHz2/D5v3LVbslZHu8+UVlpFv2CzJuX4nAmWORLE+O5/8HNzekatUI6VbMRMYdK5F2yXSkqPLps4ezZILf0N5It3wGMu1dj5S1q8Esdu8/hFadeqNU5TrIUawCNmz5A7ozDCNebrqQAOY1qVClrHrzGT9yKj4rXRe7d+zHxDmjkCZdgN3n803q+rWbGD/yJxw/csruc/5b/VN06NEK/xs2CRU/qoEeQd/hP1XLon33Z4tjxafSlT/GN31aYfromWhUvjkO7jqEYTMGIyCtv93n5y6cC7u27EHHel3RuEJL7P1jP4ZMG4B3sr8d+ZzbN2/j59Ez0aJyazQo0wwr5q5GtxGdogU68ali1fLoOaAT/jdiEiqWqok/d+zF1LljkTZdYIzLxV+/dkM9/9jhk3af4+aWBL8sHI90GdLi60YdUbpQFXRt1w8hl68kyDE4wvlEpSuXQrs+rTBt9Aw0KN8MB3YdxIgZ38d4TuV5ek51qNcFDSu0UOfU0GkD8G60c2oGmlVuhXplmqpzqvuIztECnfjE31O3/h3U77fqJ3Wwe8c+TJozOub2eHpOjf8h5vaoVP1TdOzRGj8OnYgKxb5At3YR7dGhR+sEOYbk5T6GT8evcGvKLFyq3RIP9x2C/5hBcA203xaU+vuecC+YF9f6DkfwZ40Q2m0Aws4+W7TPxd0dYcGXcWP0ZIRdNdd6ZffvP8B7b2dBt/Zfw5GGkMLj4eaQAczHH3+M1q1bq1uqVKng6+uLHj16REZsM2bMUMtpp0yZEoGBgahduzauXLkSbXhnw4YN6nnJkydH0aJFceLECZvXWbp0qXrc3d0dfn5+qFbtWdT+oteI6v79+6hYsaJasvv69etq29SpU5EtWzb1899//32MHTsWCa1hy9pYMGsJfp25BGdOncOgniPwT3AIajX83O7zgy9cxsAew7Fk3krcvX3X7nPy5M+JvbsOYvnCNer5v2/aiRWL1iJH7mwJcgw1m32B5XNWYdnslfj7r/MY1ft/uHLpCj6rX9nu8/n4rHFzcfzACVw8G4wJg6eorx+VLRL5nH3bD2DL6m3q5wX/fQnzpyzE6WNnnnsV/qqafl0P82YuwtwZi3D65Fl8130oLl/6B3Ua17D7/OALl9Cv2xAsnLscd27bX+TvizqfIVUqL7SoF4Q9u/Yj+OJl7N65D8eO2A94XpUjnE9Uq9kXWDZnZeQ5NfLpOVUthnOKj88cNwfHnp5T4wdPxgV1ThW1Oac2W51T86YswOljp/FhwRf0JryCRi3rqLaYP2MJTp86h4E9ItqDvTL28Pc7oPtwLJ63Andiao8CubB31wHb9li4BjkTqD0861TH3cWrcXfxKhWE3Bg2Dk9CriDl55XsPt+9aAG458uFK2264cGuvXhyOQSPjpzAw4NHI5/z6OgJ3Bw5Ef+u3QQ8fgwzKV6kANo2b4CyHxdL7F0Rr6sH5ueff0aSJEmwc+dOjB49Gj/88AMmT56sHnv06BG+++47HDhwAIsXL8bZs2fRsGHDaD+je/fuGD58OHbv3q1+VuPGjSMfW7FihQpYGHTs27cvMtixiO1r0K1bt1CuXDn1f/hzfHx8MGnSJPX6AwYMwLFjxzBw4ED07NlTHVdC4dV59g/fV29A1vg936Re1p6d+9XPzZnnA/V9+kzpUKJ0UWxe9zviWxK3JHgv17vYtXm3zXZ+nyN/9lj9DAavHm964PbNmFf6zfdRHmTMmh77dxxEQrVFjg+zYetv22228/t8BT586Z9b5tOS2Lf7oBpC+vPYRqzetgBfBzXBG2/EfyenI5xPzzundm7ejZz5c8T6nEquzqnbMT4n/0d5kTFrBuxLwHMqoj122GzftmnHq7XHDrZHNuTKE/H3lSFTOpQsUwyb1m1DvEuSBEmzvYv7O2zb4v72PUj2YcT5EFXyEkXw8OhJeDaoiXSr5yDtomlI1a45XBJ4yFHEzHCyIaQ4z0LKkCGDClr4xvHee+/h0KFD6vtmzZrZBCJZsmRRAU7BggVx9+5dvPnmm5GPMXgoWbKkut+lSxcVrDx48ED1iPCxL7/8En379o18/ocfPvtgie1rhISEoGbNmsiaNStmz56tumyJwQ+DJ0uvTubMmXH06FFMmDABDRo0QELw9kmlArVrVyN6gCyuXb0GP3/fl/65Kxevg4+vN2Yum6zag2+ks6b+iklj4j8YS+XjhSRJXHE99IbN9huhN+Dr7xOrn1GrRQ14JHfHhmWbbLanSJkCi/fMQ9KkbnjyJBzDu43En1v3ICF4+3qrtgi9YtudHXr1msp1eVkZ30qP9BnSYvGvK9Hoy1Z4K0smFcy4uibBmGETEJ8c4Xx60Tnl4+8dq59R+znn1NI98yPPqWEJeU49bY/QaO1xHX7+L39OrVi8Fj5+3pi1/Fl7zPxpPiaOjv/2cE3lBZckrgi/ZtsWT67fgKuv/b/vJOnTwD13DhiPHuFqh954I5UXfLu2hauXJ671HRbv+yheLFyj4CNRAhgOxfCPyaJIkSIqIHjy5AkOHjyIPn36YP/+/Wq4Jjw8Yk76+fPn8cEHz6L4XLmeXZWkSZNGfeUwUMaMGdX/ZTAUE/bKxOY1ypQpgwIFCmDevHlwdXVV265evYoLFy6gSZMmNq8RFhYGLy+vGF/z4cOH6mYt3AjHGy5xu7qOFtm6uLxStFuwaF60CGqMfp2/x8G9h5ExcwY1Dn81JBTjRkxBQnjZYyhT5RM07lAfXRr3xM1rtjPE/r37LxqWa4bkKTyQ76O8aNP7a1w6f1kNBSSUqPvs8optwXMhNPQ6ugX1U+fk4QPHEBCYGs1bN4j3AMaRzif7x6FKir5Q2SqfoEmHBujcuAdu2DmnGpRrCo8UHqoHpm3vrxF8/tJrPafYHnil9siHlkGN0LfzYBzYcxiZMmdA9wEdVXuMTai/b8ThGJ6eb6HdB8G4e09tuj5iPFIP6YXrg0fDePgoQfZRiHivA8MeFA7X8MY8ldSpU6ugonz58moIx5qbm1vkfUswZAlEPDw8YnyNe/fuxfo12KuzYMEC1buSM2dOm9fgMFKhQoVsnm8JcuwZNGiQTY8Q+SZPA78U6RAbN67fVEFS1KtjXz+faFfRcdG2S0ssnb9Sjb3TyWOn4ZHcA/2GdVPJgfHZFXjz+i2EhT2Bb2rbqzFv31S4ftX2qs1e8m/X4R3Ro0Vf7N66N9rj3M/gc5fU/VNHTuOttzOiXuvaCfJhc+PaDdUWUXtb2BZRe2Xi4krIVTwOC4s8x+ivk2fgH5haXTk/fhyG+OII59PzzynvWJxTpdBt+Lfo3qIv/ozhnLpoc05lQv3WdRLmnHraHpzpZc3Xz1v17L2sdl1bqpwl5tVYt8d3w7tjXDy3x5Obt2CEPYnW2+LqnUr1wtj9P6HX8eRqaGTwQo/PnofLG2/A1T81wi4Ex9v+idgxNErAjQ9xHqDfsWNHtO/feecdHD9+HKGhoRg8eDCKFy+ukmOfl1wbE/bOMF/Fnri8Bp/DIaHSpUurIIYCAgKQLl06nDlzBm+//bbNjUNJMenatavKp7G++SSP6DmKDX54HTlwHEVL2gZNRUsWxL4/X35c3sPD3eYDk8KfPFEXTda9ZPEh7HEYThw8iQIl8tls5/eHdx95bs8LZ4D0aTUA2zfY5mzEiN3lSZ8FufGJbcHekY8+Lmyznd/v+fPlP9x279qPtzJnsPm9Z86aCSH/XInX4MVRzifbc8p2xlnBEvlwaPfh5/a89BzRGb1b9ccfG2zfj2LC/edwUkKIqT2KlSz0Su3hrtrD9gOJw2EJ0h5hYXh07CQ8Ctn+fbsXzoeHB54l5Vp7eOAIXP184eLhHrnNLWN6GE+e4MmVq/G7fyJWDMmBeT4OwbRv3x4tWrTA3r17MWbMGDWExOEf5pnw+5YtW+Lw4cMq3ySuevfurYIO5q4wF4ZXNqtWrUKnTp3i/BrDhg1TQ1uffPKJmgHFgIfDT23btoWnpycqVKighoaYTHzjxg11XPYkS5ZM3azFdfho2vhZ+P5/fXF4/1Hs330INep9hjTpAzHn5wXqcU5V9U+TGl1a94n8P+/neFd95dAKx8L5/eNHj9XMGfpt7VY1G+XYoRM4sPcIMmVOr66iN67ZGu2DKD7MnTQfPUd1VbOKDu85iip1/4uAdAFY9Msy9XjLLk3hl8YP/b8ZHBm89BzVBSN7/4gje4/CJ3VEXsPDB49w707EVVu91rVw/MBJNVuESZ1FShdChc/LYVjXkUgok8f+ghHjBuDQvqPYu/sAatWvjrTp0mDW1Pnq8W97tkVgGn90+LpH5P/JluM99TX5m8lVngi/f/z4Mf46cUZtn/nTPDRoVgu9B3XGz5Nm460sGdEqqCmmTZqVIMfgCOcTzZ40H72fnlOH9hxB1Sjn1FddmiJ1mtTo982gyOCl16iu+KH3jzgcwzlVv3VtNUuJ5xR7v4qULqzOqSFdf0BCmTp+Job8r58Kjvf/eRA16ldT7TF7WkR7cHp6QKA/OrXuHfl/slm3hzqn3sUj6/ZYsxWNvrK0R8SQHntlNq7ZkiDtcXvmAvh91xkPj51UM4lSVquIJIH+uLMgoi1StW4CV38/XOv1vfr+3qoN8GpaB759vsWt8T/jDW8veLdrjrtL1jwbPkqSRNWCUdySqP/v9m5WGPfvI+xCRA9ZYvn33/s4f/HZPgRfCsHxk6fh5ZkSaZ4zddzMwp2sBybOQ0j169dXU5OZOMthlzZt2qB58+bqimDatGno1q2bSqzNmzevCiAqV7Y/HfJ5U7Xnz5+vAhP2ojDQKFGihHqMQ0ZxfQ0mGFsHMU2bNlXTt4cOHaqCohQpUqghpnbt2iEhrVqyTiUtturQVA1fnDp+Gi1qtcOli/9EHFuAX7Q6JIs3zoy8nyP3B6ouBMfxS+evoraNGxHRjfxN169UvgXrfPBDaOTAhJkWvmHpJnh6e6JRUH2VuHvmxDlV4yUkOEQ97hvgY1O/gwEOg5KOA9upm8XKeasxIGiIuu+e3AMdBn2jhloePniIv09fQL+2A9VrJZQVi9fA28cLbb9tjtQBqXHy2F9o/GUrNfWZ/O20xcrN8yLv58qdHVW/qIiL54NRPM9/1LbLl0JQv3pLVeRu1Zb5+OfyFUydOBPjR01NkGNwhPOJNiz9DV7enmhsdU6xxgunIJNvgK/NOVW1biV1Tn07sJ26WayYtxr9gyI+WN2Tu+PbQe2szqnz6KPOqd8S7DiYAJ3KO6I9eP6cPH4azWp9Y9MeDGisLfntWXCbM/cHqPx5BVw8fwmf5It4P2OeC9ujXTfr9tiCEQMSpj041fm6lydSNasLVz8fPDp9DlfadsOTp7WMuI0BjYVx/wFCvu4Mn06tEThjLMJv3ca/6zbj5thn57xral+knfMsB8yrfg11e7D7AEKad0BiOnz8FBq36Rz5/ZAxE9XXKhXKYECPxN03ETsuRhwGUhlc5M6dGyNHJtzVsS7e90+4olivk69bSjiCSw/MVSTrZSRzdYzpp95uz2YD6urao5in+utkXYZUcARp10YEFzpz88uS4K/h5xnRq/eqQm8nTP2q+CaLOQohhBAOINzJplHLUgJCCCGE0E6cemCYQyKEEEII8zGcrAdGhpCEEEIIBxDuZLOQZAhJCCGEENqRHhghhBDCARgyhCSEEEII3YQ7WQAjQ0hCCCGE0I4MIQkhhBAOwHCyJF4JYIQQQggHEO5kQ0gSwAghhBAOwHCyAEZyYIQQQgihHemBEUIIIRyAITkwQgghhNCNIUNIQgghhBDmJkNIQgghhAMwnKwHRgIYIYQQwgEYcC4yC0kIIYQQ+jGEKT148MDo3bu3+qozRzgORzgGkuMwD2kL83CUtnBGLvwnsYMoEd3t27fh5eWFW7duwdPTU9tfkSMchyMcA8lxmIe0hXk4Sls4IxlCEkIIIYR2JIARQgghhHYkgBFCCCGEdiSAMalkyZKhd+/e6qvOHOE4HOEYSI7DPKQtzMNR2sIZSRKvEEIIIbQjPTBCCCGE0I4EMEIIIYTQjgQwQgghhNCOBDBCCCGE0I4EMEJEERYWhr59++LChQvyuxFCCJOSAMZEpk+fjocPH0bb/ujRI/WYeD2SJEmCoUOH4smTJ/IrN4nw8HCcPHkS27Ztw5YtW2xuOrl58yYmT56Mrl274vr162rb3r17ERwcDF2cPn0aPXr0QK1atXDlyhW1bfXq1Thy5Ehi75pwMjKN2kRcXV1x+fJl+Pv722y/du2a2qbDB6q3tzdcXFyibec2d3d3vP3222jYsCEaNWoEM6tataq6cV91tnTpUrvbrdsjc+bMMLMdO3agdu3a+Pvvv7n4bLTj0OHvgg4ePIgyZcqodXfOnTuHEydOIEuWLOjZs6c6Nh0uUjZv3owKFSqgWLFiKng8duyYOoYhQ4Zg165d+PXXX6GTX375BePHj8fZs2exfft2ZMqUCSNHjlR/E1WqVEns3RMvkORFTxCvD9+c7X34X7x4Ub3p6aBXr14YMGCAepMrWLCgOqY///xTXaG1atVKvVF89dVXapimWbNmMCvuP6+SDx8+jHz58iFFihQ2j1euXBk6YBDGc8reB7/lfPvoo4+wePFiFXyaUcuWLZE/f36sWLECadKksfs3ooP27durgJgf9ilTprQ51xig6aBLly7o37+/OhbrYyhVqhRGjRoFnYwbN069X7Vr1069Z1kC4VSpUqkgRgIYDST2ctjCMHLnzm3kyZPHeOONN4ycOXOq+5Zbrly5jJQpUxpffPGFFr+qatWqGePGjYu2ffz48eoxGj16tJEjRw7DzFxcXGK8sZ10sX79eqNQoULq6+3bt9WN9wsXLmysWLHC2LZtm5E9e3ajcePGhlklT57cOHXqlKE7T09P46+//lL333zzTeP06dPq/rlz54xkyZIZOkiRIoVx5syZaMdw9uxZbY7BIlu2bMaiRYuiHcuhQ4cMX1/fRN47ERvSA2OSq2Tav38/ypcvjzfffDPysaRJk+Ktt95C9erVoYM1a9bg+++/j7a9dOnS6NChg7r/n//8R13JmT3nwhF88803mDhxIooWLWrTFhw+at68ucpb4NVm48aNYVaFChXCX3/9pYa7dMbf+e3bt6Nt51BS6tSpoQP2TnCYO+qw4759+5AuXTrohL3BefLkibadSwrcu3cvUfZJxI0EMCbAdTiIgUrNmjXVG52ufHx8sGzZMgQFBdls5zY+RnxzsO5+NrsHDx5o2yZMuPT09Iy2ndvOnDmj7r/zzjsIDQ2FWbVp00YFv//88w9y5swJNzc3m8dz5coFHXBIol+/fpg3b576nkNh58+fV8G8LhcoHOrq3Lkz5s+fr/afgf7vv/+Ojh07on79+tAJgzBeNDLvxdqqVavwwQcfJNp+iTiIVT+NELE0ceJEw9XV1ahUqZLx3XffGf379zcqV65sJEmSxJg8ebJ6zrBhw4waNWqY+ncaFhZm9OvXz0ibNq06Hkv3co8ePSKPQwfFihUzPv30U+PKlSuR23if24oXL66+X7dunfHOO+8YZhXTMJ5uw3m3bt1S7ZEqVSp1TmXIkMFwc3MzSpQoYdy9e9fQwaNHj4zatWtH/v65/7xft25d9Tejk59++slIly6dMWfOHDU0Nnv2bPV+ZbkvzE9mIZkIk8h++OEHdYXGKzNOn7ZmmXZpdrwi+/HHH1XXOBNF33//fXUVbT2MYXa8Uv7555/VVyYbM5mXsy3YNmwjzljQAduAV/7sLs+QIUPkVT+PZcmSJXj33XdVAu+dO3dQr149mBFn6DxP1Ctos9u4caOaOs3ei7x586qZSTr27HHYiMfAYRj24ulo0qRJKinZUvOJw2B9+vRBkyZNEnvXRCxIAGMizIhnjQhm+HNqZffu3dV0S37A8LG2bdsm9i46DeZbTJgwQeWLcLjrwIED6kP/+PHjKFKkCG7cuAFdMIhkbhLrqFgCyrJly+KNN6QMlHBerMnDnB7iECqDMUsJC0fIuXIGEsCYSNasWTF69GhUrFhRfWhyfNayjbUwZs2aBV16khh0sUYEr/g5nsxpx6xzowsPDw8VrPDq3jqAOXr0qJoefvfu3cTeRaeia70O/u3GllkvUHhBFVsjRoyALtgjzN6wqPlt7LXkhQvLVwhzkyReE7EkKRJnIt26dUvd/+9//6t6ZHTAKxfOMmJl0ffee09d8fPKn8MXrOPBgEwH2bNnx9atW6MNTzB50d7MBTPbsGGDurFqatTZVT/99BPMTud6HRxujA0G+mYNYDhUFBu61edh3SPOAF2+fLmqvk286Prkk09Qo0aNxN49EQsSwJhI+vTp1RTFjBkzqu7LtWvXqjFyFoLj1D4d8E2YQQp7jCyzjlhJuG7duuoxBjG6zAxjTggDMX7oL1y4UF2ZsVoq3/B0wTWdmMfDQnC6FoEbM2aMylXgh83gwYMjt/OYOPvFzNhjpDsWqGNAr1MPamwsWLBADaVyZtXcuXNVSQH2vNSpU0erniSnlthZxOKZzp07GwMGDFD358+fr2buvP3220bSpEnVYzpg0bGDBw9G275//36V3a+T1atXqxki3G8PDw81g2TNmjWGTgIDA43p06cbOnN3d1fF3qIWHDt58qR6TBd9+/Y17t27F237v//+qx4zK84yssxiy5w5sxEaGmo4ips3b6pCotWrVzf8/f2Njh07JvYuiTiQHBgT27lzp5rRw94YXUrXs9eFPRRRZxzxOCpVqqTNTCpH4evrq9ao0WXozh7mUA0aNEgNFVnnIzG/hDPF9uzZAx3outYZz6GVK1eqgoJM/A4JCdGm8F5U9goJcuieM8E4VG/dw2evfpIwFxlCMhG+SQcEBERWReUbBm/MU2B1WxaQMju+CbDC65QpU1SyqyUQ43o2ugRhjqRp06Yq+VuXHCp7vv32W7WOFgsKMqeKAdns2bPV3wtn7em+1hkDMstwqxmxyF7JkiUjhyA5dBfTcJKlOKJZMW/KXhuwbZgkzpmHlnYya0ApnpEAxkT4x2NvphHHn7/88kstAhheFTdo0EBNNbZUTH38+LG6embCpS4cZVVtfuhzKYH169erirVRq9jqMNbP3zEX/+zUqRP+/fdflbPAeh3MzeDfhS7nEm+su2N9XvFDkjPaGOCbFc+fatWqqQR95rGxLpJOlbSt/fbbb4m9CyIeyRCSifCDkVnwUdcZ4VUNu9H5YaQLvtnxWHg1w33XraYCZ4/EtKo2l0lgcian9jLB1MyranOV4Jjwg5TTSHUStV6HDjjMxfOHPasM4q1XlresdcaA36wOHjyIHDlyqOEjBpO8SNE1gBGORXpgTIRTjZkrEjWA4ba0adPCrF5UJ2LTpk1aXfHTtm3bVIXOqFfG7CXj7DDOYGCPBt/MzRzAOMoVJ3tgeB6xAix7YOjSpUsqT8F68VMzYo8k8e+6WLFikVN2dcGyAZbcnc2bN0erEK479urZq3yuyxpbzkyvvyQnyFdgrQsOubAWAbF+B7vOLSs561AngkmV7BpnHRhiHRiOmefLlw+6cJRVtR0BlxL49NNP1YfMw4cP1dRX9gAMGTJE9Uoyd0EH3Gf2SlpqPXEph6lTp6oeSpavZ2+MWfNG2OPIAIaVwR1lpfarV6+qHiUu3miP5MCYnwQwJsJAhbN0vv7668irAQ4rMfela9eu0OEqnz0sfKNmtznH/oll9/lGUbx4cehC51W1ma8wbdo01Tvx2WefPbf2C+vbmN0333yjEkeZ7MoZMRY8Ngb9umjRooUKeBnAcFiYK8+zrVgckb0AZs0Rc6QkXmu8WOR7E2tWcah10aJFaoYVe16HDx+e2LsnYkECGBPhmwOv+jljhFdqLGfPRdJ0KWJH/MPnEIsleCHe55tCuXLlTN2TZI1t8NVXX6ngjDkwbBvOfuF0UssV/7p169Qbu9kwx8IStFjWerFHl6J2HM7jMGrUHgpWSWahQV2wJzJ37tzqPoMWnjtM2uexMRnZrAGMIyXxWmP+F3vBChQooPJ7eD6xd4+BP2e4cUkXYW4SwJgQx/T5R6VrnQVexXDmlDWWseeKx7rgmzS79rmqNnspLIsgMgfAUuPGrMEYhyWsh7xYBTmm6ck64JCFve58rlWj0wcpzyHL8AtnhbHkgCX3jcnJZsYhPMvwMHvEdPq9x4Q9qJZkcPaqckiJs8TYQ8bVwoUG4lL1TogXqVevnpExY0ZVSfjChQvqxvtvvfWWUb9+ffkFvmZeXl7GsmXLom0PCgpSVXp1UKNGDaNZs2aRlXjPnDlj3Llzx/jkk0+Mhg0bGrooVaqU+htgZWQ3Nzfj1KlTavumTZuMTJkyJfbuOZ38+fOrattUpUoV9d518eJFo1OnTkaWLFkSe/dELMg0ahGvOJbP9WlYfI/JyMRZF02aNMHQoUORIkUKU/ceWapv2qvYaU2XKp2c9s3hiaVLl6JEiRJqW5s2bdQsKnahs1fJ7DjbiDkKzLs4deqUysHgVz8/P2zZskWbKdWcjsx1dpiMzJl7XG/L0h6sxqvLavMsJ8AhMHszd3TIqbKYOXOmeo9iPSdORChfvrxqBw5VMoeMOUrC3CSAEQnWPcspr+w2Zw0YMwcu9kq9c0w8poqdulXpnDNnjkoMZ24SA0uO+zO3h93lurh//76qvsuufQ7DcJFTBgPME9MdZ1Lx3ItaZNCs51L9+vVVPhtzwPiVwSTL8TOp2nr4UseLr+PHj6vFdBkcC/OTAEaIp5jfYqnTwVlUzE2IOtuCH5688rTU9tDFuHHj1IwqrmHD4EW3woLCHFgbhbOpuLSDZV0q1rfhNs5S4urnQrwuEsAI4UAL7z2vsOCvv/6qipJZL+yoQ2HB6dOnP/dx9gjogOcMKzzPmzfP7vCLDgudsif1yJEjqnoweykYDDPplbMmWbuKfzO6+Pzzz9VwZNRaThzq5oxDDpMJc5NZSELEYeE9rlvD2jxmFrWwoAUDF+b2WB7XZRo1Z71YY94Cu/uZq5A8eXJtAhj2TnDxSQaYnKbfvXt3VRhu8eLF6NWrF3TA2TqW2YRcj+rw4cMqgLl586ZqE916XC15SFFnXA0bNixR9knEjQQwQtjpveCHOz9k+AFpfQXNlbUttTzMylGWD7BgsbGomHfBOj26TAW3JI1OmjRJ1RdhMFOrVi0VVHJYhsXUWGPF7FiMkrkvDFpq1Kihgksmg3Mbp+zrhBcj9qofMxfpRUn8whwkgBHCiqV3gj0whw4dsnmD4/0PP/xQzbISiYsFHgcPHqxq3DDxUgdMdLUsI8BaT7du3VL3WQ+GwbIOWBfJsqgsq4Pzw56FBlnoTpdjsOAClXPnzo3W+8VEZdaAEuYnAYwQdnovuPTBqFGjtJku7ax5SpxirYv06dOrHBHOcmESNWeFcTYVpyXrUm2bU74//vhjVUWYs9i4/AlvOmLAxWUSOFvSeu05znaT/Bc9SBKvEMLUWMPGGnvHGAiwN4AzxWJajM9smCzKgLhbt24qoZpDSEyGZUIvZ4ixR8nsONuIuSNcFiEwMFAFMrwxqNGhplBUK1aswMCBA7F//341JZ/DecyLMeMSISI6CWCEEKbGmjzWmJ/E6eC8aubaW5y+qyPmU3EdJPbGVK5cGTrhcNimTZvUzRLQcHaeTrOQhP5kCEkIYWqW9YN0xwUCAwIC0LhxY/V9oUKF1I3FBbmIK1ed1wVrwHCRVt64YChrJ7FHRkeczs612qKeZxzqE+YmPTBCCFOLqa6NPWaua8PhIi4XYFkM1Lonhss9nD17FmbHIIs9LixgxyRYLk/B4RZ+fd7K52bEmWwMJv/44w/tq207K+mBEUKYfmYYV0HmB8p7772ntnHIgkm8TIK1MHtdGw672Bvu4nCYLkMvLPLG/WWeSJUqVZAtWzboimsgsedo+fLlql3Mfv6I6CSAEUKYWqVKldSQBZd34JCFpTYMZ4qxLkmHDh2gAyYcM+eFpfetcVvatGmhSzDJHhjmvjD/iEGkJYmXN50CGibuMjDWMflYRJAhJCGEqbHiK6ccZ8+e3WY7q8ByMUFdplIzz4U9GLxZT9vlNGQGYayrohsOJY0cORIzZsxQOSQ6DbsUKFBALe3w0UcfJfauiJckPTBCCFNjVdSQkJBoAQwTLy1l7XXAQIXrHXFlcMs6SFyWgnklOgUv7IWxzEDaunWrah9Wpy5VqhR0woCSbcJp1CwwGHU1cKkBZX7SAyOEMDWudcRhCw5ZFC5cWG1j6X0uI8DkUQ4t6VbCnosfsu4IKwrrUsSOOITH/WdFasuwEdtAxw97y/T8qLkvksSrDwlghBCmxkUCuXwDpxtzIUdi8mWTJk3UcAxXSBavBxNedQ1YomJQ/DxSzM78JIARQmjh3r17quw7r5BZ/E0CFyGcmwQwQgghnLqHj8s5WPKSLLisgDA3SeIVQgjhdK5evaqm4se0lpZOM6qcle0iI0IIIYQTaNeunaonxIRwJlSvXr1aJYQzsTrqAqLCnKQHRgghhNPZuHEjlixZourBcEZSpkyZULZsWZWgzHWrKlasmNi7KF5AemCEEEI4ZVI4V9AmHx8fNaRErAmzd+/eRN47ERsSwAghhHA6XFfrxIkT6j4L8U2YMAHBwcEYP3683TWrhPnILCQhhBBOZ+bMmaquEBd1ZHXh8uXLIzQ0FEmTJlW5MDVr1kzsXRQvIAGMEEIIp8baQvfv38fx48eRMWNG+Pn5JfYuiViQISQhhBBOacqUKciRI4dak4rLJHDZisWLFyf2bolYkllIQgghnE7Pnj3VatRt2rRBkSJF1Lbt27cjKCgI586dQ//+/RN7F8ULyBCSEEIIp8NhojFjxqBWrVo222fPnq2CGubDCHOTISQhhBBOh5V28+fPH217vnz5EBYWlij7JOJGAhghhBBOp27duhg3bly07RMnTkSdOnUSZZ9E3MgQkhBCCKfDYaLp06cjQ4YMKFy4sNrGZQUuXLigknnd3NwinztixIhE3FMREwlghBBCOJ1SpUrF6nkuLi5q2QFhPhLACCGEEEI7kgMjhBBCCO1IACOEEEII7UgAI4QQQgjtSAAjhBBCCO1IACOEEEII7UgAI4QQQgjtSAAjhBBCCO1IACOEEEII6Ob/ae2h5WfFkRwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "N = len(ans)\n",
    "sim = np.zeros((N, N))\n",
    "for i in range(N):\n",
    "    for j in range(N):\n",
    "        sim[i,j] = ans[i].similarity(ans[j])\n",
    "\n",
    "labels = words.split(\" \")\n",
    "sns.heatmap(sim, annot=True, xticklabels=labels, yticklabels=labels);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32a7f9b8-9950-4cdc-9f8e-3a0206abdc18",
   "metadata": {},
   "source": [
    "Word embeddings are a really useful primitive in NLP - and in fact, modern large language models start with the following two steps:\n",
    "1. Tokenization - break the input into (roughly word-sized) tokens\n",
    "2. Embedding - convert each token into a vector.\n",
    "\n",
    "Everything between there and the conversion back to output tokens happens on the embedding vectors."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5be5856-a63a-430b-9491-a90f891c610b",
   "metadata": {},
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "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": "3.12.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}