HMM_Даша_Андрей_ДЗ_3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# **Реализовать и протестировать для русского языка HMM теггер**"
      ],
      "metadata": {
        "id": "83NCXb4Zks6S"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install numpy\n",
        "!pip install lxml\n",
        "!pip install bs4\n",
        "!pip install nltk\n",
        "!pip install pandas\n",
        "!pip install sklearn"
      ],
      "metadata": {
        "id": "AyD6nf0ok-p8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk, re, pprint\n",
        "import numpy as np\n",
        "from bs4 import BeautifulSoup as bs\n",
        "import lxml\n",
        "import pandas as pd\n",
        "import requests\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import pprint, time\n",
        "import random\n",
        "from sklearn.model_selection import train_test_split\n",
        "from nltk.tokenize import word_tokenize"
      ],
      "metadata": {
        "id": "ZgHPb-DOF65v"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "list_tagged = []\n",
        "\n",
        "with open(\"/content/annot.opcorpora.no_ambig_strict.xml\", \"r\") as file:\n",
        "    list_tagged = file.readlines()\n",
        "\n",
        "list_tagged = \"\".join(list_tagged)\n",
        "list_tagged = bs(list_tagged, \"lxml\")\n"
      ],
      "metadata": {
        "id": "9O8ggsVxqXrn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from bs4 import BeautifulSoup\n",
        "soup = BeautifulSoup(list_tagged)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-nx6D0O2358n",
        "outputId": "cf20f854-df78-41f3-e0e4-2355443cbdf2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.9/dist-packages/bs4/builder/__init__.py:545: XMLParsedAsHTMLWarning: It looks like you're parsing an XML document using an HTML parser. If this really is an HTML document (maybe it's XHTML?), you can ignore or filter this warning. If it's XML, you should know that using an XML parser will be more reliable. To parse this document as XML, make sure you have the lxml package installed, and pass the keyword argument `features=\"xml\"` into the BeautifulSoup constructor.\n",
            "  warnings.warn(\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "paragpahs = soup.find_all('paragraph')"
      ],
      "metadata": {
        "id": "G0Ps4Rkj6KA7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sentences = [paragrapg.find_all('sentence') for paragrapg in paragpahs]"
      ],
      "metadata": {
        "id": "x4FIfROX6QNm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tokens = []\n",
        "for sent in sentences:\n",
        "  for s in sent:\n",
        "    tokens.append(s.find_all('tokens'))"
      ],
      "metadata": {
        "id": "FfanEK-S6iUH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "wordbank = []\n",
        "for sent in tokens:\n",
        "  c_tokens = sent[0].find_all('token')\n",
        "  c_sent = []\n",
        "  for token in c_tokens:\n",
        "    text = token['text']\n",
        "    pos = token.find('g')['v']\n",
        "    c_sent.append((text, pos))\n",
        "  wordbank.append(c_sent)"
      ],
      "metadata": {
        "id": "G_Y235R57agj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "wordbank[:10]"
      ],
      "metadata": {
        "id": "x8BqATib8Oqm",
        "outputId": "4d5094a0-e14e-4477-b31e-05c00dac8a1f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('«', 'PNCT'),\n",
              "  ('Школа', 'NOUN'),\n",
              "  ('злословия', 'NOUN'),\n",
              "  ('»', 'PNCT'),\n",
              "  ('учит', 'VERB'),\n",
              "  ('прикусить', 'INFN'),\n",
              "  ('язык', 'NOUN')],\n",
              " [('Сохранится', 'VERB'),\n",
              "  ('ли', 'PRCL'),\n",
              "  ('градус', 'NOUN'),\n",
              "  ('дискуссии', 'NOUN'),\n",
              "  ('в', 'PREP'),\n",
              "  ('новом', 'ADJF'),\n",
              "  ('сезоне', 'NOUN'),\n",
              "  ('?', 'PNCT')],\n",
              " [('Великолепная', 'ADJF'),\n",
              "  ('«', 'PNCT'),\n",
              "  ('Школа', 'NOUN'),\n",
              "  ('злословия', 'NOUN'),\n",
              "  ('»', 'PNCT'),\n",
              "  ('вернулась', 'VERB'),\n",
              "  ('в', 'PREP'),\n",
              "  ('эфир', 'NOUN'),\n",
              "  ('после', 'PREP'),\n",
              "  ('летних', 'ADJF'),\n",
              "  ('каникул', 'NOUN'),\n",
              "  ('в', 'PREP'),\n",
              "  ('новом', 'ADJF'),\n",
              "  ('формате', 'NOUN'),\n",
              "  ('.', 'PNCT')],\n",
              " [('Потом', 'ADVB'),\n",
              "  ('проект', 'NOUN'),\n",
              "  ('переехал', 'VERB'),\n",
              "  ('с', 'PREP'),\n",
              "  ('«', 'PNCT'),\n",
              "  ('Культуры', 'NOUN'),\n",
              "  ('»', 'PNCT'),\n",
              "  ('на', 'PREP'),\n",
              "  ('НТВ', 'NOUN'),\n",
              "  ('.', 'PNCT')],\n",
              " [('Это', 'NPRO'),\n",
              "  ('помимо', 'PREP'),\n",
              "  ('явных', 'ADJF'),\n",
              "  ('перемен', 'NOUN'),\n",
              "  ('в', 'PREP'),\n",
              "  ('виде', 'NOUN'),\n",
              "  ('тут', 'ADVB'),\n",
              "  ('же', 'PRCL'),\n",
              "  ('появившихся', 'PRTF'),\n",
              "  ('рекламных', 'ADJF'),\n",
              "  ('блоков', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('отсутствовавших', 'PRTF'),\n",
              "  ('на', 'PREP'),\n",
              "  ('«', 'PNCT'),\n",
              "  ('Культуре', 'NOUN'),\n",
              "  ('»', 'PNCT'),\n",
              "  (',', 'PNCT'),\n",
              "  ('позволило', 'VERB'),\n",
              "  (',', 'PNCT'),\n",
              "  ('с', 'PREP'),\n",
              "  ('одной', 'ADJF'),\n",
              "  ('стороны', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('расширить', 'INFN'),\n",
              "  ('круг', 'NOUN'),\n",
              "  ('гостей', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('с', 'PREP'),\n",
              "  ('другой', 'ADJF'),\n",
              "  ('–', 'PNCT'),\n",
              "  ('изменить', 'INFN'),\n",
              "  ('тон', 'NOUN'),\n",
              "  ('разговора', 'NOUN'),\n",
              "  ('.', 'PNCT')],\n",
              " [('Народные', 'ADJF'), ('любимцы', 'NOUN')],\n",
              " [('Продлится', 'VERB'),\n",
              "  ('она', 'NPRO'),\n",
              "  ('55', 'NUMB'),\n",
              "  ('дней', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('до', 'PREP'),\n",
              "  ('5', 'NUMB'),\n",
              "  ('ноября', 'NOUN'),\n",
              "  ('.', 'PNCT')],\n",
              " [('За', 'PREP'),\n",
              "  ('кота', 'NOUN'),\n",
              "  ('–', 'PNCT'),\n",
              "  ('ответишь', 'VERB'),\n",
              "  ('!', 'PNCT')],\n",
              " [('Шпаги', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('деньги', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('биржа', 'NOUN')],\n",
              " [('Звери', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('деньги', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('наркота', 'NOUN')]]"
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "list_tagged = wordbank"
      ],
      "metadata": {
        "id": "LoLoXmR99zWX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "random.seed(1234)\n",
        "train_set, test_set = train_test_split(list_tagged, test_size=0.3)\n",
        "\n",
        "print(len(train_set))\n",
        "print(len(test_set))\n",
        "print(train_set[:5])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hEqlkImwFsq4",
        "outputId": "d2151294-306e-48f7-f3ab-d5099f3d8bb6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "8078\n",
            "3462\n",
            "[[('«', 'PNCT'), ('Я', 'NPRO'), ('очень', 'ADVB'), ('по', 'PREP'), ('отцу', 'NOUN'), ('тосковал', 'VERB'), (',', 'PNCT'), ('даже', 'PRCL'), ('плакал', 'VERB'), ('потихоньку', 'ADVB'), ('.', 'PNCT')], [('Демография', 'NOUN')], [('Всем', 'ADJF'), ('миром', 'NOUN')], [('Стивен', 'NOUN'), ('Хант', 'NOUN'), ('отчасти', 'ADVB'), ('поддерживает', 'VERB'), ('эту', 'ADJF'), ('позицию', 'NOUN'), (',', 'PNCT'), ('отчасти', 'ADVB'), ('оппонирует', 'VERB'), ('своим', 'ADJF'), ('коллегам', 'NOUN'), ('по', 'PREP'), ('цеху', 'NOUN'), ('.', 'PNCT')], [('Впервые', 'ADVB'), ('за', 'PREP'), ('последние', 'ADJF'), ('годы', 'NOUN'), ('в', 'PREP'), ('стране', 'NOUN'), (',', 'PNCT'), ('которая', 'ADJF'), ('является', 'VERB'), ('образцом', 'NOUN'), ('соблюдения', 'NOUN'), ('гражданских', 'ADJF'), ('свобод', 'NOUN'), (',', 'PNCT'), ('журналисты', 'NOUN'), ('должны', 'ADJS'), ('были', 'VERB'), ('обратиться', 'INFN'), ('к', 'PREP'), ('полицейской', 'ADJF'), ('защите', 'NOUN'), ('от', 'PREP'), ('угроз', 'NOUN'), (',', 'PNCT'), ('которым', 'ADJF'), ('они', 'NPRO'), ('подверглись', 'VERB'), ('из-за', 'PREP'), ('карикатур', 'NOUN'), ('.', 'PNCT')]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_tagged_words = [tup for sent in train_set for tup in sent]\n",
        "len(train_tagged_words)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V5e8AVDcHbrL",
        "outputId": "6afcdb67-3f67-42f8-c9aa-ad4117a2b440"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "53751"
            ]
          },
          "metadata": {},
          "execution_count": 113
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tokens = [pair[0] for pair in train_tagged_words]\n",
        "tokens[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6k0PCAy6kEst",
        "outputId": "576a920a-b778-4412-c735-6c32070313df"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['«',\n",
              " 'Я',\n",
              " 'очень',\n",
              " 'по',\n",
              " 'отцу',\n",
              " 'тосковал',\n",
              " ',',\n",
              " 'даже',\n",
              " 'плакал',\n",
              " 'потихоньку']"
            ]
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "V = set(tokens)\n",
        "print(len(V))\n",
        "\n",
        "T = set([pair[1] for pair in train_tagged_words])\n",
        "print(len(T))\n",
        "\n",
        "print(T)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oiepQjUXkLLN",
        "outputId": "cccb563b-cb97-4707-ac90-5f0b7d60a46b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "18065\n",
            "26\n",
            "{'SYMB', 'GRND', 'PNCT', 'COMP', 'CONJ', 'PRTS', 'NOUN', 'VERB', 'LATN', 'ADJS', 'INTJ', 'NUMR', 'TIME', 'ADJF', 'GREK', 'ROMN', 'PRCL', 'PRED', 'PRTF', 'NPRO', 'HANI', 'INFN', 'NUMB', 'DATE', 'ADVB', 'PREP'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "P(w/t)"
      ],
      "metadata": {
        "id": "Kj7bxm42kY0Q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "t = len(T)\n",
        "v = len(V)\n",
        "w_given_t = np.zeros((t, v))"
      ],
      "metadata": {
        "id": "cPDn0jc3kde6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def word_given_tag(word, tag, train_bag = train_tagged_words):\n",
        "    tag_list = [pair for pair in train_bag if pair[1]==tag]\n",
        "    count_tag = len(tag_list)\n",
        "    w_given_tag_list = [pair[0] for pair in tag_list if pair[0]==word]\n",
        "    count_w_given_tag = len(w_given_tag_list)\n",
        "\n",
        "    return (count_w_given_tag, count_tag)"
      ],
      "metadata": {
        "id": "cS_7I55JkhSm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def t2_given_t1(t2, t1, train_bag = train_tagged_words):\n",
        "    tags = [pair[1] for pair in train_bag]\n",
        "    count_t1 = len([t for t in tags if t==t1])\n",
        "    count_t2_t1 = 0\n",
        "    for index in range(len(tags)-1):\n",
        "        if tags[index]==t1 and tags[index+1] == t2:\n",
        "            count_t2_t1 += 1\n",
        "    return (count_t2_t1, count_t1)"
      ],
      "metadata": {
        "id": "dsN7c-5umAQA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tags_matrix = np.zeros((len(T), len(T)), dtype='float32')\n",
        "for i, t1 in enumerate(list(T)):\n",
        "    for j, t2 in enumerate(list(T)):\n",
        "        tags_matrix[i, j] = t2_given_t1(t2, t1)[0]/t2_given_t1(t2, t1)[1]"
      ],
      "metadata": {
        "id": "yQHrI6uBmDdG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tags_matrix"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MYZ_PndTmLTb",
        "outputId": "48a703a1-a78b-4146-eaa7-5c644f37823c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[5.36912754e-02, 0.00000000e+00, 1.74496651e-01, 0.00000000e+00,\n",
              "        6.71140943e-03, 0.00000000e+00, 8.05369094e-02, 6.71140943e-03,\n",
              "        2.68456370e-01, 0.00000000e+00, 6.71140943e-03, 0.00000000e+00,\n",
              "        0.00000000e+00, 2.01342274e-02, 6.71140924e-02, 0.00000000e+00,\n",
              "        6.71140943e-03, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 2.81879187e-01, 0.00000000e+00,\n",
              "        6.71140943e-03, 2.01342274e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 1.42857149e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 3.19327742e-01, 8.40336177e-03,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 9.24369767e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        8.40336177e-03, 0.00000000e+00, 0.00000000e+00, 8.40336177e-03,\n",
              "        0.00000000e+00, 5.88235296e-02, 1.68067235e-02, 0.00000000e+00,\n",
              "        1.68067235e-02, 3.27731103e-01],\n",
              "       [3.57625377e-03, 7.07303500e-03, 1.65858701e-01, 2.70205829e-03,\n",
              "        4.15640138e-02, 4.76833805e-03, 2.65040129e-01, 9.52872932e-02,\n",
              "        4.37097661e-02, 3.33783682e-03, 9.93403792e-03, 6.35778415e-04,\n",
              "        1.58944604e-04, 8.36843327e-02, 6.35778415e-04, 1.74839073e-03,\n",
              "        2.23317165e-02, 2.54311366e-03, 8.18564743e-03, 4.81602177e-02,\n",
              "        2.78153061e-03, 8.90089758e-03, 5.10212183e-02, 1.58944604e-04,\n",
              "        4.02129851e-02, 8.59095603e-02],\n",
              "       [0.00000000e+00, 2.40000002e-02, 4.48000014e-01, 0.00000000e+00,\n",
              "        1.60000008e-02, 0.00000000e+00, 9.60000008e-02, 1.19999997e-01,\n",
              "        8.00000038e-03, 0.00000000e+00, 0.00000000e+00, 8.00000038e-03,\n",
              "        0.00000000e+00, 1.12000003e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        1.60000008e-02, 8.00000038e-03, 0.00000000e+00, 2.40000002e-02,\n",
              "        0.00000000e+00, 2.40000002e-02, 8.00000038e-03, 0.00000000e+00,\n",
              "        8.00000038e-03, 7.99999982e-02],\n",
              "       [9.88142332e-04, 9.88142332e-04, 1.78853750e-01, 9.88142285e-03,\n",
              "        1.48221347e-02, 3.95256933e-03, 2.27272734e-01, 1.03754938e-01,\n",
              "        1.08695654e-02, 6.91699609e-03, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 1.35375500e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        4.94071133e-02, 5.92885353e-03, 3.95256933e-03, 6.81818202e-02,\n",
              "        0.00000000e+00, 2.66798418e-02, 7.90513866e-03, 0.00000000e+00,\n",
              "        6.91699609e-02, 7.50988126e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 1.45552561e-01, 0.00000000e+00,\n",
              "        2.15633418e-02, 0.00000000e+00, 2.29110509e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 1.10512130e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        1.61725059e-02, 0.00000000e+00, 5.39083546e-03, 0.00000000e+00,\n",
              "        0.00000000e+00, 2.15633418e-02, 8.89487863e-02, 0.00000000e+00,\n",
              "        2.15633418e-02, 3.39622647e-01],\n",
              "       [9.84009821e-04, 4.92004910e-04, 3.66605163e-01, 1.10701111e-03,\n",
              "        1.83886830e-02, 4.98154992e-03, 2.16605172e-01, 1.00984007e-01,\n",
              "        1.37761375e-02, 4.92004910e-03, 7.38007366e-04, 1.10701111e-03,\n",
              "        6.15006138e-05, 6.15006164e-02, 3.07503069e-04, 2.02952023e-03,\n",
              "        1.70356706e-02, 3.01353005e-03, 2.09102081e-03, 1.09471092e-02,\n",
              "        0.00000000e+00, 6.51906524e-03, 3.56088579e-02, 0.00000000e+00,\n",
              "        1.63591634e-02, 1.13837637e-01],\n",
              "       [0.00000000e+00, 2.17013891e-04, 2.35894099e-01, 5.20833349e-03,\n",
              "        8.46354198e-03, 3.34201381e-02, 2.48263896e-01, 2.17013891e-04,\n",
              "        7.37847202e-03, 8.02951399e-03, 0.00000000e+00, 1.51909725e-03,\n",
              "        0.00000000e+00, 8.59375000e-02, 2.17013891e-04, 2.17013891e-04,\n",
              "        1.95312500e-02, 0.00000000e+00, 1.95312500e-03, 4.38368060e-02,\n",
              "        0.00000000e+00, 8.26822892e-02, 2.17013881e-02, 0.00000000e+00,\n",
              "        3.58072929e-02, 1.59505203e-01],\n",
              "       [3.38541679e-02, 0.00000000e+00, 3.81510407e-01, 0.00000000e+00,\n",
              "        9.11458302e-03, 3.90625000e-03, 3.25520821e-02, 5.98958321e-02,\n",
              "        3.84765625e-01, 6.51041686e-04, 6.51041686e-04, 0.00000000e+00,\n",
              "        0.00000000e+00, 9.76562500e-03, 1.30208337e-03, 2.60416674e-03,\n",
              "        7.81250000e-03, 0.00000000e+00, 0.00000000e+00, 5.85937500e-03,\n",
              "        0.00000000e+00, 0.00000000e+00, 3.32031250e-02, 0.00000000e+00,\n",
              "        5.20833349e-03, 2.73437500e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 4.40559447e-01, 3.49650346e-03,\n",
              "        2.79720277e-02, 3.49650346e-03, 8.74125883e-02, 5.94405606e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 7.34265745e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 6.99300691e-03,\n",
              "        0.00000000e+00, 1.36363640e-01, 1.39860138e-02, 0.00000000e+00,\n",
              "        1.04895104e-02, 1.36363640e-01],\n",
              "       [0.00000000e+00, 0.00000000e+00, 9.30069923e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 1.39860138e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 1.39860138e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 4.19580415e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00],\n",
              "       [0.00000000e+00, 0.00000000e+00, 6.57894760e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 7.36842096e-01, 2.63157897e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 1.18421055e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        1.31578948e-02, 0.00000000e+00, 2.63157897e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        1.31578948e-02, 0.00000000e+00],\n",
              "       [0.00000000e+00, 0.00000000e+00, 2.50000000e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 5.00000000e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 2.50000000e-01],\n",
              "       [2.16309752e-04, 2.16309752e-04, 8.06835368e-02, 2.16309752e-04,\n",
              "        9.30131972e-03, 2.16309750e-03, 7.50594854e-01, 2.37940736e-02,\n",
              "        1.29785854e-03, 2.16309752e-04, 0.00000000e+00, 1.08154875e-03,\n",
              "        0.00000000e+00, 8.78217593e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        6.05667336e-03, 6.48929272e-04, 2.37940741e-03, 4.97512426e-03,\n",
              "        0.00000000e+00, 4.32619505e-04, 1.94678782e-03, 0.00000000e+00,\n",
              "        2.81202677e-03, 2.31451429e-02],\n",
              "       [2.50000000e-01, 0.00000000e+00, 5.35714269e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 7.14285746e-02,\n",
              "        1.42857149e-01, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00],\n",
              "       [0.00000000e+00, 1.44927539e-02, 3.04347813e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 4.34782594e-01, 2.89855078e-02,\n",
              "        0.00000000e+00, 1.44927539e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 7.24637657e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 4.34782617e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        4.34782617e-02, 4.34782617e-02],\n",
              "       [0.00000000e+00, 9.57854372e-03, 5.93869723e-02, 1.34099619e-02,\n",
              "        6.70498097e-03, 1.53256701e-02, 6.41762465e-02, 5.24904191e-01,\n",
              "        9.57854383e-04, 2.49042138e-02, 0.00000000e+00, 9.57854383e-04,\n",
              "        0.00000000e+00, 3.25670503e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        2.96934862e-02, 1.53256701e-02, 7.66283507e-03, 3.06513403e-02,\n",
              "        0.00000000e+00, 5.65134101e-02, 3.83141753e-03, 0.00000000e+00,\n",
              "        4.78927195e-02, 5.55555560e-02],\n",
              "       [0.00000000e+00, 7.14285718e-03, 2.71428585e-01, 0.00000000e+00,\n",
              "        1.42857144e-02, 0.00000000e+00, 1.00000001e-01, 5.71428575e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 6.42857179e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        3.57142873e-02, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 3.71428579e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        4.28571440e-02, 3.57142873e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 8.78048763e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 4.04878050e-01, 4.87804879e-03,\n",
              "        1.95121951e-02, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 1.17073171e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        9.75609757e-03, 9.75609757e-03, 0.00000000e+00, 4.87804879e-03,\n",
              "        0.00000000e+00, 1.95121951e-02, 4.87804879e-03, 0.00000000e+00,\n",
              "        9.75609757e-03, 3.07317078e-01],\n",
              "       [0.00000000e+00, 0.00000000e+00, 1.67943805e-01, 5.10855671e-03,\n",
              "        9.57854372e-03, 6.38569612e-03, 7.47126415e-02, 3.58237535e-01,\n",
              "        1.91570877e-03, 3.00127715e-02, 0.00000000e+00, 6.38569589e-04,\n",
              "        0.00000000e+00, 3.32056209e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        9.64240134e-02, 1.08556831e-02, 1.27713918e-03, 4.08684537e-02,\n",
              "        0.00000000e+00, 2.80970633e-02, 1.91570877e-03, 0.00000000e+00,\n",
              "        6.32183924e-02, 6.96040839e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 1.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00],\n",
              "       [0.00000000e+00, 0.00000000e+00, 2.49718785e-01, 4.49943775e-03,\n",
              "        1.01237344e-02, 7.87401572e-03, 2.46344209e-01, 7.87401572e-03,\n",
              "        5.62429707e-03, 4.49943775e-03, 0.00000000e+00, 1.12485944e-03,\n",
              "        0.00000000e+00, 1.09111361e-01, 0.00000000e+00, 0.00000000e+00,\n",
              "        2.69966256e-02, 5.62429707e-03, 3.37457820e-03, 2.02474687e-02,\n",
              "        0.00000000e+00, 8.99887551e-03, 7.87401572e-03, 0.00000000e+00,\n",
              "        3.93700786e-02, 2.40719914e-01],\n",
              "       [9.35608149e-03, 0.00000000e+00, 4.93670881e-01, 0.00000000e+00,\n",
              "        2.75178859e-03, 5.50357741e-04, 4.10566866e-01, 6.60429290e-03,\n",
              "        8.25536624e-03, 0.00000000e+00, 5.50357741e-04, 0.00000000e+00,\n",
              "        0.00000000e+00, 1.54100163e-02, 1.10071548e-03, 0.00000000e+00,\n",
              "        1.10071548e-03, 0.00000000e+00, 0.00000000e+00, 2.75178859e-03,\n",
              "        0.00000000e+00, 0.00000000e+00, 1.43093010e-02, 0.00000000e+00,\n",
              "        3.30214645e-03, 2.97193173e-02],\n",
              "       [0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 5.00000000e-01, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "        0.00000000e+00, 5.00000000e-01],\n",
              "       [0.00000000e+00, 3.01659130e-03, 2.02111617e-01, 8.29562638e-03,\n",
              "        1.65912528e-02, 1.58371050e-02, 8.37104097e-02, 2.12669685e-01,\n",
              "        1.50829565e-03, 3.01659126e-02, 7.54147826e-04, 3.77073907e-03,\n",
              "        0.00000000e+00, 9.87933651e-02, 0.00000000e+00, 0.00000000e+00,\n",
              "        5.95776774e-02, 6.78733038e-03, 7.54147815e-03, 6.78733066e-02,\n",
              "        0.00000000e+00, 2.79034693e-02, 2.18702871e-02, 0.00000000e+00,\n",
              "        4.60030176e-02, 8.52187052e-02],\n",
              "       [4.22386482e-04, 0.00000000e+00, 1.05596622e-02, 0.00000000e+00,\n",
              "        0.00000000e+00, 0.00000000e+00, 6.10137284e-01, 0.00000000e+00,\n",
              "        9.50369611e-03, 0.00000000e+00, 0.00000000e+00, 6.12460403e-03,\n",
              "        2.11193241e-04, 2.40126714e-01, 0.00000000e+00, 1.90073915e-03,\n",
              "        2.11193241e-04, 0.00000000e+00, 3.59028508e-03, 5.36430851e-02,\n",
              "        0.00000000e+00, 0.00000000e+00, 5.82893342e-02, 0.00000000e+00,\n",
              "        4.22386499e-03, 1.05596625e-03]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tags_df = pd.DataFrame(tags_matrix, columns = list(T), index=list(T))"
      ],
      "metadata": {
        "id": "usItCbanmNAp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "tags_df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 958
        },
        "id": "YIDpB1AbmN6v",
        "outputId": "10652839-0c1b-4f9e-a6aa-6214802ff1d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          SYMB      GRND      PNCT      COMP      CONJ      PRTS      NOUN  \\\n",
              "SYMB  0.053691  0.000000  0.174497  0.000000  0.006711  0.000000  0.080537   \n",
              "GRND  0.000000  0.000000  0.142857  0.000000  0.000000  0.000000  0.319328   \n",
              "PNCT  0.003576  0.007073  0.165859  0.002702  0.041564  0.004768  0.265040   \n",
              "COMP  0.000000  0.024000  0.448000  0.000000  0.016000  0.000000  0.096000   \n",
              "CONJ  0.000988  0.000988  0.178854  0.009881  0.014822  0.003953  0.227273   \n",
              "PRTS  0.000000  0.000000  0.145553  0.000000  0.021563  0.000000  0.229111   \n",
              "NOUN  0.000984  0.000492  0.366605  0.001107  0.018389  0.004982  0.216605   \n",
              "VERB  0.000000  0.000217  0.235894  0.005208  0.008464  0.033420  0.248264   \n",
              "LATN  0.033854  0.000000  0.381510  0.000000  0.009115  0.003906  0.032552   \n",
              "ADJS  0.000000  0.000000  0.440559  0.003497  0.027972  0.003497  0.087413   \n",
              "INTJ  0.000000  0.000000  0.930070  0.000000  0.000000  0.000000  0.013986   \n",
              "NUMR  0.000000  0.000000  0.065789  0.000000  0.000000  0.000000  0.736842   \n",
              "TIME  0.000000  0.000000  0.250000  0.000000  0.000000  0.000000  0.500000   \n",
              "ADJF  0.000216  0.000216  0.080684  0.000216  0.009301  0.002163  0.750595   \n",
              "GREK  0.250000  0.000000  0.535714  0.000000  0.000000  0.000000  0.000000   \n",
              "ROMN  0.000000  0.014493  0.304348  0.000000  0.000000  0.000000  0.434783   \n",
              "PRCL  0.000000  0.009579  0.059387  0.013410  0.006705  0.015326  0.064176   \n",
              "PRED  0.000000  0.007143  0.271429  0.000000  0.014286  0.000000  0.100000   \n",
              "PRTF  0.000000  0.000000  0.087805  0.000000  0.000000  0.000000  0.404878   \n",
              "NPRO  0.000000  0.000000  0.167944  0.005109  0.009579  0.006386  0.074713   \n",
              "HANI  0.000000  0.000000  1.000000  0.000000  0.000000  0.000000  0.000000   \n",
              "INFN  0.000000  0.000000  0.249719  0.004499  0.010124  0.007874  0.246344   \n",
              "NUMB  0.009356  0.000000  0.493671  0.000000  0.002752  0.000550  0.410567   \n",
              "DATE  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.500000   \n",
              "ADVB  0.000000  0.003017  0.202112  0.008296  0.016591  0.015837  0.083710   \n",
              "PREP  0.000422  0.000000  0.010560  0.000000  0.000000  0.000000  0.610137   \n",
              "\n",
              "          VERB      LATN      ADJS  ...      PRCL      PRED      PRTF  \\\n",
              "SYMB  0.006711  0.268456  0.000000  ...  0.006711  0.000000  0.000000   \n",
              "GRND  0.008403  0.000000  0.000000  ...  0.008403  0.000000  0.000000   \n",
              "PNCT  0.095287  0.043710  0.003338  ...  0.022332  0.002543  0.008186   \n",
              "COMP  0.120000  0.008000  0.000000  ...  0.016000  0.008000  0.000000   \n",
              "CONJ  0.103755  0.010870  0.006917  ...  0.049407  0.005929  0.003953   \n",
              "PRTS  0.000000  0.000000  0.000000  ...  0.016173  0.000000  0.005391   \n",
              "NOUN  0.100984  0.013776  0.004920  ...  0.017036  0.003014  0.002091   \n",
              "VERB  0.000217  0.007378  0.008030  ...  0.019531  0.000000  0.001953   \n",
              "LATN  0.059896  0.384766  0.000651  ...  0.007812  0.000000  0.000000   \n",
              "ADJS  0.059441  0.000000  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "INTJ  0.000000  0.000000  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "NUMR  0.026316  0.000000  0.000000  ...  0.013158  0.000000  0.026316   \n",
              "TIME  0.000000  0.000000  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "ADJF  0.023794  0.001298  0.000216  ...  0.006057  0.000649  0.002379   \n",
              "GREK  0.071429  0.142857  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "ROMN  0.028986  0.000000  0.014493  ...  0.000000  0.000000  0.000000   \n",
              "PRCL  0.524904  0.000958  0.024904  ...  0.029693  0.015326  0.007663   \n",
              "PRED  0.057143  0.000000  0.000000  ...  0.035714  0.000000  0.000000   \n",
              "PRTF  0.004878  0.019512  0.000000  ...  0.009756  0.009756  0.000000   \n",
              "NPRO  0.358238  0.001916  0.030013  ...  0.096424  0.010856  0.001277   \n",
              "HANI  0.000000  0.000000  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "INFN  0.007874  0.005624  0.004499  ...  0.026997  0.005624  0.003375   \n",
              "NUMB  0.006604  0.008255  0.000000  ...  0.001101  0.000000  0.000000   \n",
              "DATE  0.000000  0.000000  0.000000  ...  0.000000  0.000000  0.000000   \n",
              "ADVB  0.212670  0.001508  0.030166  ...  0.059578  0.006787  0.007541   \n",
              "PREP  0.000000  0.009504  0.000000  ...  0.000211  0.000000  0.003590   \n",
              "\n",
              "          NPRO      HANI      INFN      NUMB      DATE      ADVB      PREP  \n",
              "SYMB  0.000000  0.000000  0.000000  0.281879  0.000000  0.006711  0.020134  \n",
              "GRND  0.008403  0.000000  0.058824  0.016807  0.000000  0.016807  0.327731  \n",
              "PNCT  0.048160  0.002782  0.008901  0.051021  0.000159  0.040213  0.085910  \n",
              "COMP  0.024000  0.000000  0.024000  0.008000  0.000000  0.008000  0.080000  \n",
              "CONJ  0.068182  0.000000  0.026680  0.007905  0.000000  0.069170  0.075099  \n",
              "PRTS  0.000000  0.000000  0.021563  0.088949  0.000000  0.021563  0.339623  \n",
              "NOUN  0.010947  0.000000  0.006519  0.035609  0.000000  0.016359  0.113838  \n",
              "VERB  0.043837  0.000000  0.082682  0.021701  0.000000  0.035807  0.159505  \n",
              "LATN  0.005859  0.000000  0.000000  0.033203  0.000000  0.005208  0.027344  \n",
              "ADJS  0.006993  0.000000  0.136364  0.013986  0.000000  0.010490  0.136364  \n",
              "INTJ  0.041958  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  \n",
              "NUMR  0.000000  0.000000  0.000000  0.000000  0.000000  0.013158  0.000000  \n",
              "TIME  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.250000  \n",
              "ADJF  0.004975  0.000000  0.000433  0.001947  0.000000  0.002812  0.023145  \n",
              "GREK  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  \n",
              "ROMN  0.043478  0.000000  0.000000  0.000000  0.000000  0.043478  0.043478  \n",
              "PRCL  0.030651  0.000000  0.056513  0.003831  0.000000  0.047893  0.055556  \n",
              "PRED  0.000000  0.000000  0.371429  0.000000  0.000000  0.042857  0.035714  \n",
              "PRTF  0.004878  0.000000  0.019512  0.004878  0.000000  0.009756  0.307317  \n",
              "NPRO  0.040868  0.000000  0.028097  0.001916  0.000000  0.063218  0.069604  \n",
              "HANI  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  \n",
              "INFN  0.020247  0.000000  0.008999  0.007874  0.000000  0.039370  0.240720  \n",
              "NUMB  0.002752  0.000000  0.000000  0.014309  0.000000  0.003302  0.029719  \n",
              "DATE  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000  0.500000  \n",
              "ADVB  0.067873  0.000000  0.027903  0.021870  0.000000  0.046003  0.085219  \n",
              "PREP  0.053643  0.000000  0.000000  0.058289  0.000000  0.004224  0.001056  \n",
              "\n",
              "[26 rows x 26 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5df812e4-d6f7-41b6-9c3f-caff894569de\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>SYMB</th>\n",
              "      <th>GRND</th>\n",
              "      <th>PNCT</th>\n",
              "      <th>COMP</th>\n",
              "      <th>CONJ</th>\n",
              "      <th>PRTS</th>\n",
              "      <th>NOUN</th>\n",
              "      <th>VERB</th>\n",
              "      <th>LATN</th>\n",
              "      <th>ADJS</th>\n",
              "      <th>...</th>\n",
              "      <th>PRCL</th>\n",
              "      <th>PRED</th>\n",
              "      <th>PRTF</th>\n",
              "      <th>NPRO</th>\n",
              "      <th>HANI</th>\n",
              "      <th>INFN</th>\n",
              "      <th>NUMB</th>\n",
              "      <th>DATE</th>\n",
              "      <th>ADVB</th>\n",
              "      <th>PREP</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>SYMB</th>\n",
              "      <td>0.053691</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.174497</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.006711</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.080537</td>\n",
              "      <td>0.006711</td>\n",
              "      <td>0.268456</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.006711</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.281879</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.006711</td>\n",
              "      <td>0.020134</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>GRND</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.142857</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.319328</td>\n",
              "      <td>0.008403</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.008403</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.008403</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.058824</td>\n",
              "      <td>0.016807</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.016807</td>\n",
              "      <td>0.327731</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PNCT</th>\n",
              "      <td>0.003576</td>\n",
              "      <td>0.007073</td>\n",
              "      <td>0.165859</td>\n",
              "      <td>0.002702</td>\n",
              "      <td>0.041564</td>\n",
              "      <td>0.004768</td>\n",
              "      <td>0.265040</td>\n",
              "      <td>0.095287</td>\n",
              "      <td>0.043710</td>\n",
              "      <td>0.003338</td>\n",
              "      <td>...</td>\n",
              "      <td>0.022332</td>\n",
              "      <td>0.002543</td>\n",
              "      <td>0.008186</td>\n",
              "      <td>0.048160</td>\n",
              "      <td>0.002782</td>\n",
              "      <td>0.008901</td>\n",
              "      <td>0.051021</td>\n",
              "      <td>0.000159</td>\n",
              "      <td>0.040213</td>\n",
              "      <td>0.085910</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>COMP</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.024000</td>\n",
              "      <td>0.448000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.016000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.096000</td>\n",
              "      <td>0.120000</td>\n",
              "      <td>0.008000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.016000</td>\n",
              "      <td>0.008000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.024000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.024000</td>\n",
              "      <td>0.008000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.008000</td>\n",
              "      <td>0.080000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>CONJ</th>\n",
              "      <td>0.000988</td>\n",
              "      <td>0.000988</td>\n",
              "      <td>0.178854</td>\n",
              "      <td>0.009881</td>\n",
              "      <td>0.014822</td>\n",
              "      <td>0.003953</td>\n",
              "      <td>0.227273</td>\n",
              "      <td>0.103755</td>\n",
              "      <td>0.010870</td>\n",
              "      <td>0.006917</td>\n",
              "      <td>...</td>\n",
              "      <td>0.049407</td>\n",
              "      <td>0.005929</td>\n",
              "      <td>0.003953</td>\n",
              "      <td>0.068182</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.026680</td>\n",
              "      <td>0.007905</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.069170</td>\n",
              "      <td>0.075099</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PRTS</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.145553</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.021563</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.229111</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.016173</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.005391</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.021563</td>\n",
              "      <td>0.088949</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.021563</td>\n",
              "      <td>0.339623</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>NOUN</th>\n",
              "      <td>0.000984</td>\n",
              "      <td>0.000492</td>\n",
              "      <td>0.366605</td>\n",
              "      <td>0.001107</td>\n",
              "      <td>0.018389</td>\n",
              "      <td>0.004982</td>\n",
              "      <td>0.216605</td>\n",
              "      <td>0.100984</td>\n",
              "      <td>0.013776</td>\n",
              "      <td>0.004920</td>\n",
              "      <td>...</td>\n",
              "      <td>0.017036</td>\n",
              "      <td>0.003014</td>\n",
              "      <td>0.002091</td>\n",
              "      <td>0.010947</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.006519</td>\n",
              "      <td>0.035609</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.016359</td>\n",
              "      <td>0.113838</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>VERB</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000217</td>\n",
              "      <td>0.235894</td>\n",
              "      <td>0.005208</td>\n",
              "      <td>0.008464</td>\n",
              "      <td>0.033420</td>\n",
              "      <td>0.248264</td>\n",
              "      <td>0.000217</td>\n",
              "      <td>0.007378</td>\n",
              "      <td>0.008030</td>\n",
              "      <td>...</td>\n",
              "      <td>0.019531</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.001953</td>\n",
              "      <td>0.043837</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.082682</td>\n",
              "      <td>0.021701</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.035807</td>\n",
              "      <td>0.159505</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LATN</th>\n",
              "      <td>0.033854</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.381510</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.009115</td>\n",
              "      <td>0.003906</td>\n",
              "      <td>0.032552</td>\n",
              "      <td>0.059896</td>\n",
              "      <td>0.384766</td>\n",
              "      <td>0.000651</td>\n",
              "      <td>...</td>\n",
              "      <td>0.007812</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.005859</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.033203</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.005208</td>\n",
              "      <td>0.027344</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ADJS</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.440559</td>\n",
              "      <td>0.003497</td>\n",
              "      <td>0.027972</td>\n",
              "      <td>0.003497</td>\n",
              "      <td>0.087413</td>\n",
              "      <td>0.059441</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.006993</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.136364</td>\n",
              "      <td>0.013986</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.010490</td>\n",
              "      <td>0.136364</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>INTJ</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.930070</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.013986</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.041958</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>NUMR</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.065789</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.736842</td>\n",
              "      <td>0.026316</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.013158</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.026316</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.013158</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>TIME</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.250000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.250000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ADJF</th>\n",
              "      <td>0.000216</td>\n",
              "      <td>0.000216</td>\n",
              "      <td>0.080684</td>\n",
              "      <td>0.000216</td>\n",
              "      <td>0.009301</td>\n",
              "      <td>0.002163</td>\n",
              "      <td>0.750595</td>\n",
              "      <td>0.023794</td>\n",
              "      <td>0.001298</td>\n",
              "      <td>0.000216</td>\n",
              "      <td>...</td>\n",
              "      <td>0.006057</td>\n",
              "      <td>0.000649</td>\n",
              "      <td>0.002379</td>\n",
              "      <td>0.004975</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000433</td>\n",
              "      <td>0.001947</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.002812</td>\n",
              "      <td>0.023145</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>GREK</th>\n",
              "      <td>0.250000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.535714</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.142857</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ROMN</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.014493</td>\n",
              "      <td>0.304348</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.434783</td>\n",
              "      <td>0.028986</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.014493</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.043478</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.043478</td>\n",
              "      <td>0.043478</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PRCL</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.009579</td>\n",
              "      <td>0.059387</td>\n",
              "      <td>0.013410</td>\n",
              "      <td>0.006705</td>\n",
              "      <td>0.015326</td>\n",
              "      <td>0.064176</td>\n",
              "      <td>0.524904</td>\n",
              "      <td>0.000958</td>\n",
              "      <td>0.024904</td>\n",
              "      <td>...</td>\n",
              "      <td>0.029693</td>\n",
              "      <td>0.015326</td>\n",
              "      <td>0.007663</td>\n",
              "      <td>0.030651</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.056513</td>\n",
              "      <td>0.003831</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.047893</td>\n",
              "      <td>0.055556</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PRED</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.007143</td>\n",
              "      <td>0.271429</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.014286</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.100000</td>\n",
              "      <td>0.057143</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.035714</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.371429</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.042857</td>\n",
              "      <td>0.035714</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PRTF</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.087805</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.404878</td>\n",
              "      <td>0.004878</td>\n",
              "      <td>0.019512</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.009756</td>\n",
              "      <td>0.009756</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.004878</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.019512</td>\n",
              "      <td>0.004878</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.009756</td>\n",
              "      <td>0.307317</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>NPRO</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.167944</td>\n",
              "      <td>0.005109</td>\n",
              "      <td>0.009579</td>\n",
              "      <td>0.006386</td>\n",
              "      <td>0.074713</td>\n",
              "      <td>0.358238</td>\n",
              "      <td>0.001916</td>\n",
              "      <td>0.030013</td>\n",
              "      <td>...</td>\n",
              "      <td>0.096424</td>\n",
              "      <td>0.010856</td>\n",
              "      <td>0.001277</td>\n",
              "      <td>0.040868</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.028097</td>\n",
              "      <td>0.001916</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.063218</td>\n",
              "      <td>0.069604</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>HANI</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>INFN</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.249719</td>\n",
              "      <td>0.004499</td>\n",
              "      <td>0.010124</td>\n",
              "      <td>0.007874</td>\n",
              "      <td>0.246344</td>\n",
              "      <td>0.007874</td>\n",
              "      <td>0.005624</td>\n",
              "      <td>0.004499</td>\n",
              "      <td>...</td>\n",
              "      <td>0.026997</td>\n",
              "      <td>0.005624</td>\n",
              "      <td>0.003375</td>\n",
              "      <td>0.020247</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.008999</td>\n",
              "      <td>0.007874</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.039370</td>\n",
              "      <td>0.240720</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>NUMB</th>\n",
              "      <td>0.009356</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.493671</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.002752</td>\n",
              "      <td>0.000550</td>\n",
              "      <td>0.410567</td>\n",
              "      <td>0.006604</td>\n",
              "      <td>0.008255</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.001101</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.002752</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.014309</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.003302</td>\n",
              "      <td>0.029719</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>DATE</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ADVB</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.003017</td>\n",
              "      <td>0.202112</td>\n",
              "      <td>0.008296</td>\n",
              "      <td>0.016591</td>\n",
              "      <td>0.015837</td>\n",
              "      <td>0.083710</td>\n",
              "      <td>0.212670</td>\n",
              "      <td>0.001508</td>\n",
              "      <td>0.030166</td>\n",
              "      <td>...</td>\n",
              "      <td>0.059578</td>\n",
              "      <td>0.006787</td>\n",
              "      <td>0.007541</td>\n",
              "      <td>0.067873</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.027903</td>\n",
              "      <td>0.021870</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.046003</td>\n",
              "      <td>0.085219</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PREP</th>\n",
              "      <td>0.000422</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.010560</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.610137</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.009504</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000211</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.003590</td>\n",
              "      <td>0.053643</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.058289</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.004224</td>\n",
              "      <td>0.001056</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>26 rows × 26 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5df812e4-d6f7-41b6-9c3f-caff894569de')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-5df812e4-d6f7-41b6-9c3f-caff894569de button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-5df812e4-d6f7-41b6-9c3f-caff894569de');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(18, 12))\n",
        "sns.heatmap(tags_df)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 861
        },
        "id": "Y5i5I-nfmRWB",
        "outputId": "da64f8d3-e218-424e-a1b9-ca2c450cb237"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1200 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tags_frequent = tags_df[tags_df>0.5]\n",
        "plt.figure(figsize=(18, 12))\n",
        "sns.heatmap(tags_frequent)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 861
        },
        "id": "3-COZRFMmV-C",
        "outputId": "c91091f5-2f14-4661-d80e-60210572a161"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1200 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABTgAAAPNCAYAAACtZgVrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACk9UlEQVR4nOzdebyXZZ0//tdHlsNyBBTU44K5nABFXL5apk5ukbgE4lcNFwpEmRm3ZDCzYwVIycHKPdMc2fy2iIyDuZQbRtlkmRqTToiamZOCpoPggB6F8/n90cPz6xMg2/l4uOn5nMf1GM91X/d9vz/x3+vxvq67VC6XywEAAAAAKKAt2roAAAAAAIANJeAEAAAAAApLwAkAAAAAFJaAEwAAAAAoLAEnAAAAAFBYAk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCEnACAAAAAIUl4AQAAAAACkvACQAAAABstJ/97GcZPHhwdthhh5RKpdxxxx1rvWfu3Ln5P//n/6Smpib19fWZPn36er9XwAkAAAAAbLRly5Zln332yfXXX79O6//whz/kuOOOyxFHHJF58+ZlzJgxOeuss3Lfffet13tL5XK5vCEFAwAAAACsTqlUyuzZszN06NA1rrn44otzzz335KmnnmqZO+WUU/LGG2/k3nvvXed36eAEAAAAAFarqakpS5curRhNTU2t8uxHHnkkAwcOrJgbNGhQHnnkkfV6TvtWqQYAAAAAWK13X3u+rUvYYI3fuiWXXnppxdz48eMzYcKEjX72okWLst1221XMbbfddlm6dGneeuutdO7ceZ2eI+AEAAAAAFaroaEhY8eOrZirqalpo2pWT8AJAAAAAKxWTU1N1QLNurq6vPLKKxVzr7zySrp167bO3ZuJMzgBAAAAgDZw0EEHZc6cORVzDzzwQA466KD1eo4OTgAAAACopuaVbV3BB+J///d/89xzz7X8/Yc//CHz5s3L1ltvnZ133jkNDQ156aWXcssttyRJ/vmf/znf+ta38oUvfCGjRo3KQw89lNtuuy333HPPer1XBycAAAAAsNEee+yx7Lffftlvv/2SJGPHjs1+++2XcePGJUkWLlyYF198sWX9rrvumnvuuScPPPBA9tlnn1xxxRW5+eabM2jQoPV6b6lcLpdb72cAAAAAAH/t3VefbesSNliHbT/c1iWslS3qAAAAAFBN5ea2rmCzZos6AAAAAFBYAk4AAAAAoLAEnAAAAABAYTmDEwAAAACqqdkZnNW00R2cf/7zn3P22Wdn5513Tk1NTerq6jJo0KD89Kc/TV1dXSZNmrTKPZ/+9KfzsY99LCtXrsyECRNSKpVy9NFHr7LuG9/4RkqlUg4//PCWuffWvze6d++ej3/84/npT3+6sT8FAAAAACiYjQ44TzzxxPzmN7/JjBkz8swzz+TOO+/M4YcfniVLluSmm27KpZdemieffLJl/axZs3L33XdnxowZadeuXZJk++23z09+8pP86U9/qnj21KlTs/POO6/yzv79+2fhwoVZuHBhHnnkkXz4wx/Opz71qSxZsmRjfw4AAAAAUCAbtUX9jTfeyMMPP5y5c+fmsMMOS5J86EMfykc/+tGWNaeddlpGjBiRX/3qV3njjTdy7rnnZvLkyenbt2/Lmm233Tb7779/ZsyYkS996UtJkl/84hd57bXXcvLJJ+d3v/tdZdHt26euri5JUldXl4kTJ2batGl55pln8pGPfGRjfhIAAAAAUCAb1cFZW1ub2tra3HHHHWlqalrtmmuuuSavv/56vvrVr+acc87JXnvtlfPPP3+VdaNGjcr06dNb/p46dWpOP/30dOzY8X1raGpqyrRp09KjR4+K0BQAAAAANgXlcnNhRxFsVMDZvn37TJ8+PTNmzEiPHj1yyCGH5JJLLslvf/vbljXdunXLtGnTMmnSpNx///2ZNm1aSqXSKs/61Kc+laVLl+ZnP/tZli1blttuuy2jRo1a7XuffPLJlnC1c+fO+eY3v5kf/OAH6dat28b8HAAAAACgYFrlDM6XX345d955Z44++ujMnTs3/+f//J+KbswjjzwyH/vYx/KZz3wmH/rQh1b7nA4dOmT48OGZNm1aZs2alT59+mTvvfde7dq+fftm3rx5mTdvXh5//PGcffbZOfnkk/PYY4+tsc6mpqYsXbq0Yqyp6xQAAAAAKIaNDjiTpFOnTvnkJz+Zr3zlK/nFL36RkSNHZvz48RVr2rdvn/bt3//Iz1GjRmXWrFm5/vrr19i9mSQdO3ZMfX196uvrs99++2Xy5MnZcccdc/XVV6/xnsbGxnTv3r1iNDY2rtfvBAAAAID11txc3FEArRJw/q0999wzy5YtW+/7+vfvn/79++epp57Kaaedtl73tmvXLm+99dYarzc0NGTJkiUVo6GhYb1rBAAAAAA2HRv1FfXXX389J598ckaNGpW99947W265ZR577LF8/etfz/HHH79Bz3zooYfy7rvvpkePHmtcs2LFiixatChJ8uabb2bmzJn53e9+l4svvniN99TU1KSmpmaDagIAAAAANk0bFXDW1tbmwAMPzFVXXZXf//73effdd9O7d++MHj06l1xyyQY9s2vXrmtd81//9V/ZfvvtkyRdunTJ7rvvnhtuuCGf/exnN+idAAAAAEAxlcrlcrmtiwAAAACAzdU7//2fbV3CBuvYe5+2LmGtqnIGJwAAAADAB0HACQAAAAAUloATAAAAACisjfrIEAAAAACwFs0r27qCzZoOTgAAAACgsAScAAAAAEBh2aIOAAAAANVUbm7rCjZrOjgBAAAAgMIScAIAAAAAhSXgBAAAAAAKyxmcAAAAAFBNzc7grCYdnAAAAABAYQk4AQAAAIDCskUdAAAAAKqoXLZFvZp0cAIAAAAAhSXgBAAAAAAKS8AJAAAAABSWMzgBAAAAoJqancFZTTo4AQAAAIDCEnACAAAAAIUl4AQAAAAACssZnAAAAABQTWVncFaTDk4AAAAAoLAEnAAAAABAYdmiDgAAAADV1LyyrSvYrG1wB+eiRYtywQUXpL6+Pp06dcp2222XQw45JDfccEOWL1+eJNlll11SKpVSKpXSpUuXDBgwIDfffHPFc+bOnZtSqZT+/ftn5crKf+wePXpk+vTpLX//9fM6d+6cXXbZJZ/+9Kfz0EMPbejPAAAAAAAKbIMCzueffz777bdf7r///kyaNCm/+c1v8sgjj+QLX/hC7r777jz44IMtaydOnJiFCxfmqaeeyvDhwzN69Oj8+Mc/Xu0zb7nllrW++73nLViwILfcckt69OiRgQMH5rLLLtuQnwIAAAAAFNgGbVE/55xz0r59+zz22GPp2rVry/xuu+2W448/PuVyuWVuyy23TF1dXZLk4osvzte//vU88MADOeaYYyqeef7552f8+PE57bTTUlNTs8Z3//Xzdt555xx66KHZfvvtM27cuJx00knp27fvhvwkAAAAAKCA1ruD8/XXX8/999+fc889tyLc/GulUmmVuebm5tx+++1ZvHhxOnbsuMr1MWPGZMWKFbnuuuvWt6RccMEFKZfL+eEPf7je9wIAAABAVZWbizsKYL0Dzueeey7lcnmVTslevXqltrY2tbW1ufjii1vmL7744tTW1qampiYnnXRSttpqq5x11lmrPLdLly4ZP358Ghsbs2TJkvWqaeutt862226bF154YX1/DgAAAABQYBv8kaG/9eijj2bevHnp379/mpqaWuYvuuiizJs3Lw899FAOPPDAXHXVVamvr1/tM84888z07Nkzl19++Xq/v1wur7Zz9D1NTU1ZunRpxfjrOgEAAACA4lnvgLO+vj6lUikLFiyomN9tt91SX1+fzp07V8z36tUr9fX1+fjHP55Zs2blc5/7XH73u9+t9tnt27fPZZddlmuuuSYvv/zyOtf0+uuv589//nN23XXXNa5pbGxM9+7dK0ZjY+M6vwMAAAAANkhzc3FHAax3wNmzZ8988pOfzLe+9a0sW7Zsve7t3bt3hg0bloaGhjWuOfnkk9O/f/9ceuml6/zca665JltssUWGDh26xjUNDQ1ZsmRJxXi/OgAAAACATd8GfUX929/+dg455JAccMABmTBhQvbee+9sscUW+fWvf52nn346+++//xrvveCCC7LXXnvlscceywEHHLDaNZMnT86gQYNWe+3NN9/MokWL8u677+YPf/hDvvvd7+bmm29OY2PjGre+J0lNTc37fp0dAAAAACieDTqDc/fdd89vfvObDBw4MA0NDdlnn31ywAEH5LrrrsvnP//5fPWrX13jvXvuuWeOOuqojBs3bo1rjjzyyBx55JFZsWLFKtfGjRuX7bffPvX19fnMZz6TJUuWZM6cORUfNgIAAAAA/j6UyuVyua2LAAAAAIDNVdNTD7R1CRusZq9PtnUJa9VqX1EHAAAAAPigCTgBAAAAgMIScAIAAAAAhbVBX1EHAAAAANZRc3NbV7BZ08EJAAAAABSWgBMAAAAAKCxb1AEAAACgisrllW1dwmZNBycAAAAAUFgCTgAAAACgsAScAAAAAEBhOYMTAAAAAKqp3NzWFWzWdHACAAAAAIUl4AQAAAAACssWdQAAAACopmZb1KtJBycAAAAAUFgCTgAAAACgsAScAAAAAEBhOYMTAAAAAKqp7AzOatLBCQAAAAAUloATAAAAACgsAScAAAAAUFjO4AQAAACAampe2dYVbNZ0cAIAAAAAhSXgBAAAAAAKq1UCzpEjR6ZUKqVUKqVjx46pr6/PxIkTs2LFisydOzelUin9+/fPypWV7bg9evTI9OnTK+Z+85vf5OSTT852222XTp065cMf/nBGjx6dZ555JhMmTGh5z5oGAAAAAGxSys3FHQXQah2cRx99dBYuXJhnn302F154YSZMmJBvfOMbLdeff/753HLLLe/7jLvvvjsf+9jH0tTUlO9973uZP39+vvvd76Z79+75yle+ks9//vNZuHBhy9hpp50yceLEijkAAAAA4O9Hq31kqKamJnV1dUmSs88+O7Nnz86dd96Zgw46KEly/vnnZ/z48TnttNNSU1Ozyv3Lly/PGWeckWOPPTazZ89umd91111z4IEH5o033khtbW1qa2tbrrVr1y5bbrlly3sBAAAAgL8vVTuDs3PnznnnnXda/h4zZkxWrFiR6667brXr77vvvrz22mv5whe+sNrrPXr0qEaZAAAAAECBtXrAWS6X8+CDD+a+++7LkUce2TLfpUuXjB8/Po2NjVmyZMkq9z377LNJkn79+rV2SQAAAADQdpqbizsKoNUCzrvvvju1tbXp1KlTjjnmmAwbNiwTJkyoWHPmmWemZ8+eufzyy1e5v1wut1Ypq9XU1JSlS5dWjKampqq+EwAAAACorlYLOI844ojMmzcvzz77bN56663MmDEjXbt2rVjTvn37XHbZZbnmmmvy8ssvV1zr06dPkuTpp59urZIqNDY2pnv37hWjsbGxKu8CAAAAAD4YrRZwdu3aNfX19dl5553Tvv2av1108sknp3///rn00ksr5o866qj06tUrX//611d73xtvvLFR9TU0NGTJkiUVo6GhYaOeCQAAAAC0rVb7ivr6mDx5cgYNGlQx17Vr19x88805+eSTM2TIkHzuc59LfX19Xnvttdx222158cUXc+utt27wO2tqalb79XYAAAAAqKpyMc6yLKqqfUX9/Rx55JE58sgjs2LFior5448/Pr/4xS/SoUOHnHbaaenXr19OPfXULFmyJF/72tfaolQAAAAAYBNWKlf76z4AAAAA8Hfs7Ud+0NYlbLBOB53a1iWsVZtsUQcAAACAvxvNtqhXU5tsUQcAAAAAaA0CTgAAAACgsAScAAAAAEBhOYMTAAAAAKrJGZxVpYMTAAAAACgsAScAAAAAUFi2qAMAAABAFZXLK9u6hM2aDk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCcgYnAAAAAFRTc3NbV7BZ08EJAAAAABSWgBMAAAAAKCwBJwAAAABQWM7gBAAAAIBqKjuDs5p0cAIAAAAAhSXgBAAAAAAKyxZ1AAAAAKimZlvUq0kHJwAAAABQWAJOAAAAAKCwBJwAAAAAQGE5gxMAAAAAqqnsDM5q0sEJAAAAABSWgBMAAAAAKKyNCjgXLVqU888/P7vttltqamrSu3fvDB48OHPmzGlZ84tf/CLHHntsttpqq3Tq1CkDBgzIlVdemZUrV1Y8q1QqpVQq5Ze//GXFfFNTU3r27JlSqZS5c+eusr5UKqV79+455JBD8tBDD23MzwEAAACA1tfcXNxRABsccL7wwgvZf//989BDD+Ub3/hGnnzyydx777054ogjcu655yZJZs+encMOOyw77bRTfvKTn+Tpp5/OBRdckK997Ws55ZRTUi6XK57Zu3fvTJs2rWJu9uzZqa2tXW0N06ZNy8KFC/Mf//Ef6dWrVz71qU/l+eef39CfBAAAAAAUzAYHnOecc05KpVIeffTRnHjiienTp0/69++fsWPH5pe//GWWLVuW0aNHZ8iQIbnpppuy7777ZpdddslZZ52VGTNm5N/+7d9y2223VTxzxIgRufXWW/PWW2+1zE2dOjUjRoxYbQ09evRIXV1d9tprr9xwww1566238sADD2zoTwIAAAAACmaDAs7/+Z//yb333ptzzz03Xbt2XeV6jx49cv/99+f111/P5z//+VWuDx48OH369MkPfvCDivn9998/u+yyS26//fYkyYsvvpif/exn+cxnPrPWmjp37pwkeeeddzbkJwEAAAAABbRBAedzzz2Xcrmcfv36rXHNM888kyTZY489Vnu9X79+LWv+2qhRozJ16tQkyfTp03Psscdmm222ed96li9fni9/+ctp165dDjvssHX9GQAAAABQfeXm4o4CaL8hN/3t2ZmttTZJhg8fni9+8Yt5/vnnM3369Fx77bVrXHvqqaemXbt2eeutt7LNNttkypQp2XvvvVe7tqmpKU1NTRVzNTU1qampWa/6AAAAAIBNxwZ1cH74wx9OqVTK008/vcY1ffr0SZLMnz9/tdfnz5/fsuav9ezZM5/61Kdy5pln5u23384xxxyzxndcddVVmTdvXhYtWpRFixat8azOJGlsbEz37t0rRmNj4xrXAwAAAACbvg0KOLfeeusMGjQo119/fZYtW7bK9TfeeCNHHXVUtt5661xxxRWrXL/zzjvz7LPP5tRTT13t80eNGpW5c+fms5/9bNq1a7fGOurq6lJfX7/WLexJ0tDQkCVLllSMhoaGtd4HAAAAAGy6NmiLepJcf/31OeSQQ/LRj340EydOzN57750VK1bkgQceyA033JD58+fnO9/5Tk455ZT84z/+Y84777x069Ytc+bMyUUXXZSTTjopn/70p1f77KOPPjp//vOf061btw3+YX/LdnQAAAAA2kRzMc6yLKoNDjh32223PPHEE7nsssty4YUXZuHChdlmm22y//7754YbbkiSnHTSSfnJT36Syy67LB//+Mfz9ttv58Mf/nC+9KUvZcyYMSmVSqt9dqlUSq9evTa0NAAAAADg70SpvL5fAQIAAAAA1tlbP17zR7Q3dZ2P+Vxbl7BWG9zBCQAAAACsA1vUq2qDPjIEAAAAALApEHACAAAAAIUl4AQAAAAACssZnAAAAABQTWVncFaTDk4AAAAAoLAEnAAAAABAYdmiDgAAAADV1GyLejXp4AQAAAAACkvACQAAAAAUloATAAAAACgsZ3ACAAAAQDWVncFZTTo4AQAAAIDCEnACAAAAAIUl4AQAAAAACssZnAAAAABQTc3O4KwmHZwAAAAAQGEJOAEAAACAwrJFHQAAAACqqWyLejXp4AQAAAAACkvACQAAAAAUloATAAAAACgsZ3ACAAAAQDU1O4OzmnRwAgAAAACFVZWAc9GiRTn//POz2267paamJr17987gwYMzZ86cljW/+MUvcuyxx2arrbZKp06dMmDAgFx55ZVZuXJlxbNKpVI6deqUP/7xjxXzQ4cOzciRI1v+HjlyZIYOHVqNnwMAAAAAbKJaPeB84YUXsv/+++ehhx7KN77xjTz55JO59957c8QRR+Tcc89NksyePTuHHXZYdtppp/zkJz/J008/nQsuuCBf+9rXcsopp6RcLlc8s1QqZdy4ca1dKgAAAABUX3NzcUcBtPoZnOecc05KpVIeffTRdO3atWW+f//+GTVqVJYtW5bRo0dnyJAhuemmm1qun3XWWdluu+0yZMiQ3HbbbRk2bFjLtfPOOy9XXnllLrroouy1116tXTIAAAAAUFCt2sH5P//zP7n33ntz7rnnVoSb7+nRo0fuv//+vP766/n85z+/yvXBgwenT58++cEPflAxf8ghh+RTn/pUvvjFL7ZmuQAAAABAwbVqwPncc8+lXC6nX79+a1zzzDPPJEn22GOP1V7v169fy5q/1tjYmHvvvTcPP/xw6xQLAAAAABReq25R/9uzM1trbZLsueee+exnP5svfvGL+Y//+I/1LS1NTU1pamqqmKupqUlNTc16PwsAAAAA1tl65mCsn1bt4Pzwhz+cUqmUp59+eo1r+vTpkySZP3/+aq/Pnz+/Zc3fuvTSS/PEE0/kjjvuWO/aGhsb071794rR2Ni43s8BAAAAADYdrRpwbr311hk0aFCuv/76LFu2bJXrb7zxRo466qhsvfXWueKKK1a5fuedd+bZZ5/Nqaeeutrn9+7dO+edd14uueSSrFy5cr1qa2hoyJIlSypGQ0PDej0DAAAAANi0tGrAmSTXX399Vq5cmY9+9KO5/fbb8+yzz2b+/Pm59tprc9BBB6Vr1675zne+kx/+8If5x3/8x/z2t7/NCy+8kClTpmTkyJE56aST8ulPf3qNz29oaMjLL7+cBx98cL3qqqmpSbdu3SqG7ekAAAAAUGytHnDutttueeKJJ3LEEUfkwgsvzF577ZVPfvKTmTNnTm644YYkyUknnZSf/OQnefHFF/Pxj388ffv2zVVXXZUvfelLufXWW1Mqldb4/K233joXX3xx3n777Yr55ubmtG/fqkeKAgAAAMDGa24u7iiAUnl9v/aziTr66KNTX1+fb33rW21dCgAAAAC0eOsH49u6hA3W+dRL27qEtWr1Ds4P2uLFi3P33Xdn7ty5GThwYFuXAwAAAAB8gAq/p3vUqFH59a9/nQsvvDDHH398W5cDAAAAAJUKstW7qAofcM6ePbutSwAAAAAA2kjht6gDAAAAAH+/BJwAAAAAQGEVfos6AAAAAGzSys7grCYdnAAAAABAYQk4AQAAAIDCEnACAAAAAIXlDE4AAAAAqKZmZ3BWkw5OAAAAAKCwBJwAAAAAQGHZog4AAAAA1VQut3UFmzUdnAAAAABAYQk4AQAAAIDCEnACAAAAAIXlDE4AAAAAqKbm5rauYLOmgxMAAAAAKCwBJwAAAABQWLaoAwAAAEA12aJeVTo4AQAAAIDCEnACAAAAAIUl4AQAAAAACssZnAAAAABQTWVncFaTDk4AAAAAoLBaPeAcOXJkSqVSSqVSOnbsmPr6+kycODErVqzI3LlzW66VSqVss802OfbYY/Pkk08mScW11Y0JEyYkSWbPnp2Pfexj6d69e7bccsv0798/Y8aMae2fAgAAAACsh+uvvz677LJLOnXqlAMPPDCPPvroGte+++67mThxYnbfffd06tQp++yzT+699971fmdVtqgfffTRmTZtWpqamvKjH/0o5557bjp06JCDDjooSbJgwYJ069YtL7/8ci666KIcd9xxee6557Jw4cKWZ8ycOTPjxo3LggULWuZqa2szZ86cDBs2LJdddlmGDBmSUqmU3/3ud3nggQeq8VMAAAAAgHUwc+bMjB07NjfeeGMOPPDAXH311Rk0aFAWLFiQbbfddpX1X/7yl/Pd7343//qv/5p+/frlvvvuywknnJBf/OIX2W+//db5vaVyuVxuzR8ycuTIvPHGG7njjjta5o466qi8+eabaWxszBFHHJHFixenR48eSZK77rorQ4YMyX/+539m7733brln+vTpGTNmTN54442K548ZMyb/+Z//mZ/85CetWTYAAAAAVMXym/6lrUvYYF3+8ap1XnvggQfmIx/5SL71rW8lSZqbm9O7d++cf/75+eIXv7jK+h122CFf+tKXcu6557bMnXjiiencuXO++93vrvN7P5AzODt37px33nlnlfklS5bk1ltvTZJ07NhxnZ5VV1eX//qv/8pTTz3VqjUCAAAAABvmnXfeyeOPP56BAwe2zG2xxRYZOHBgHnnkkdXe09TUlE6dOlXMde7cOT//+c/X691V/Yp6uVzOnDlzct999+X8889vmd9pp52SJMuWLUuSDBkyJP369VunZ55//vl5+OGHM2DAgHzoQx/Kxz72sRx11FE5/fTTU1NT0/o/AgAAAAD+TjU1NaWpqalirqamZpUc7rXXXsvKlSuz3XbbVcxvt912efrpp1f77EGDBuXKK6/MoYcemt133z1z5szJv//7v2flypXrVWNVOjjvvvvu1NbWplOnTjnmmGMybNiwlg8EJcnDDz+cxx9/PNOnT0+fPn1y4403rvOzu3btmnvuuSfPPfdcvvzlL6e2tjYXXnhhPvrRj2b58uVrvK+pqSlLly6tGH/7jwMAAAAAra65ubCjsbEx3bt3rxiNjY2t8j/LNddckw9/+MPp169fOnbsmPPOOy9nnHFGtthi/SLLqgScRxxxRObNm5dnn302b731VmbMmJGuXbu2XN91113Tt2/fjBgxImeddVaGDRu23u/Yfffdc9ZZZ+Xmm2/OE088kd/97neZOXPmGtdX8x8DAAAAADZHDQ0NWbJkScVoaGhYZV2vXr3Srl27vPLKKxXzr7zySurq6lb77G222SZ33HFHli1blj/+8Y95+umnU1tbm9122229aqxKwNm1a9fU19dn5513Tvv2778L/txzz81TTz2V2bNnb/D7dtlll3Tp0qVly/vqrOs/BgAAAADwFzU1NenWrVvFWN0xkR07dsz++++fOXPmtMw1Nzdnzpw5Oeigg973HZ06dcqOO+6YFStW5Pbbb8/xxx+/XjVW9QzOddGlS5eMHj0648ePz9ChQ1Mqld53/YQJE7J8+fIce+yx+dCHPpQ33ngj1157bd5999188pOfXON9qzsbAAAAAABoHWPHjs2IESNywAEH5KMf/WiuvvrqLFu2LGeccUaS5LOf/Wx23HHHll3Vv/rVr/LSSy9l3333zUsvvZQJEyakubk5X/jCF9brvW0ecCbJeeedlyuvvDKzZs3Kpz/96fdde9hhh+X666/PZz/72bzyyivZaqutst9+++X+++9P3759P6CKAQAAAGAdlZvbuoIPxLBhw/LnP/8548aNy6JFi7Lvvvvm3nvvbfnw0Isvvlhxvubbb7+dL3/5y3n++edTW1ubY489Nv/v//2/9OjRY73eWyqXy+XW/CEAAAAAwP9v+Q3nt3UJG6zL2de1dQlrVZUzOAEAAAAAPgibxBZ1AAAAANhsNdtAXU06OAEAAACAwhJwAgAAAACFJeAEAAAAAArLGZwAAAAAUE3NzW1dwWZNBycAAAAAUFgCTgAAAACgsAScAAAAAEBhOYMTAAAAAKrJGZxVpYMTAAAAACgsAScAAAAAUFi2qAMAAABANZXLbV3BZk0HJwAAAABQWAJOAAAAAKCwBJwAAAAAQGE5gxMAAAAAqqm5ua0r2Kzp4AQAAAAACkvACQAAAAAUli3qAAAAAFBNzeW2rmCzpoMTAAAAACgsAScAAAAAUFgCTgAAAACgsJzBCQAAAADVVG5u6wo2azo4AQAAAIDC2qCAc+TIkSmVSpk8eXLF/B133JFSqdTy98qVK3PVVVdlwIAB6dSpU7baaqscc8wx+Y//+I+K+yZMmJB99913lfe88MILKZVKmTdvXpJk7ty5KZVK6d+/f1auXFmxtkePHpk+ffqG/BwAAAAAoKA2uIOzU6dOufzyy7N48eLVXi+XyznllFMyceLEXHDBBZk/f37mzp2b3r175/DDD88dd9yxoa/O888/n1tuuWWD7wcAAAAANg8bHHAOHDgwdXV1aWxsXO312267Lf/2b/+WW265JWeddVZ23XXX7LPPPrnpppsyZMiQnHXWWVm2bNkGvfv888/P+PHj09TUtKHlAwAAAMAHo7lc3FEAGxxwtmvXLpMmTcp1112XP/3pT6tc//73v58+ffpk8ODBq1y78MIL8/rrr+eBBx7YoHePGTMmK1asyHXXXbdB9wMAAAAAm4eN+sjQCSeckH333Tfjx49f5dozzzyTPfbYY7X3vTf/zDPPbNB7u3TpkvHjx6exsTFLlixZp3uampqydOnSiqEDFAAAAACKbaO/on755ZdnxowZmT9//irXyuXqtbGeeeaZ6dmzZy6//PJ1Wt/Y2Jju3btXjDVtrwcAAACA1lJubi7sKIKNDjgPPfTQDBo0KA0NDRXzffr0WW3omaRlvk+fPkmSbt26rbYT84033kiSdO/efZVr7du3z2WXXZZrrrkmL7/88lrrbGhoyJIlSyrG39YMAAAAABTLRgecSTJ58uTcddddeeSRR1rmTjnllDz77LO56667Vll/xRVXpGfPnvnkJz+ZJOnbt2/+9Kc/5ZVXXqlY98QTT6RTp07ZeeedV/vek08+Of3798+ll1661hpramrSrVu3ilFTU7M+PxMAAAAA2MS0b42HDBgwIKeffnquvfbalrlTTjkls2bNyogRI/KNb3wjn/jEJ7J06dJcf/31ufPOOzNr1qx07do1STJo0KD07ds3p556ar72ta+lrq4uTzzxRL785S/nggsuSLt27db47smTJ2fQoEGt8TMAAAAAgIJplQ7OJJk4cWKa/2pffqlUym233ZZLLrkkV111Vfr27ZuPf/zj+eMf/5i5c+dm6NChLWvbt2+f+++/PzvvvHNOPfXU7LXXXhk/fnwuuOCCfPWrX33f9x555JE58sgjs2LFitb6KQAAAADQeprLxR0FUCpX80tAAAAAAPB3btlln23rEjZY1y/d0tYlrFWrdXACAAAAAHzQWuUMTgAAAABgDcrNa1/DBtPBCQAAAAAUloATAAAAACgsAScAAAAAUFjO4AQAAACAamout3UFmzUdnAAAAABAYQk4AQAAAIDCEnACAAAAAIXlDE4AAAAAqKbm5rauYLOmgxMAAAAAKCwBJwAAAABQWLaoAwAAAEA1NZfbuoLNmg5OAAAAAKCwBJwAAAAAQGEJOAEAAACAwnIGJwAAAABUU7m5rSvYrOngBAAAAAAKS8AJAAAAABSWgBMAAAAAKCxncAIAAABANTWX27qCzZoOTgAAAACgsAScAAAAAEBh2aIOAAAAAFVUbm5u6xI2axvVwTl48OAcffTRq7328MMPp1Qq5be//W1KpdJqxy9/+cskyfTp01vmtthii2y//fYZNmxYXnzxxYpnHn744RX3b7fddjn55JPzxz/+cWN+BgAAAABQUBsVcJ555pl54IEH8qc//WmVa9OmTcsBBxyQbt26JUkefPDBLFy4sGLsv//+Leu7deuWhQsX5qWXXsrtt9+eBQsW5OSTT17luaNHj87ChQvz8ssv54c//GH++7//O8OHD9+YnwEAAAAAFNRGBZyf+tSnss0222T69OkV8//7v/+bWbNm5cwzz2yZ69mzZ+rq6ipGhw4dWq6XSqXU1dVl++23z8EHH5wzzzwzjz76aJYuXVrx7C5durSs+9jHPpbzzjsvTzzxxMb8DAAAAACgoDYq4Gzfvn0++9nPZvr06SmX///P3c+aNSsrV67MqaeeukHPffXVVzN79uy0a9cu7dq1W+O6//mf/8ltt92WAw88cIPeAwAAAABV11wu7iiAjf6K+qhRo/L73/8+P/3pT1vmpk2blhNPPDHdu3dvmTv44INTW1tbMf7akiVLUltbm65du2a77bbLT37yk5x77rnp2rVrxbpvf/vbLet69uyZBQsWZOrUqRv7MwAAAACAAtror6j369cvBx98cKZOnZrDDz88zz33XB5++OFMnDixYt3MmTOzxx57rPE5W265ZZ544om8++67+fGPf5zvfe97ueyyy1ZZd/rpp+dLX/pSkuSVV17JpEmTctRRR+Xxxx/PlltuucbnNzU1pampqWKupqYmNTU16/NzAQAAAIBNyEZ3cCZ/+djQ7bffnjfffDPTpk3L7rvvnsMOO6xiTe/evVNfX18xKgrZYovU19dnjz32yNixY/Oxj30sZ5999irv6t69e8v9hxxySKZMmZJnn302M2fOfN8aGxsb071794rR2Ni48T8eAAAAAN5PW28zt0V97T796U9niy22yPe///3ccsstGTVqVEql0kY984tf/GJmzpy51g8IvXdG51tvvfW+6xoaGrJkyZKK0dDQsFE1AgAAAABta6O3qCdJbW1thg0bloaGhixdujQjR45cZc3rr7+eRYsWVcz16NEjnTp1Wu0ze/funRNOOCHjxo3L3Xff3TK/fPnylue88sor+epXv5pOnTrlqKOOet8abUcHAAAAgM1Pq3RwJn/Zpr548eIMGjQoO+ywwyrXBw4cmO23375i3HHHHe/7zH/5l3/JPffck0cffbRl7l//9V9b7j/iiCPy2muv5Uc/+lH69u3bWj8FAAAAACiIUrlcLsZmegAAAAAooP/9/PFtXcIGq/3mD9u6hLVqtQ5OAAAAAIAPmoATAAAAACgsAScAAAAAUFit8hV1AAAAAGANmn0Cp5p0cAIAAAAAhSXgBAAAAAAKyxZ1AAAAAKiisi3qVaWDEwAAAAAoLAEnAAAAAFBYAk4AAAAAoLCcwQkAAAAA1eQMzqrSwQkAAAAAFJaAEwAAAAAoLFvUAQAAAKCampvbuoLNmg5OAAAAAKCwBJwAAAAAQGEJOAEAAACAwnIGJwAAAABUU3O5rSvYrOngBAAAAAAKS8AJAAAAABSWgBMAAAAAKCxncAIAAABANTmDs6p0cAIAAAAAhSXgBAAAAAAKyxZ1AAAAAKiictkW9Wpq1Q7OkSNHZujQoe+75k9/+lM6duyYvfbaq2VuwoQJKZVK7zvee36pVMrkyZMrnnnHHXe0rAEAAAAA/n584FvUp0+fnk9/+tNZunRpfvWrXyVJPv/5z2fhwoUtY6eddsrEiRMr5t7TqVOnXH755Vm8ePEHXToAAAAAsIn5QAPOcrmcadOm5TOf+UxOO+20TJkyJUlSW1uburq6ltGuXbtsueWWFXPvGThwYOrq6tLY2PhBlg4AAAAAbII+0IDzJz/5SZYvX56BAwdm+PDhufXWW7Ns2bL1eka7du0yadKkXHfddfnTn/5UpUoBAAAAoJU0l4s7CuADDTinTJmSU045Je3atctee+2V3XbbLbNmzVrv55xwwgnZd999M378+HW+p6mpKUuXLq0YTU1N6/1uAAAAAGDT8YEFnG+88Ub+/d//PcOHD2+ZGz58eMs29fV1+eWXZ8aMGZk/f/46rW9sbEz37t0rhm3uAAAAAFBs7T+oF33/+9/P22+/nQMPPLBlrlwup7m5Oc8880z69OmzXs879NBDM2jQoDQ0NGTkyJFrXd/Q0JCxY8dWzNXU1KzXOwEAAABgvRVkq3dRfWAB55QpU3LhhReuEkaec845mTp1aiZPnrzez5w8eXL23Xff9O3bd61ra2pqBJoAAAAAsJlp9YBzyZIlmTdvXsXcm2++mSeeeCLf+9730q9fv4prp556aiZOnJivfe1rad9+/coZMGBATj/99Fx77bUbWzYAAAAAUECtfgbn3Llzs99++1WMqVOnZs8991wl3Ez+8sGgV199NT/60Y826H0TJ05Mc3PzxpYNAAAAABRQqVwuOwQAAAAAAKpkyRkD27qEDdZ92oNtXcJafWBfUQcAAAAAaG0CTgAAAACgsAScAAAAAEBhtfpX1AEAAACAv9LsEzjVpIMTAAAAACgsAScAAAAAUFi2qAMAAABANTW3dQGbNx2cAAAAAEBhCTgBAAAAgMIScAIAAAAAheUMTgAAAACoonJzua1L2Kzp4AQAAAAACkvACQAAAAAUloATAAAAACgsZ3ACAAAAQDU5g7OqdHACAAAAAIUl4AQAAAAACssWdQAAAACopua2LmDzpoMTAAAAACgsAScAAAAAUFgCTgAAAACgsJzBCQAAAABVVG4ut3UJmzUdnAAAAABAYQk4AQAAAIDCqmrA+cgjj6Rdu3Y57rjjKuZfeOGFlEqllrHlllumf//+Offcc/Pss89WrJ0+fXp69OjR8vfKlSszefLk9OvXL507d87WW2+dAw88MDfffHM1fwoAAAAAbJjmAo8CqOoZnFOmTMn555+fKVOm5OWXX84OO+xQcf3BBx9M//79s3z58jz55JO55pprss8+++Suu+7KJz7xidU+89JLL813vvOdfOtb38oBBxyQpUuX5rHHHsvixYur+VMAAAAAgE1Q1QLO//3f/83MmTPz2GOPZdGiRZk+fXouueSSijU9e/ZMXV1dkmS33XbL4MGD84lPfCJnnnlmfv/736ddu3arPPfOO+/MOeeck5NPPrllbp999qnWzwAAAAAANmFV26J+2223pV+/funbt2+GDx+eqVOnplx+/y9GbbHFFrngggvyxz/+MY8//vhq19TV1eWhhx7Kn//852qUDQAAAAAUSNUCzilTpmT48OFJkqOPPjpLlizJT3/607Xe169fvyR/Oadzda688sr8+c9/Tl1dXfbee+/88z//c3784x+3Wt0AAAAA0JrKzeXCjiKoSsC5YMGCPProozn11FOTJO3bt8+wYcMyZcqUtd77XpdnqVRa7fU999wzTz31VH75y19m1KhRefXVVzN48OCcddZZ7/vcpqamLF26tGI0NTWt5y8DAAAAADYlVQk4p0yZkhUrVmSHHXZI+/bt0759+9xwww25/fbbs2TJkve9d/78+UmSXXfddY1rtthii3zkIx/JmDFj8u///u+ZPn16pkyZkj/84Q9rvKexsTHdu3evGI2NjRv2AwEAAACATUKrf2RoxYoVueWWW3LFFVfkqKOOqrg2dOjQ/OAHP8jRRx+92nubm5tz7bXXZtddd81+++23zu/cc889kyTLli1b45qGhoaMHTu2Yq6mpmad3wEAAAAAbHpaPeC8++67s3jx4px55pnp3r17xbUTTzwxU6ZMaQk4X3/99SxatCjLly/PU089lauvvjqPPvpo7rnnntV+QT1JTjrppBxyyCE5+OCDU1dXlz/84Q9paGhInz59Ws7vXJ2amhqBJgAAAAAfvOa2LmDz1upb1KdMmZKBAweuEm4mfwk4H3vssSxdujRJMnDgwGy//fYZMGBAvvjFL2aPPfbIb3/72xxxxBEt9zQ3N6d9+/8/hx00aFDuuuuuDB48OH369MmIESPSr1+/3H///RXrAAAAAIDNX6n83ld9NlGTJ0/Od7/73Tz11FNtXQoAAAAArLf/Of6wti5hg239w5+2dQlrtcm2PC5fvjxPP/10pk2blmOOOaatywEAAACADVK2Rb2qqvIV9dZw0003ZeDAgdlnn30ybty4ti4HAAAAANgEbfJb1AEAAACgyF4fXNwt6j3v2vS3qG+yHZwAAAAAAGuzyZ7BCQAAAACbBWdwVpUOTgAAAACgsAScAAAAAEBh2aIOAAAAAFVUtkW9qnRwAgAAAACFJeAEAAAAAApLwAkAAAAAFJYzOAEAAACgmpzBWVU6OAEAAACAwtLBSat66+4r27qEddb5U2PbugQAAAAANpIOTgAAAACgsHRwAgAAAEAVlZ3BWVU6OAEAAACAwhJwAgAAAACFZYs6AAAAAFSRLerVpYMTAAAAACgsAScAAAAAUFgCTgAAAACgsJzBCQAAAABV5AzO6tLBCQAAAAAUloATAAAAACisDzTgHDlyZIYOHdry36VSKZMnT65Yc8cdd6RUKlWsWdPYZZddkiSHH354xowZ8wH+EgAAAABYR+VScUcBtGkHZ6dOnXL55Zdn8eLFq71+zTXXZOHChS0jSaZNm9by969//esPslwAAAAAYBPTpgHnwIEDU1dXl8bGxtVe7969e+rq6lpGkvTo0aPl72222eaDLBcAAAAA2MS0acDZrl27TJo0Kdddd13+9Kc/tWUpAAAAAEABtflHhk444YTsu+++GT9+fFXf09TUlKVLl1aMpqamqr4TAAAAAMrNxR1F0OYBZ5JcfvnlmTFjRubPn1+1dzQ2NqZ79+4VY01b4wEAAACAYtgkAs5DDz00gwYNSkNDQ9Xe0dDQkCVLllSMar4PAAAAAKi+9m1dwHsmT56cfffdN3379q3K82tqalJTU1OVZwMAAAAAbWOTCTgHDBiQ008/Pddee21blwIAAAAArabcXGrrEjZrm8QW9fdMnDgxzc3rf3ppc3Nz2rffZLJaAAAAAOAD8oGmgtOnT1/tf79nl112ed8vm5fL5dXOv/rqq6mrq9vY8gAAAACAgil02+Orr76aH//4x1mwYEE+8YlPtHU5AAAAALCK8vpvWGY9FDrgPProo7N48eJce+212W+//dq6HAAAAADgA1bogPOJJ55o6xIAAAAAgDa0SX1kCAAAAABgfRS6gxMAAAAANnXlcqmtS9is6eAEAAAAAApLwAkAAAAAtIrrr78+u+yySzp16pQDDzwwjz766Puuv/rqq9O3b9907tw5vXv3zr/8y7/k7bffXq932qIOAAAAAFVUbm7rCj4YM2fOzNixY3PjjTfmwAMPzNVXX51BgwZlwYIF2XbbbVdZ//3vfz9f/OIXM3Xq1Bx88MF55plnMnLkyJRKpVx55ZXr/F4dnAAAAADARrvyyiszevTonHHGGdlzzz1z4403pkuXLpk6depq1//iF7/IIYccktNOOy277LJLjjrqqJx66qlr7fr8WwJOAAAAAGC1mpqasnTp0orR1NS0yrp33nknjz/+eAYOHNgyt8UWW2TgwIF55JFHVvvsgw8+OI8//nhLoPn888/nRz/6UY499tj1qlHACQAAAACsVmNjY7p3714xGhsbV1n32muvZeXKldluu+0q5rfbbrssWrRotc8+7bTTMnHixPzDP/xDOnTokN133z2HH354LrnkkvWq0RmcAAAAAFBF5eZSW5ewwRoaGjJ27NiKuZqamlZ59ty5czNp0qR8+9vfzoEHHpjnnnsuF1xwQb761a/mK1/5yjo/R8BJq+r8qbFrXwSbiB9td0pbl7DOjn3l1rYuAQAAgL9DNTU16xRo9urVK+3atcsrr7xSMf/KK6+krq5utfd85StfyWc+85mcddZZSZIBAwZk2bJl+cd//Md86UtfyhZbrNvmc1vUAQAAAICN0rFjx+y///6ZM2dOy1xzc3PmzJmTgw46aLX3LF++fJUQs127dkmScrm8zu/WwQkAAAAAbLSxY8dmxIgROeCAA/LRj340V199dZYtW5YzzjgjSfLZz342O+64Y8sZnoMHD86VV16Z/fbbr2WL+le+8pUMHjy4JehcFwJOAAAAAKii9WhGLLRhw4blz3/+c8aNG5dFixZl3333zb333tvy4aEXX3yxomPzy1/+ckqlUr785S/npZdeyjbbbJPBgwfnsssuW6/3lsrr0+8JsBlxBicAAAAfhBcP+ERbl7DBdn5sztoXtTFncAIAAAAAhWWLOgAAAABUUbm51NYlbNZ0cAIAAAAAhSXgBAAAAAAKS8AJAAAAABSWMzgBAAAAoIqcwVldOjgBAAAAgMJa74Bz5MiRKZVKmTx5csX8HXfckVLpL2n09OnT06NHj9XeXyqVcscddyRJXnjhhZRKpbRr1y4vvfRSxbqFCxemffv2KZVKeeGFFyrWvze23nrrHHbYYXn44YfX92cAAAAAAJuBDerg7NSpUy6//PIsXry4VYrYcccdc8stt1TMzZgxIzvuuONq1z/44INZuHBhfvazn2WHHXbIpz71qbzyyiutUgsAAAAAUBwbFHAOHDgwdXV1aWxsbJUiRowYkWnTplXMTZs2LSNGjFjt+p49e6auri577bVXLrnkkixdujS/+tWvWqUWAAAAAGhN5XJxRxFsUMDZrl27TJo0Kdddd13+9Kc/bXQRQ4YMyeLFi/Pzn/88SfLzn/88ixcvzuDBg9/3vrfeequl87Njx44bXQcAAAAAUCwb/JGhE044Ifvuu2/Gjx+/0UV06NAhw4cPz9SpU5MkU6dOzfDhw9OhQ4fVrj/44INTW1ubrl275pvf/Gb233//fOITn9joOgAAAACAYtmor6hffvnlmTFjRubPn7/RhYwaNSqzZs3KokWLMmvWrIwaNWqNa2fOnJnf/OY3uf3221NfX5/p06evMQx9T1NTU5YuXVoxmpqaNrpuAAAAAHg/5eZSYUcRbFTAeeihh2bQoEFpaGiomO/WrVuWLVuW5ubmivk33ngjSdK9e/dVnjVgwID069cvp556avbYY4/stddea3xv79698+EPfzgnnHBCJk2alBNOOGGtYWVjY2O6d+9eMVrrDFEAAAAAoG1sVMCZJJMnT85dd92VRx55pGWub9++WbFiRebNm1ex9oknnkiS9OnTZ7XPGjVqVObOnfu+3Zt/66STTkr79u3z7W9/+33XNTQ0ZMmSJRXjb4NZAAAAAKBYNjrgHDBgQE4//fRce+21LXP9+/fPUUcdlVGjRmXOnDn5wx/+kHvvvTfnnHNOhg0blh133HG1zxo9enT+/Oc/56yzzlrn95dKpXzuc5/L5MmTs3z58jWuq6mpSbdu3SpGTU3Nuv9QAAAAAGCTs9EBZ5JMnDhxle3oM2fOzGGHHZZ/+qd/Sv/+/fO5z30uxx9/fG6++eY1Pqd9+/bp1atX2rdvv17vHzFiRN59991861vf2qD6AQAAAKBayuVSYUcRlMrlcrmtiwBoCz/a7pS2LmGdHfvKrW1dAgAAABvo93sNausSNtjuT93X1iWsVat0cAIAAAAAtIX12wsOAAAAAKyXcvPa17DhdHACAAAAAIUl4AQAAAAACkvACQAAAAAUljM4AQAAAKCKmsulti5hs6aDEwAAAAAoLAEnAAAAAFBYAk4AAAAAoLCcwQkAAAAAVVR2BmdV6eAEAAAAAApLwAkAAAAAFJYt6gAAAABQReVmW9SrScAJ/N069pVb27oEAAAAYCPZog4AAAAAFJaAEwAAAAAoLFvUAQAAAKCKyuW2rmDzpoMTAAAAACgsAScAAAAAUFi2qAMAAABAFZWbS21dwmZNBycAAAAAUFgCTgAAAACgsAScAAAAAEBhOYMTAAAAAKqouewMzmrSwQkAAAAAFFarB5ylUul9x4QJE/LCCy+kVCpl3rx5SdLyd7t27fLSSy9VPG/hwoVp3759SqVSXnjhhYr1qxu//OUvW/snAQAAAACbqFbfor5w4cKW/545c2bGjRuXBQsWtMzV1tbmtddeW+29O+64Y2655ZY0NDS0zM2YMSM77rhjXnzxxVXWP/jgg+nfv3/FXM+ePTf2JwAAAAAABdHqHZx1dXUto3v37imVShVztbW1a7x3xIgRmTZtWsXctGnTMmLEiNWu79mzZ8Wz6+rq0qFDh1b9PQAAAACwMcrlUmFHEWxSZ3AOGTIkixcvzs9//vMkyc9//vMsXrw4gwcPbuPKAAAAAIBN0SYVcHbo0CHDhw/P1KlTkyRTp07N8OHD19iVefDBB6e2trZiAAAAAAB/P1r9DM6NNWrUqBx88MGZNGlSZs2alUceeSQrVqxY7dqZM2dmjz32WKfnNjU1pampqWKupqYmNTU1G10zAAAAAKxJudzWFWzeNqkOziQZMGBA+vXrl1NPPTV77LFH9tprrzWu7d27d+rr6yvGmjQ2NqZ79+4Vo7GxsRo/AQAAAAD4gGxyHZzJX7o4zznnnNxwww2t9syGhoaMHTu2Yk73JgAAAAAU2yYZcI4ePTonn3xyevTo8b7rXn/99SxatKhirkePHunUqdMqa21HBwAAAIDNzyYZcLZv3z69evVa67qBAweuMveDH/wgp5xySjXKAgAAAID11lwutXUJm7VSueyYUwAAAAColnkfGtLWJWywff94Z1uXsFab3EeGAAAAAADW1Sa5RR0AAAAANhdlW9SrSgcnAAAAAFBYAk4AAAAAoLAEnAAAAABAYTmDEwAAAACqqFxu6wo2bzo4AQAAAIDCEnACAAAAAIUl4AQAAAAACssZnAAAAABQRc3lUluXsFnTwQkAAAAAFJaAEwAAAAAoLFvUgb9bP6s7ua1LWGeHLprV1iUAAACwgcq2qFeVDk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCcgYnAAAAAFRRszM4q0oHJwAAAABQWAJOAAAAAKCwbFEHAAAAgCoqt3UBmzkdnAAAAABAYQk4AQAAAIDCEnACAAAAAIXlDE4AAAAAqKLmcqmtS9isVb2D85FHHkm7du1y3HHHVcy/8MILKZVKLWPLLbdM//79c+655+bZZ5+tWDt9+vT06NGj4u+/vve9cfPNN1f75wAAAAAAm5Cqd3BOmTIl559/fqZMmZKXX345O+ywQ8X1Bx98MP3798/y5cvz5JNP5pprrsk+++yTu+66K5/4xCfW+Nxu3bplwYIFFXPdu3evym8AAAAAADZNVQ04//d//zczZ87MY489lkWLFmX69Om55JJLKtb07NkzdXV1SZLddtstgwcPzic+8YmceeaZ+f3vf5927dqt9tmlUqnlPgAAAADg71NVt6jfdttt6devX/r27Zvhw4dn6tSpKZfL71/QFlvkggsuyB//+Mc8/vjj1SwPAAAAAKquXC4VdhRBVQPOKVOmZPjw4UmSo48+OkuWLMlPf/rTtd7Xr1+/JH85p3NNlixZktra2pahmxMAAAAA/v5UbYv6ggUL8uijj2b27Nl/eVH79hk2bFimTJmSww8//H3vfa/Ls1Rac0q85ZZb5oknnmj5e4st3j+rbWpqSlNTU8VcTU1Nampq3vc+AAAAAGDTVbWAc8qUKVmxYkXFR4XK5XJqamryrW99633vnT9/fpJk1113XeOaLbbYIvX19etcT2NjYy699NKKufHjx2fChAnr/AwAAAAAWF/NbV3AZq4qAeeKFStyyy235IorrshRRx1VcW3o0KH5wQ9+kKOPPnq19zY3N+faa6/Nrrvumv3226/VampoaMjYsWMr5nRvAgAAAECxVSXgvPvuu7N48eKceeaZ6d69e8W1E088MVOmTGkJOF9//fUsWrQoy5cvz1NPPZWrr746jz76aO655541fkF9Q9iODgAAAACbn6oEnFOmTMnAgQNXCTeTvwScX//617N06dIkycCBA5MkXbp0yYc+9KEcccQRuemmmyq2nzc3N6d9+6rtpgcAAAAACqoqqeFdd921xmsf/ehHWz4i9N7/X5tXX3214ivpI0eOzMiRIzeqRgAAAAD4IJSz5g9ps/E26bbI5cuX5+mnn860adNyzDHHtHU5AAAAAMAmZou2LuD93HTTTRk4cGD22WefjBs3rq3LAQAAAAA2MZt0B+eYMWMyZsyYti4DAAAAANhEbdIBJwAAAAAUXfO6fYaGDbRJb1EHAAAAAHg/Ak4AAAAAoLBsUQcAAACAKmpOqa1L2Kzp4AQAAAAACkvACQAAAAAUloATAAAAACgsZ3ACAAAAQBWVncFZVTo4AQAAAIDCEnACAAAAAIVlizqtqnPnD7V1Cevsrbf+2NYl0MYOXTSrrUsAAADg70BzWxewmdPBCQAAAAAUloATAAAAACgsAScAAAAAUFjO4AQAAACAKiqn1NYlbNZ0cAIAAAAAhSXgBAAAAAAKS8AJAAAAABSWMzgBAAAAoIqa27qAzZwOTgAAAACgsAScAAAAAEBh2aIOAAAAAFVki3p1tVoH56JFi3LBBRekvr4+nTp1ynbbbZdDDjkkN9xwQ5YvX54k2WWXXVIqlVIqldKlS5cMGDAgN998c8Vz5s6d27Lmb8eiRYuSJBMmTMi+++5bcd/DDz+cHj16ZMyYMSmXy631swAAAACATVirdHA+//zzOeSQQ9KjR49MmjQpAwYMSE1NTZ588sncdNNN2XHHHTNkyJAkycSJEzN69OgsX748s2bNyujRo7PjjjvmmGOOqXjmggUL0q1bt4q5bbfddrXvv+eee3LyySfni1/8YsaNG9caPwkAAAAAKIBWCTjPOeectG/fPo899li6du3aMr/bbrvl+OOPr+io3HLLLVNXV5ckufjii/P1r389DzzwwCoB57bbbpsePXqs9d3f//73c8YZZ+SKK67Ieeed1xo/BwAAAAAoiI0OOF9//fXcf//9mTRpUkW4+ddKpdIqc83NzZk9e3YWL16cjh07btC7r7/++owdOzZTp07N6aefvkHPAAAAAIBqKmfVbIzWs9FncD733HMpl8vp27dvxXyvXr1SW1ub2traXHzxxS3zF198cWpra1NTU5OTTjopW221Vc4666xVnrvTTju13F9bW5v+/ftXXJ8/f37OO++83HDDDcJNAAAAAPg7VbWvqD/66KNpbm7O6aefnqamppb5iy66KCNHjszChQtz0UUX5Zxzzkl9ff0q9z/88MPZcsstW/7u0KFDxfWddtopPXr0yDe+8Y0cc8wx2X777d+3nqampoo6kqSmpiY1NTUb8vMAAAAAgE3ARndw1tfXp1QqZcGCBRXzu+22W+rr69O5c+eK+V69eqW+vj4f//jHM2vWrHzuc5/L7373u1Weu+uuu6a+vr5lfOhDH6q4vuWWW+bBBx9M165dc8QRR2ThwoXvW2djY2O6d+9eMRobGzfwVwMAAADAumkuFXcUwUYHnD179swnP/nJfOtb38qyZcvW697evXtn2LBhaWho2KB3b7XVVnnwwQfTrVu3HH744Xn55ZfXuLahoSFLliypGBv6XgAAAABg07DRAWeSfPvb386KFStywAEHZObMmZk/f34WLFiQ7373u3n66afTrl27Nd57wQUX5K677spjjz1WMf/qq69m0aJFFePdd99d5f4ePXrkgQceyFZbbfW+IWdNTU26detWMWxPBwAAAIBia5WAc/fdd89vfvObDBw4MA0NDdlnn31ywAEH5LrrrsvnP//5fPWrX13jvXvuuWeOOuqojBs3rmK+b9++2X777SvG448/vtpndO/ePffff3969eqVww47LC+99FJr/CwAAAAAYBNXKpfL5bYugs1H584fWvuiTcRbb/2xrUsAAAAA/g78sO60ti5hgx2/6PttXcJatUoHJwAAAABAWxBwAgAAAACFJeAEAAAAAAqrfVsXAAAAAACbMx/AqS4dnAAAAABAYQk4AQAAAIDCskUdAAAAAKqoua0L2Mzp4AQAAAAACkvACQAAAAAUloATAAAAACgsZ3ACAAAAQBU1l0ptXcJmTQcnAAAAAFBYOjhpVW+99ce2LgEAAACAvyMCTgAAAACoonJbF7CZs0UdAAAAACgsAScAAAAAUFgCTgAAAACgsJzBCQAAAABV1NzWBWzmdHACAAAAAIUl4AQAAAAACkvACQAAAAAUljM4AQAAAKCKmkttXcHmTQcnAAAAAFBYAk4AAAAAoLDWK+AcOXJkSqVSSqVSOnTokF133TVf+MIX8vbbb1esu/vuu3PYYYdlyy23TJcuXfKRj3wk06dPr1jzwgsvpFQqpV27dnnppZcqri1cuDDt27dPqVTKCy+8ULF+2223zZtvvlmxft99982ECRPW56cAAAAAwAeiOaXCjiJY7w7Oo48+OgsXLszzzz+fq666Kt/5zncyfvz4luvXXXddjj/++BxyyCH51a9+ld/+9rc55ZRT8s///M/5/Oc/v8rzdtxxx9xyyy0VczNmzMiOO+642ve/+eab+eY3v7m+ZQMAAAAAm6H1DjhrampSV1eX3r17Z+jQoRk4cGAeeOCBJMl///d/58ILL8yYMWMyadKk7Lnnnqmvr8+FF16Yb3zjG7niiivyq1/9quJ5I0aMyLRp0yrmpk2blhEjRqz2/eeff36uvPLKvPrqq+tbOgAAAACwmdmoMzifeuqp/OIXv0jHjh2TJP/2b/+Wd999d7Wdmv/0T/+U2tra/OAHP6iYHzJkSBYvXpyf//znSZKf//znWbx4cQYPHrzad5566qmpr6/PxIkTN6Z0AAAAAGAzsN4B5913353a2tp06tQpAwYMyKuvvpqLLrooSfLMM8+ke/fu2X777Ve5r2PHjtltt93yzDPPVMx36NAhw4cPz9SpU5MkU6dOzfDhw9OhQ4fVvr9UKmXy5Mm56aab8vvf/359ywcAAACAD1S5wKMI1jvgPOKIIzJv3rz86le/yogRI3LGGWfkxBNP3KgiRo0alVmzZmXRokWZNWtWRo0a9b7rBw0alH/4h3/IV77ylXV+R1NTU5YuXVoxmpqaNqpuAAAAAKBtrXfA2bVr19TX12efffbJ1KlT86tf/SpTpkxJkvTp0ydLlizJyy+/vMp977zzTn7/+9+nT58+q1wbMGBA+vXrl1NPPTV77LFH9tprr7XWMXny5MycOTO/+c1v1qnuxsbGdO/evWI0Njau070AAAAAwKZpo87g3GKLLXLJJZfky1/+ct56662ceOKJ6dChQ6644opV1t54441ZtmxZTj311NU+a9SoUZk7d+5auzff89GPfjT/9//+33zxi19cp/UNDQ1ZsmRJxWhoaFinewEAAACATVP7jX3AySefnIsuuijXX399Pv/5z+frX/96LrzwwnTq1Cmf+cxn0qFDh/zwhz/MJZdckgsvvDAHHnjgap8zevTonHzyyenRo8c6v/uyyy5L//7907792n9GTU1Nampq1vnZAAAAANAamkttXcHmbaM6OJOkffv2Oe+88/L1r389y5Yty5gxYzJ79uw8/PDDOeCAA7LXXnvl+9//fm644YZ885vffN/n9OrVa53Cyvf06dMno0aNyttvv72xPwMAAAAAKKBSuVwuygeRAAAAAKBwbtlxeFuXsME++9J327qEtdroLeoAAAAAwJo1t3UBm7mN3qIOAAAAANBWBJwAAAAAQGEJOAEAAACAwnIGJwAAAABUkS98V5cOTgAAAACgsAScAAAAAEBh2aIOAAAAAFXUXGrrCjZvOjgBAAAAgMIScAIAAAAAhSXgBAAAAAAKyxmcAAAAAFBFzW1dwGZOBycAAAAAUFg6OAEKoH3HHdu6hHW24p2X2roEAAAA/o7o4AQAAAAACksHJwAAAABUkTM4q0sHJwAAAABQWAJOAAAAAKCwbFEHAAAAgCoql9q6gs2bDk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCcgYnAAAAAFRRc1sXsJnTwQkAAAAAtIrrr78+u+yySzp16pQDDzwwjz766BrXHn744SmVSquM4447br3eKeAEAAAAADbazJkzM3bs2IwfPz5PPPFE9tlnnwwaNCivvvrqatf/+7//exYuXNgynnrqqbRr1y4nn3zyer231QPOkSNHtqStHTt2TH19fSZOnJgVK1Zk7ty5FWnsNttsk2OPPTZPPvlkxTPeeeedfP3rX88+++yTLl26pFevXjnkkEMybdq0vPvuuy3vGTp0aGuXDwAAAACtqrnAY31ceeWVGT16dM4444zsueeeufHGG9OlS5dMnTp1teu33nrr1NXVtYwHHnggXbp0We+AsypncB599NGZNm1ampqa8qMf/SjnnntuOnTokIMOOihJsmDBgnTr1i0vv/xyLrroohx33HF57rnn0rFjx7zzzjsZNGhQ/vM//zNf/epXc8ghh6Rbt2755S9/mW9+85vZb7/9su+++1ajbAAAAABgA7zzzjt5/PHH09DQ0DK3xRZbZODAgXnkkUfW6RlTpkzJKaeckq5du67Xu6sScNbU1KSuri5JcvbZZ2f27Nm58847WwLObbfdNj169EhdXV3GjBmTIUOG5Omnn87ee++dq6++Oj/72c/y2GOPZb/99mt55m677ZaTTz4577zzTjVKBgAAAAD+RlNTU5qamirmampqUlNTUzH32muvZeXKldluu+0q5rfbbrs8/fTTa33Po48+mqeeeipTpkxZ7xo/kDM4O3fuvNpgcsmSJbn11luTJB07dkySfO9738vAgQMrws33dOjQYb0TXAAAAABgwzQ2NqZ79+4Vo7GxsdXfM2XKlAwYMCAf/ehH1/veqnRwvqdcLmfOnDm57777cv7557fM77TTTkmSZcuWJUmGDBmSfv36JUmeffbZHH744dUsCwAAAAA+MOW2LmAjNDQ0ZOzYsRVzf9u9mSS9evVKu3bt8sorr1TMv/LKKy07vddk2bJlufXWWzNx4sQNqrEqHZx33313amtr06lTpxxzzDEZNmxYJkyY0HL94YcfzuOPP57p06enT58+ufHGG1uulcvV+SdvamrK0qVLK8bfttcCAAAAAP+/mpqadOvWrWKsLuDs2LFj9t9//8yZM6dlrrm5OXPmzGk5tnJNZs2alaampgwfPnyDaqxKwHnEEUdk3rx5efbZZ/PWW29lxowZFVvLd9111/Tt2zcjRozIWWedlWHDhrVc69Onzzrty19fH1Q7LQAAAAD8PRo7dmz+9V//NTNmzMj8+fNz9tlnZ9myZTnjjDOSJJ/97GcrPkL0nilTpmTo0KHp2bPnBr23KgFn165dU19fn5133jnt27//Lvhzzz03Tz31VGbPnp0kOe200/Lggw/mN7/5zSpr33333ZZt7euroaEhS5YsqRir+x8UAAAAAFh/w4YNyze/+c2MGzcu++67b+bNm5d777235cNDL774YhYuXFhxz4IFC/Lzn/88Z5555ga/t6pncK6LLl26ZPTo0Rk/fnyGDh2aMWPG5J577sknPvGJfPWrX80//MM/ZMstt8xjjz2Wyy+/PFOmTMm+++6b5C8fKZo3b17F83r27JnevXuv8p7Vfd0JAAAAAKqtudTWFXxwzjvvvJx33nmrvTZ37txV5vr27bvRR1Z+IF9RX5vzzjsv8+fPz6xZs1JTU5MHHnggX/jCF/Kd73wnH/vYx/KRj3wk1157bT73uc9lr732arlv7ty52W+//SrGpZde2oa/BAAAAAD4IJXK1fqqDwCtpn3HHdu6hHW24p2X2roEAACATco1O2/Yx3M2BRe8+N22LmGt2nyLOgAAAABszprbuoDN3CaxRR0AAAAAYEMIOAEAAACAwhJwAgAAAACF5QxOAAAAAKgiZ3BWlw5OAAAAAKCwBJwAAAAAQGHZog4AAAAAVVRu6wI2czo4AQAAAIDCEnACAAAAAIUl4AQAAAAACssZnAAAAABQRc2ltq5g8ybgBCiAFe+81NYlAAAAwCbJFnUAAAAAoLAEnAAAAABAYdmiDgAAAABV1NzWBWzmdHACAAAAAIUl4AQAAAAACssWdQAAAACoonJbF7CZ08EJAAAAABSWgBMAAAAAKCwBJwAAAABQWM7gBAAAAIAqanYKZ1Xp4AQAAAAACqtVAs6RI0emVCqlVCqlY8eOqa+vz8SJE7NixYrMnTu35VqpVMo222yTY489Nk8++eQan/HX4+ijj25Zs8suu7TMd+7cObvssks+/elP56GHHmqNnwEAAAAAFEyrdXAeffTRWbhwYZ599tlceOGFmTBhQr7xjW+0XF+wYEEWLlyY++67L01NTTnuuOPyzjvvrPYZfz1+8IMfVKyZOHFiFi5cmAULFuSWW25Jjx49MnDgwFx22WWt9VMAAAAAoNU0F3gUQaudwVlTU5O6urokydlnn53Zs2fnzjvvzEEHHZQk2XbbbdOjR4/U1dVlzJgxGTJkSJ5++unsvffeq33Gmmy55ZYta3beeecceuih2X777TNu3LicdNJJ6du3b2v9JAAAAABgE1e1Mzg7d+68SodmkixZsiS33nprkqRjx46t8q4LLrgg5XI5P/zhD1vleQAAAABAMbR6wFkul/Pggw/mvvvuy5FHHtkyv9NOO6W2tjY9evTI97///QwZMiT9+vWruPfuu+9ObW1txZg0adJa37n11ltn2223zQsvvNDaPwcAAAAA2IS12hb198LJd999N83NzTnttNMyYcKE/PrXv06SPPzww+nSpUt++ctfZtKkSbnxxhtXecYRRxyRG264oWJu6623Xqf3l8vllEqlNV5vampKU1NTxVxNTU1qamrW6fkAAAAAsCHKbV3AZq7VAs73wsmOHTtmhx12SPv2lY/edddd06NHj/Tt2zevvvpqhg0blp/97GcVa7p27Zr6+vr1fvfrr7+eP//5z9l1113XuKaxsTGXXnppxdz48eMzYcKE9X4fAAAAALBpaLUt6u+FkzvvvPMq4ebfOvfcc/PUU09l9uzZrfLua665JltssUWGDh26xjUNDQ1ZsmRJxWhoaGiV9wMAAAAAbaPVOjjXR5cuXTJ69OiMHz8+Q4cObdla3tTUlEWLFlUW2L59evXq1fL3m2++mUWLFuXdd9/NH/7wh3z3u9/NzTffnMbGxvft/rQdHQAAAAA2P1X7ivranHfeeZk/f35mzZrVMnfvvfdm++23rxj/8A//UHHfuHHjsv3226e+vj6f+cxnsmTJksyZMycXX3zxB/0TAAAAAGCtmgs8iqBULpedcwoAAAAAVTLhQ6e3dQkbbMIfv9fWJaxVm3VwAgAAAABsrDY5gxMAAAAA/l40l9q6gs2bDk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCcgYnAAAAAFRRc8ptXcJmTQcnAAAAAFBYAk4AAAAAoLAEnAAAAABAYTmDEwAAAACqyAmc1aWDEwAAAAAoLAEnAAAAAFBYtqgDAAAAQBU1t3UBmzkdnAAAAABAYQk4AQAAAIDCEnACAAAAAIXlDE4AAAAAqKLmlNu6hM2aDk4AAAAAoLAEnAAAAABAYdmiDgAAAABVZIN6dengBAAAAAAKS8AJAAAAABSWgBMAAAAAKCxncAIAAABAFTW3dQGbOR2cAAAAAEBhVSXgHDlyZEqlUkqlUjp27Jj6+vpMnDgxK1asyNy5c1uulUqlbLPNNjn22GPz5JNPJknFtdWNCRMm5IUXXljtteHDh1fj5wAAAAAAm6iqbVE/+uijM23atDQ1NeVHP/pRzj333HTo0CEHHXRQkmTBggXp1q1bXn755Vx00UU57rjj8txzz2XhwoUtz5g5c2bGjRuXBQsWtMzV1tbmtddeS5I8+OCD6d+/f8u1zp07V+vnAAAAAACboKoFnDU1Namrq0uSnH322Zk9e3buvPPOloBz2223TY8ePVJXV5cxY8ZkyJAhefrpp7P33nu3PKN79+4plUotz3nPewFnz549V7kGAAAAAJuS5pTbuoTN2gd2Bmfnzp3zzjvvrDK/ZMmS3HrrrUmSjh07flDlAAAAAACbgap/Rb1cLmfOnDm57777cv7557fM77TTTkmSZcuWJUmGDBmSfv36rdezDz744Gyxxf+f0T788MPZb7/9WqFqAAAAAKAIqhZw3n333amtrc27776b5ubmnHbaaZkwYUJ+/etfJ/lLGNmlS5f88pe/zKRJk3LjjTeu9ztmzpyZPfbYo+Xv3r17r3FtU1NTmpqaKuZqampSU1Oz3u8FAAAAgHVlg3p1VS3gPOKII3LDDTekY8eO2WGHHdK+feWrdt111/To0SN9+/bNq6++mmHDhuVnP/vZer2jd+/eqa+vX6e1jY2NufTSSyvmxo8fnwkTJqzXOwEAAACATUfVzuDs2rVr6uvrs/POO68Sbv6tc889N0899VRmz55drXLS0NCQJUuWVIyGhoaqvQ8AAAAAqL4P7CND76dLly4ZPXp0xo8fn3K5Ok27NTU16datW8WwPR0AAAAAim2TCDiT5Lzzzsv8+fMza9asti4FAAAAAFpNc4FHEZTK1WqZBAAAAABywS6ntHUJG+yaF25t6xLWapPp4AQAAAAAWF9V+4o6AAAAAJCUYwN1NengBAAAAAAKS8AJAAAAABSWgBMAAAAAKCxncAIAAABAFTW3dQGbOR2cAAAAAEBhCTgBAAAAgMIScAIAAAAAheUMTgAAAACoouaU27qEzZoOTgAAAACgsAScAAAAAEBh2aIOAAAAAFVkg3p16eAEAAAAAApLwAkAAAAAFJaAEwAAAAAoLGdwAgAAAEAVNTuFs6p0cAIAAAAAhSXgBAAAAAAKyxZ1AAAAAKii5rYuYDOngxMAAAAAKCwBJwAAAABQWAJOAAAAAKCwnMEJAAAAAFVUTrmtS9is6eAEAAAAAAprowLOkSNHplQqZfLkyRXzd9xxR0qlUpJk7ty5KZVKLWO77bbLiSeemOeff75l/S677NJyvUuXLhkwYEBuvvnmVd63cuXKXHXVVRkwYEA6deqUrbbaKsccc0z+4z/+Y2N+BgAAAABQUBvdwdmpU6dcfvnlWbx48fuuW7BgQV5++eXMmjUr//Vf/5XBgwdn5cqVLdcnTpyYhQsX5qmnnsrw4cMzevTo/PjHP265Xi6Xc8opp2TixIm54IILMn/+/MydOze9e/fO4YcfnjvuuGNjfwoAAAAAUDAbHXAOHDgwdXV1aWxsfN912267bbbffvsceuihGTduXH73u9/lueeea7m+5ZZbpq6uLrvttlsuvvjibL311nnggQdart922235t3/7t9xyyy0566yzsuuuu2afffbJTTfdlCFDhuSss87KsmXLNvbnAAAAAECrai7wKIKNDjjbtWuXSZMm5brrrsuf/vSndbqnc+fOSZJ33nlnlWvNzc25/fbbs3jx4nTs2LFl/vvf/3769OmTwYMHr3LPhRdemNdff70iEAUAAAAANn+t8pGhE044Ifvuu2/Gjx+/1rULFy7MN7/5zey4447p27dvy/zFF1+c2tra1NTU5KSTTspWW22Vs846q+X6M888kz322GO1z3xv/plnnlnje5uamrJ06dKK0dTUtK4/EQAAAADYBLXaV9Qvv/zyzJgxI/Pnz1/t9Z122ildu3bNDjvskGXLluX222+v6NC86KKLMm/evDz00EM58MADc9VVV6W+vr7iGeVyeYPra2xsTPfu3SvG2rbVAwAAAMDGKhf4/4qgfWs96NBDD82gQYPS0NCQkSNHrnL94YcfTrdu3bLttttmyy23XOV6r169Ul9fn/r6+syaNSsDBgzIAQcckD333DNJ0qdPnzWGp+/N9+nTZ431NTQ0ZOzYsRVzNTU16/rzAAAAAIBNUKt1cCbJ5MmTc9ddd+WRRx5Z5dquu+6a3XfffbXh5t/q3bt3hg0bloaGhpa5U045Jc8++2zuuuuuVdZfccUV6dmzZz75yU+u8Zk1NTXp1q1bxRBwAgAAAECxtWrAOWDAgJx++um59tprN/pZF1xwQe6666489thjSf4ScJ5wwgkZMWJEpkyZkhdeeCG//e1v80//9E+58847c/PNN6dr164b/V4AAAAAoDhaNeBMkokTJ6a5eeM/Ir/nnnvmqKOOyrhx45IkpVIpt912Wy655JJcddVV6du3bz7+8Y/nj3/8Y+bOnZuhQ4du9DsBAAAAoLU1F3gUQam8MV/uAQAAAADe14hdTmzrEjbYjBdub+sS1qrVOzgBAAAAAD4orfYVdQAAAABgVc02UFeVDk4AAAAAoLAEnAAAAABAYQk4AQAAAIDCcgYnAAAAAFSREzirSwcnAAAAAFBYAk4AAAAAoLAEnAAAAABAYTmDEwAAAACqqNkpnFWlgxMAAAAAKCwBJwAAAABQWLao06refe35ti5hnXXotVtblwAAAAD8HSjbol5VOjgBAAAAgMIScAIAAAAAhSXgBAAAAAAKyxmcAAAAAFBFzW1dwGZOBycAAAAAUFgCTgAAAACgsAScAAAAAEBhOYMTAAAAAKqoOeW2LmGzpoMTAAAAACgsAScAAAAAUFi2qAMAAABAFZVtUa+qVu/gHDlyZIYOHbrK/Ny5c1MqlfLGG29UzPfr1y81NTVZtGjRKvccfvjhKZVKufXWWyvmr7766uyyyy4tf0+fPj09evRoheoBAAAAgCJp0y3qP//5z/PWW2/lpJNOyowZM1a7plOnTvnyl7+cd9999wOuDgAAAADY1LVpwDllypScdtpp+cxnPpOpU6euds2pp56aN954I//6r//6AVcHAAAAAGzq2izgfPPNNzNr1qwMHz48n/zkJ7NkyZI8/PDDq6zr1q1bvvSlL2XixIlZtmxZG1QKAAAAABuuucCjCKoScN59992pra2tGMccc0zFmltvvTUf/vCH079//7Rr1y6nnHJKpkyZstrnnXPOOenUqVOuvPLKapQLAAAAABRUVQLOI444IvPmzasYN998c8WaqVOnZvjw4S1/Dx8+PLNmzcqbb765yvNqamoyceLEfPOb38xrr722QTU1NTVl6dKlFaOpqWmDngUAAAAAbBqqEnB27do19fX1FWPHHXdsuf673/0uv/zlL/OFL3wh7du3T/v27fOxj30sy5cvX+WL6e8ZPnx4PvShD+VrX/vaBtXU2NiY7t27V4zGxsYNehYAAAAArKtyuVzYUQTt2+KlU6ZMyaGHHprrr7++Yn7atGmZMmVKRo8evco9W2yxRRobG/N//+//zdlnn73e72xoaMjYsWMr5mpqatb7OQAAAADApuMDDzjffffd/L//9/8yceLE7LXXXhXXzjrrrFx55ZX5r//6r/Tv33+Ve4877rgceOCB+c53vpPttttuvd5bU1Mj0AQAAACAzcwH/hX1n/3sZ3n99ddzwgknrHJtjz32yB577LHGjw0lyeWXX5633367miUCAAAAAAVRKhdlMz2F8O5rz7d1CeusQ6/d2roEAAAA4O/A8Tt/qq1L2GA/fPHuti5hrT7wDk4AAAAAgNYi4AQAAAAACkvACQAAAAAU1gf+FXUAAAAA+HvS3NYFbOZ0cAIAAAAAhSXgBAAAAAAKyxZ1AAAAAKiicsptXcJmTQcnAAAAAFBYAk4AAAAAoLAEnAAAAABAYTmDEwAAAACqqNkZnFWlgxMAAAAAKCwBJwAAAABQWLao06o69NqtrUsAAAAA2KSUy7aoV5MOTgAAAACgsAScAAAAAEBhCTgBAPj/2rvz8CiK/I/jn8kNBBIEIRyRwxiCAUGQBUHlNuFSkSMgEWIQXRYwCKJEwUAEAgKisCwemwPWA4gsiKCIICwouMgpuCHigWcQxRgUJJCkf3/wm9kMM0lmJjOEcd8vnn4e0l3d/Z0+qmpqqqsBAAAAr8UYnAAAAAAAAIAHlVR1AH9w9OAEAAAAAAAA4LVo4AQAAAAAAADgtWjgBAAAAAAAAOC1GIMTAAAAAAAA8CBDRlWH8IdGD04AAAAAAAAAXosGTgAAAAAAAABei0fUAQAAAAAAAA8q4RF1j/JYD86EhATdddddlv+bTCbNnTvXKs26detkMpksf2/fvl0mk8lmmjZtmtXy6OhoFRcXW20rNDRUWVlZnvo4AAAAAAAAAK5Al+0R9aCgIM2bN0/5+fkVps3NzVVeXp5lmjp1qtXyL774QitWrPBUqAAAAAAAAAC8xGVr4OzVq5fCwsKUlpZWYdp69eopLCzMMgUHB1stnzBhglJSUlRYWOipcAEAAAAAAAB4gcvWwOnr66s5c+ZoyZIl+vbbbyu1rYkTJ6qoqEhLlixxU3QAAAAAAACAZxiG4bWTN7isb1EfOHCg2rZtq5SUlHLTNW7cWMHBwZbp1KlTVsurV6+ulJQUpaWlqaCgwKF9FxYW6vTp01YTPUABAAAAAAAA73ZZGzglad68eVq+fLlycnLKTLNz504dPHjQMtWuXdsmzejRo1WnTh3NmzfPof2mpaUpJCTEanLkcXkAAAAAAAAAVy6/y73D2267TTExMUpOTlZCQoLdNM2aNVNoaGi52/Hz89Ps2bOVkJCg8ePHV7jf5ORkTZo0yWpeYGCgo2EDAAAAAAAALimRdzzq7a0uewOnJM2dO1dt27ZVixYtKrWdIUOGaP78+Zo5c2aFaQMDA2nQBAAAAAAAAP5gqqSBs3Xr1hoxYoQWL15c6W3NnTtXMTExbogKAAAAAAAAgLe57GNwmqWmpqqkpKTS2+nRo4d69OihoqIiN0QFAAAAAAAAwJuYDG953zsAAAAAAADghbo17lXVIbhs+7dbqjqEClVZD04AAAAAAAAAqCwaOAEAAAAAAAB4LRo4AQAAAAAAAHitKnmLOgAAAAAAAPC/ooRX4HgUPTgBAAAAAAAAuMXSpUvVtGlTBQUFqWPHjtqzZ0+56X/55ReNGzdODRo0UGBgoCIjI/XWW285tU96cAIAAAAAAACotFWrVmnSpEl6/vnn1bFjRz377LOKiYlRbm6u6tWrZ5P+/Pnz6t27t+rVq6fXX39djRo10ldffaXQ0FCn9msyDPrIAgAAAAAAAJ5ya6OeVR2Cy3Z+t9XhtB07dlSHDh3017/+VZJUUlKi8PBwTZgwQVOnTrVJ//zzz2v+/Pk6evSo/P39XY6RR9QBAAAAAAAA2FVYWKjTp09bTYWFhTbpzp8/r3379qlXr16WeT4+PurVq5d2795td9vr16/XzTffrHHjxql+/fpq1aqV5syZo+LiYqdipIETAAAAAAAAgF1paWkKCQmxmtLS0mzS/fTTTyouLlb9+vWt5tevX18nTpywu+0vvvhCr7/+uoqLi/XWW29p+vTpWrhwoWbNmuVUjIzBCQAAAAAAAMCu5ORkTZo0yWpeYGCgW7ZdUlKievXq6cUXX5Svr6/at2+v7777TvPnz1dKSorD26GBEwAAAAAAAPCgEnnvK3ACAwMdatCsW7eufH199cMPP1jN/+GHHxQWFmZ3nQYNGsjf31++vr6WeS1bttSJEyd0/vx5BQQEOBQjj6gDAAAAAAAAqJSAgAC1b99eW7f+96VEJSUl2rp1q26++Wa763Tp0kWfffaZSkpKLPM+/fRTNWjQwOHGTYkGTgAAAAAAAABuMGnSJL300ktavny5cnJyNHbsWJ05c0b33XefJGnkyJFKTk62pB87dqx+/vlnJSUl6dNPP9XGjRs1Z84cjRs3zqn98og6AAAAAAAAgEqLi4vTjz/+qCeffFInTpxQ27ZttWnTJsuLh77++mv5+Py3v2V4eLjeeecdPfzww7rhhhvUqFEjJSUl6bHHHnNqvybDMLx3EAAAAAAAAADgCndzo+5VHYLLdn+3rapDqBCPqAMAAAAAAADwWjRwAgAAAAAAAPBajMEJAAAAAAAAeBAjRHoWPTgBAAAAAAAAeC0aOAEAAAAAAAB4LRo4AQAAAAAAAHgtxuAEAAAAAAAAPKhEjMHpSfTgBAAAAAAAAOC1aOAEAAAAAAAA4LWcbuBMSEiQyWTS3LlzreavW7dOJpNJkpSVlaXQ0FC765tMJq1bt06SdPz4cZlMJvn6+uq7776zSpeXlyc/Pz+ZTCYdP37cKr15CggIUEREhGbNmiXDoKsvAAAAAAAArjyGF//zBi714AwKCtK8efOUn5/vliAaNWqkFStWWM1bvny5GjVqZDf9li1blJeXp2PHjmnmzJmaPXu2MjIy3BILAAAAAAAAAO/hUgNnr169FBYWprS0NLcEMWrUKGVmZlrNy8zM1KhRo+ymr1OnjsLCwtSkSRONGDFCXbp00f79+90SCwAAAAAAAADv4VIDp6+vr+bMmaMlS5bo22+/rXQQd9xxh/Lz8/X+++9Lkt5//33l5+drwIABFa67d+9e7du3Tx07dqx0HAAAAAAAAAC8i8svGRo4cKDatm2rlJSUSgfh7++v+Ph4y2PmGRkZio+Pl7+/v930nTt3VnBwsAICAtShQwcNHTpUI0eOrHQcAAAAAAAAgLsZhuG1kzfwq8zK8+bNU48ePfTII49UOpDExER17txZc+bMUXZ2tnbv3q2ioiK7aVetWqWWLVvqwoULOnLkiCZMmKDatWvbvPiotMLCQhUWFlrNCwwMVGBgYKVjBwAAAAAAAFA1XO7BKUm33XabYmJilJycbDW/Vq1aOnPmjEpKSqzm//LLL5KkkJAQm221bt1aUVFRGj58uFq2bKlWrVqVud/w8HBFRESoZcuWGjJkiCZOnKiFCxfq3LlzZa6TlpamkJAQq8ldY4gCAAAAAAAAqBqVauCUpLlz5+rNN9/U7t27LfNatGihoqIiHTx40Cqt+UVAkZGRdreVmJio7du3KzEx0akYfH19VVRUpPPnz5eZJjk5WQUFBVbTpQ2zAAAAAAAAALxLpR5Rly72vBwxYoQWL15smRcdHa3bb79diYmJWrhwoZo3b67c3FxNnDhRcXFxatSokd1tjRkzRkOGDFFoaGi5+zx16pROnDihoqIiHT58WM8995y6d++uWrVqlbkOj6MDAAAAAACgKpTIO8ay9FaVbuCUpNTUVK1atcpq3qpVq5SSkqIHH3xQ33//vRo3bqyBAwdq+vTpZQfj56e6detWuL9evXpJuthzs0GDBurbt69mz55duQ8BAAAAAAAAwOuYDG95HRIAAAAAAADghdo1uKWqQ3DZ/rz3qzqECrmlBycAAAAAAAAA++hf6FmVfskQAAAAAAAAAFQVGjgBAAAAAAAAeC0aOAEAAAAAAAB4LcbgBAAAAAAAADyoRIzB6Un04AQAAAAAAADgtWjgBAAAAAAAAOC1eEQdAAAAAAAA8CCDR9Q9ih6cAAAAAAAAALwWDZwAAAAAAAAAvBYNnAAAAAAAAAC8FmNwAgAAAAAAAB5UYjAGpyfRgxMAAAAAAACA16KBEwAAAAAAAIDXooETAAAAAAAAgNdiDE4AAAAAAADAgwwxBqcn0YMTAAAAAAAAgNeigRMAAAAAAACA1+IRdQAAAAAAAMCDSgweUfckenACAAAAAAAA8Fo0cAIAAAAAAADwWjRwAgAAAAAAAPBajMEJAAAAAAAAeJAhxuD0JHpwAgAAAAAAAPBaNHACAAAAAAAA8Fpua+BMSEiQyWSSyWSSv7+/6tevr969eysjI0MlJSU26WNiYuTr66uPPvpIknT8+HHL+mVNWVlZ2r59e5nLT5w44a6PAwAAAAAAALhFiWF47eQN3NqDMzY2Vnl5eTp+/Ljefvttde/eXUlJSerfv7+Kioos6b7++mvt2rVL48ePV0ZGhiQpPDxceXl5lmny5MmKjo62mhcXF2fZRm5urtWyvLw81atXz50fBwAAAAAAAMAVzq0vGQoMDFRYWJgkqVGjRmrXrp06deqknj17KisrS/fff78kKTMzU/3799fYsWPVqVMnPfPMM6pWrZplXUkKDg6Wn5+f1bzS6tWrp9DQUHeGDwAAAAAAAMDLeHwMzh49eqhNmzb65z//KUkyDEOZmZmKj49XVFSUIiIi9Prrr3s6DAAAAAAAAAB/QJflJUNRUVE6fvy4JGnLli06e/asYmJiJEnx8fFKT093epuNGzdWcHCwZYqOji43fWFhoU6fPm01FRYWOr1fAAAAAAAAwBmGF//zBpelgdMwDJlMJklSRkaG4uLi5Od38en44cOH64MPPtDnn3/u1DZ37typgwcPWqa33nqr3PRpaWkKCQmxmtLS0lz7QAAAAAAAAACuCG4dg7MsOTk5atasmX7++WetXbtWFy5c0LJlyyzLi4uLlZGRodmzZzu8zWbNmjk1BmdycrImTZpkNS8wMNDh9QEAAAAAAABceTzewPnee+/p8OHDevjhh/XKK6+ocePGWrdunVWazZs3a+HChUpNTZWvr69H4ggMDKRBEwAAAAAAAPiDcWsDZ2FhoU6cOKHi4mL98MMP2rRpk9LS0tS/f3+NHDlS7du31+DBg9WqVSur9cLDw5WcnKxNmzapX79+Du3r5MmTOnfunNW8OnXqyN/f322fBwAAAAAAAKisEsM7xrL0Vm5t4Ny0aZMaNGggPz8/1a5dW23atNHixYs1atQoHThwQIcOHdJLL71ks15ISIh69uyp9PR0hxs4W7RoYTNv9+7d6tSpU6U/BwAAAAAAAADvYDIMmpABAAAAAAAAT7m2bruqDsFln/+0v6pDqNBleckQAAAAAAAA8L/KEP0LPcmnqgMAAAAAAAAAAFfRwAkAAAAAAADAa9HACQAAAAAAAMBrMQYnAAAAAAAA4EGGUVLVIfyh0YMTAAAAAAAAgNeigRMAAAAAAACA1+IRdQAAAAAAAMCDSmRUdQh/aPTgBAAAAAAAAOC1aOAEAAAAAAAA4LVo4AQAAAAAAADgtRiDEwAAAAAAAPAgw2AMTk+iBycAAAAAAAAAr0UDJwAAAAAAAACvRQMnAAAAAAAAAK/FGJwAAAAAAACAB5WIMTg9iR6cAAAAAAAAALwWDZwAAAAAAAAAvBaPqAMAAAAAAAAeZBg8ou5J9OAEAAAAAAAA4LVo4AQAAAAAAADgtWjgBAAAAAAAAOC1GIMTAAAAAAAA8KASxuD0KHpwAgAAAAAAAPBabmng3L17t3x9fdWvXz+r+cePH5fJZLJMNWvWVHR0tMaNG6djx45Z0g0YMECxsbF2t71z506ZTCZ9/PHHNtsLCAhQRESEZs2axduoAAAAAAAAgP9BbmngTE9P14QJE7Rjxw59//33Nsu3bNmivLw8HTp0SHPmzFFOTo7atGmjrVu3SpJGjx6td999V99++63NupmZmbrpppt0ww032Gzv2LFjmjlzpmbPnq2MjAx3fBQAAAAAAAAAXqTSDZy//fabVq1apbFjx6pfv37KysqySVOnTh2FhYWpefPmuvPOO7VlyxZ17NhRo0ePVnFxsfr376+rr77aZt3ffvtN2dnZGj16tN3tNWnSRCNGjFCXLl20f//+yn4UAAAAAAAAwO0ML/7nDSrdwLl69WpFRUWpRYsWio+PV0ZGRoWPi/v4+CgpKUlfffWV9u3bJz8/P40cOVJZWVlW62ZnZ6u4uFjDhw8vc1t79+7Vvn371LFjx8p+FAAAAAAAAABeptINnOnp6YqPj5ckxcbGqqCgQP/6178qXC8qKkrSxXE6JSkxMVGff/651bqZmZkaNGiQQkJCrNbt3LmzgoODFRAQoA4dOmjo0KEaOXJkZT8KAAAAAAAAAC9TqQbO3Nxc7dmzx9LD0s/PT3FxcUpPT69wXXNPTZPJJOlig2fnzp0tY2l+9tln2rlzp83j6ZK0atUqHTx4UIcOHdLq1av1xhtvaOrUqeXur7CwUKdPn7aaCgsLnfq8AAAAAAAAgLMMw/DayRtUqoEzPT1dRUVFatiwofz8/OTn56dly5ZpzZo1KigoKHfdnJwcSVKzZs0s80aPHq01a9bo119/VWZmpq699lp17drVZt3w8HBFRESoZcuWGjJkiCZOnKiFCxfq3LlzZe4vLS1NISEhVlNaWpqLnxwAAAAAAADAlcDlBs6ioiKtWLFCCxcu1MGDBy3ToUOH1LBhQ7322mtlrltSUqLFixerWbNmuvHGGy3zhw4dKh8fH7366qtasWKFEhMTLT08y+Pr66uioiKdP3++zDTJyckqKCiwmpKTk5370AAAAAAAAACuKH6urrhhwwbl5+dr9OjRNmNkDho0SOnp6YqNjZUknTp1SidOnNDZs2d15MgRPfvss9qzZ482btwoX19fy3rBwcGKi4tTcnKyTp8+rYSEBLv7Nm+vqKhIhw8f1nPPPafu3burVq1aZcYbGBiowMBAVz8uAAAAAAAAgCuQyXDxYfoBAwaopKREGzdutFm2Z88edezYUYcOHVKbNm0s86tXr64mTZqoe/fuevjhhxUREWGz7u7du9W5c2f17dvXZtvHjx+3eqTd19dXDRo0UJ8+fTR79mxdffXVrnwUAAAAAAAAwGOuDmlR1SG47MeC3KoOoUIuN3ACAAAAAAAAqBgNnJ5VqZcMAQAAAAAAAEBVcnkMTgAAAAAAAAAV4wFqz6IHJwAAAAAAAACvRQMnAAAAAAAAAK9FAycAAAAAAAAAr8UYnAAAAAAAAIAHlTAGp0fRgxMAAAAAAACA16KBEwAAAAAAAIDXooETAAAAAAAAgNdiDE4AAAAAAADAgwzG4PQoenACAAAAAAAA8Fo0cAIAAAAAAADwWjyiDgAAAAAAAHhQiXhE3ZNo4ATwP6tXeExVh+CwLd+8U9UhAAAAAABwReIRdQAAAAAAAABeiwZOAAAAAAAAAF6LR9QBAAAAAAAADzIMxuD0JHpwAgAAAAAAAPBaNHACAAAAAAAA8Fo8og4AAAAAAAB4UAmPqHsUPTgBAAAAAAAAeC0aOAEAAAAAAAB4LRo4AQAAAAAAAHgtxuAEAAAAAAAAPMgQY3B6Ej04AQAAAAAAAHgttzZwJiQkyGQyyWQyKSAgQBEREUpNTVVRUZG2b99uWWYymXT11Verb9++Onz4cJnbKD3FxsZa0jRt2tQyv0aNGmrXrp2ys7Pd+VEAAAAAAAAAeAG39+CMjY1VXl6ejh07psmTJ2vGjBmaP3++ZXlubq7y8vL0zjvvqLCwUP369dP58+ftbqP09Nprr1mlSU1NVV5eng4cOKAOHTooLi5Ou3btcvfHAQAAAAAAAHAFc3sDZ2BgoMLCwtSkSRONHTtWvXr10vr16y3L69Wrp7CwMLVr104TJ07UN998o6NHj9rdRumpdu3aVmlq1qypsLAwRUZGaunSpapWrZrefPNNd38cAAAAAAAAoFJKDMNrJ2/g8TE4q1WrZtNDU5IKCgq0cuVKSVJAQECl9uHn5yd/f3+7+wEAAAAAAADwx+WxBk7DMLRlyxa988476tGjh2V+48aNFRwcrNDQUL366qu64447FBUVZbXuhg0bFBwcbDXNmTPH7n7Onz+vtLQ0FRQUWO0HAAAAAAAAwB+fn7s3aG6cvHDhgkpKSnTPPfdoxowZ+uijjyRJO3fuVPXq1fXhhx9qzpw5ev7552220b17dy1btsxq3lVXXWX192OPPaZp06bp3LlzCg4O1ty5c9WvX78y4yosLFRhYaHVvMDAQAUGBrr6UQEAAAAAAIAKGV7yqLe3cnsDp7lxMiAgQA0bNpSfn/UumjVrptDQULVo0UInT55UXFycduzYYZWmRo0aioiIKHc/U6ZMUUJCgoKDg1W/fn2ZTKZy06elpWnmzJlW81JSUjRjxgzHPxwAAAAAAACAK4rbH1E3N05ec801No2blxo3bpyOHDmitWvXOr2funXrKiIiQmFhYRU2bkpScnKyCgoKrKbk5GSn9wsAAAAAAADgyuH2HpzOqF69usaMGaOUlBTdddddlobKwsJCnThxwiqtn5+f6tat6/K+eBwdAAAAAAAA+OPx+FvUKzJ+/Hjl5OQoOzvbMm/Tpk1q0KCB1XTLLbdUYZQAAAAAAACAawwv/ucNTAajnAL4H9UrPKaqQ3DYlm/eqeoQAAAAAAAuCgwKr+oQXFZ47puqDqFCVd6DEwAAAAAAAABcVaVjcAIAAAAAAAB/dDxA7Vn04AQAAAAAAADgtWjgBAAAAAAAAOC1aOAEAAAAAAAA4LUYgxMAAAAAAADwIMbg9Cx6cAIAAAAAAADwWjRwAgAAAAAAAPBaNHACAAAAAAAA8FqMwQkAAAAAAAB4ECNwehY9OAEAAAAAAAB4LRo4AQAAAAAAAHgvA25z7tw5IyUlxTh37lxVh1IhYvUMYvUMYvUMb4nVW+I0DGL1FGL1DGL1DGL1DGL1DGJ1P2+J0zCI1VOIFagaJsMwGAbATU6fPq2QkBAVFBSoVq1aVR1OuYjVM4jVM4jVM7wlVm+JUyJWTyFWzyBWzyBWzyBWzyBW9/OWOCVi9RRiBaoGj6gDAAAAAAAA8Fo0cAIAAAAAAADwWjRwAgAAAAAAAPBaNHC6UWBgoFJSUhQYGFjVoVSIWD2DWD2DWD3DW2L1ljglYvUUYvUMYvUMYvUMYvUMYnU/b4lTIlZPIVagavCSIQAAAAAAAABeix6cAAAAAAAAALwWDZwAAAAAAAAAvBYNnAAAAAAAAAC8Fg2cAAAAAAAAALwWDZyl/Pjjjxo7dqyuueYaBQYGKiwsTDExMfrXv/6lsLAwzZkzx2adoUOHqlOnTiouLtaMGTNkMpkUGxtrk27+/PkymUzq1q2bZZ45vXkKCQnRrbfeqn/961/lxnnixAklJSUpIiJCQUFBql+/vrp06aJly5bp7NmzkqSmTZtatlu9enW1bt1af//73622s337dplMJkVHR6u4uNhqWWhoqLKysix/l95etWrV1LRpUw0dOlTvvfdeRYdVCQkJlnUDAgIUERGh1NRUFRUVORWDJB04cEBDhgxR/fr1FRQUpOuuu05jxozRp59+anM87U0VHdcJEyaoefPmCgwMVHh4uAYMGKCtW7da0uzatUt9+/ZV7dq1FRQUpNatW+uZZ56xid28vw8//NBqfmFhoerUqSOTyaTt27fbpDdfB126dHHo2Hoi7qCgIH311VdW8++66y4lJCRY/k5ISNBdd93lUHyl16noOjBPV199tfr27avDhw/bHB9704wZMyRJa9euVadOnRQSEqKaNWsqOjpaEydOdDi2uXPnWs1ft26d1XVTXFysRYsWqXXr1goKClLt2rXVp08fffDBB1brzZgxQ23btrXZz/Hjx2UymXTw4EFJzt2DpQ0YMMBuPiNJO3fulMlk0scff1zm8TJfl1lZWZZ5Pj4+atCggeLi4vT1119bbbNbt25W69evX19DhgyxuU7K48g18+233yogIECtWrWyzHP0vnb0HDpr9+7d8vX1Vb9+/azmm8+leTJfb+PGjdOxY8es0mZlZSk0NNTyd3FxsebOnauoqChVq1ZNV111lTp27GiTR7ui9HF25JiUvi/tTU2bNpV08Rpw5F6yF09FMVx6fEozmUxat26dpP8ec19fX3333XdW6fLy8uTn5yeTyaTjx49bpTdPV111lbp27aqdO3dWGLcjec6l93Nl47N3jzrLE9dr6Xyi9OTK9eru+ou96cSJE5Ls58M7d+5UaGioJk6cqEvfsVn6XvD391ezZs306KOP6ty5c1bpNmzYoK5du6pmzZqqXr26OnToYJNXu3ot1KtXT7/++qtV+rZt21rKuLJUpnw1O3/+vJ5++mm1adNG1atXV926ddWlSxdlZmbqwoULlv04W/a7M8ay8qvS5WFl6qvujNmZPOTSKT4+3qn4ystfL42zfv36GjRokL744gu7x6yse05yvA7kSNz2riNzrL/88ovV/KioKAUGBlru7dLM9ZOVK1dazX/22Wct5ZdUfjnjaKyO1jPKyp+mTZtmtdzZup+92K6U8tV8f8yaNcsmby0rbnNeW79+ffXu3VsZGRkqKSmxSR8TEyNfX1999NFHdvdtb8rKynKonKhIZctUR+vrlTme9o7r5cpja9SooXbt2ik7O9uhGIHLgQbOUgYNGqQDBw5o+fLl+vTTT7V+/Xp169ZNBQUFevHFFzVz5kyrjCA7O1sbNmzQ8uXL5evrK0lq0KCBtm3bpm+//dZq2xkZGbrmmmts9hkdHa28vDzl5eVp9+7duu6669S/f38VFBTYjfGLL77QjTfeqM2bN2vOnDk6cOCAdu/erUcffVQbNmzQli1bLGlTU1OVl5enI0eOKD4+XmPGjNHbb79td5srVqyo8PiYt5ebm6sVK1YoNDRUvXr10uzZsytcNzY2Vnl5eTp27JgmT56sGTNmaP78+U7FsGHDBnXq1EmFhYV65ZVXlJOTo5dfflkhISGaPn26HnnkEcuxzMvLU+PGjS0xm6eyHD9+XO3bt9d7772n+fPn6/Dhw9q0aZO6d++ucePGSbrYeNa1a1c1btxY27Zt09GjR5WUlKRZs2Zp2LBhNgVQeHi4MjMzreatXbtWwcHBdmPIzMxUXl6ePvjgA9WtW1f9+/e3qoBerrhNJpOefPLJcvfrqoqug9zcXOXl5emdd95RYWGh+vXrp/Pnz1udw2effVa1atWymvfII49o69atiouL06BBg7Rnzx7t27dPs2fPtnwxq0hQUJDmzZun/Px8u8sNw9CwYcOUmpqqpKQk5eTkaPv27QoPD1e3bt0slURXOHoPmo0ePVrvvvuuTT4jXbyObrrpJtWqVUuStGXLFqtjlZeXp/bt21vSm4/ld999pzVr1ig3N1dDhgyx2e6YMWOUl5en77//Xm+88Ya++eYbh7+EOSorK0tDhw7V6dOn9e9//1uSnLqvKzqHrkhPT9eECRO0Y8cOff/99zbLzcf30KFDmjNnjnJyctSmTRurHxguNXPmTC1atEhPPfWU/vOf/2jbtm164IEHbL7UuUNFx+S5556zOZbmvCgvL8/yhcKTMTirUaNGNvfL8uXL1ahRI7vpzedox44datiwofr3768ffvih3H04kue4O76y7lFneOJ6lWTz+fPy8jRixAinYvNE/cVcZpSe6tWrZ3f/GzduVExMjCZNmqRnn33WqkHCzFxGffHFF1q0aJFeeOEFpaSkWJYvWbJEd955p7p06aJ///vf+vjjjzVs2DD9+c9/tntNOHst/Prrr1qwYIH9A1gBV8tX6WLjZkxMjObOnasHHnhAu3bt0p49ezRu3DgtWbJEn3zyiUsxuTPGS7dRenrttdes0lSmvuqumJ3JQy7NA5YuXepwfI7mr7m5ufr++++VnZ2tTz75RAMGDLBqXKvonvNkHag877//vn7//XcNHjxYy5cvt5smKChI06ZNc7i+VxnOlGeX5k9Tp061Wu5s3a+y8TjC1fLr2LFjmjlzpmbPnq2MjIwK92O+r44fP663335b3bt3V1JSkvr376+ioiJLuq+//lq7du3S+PHjLdsNDw+3Oq6TJ0+2+j6dl5enuLg4yzacKScuVdky1ZH6+g033FDp42l2ufPYAwcOqEOHDoqLi9OuXbscjhPwKAOGYRhGfn6+IcnYvn17mWkSEhKMG2+80Th//rxx8uRJ4+qrrzaee+45y/KUlBSjTZs2Rv/+/Y1Zs2ZZ5n/wwQdG3bp1jbFjxxpdu3a1SV/aN998Y0gy9uzZYzeGmJgYo3HjxsZvv/1md3lJSYlhGIbRpEkTY9GiRVbLrrrqKuPhhx+2/L1t2zZDkjFlyhQjPDzcOHfunGVZSEiIkZmZafnb3vYMwzCefPJJw8fHxzh69KjdeAzDMEaNGmXceeedVvN69+5tdOrUyeEYzpw5Y9StW9e466677O4jPz/fZl5ZMdvTp08fo1GjRnaPa35+vvHbb78ZderUMe6++26b5evXrzckGStXrrTMk2RMmzbNqFWrlnH27Fmrzz19+nRDkrFt2zar9GvXrrX8/d133xmSjOeff/6yx/3II48YPj4+xuHDhy3z77zzTmPUqFGWv+2d04o4ch2UPo/m+A4dOmS1TmZmphESEmKz/aSkJKNbt25OxVQ6tv79+xtRUVHGlClTLPPXrl1rmLPJlStXGpKM9evX26x/9913G3Xq1LGcB3v3tmEYxpdffmlIMg4cOGAYhnP3YGkXLlww6tevbzz11FNW83/99VcjODjYWLZsmc2+7LF3LBcvXmxIMgoKCizzunbtaiQlJVml+8c//mFUr169zG1fqqJrpqSkxGjevLmxadMm47HHHjPGjBljN11Z97Uj59BZ5uN59OhRIy4uzpg9e7ZlWVnHt7i42OjWrZvRpEkTo6ioyDAM2+Pcpk0bY8aMGS7FVJHSx9mVY3JpXmRm7xpwNJ6KYijrnr40HvMxnzZtmnHddddZpYuMjLTkrV9++aVV+tLn6OOPPzYkGW+88YbDn6Gs+C7dvjviqwxPXa/lnR9neKL+Yq/sNyudD7/yyitGQECAsWTJkjLT28uj7r77buPGG280DMMwvv76a8Pf39+YNGmSzbrmfPPDDz80DMP1a2HKlClGcHCw8cMPP1jSt2nTxkhJSSkz7rJid6Z8nTdvnuHj42Ps37/fZtvnz5+3nDNXyn53xejo/itTX/VEzIbheB7iLEfyV3txvvLKK4Yky7Fw5J5zpg7kSNz2zqO9WBMSEoypU6cab7/9thEZGWmzTteuXY377rvPqFOnjrF06VLL/EWLFhlNmjSx/O1qPuZKmVpR/uRq3c9ebFdS+WoYhtGzZ0/jL3/5S4Vx2zv/W7duNSQZL730kmXejBkzjGHDhhk5OTlGSEiI1Xcqs7Lq3I6UE+VxR5nqSH29vO05cjzNqiqPvXDhglG9enVj6tSpDsUJeBo9OP9fcHCwgoODtW7dOhUWFtpN89xzz+nUqVN66qmn9Je//EWtWrXShAkTbNIlJiZaPV6QkZGhESNGKCAgoNwYCgsLlZmZqdDQULVo0cJm+alTp7R582aNGzdONWrUsLsNez0SSkpKtGbNGuXn59uNYeLEiSoqKtKSJUvKjc+epKQkGYahN954w6n1qlWrZvWLUUUxvPPOO/rpp5/06KOP2l3uymMnZj///LM2bdpU5nENDQ3V5s2bderUKbs9NAYMGKDIyEibX7fat2+vpk2bas2aNZIu/gq5Y8cO3XvvvRXGVK1aNUmy+VXtcsTdpUsX9e/f3+aXZk+49DowKygosDxuVNF9YxYWFqZPPvlER44ccSkWX19fzZkzR0uWLLH7S+urr76qyMhIDRgwwGbZ5MmTderUKb377rsu7dvZe9DPz08jR45UVlaWVQ/c7OxsFRcXa/jw4S7FcfLkSa1du1a+vr6WXun2/Pzzz1q9erU6duzo0n7s2bZtm86ePatevXopPj5eK1eu1JkzZ5zaRkXn0FmrV69WVFSUWrRoofj4eGVkZFT4qJCPj4+SkpL01Vdfad++fXbThIWF6b333tOPP/5Y6Rgr4u5jciXEcMcddyg/P1/vv/++pIs9fPLz8+3em6X9/vvvlp4pjuYrlzO+yvLU9eoOnqq/OGLp0qW67777lJGRofHjxzu83pEjR7Rr1y7Lfl9//XVduHDBbnn64IMPKjg42KY8dfZaGD58uOWxwspypnx95ZVX1KtXL91444026f39/cs8Z5czxspytb56qcsZsyNcyV/Lq2OWdc95sg5Ull9//VXZ2dmKj49X7969VVBQYHeIkVq1aumJJ55Qamqq0/UGZ7mzPKvM9y9PxCNVrvzau3ev9u3b53LdsEePHmrTpo3++c9/SrrYazgzM1Px8fGKiopSRESEXn/9dZe27Qp3lKmVqa9X9nhKlye/8vPzk7+/f7nfWYHLiQbO/+fn56esrCwtX75coaGh6tKlix5//HF9/PHHljS1atVSZmam5syZo82bNyszM9Nuhbx///46ffq0duzYoTNnzmj16tVKTEy0u9/Dhw9bGlerVaumBQsW6LXXXrM8XlraZ599JsMwbBo/69ata9nGY489Zpn/2GOPKTg4WIGBgRo8eLBq166t+++/32a71atXV0pKitLS0sp8NL4sV111lerVq2cZk6UihmFoy5Yteuedd9SjRw+HYzCPZxIVFeVUfI4wH9fytv3pp59Kklq2bGl3eVRUlCVNaYmJiZZHC7KystS3b19dffXV5cZz9uxZTZs2Tb6+vuratWuVxJ2WlqZNmzY5NFadK8q6Dho3bqzg4GCFhobq1Vdf1R133OHwOZ8wYYI6dOig1q1bq2nTpho2bJgyMjLK/MHCnoEDB6pt27ZWjySaffrpp2UeR/N8e8fSEa7cg4mJifr888+txuzNzMzUoEGDFBISYpnXuXNnS/5gnkorKChQcHCwatSoofr162vbtm12GyH+9re/WdLVqVNHubm5Tj02U5H09HQNGzZMvr6+atWqlZo3b+7SmD7lnUNXYjI/hh8bG6uCgoIKx0iW/ptPlZUvPvPMM/rxxx8VFhamG264QX/+85/tPn7rLu48JldCDP7+/pYvG9LFHxHj4+Pl7+9vN735HqhRo4YWLFig9u3bq2fPnpWOw93xlXWPOspT16v033zCPIWFhTkVm6fqL+YywzxFR0dbLc/JydH48eO1bNkyhx6p37Bhg4KDgy1jVZ88eVJTpkyRdDF/DwkJUYMGDWzWCwgIUPPmzW3KAGevBfN4ei+++KI+//zzCuO1x5Xy9dixYx6pX7kzRjPzOSo92Rsj/1LO1lfdGXNFLs0DDhw44NT6zuSveXl5WrBggRo1amR1P1Z0z7m7DmTvPPbp08cqzcqVK3XdddcpOjpavr6+GjZsmNLT0+1u7y9/+YuCgoL0zDPPOByDqxw93pfmT6dOnbJaXpnvX67E4whXy6+AgAB16NBBQ4cO1ciRI13ef1RUlOUe3bJli86ePauYmBhJUnx8fJnnvzwVlRNlcVeZ6mh9XXLf8bxceez58+ct12/p/QBViQbOUgYNGqTvv/9e69evV2xsrLZv36527dpZ9cbs0aOHOnXqpHvvvVdNmjSxux1z4ZCZmans7GxFRkZaja9RWosWLXTw4EEdPHhQ+/bt09ixYzVkyBDt3bvX4bj37NmjgwcPKjo62qoxZ8qUKTp48KDee+89dezYUYsWLVJERITdbYwePVp16tTRvHnzHN6vmWEYdht6Syv9paFPnz6Ki4uzGTS/vBgq+sWsMpzZtrNxxMfHa/fu3friiy+UlZVVZkO3dLHnRnBwsGrWrKk1a9YoPT29zOvG03Fff/31GjlypNt7cVZ0HezcuVP79u1TVlaWIiMj9fzzzzu87Ro1amjjxo367LPPNG3aNAUHB2vy5Mn605/+ZHl5hSPmzZun5cuXKycnx2aZJ69DZ+/BqKgode7c2VIJ/eyzz7Rz506NHj3aKt2qVasseYx5Kq1mzZo6ePCg9u7dq4ULF6pdu3Z2xykbMWKEDh48qEOHDun9999XRESEbr/9dpsXYrjil19+0T//+U+rMT1drchK5Z9DR+Xm5mrPnj2WX9f9/PwUFxfnUEzm66SsfPH666/XkSNH9OGHHyoxMVEnT57UgAED7DbguIs7jsmVFENiYqKys7N14sQJZWdnl5u3rlq1SgcOHNCaNWsUERGhrKysMr+suYuz8ZV3jzrCk9er9N98wjy5a6ytytZfdu7caRXXW2+9ZbW8cePGateunebPn1/uONxm3bt318GDB/Xvf/9bo0aN0n333adBgwZV6jM6cy1IF1+occstt2j69OlO7acy5asnyzZ3xWhmPkelpz//+c8O7d+R+qonYq7IpXnA9ddf7/Q2KspfGzdurBo1aqhhw4Y6c+aM1qxZY9Vzy5F7zp3Xib3zeOmLjcyNa2bx8fHKzs62W+8IDAxUamqqFixYoJ9++sltcZbFkfLs0vypdu3aNmkq8/3L2Xgc5Ur5dejQIa1evVpvvPFGpb47lL5HMzIyFBcXJz8/P0kXvyd98MEHTv/4U1E5YY87y1RH6+tS5Y/n5cpjzT+IVK9eXfPmzdPcuXNtXsQEVBW/qg7gShMUFKTevXurd+/emj59uu6//36lpKRYvUHaz8/PktmWJTExUR07dtSRI0fKLRjMbzkzu/HGG7Vu3To9++yzevnll63SRkREyGQyKTc312p+8+bNJf33kROzunXrKiIiQhEREcrOzlbr1q1100032a04+fn5afbs2UpISHDqEa5Tp07pxx9/VLNmzcpN1717dy1btkwBAQFq2LCh3eNXXgyRkZGSpKNHj+rmm292OD5HXHfddTKZTDp69GiZacz7z8nJUefOnW2W5+Tk2D2uderUUf/+/TV69GidO3dOffr0KbNBaNGiRerVq5dCQkIq7OXp6biliy9CiYyMdOvA8RVdB82aNbMM0XDy5EnFxcVpx44dTu3j2muv1bXXXqv7779fTzzxhCIjI7Vq1Srdd999Dq1/2223KSYmRsnJyVb3fWRkZJkVR/N88/GuVauW3V/jzS+RufQXW8m1e3D06NGaMGGCli5dqszMTF177bU2vX7Dw8PL/GFDuvg4jXl5y5Yt9fnnn2vs2LH6xz/+YZUuJCTEki4iIkLp6elq0KCBVq1aVemGuVdffVXnzp2zegzHMAyVlJTo008/tRxXR5V1Dp2Rnp6uoqIiNWzY0CqmwMBA/fWvfy13XfP1UF6+6OPjow4dOqhDhw6aOHGiXn75Zd1777164oknKsxPXeGOY+KpGGrVqqUzZ86opKREPj7//d21vPuldevWioqK0vDhw9WyZUu1atWqzIbB8PBwXXfddbruuutUVFSkgQMH6siRIwoMDHTnx6tUfOXdo464HNdrZWL0VP3FXGaUpWbNmtqyZYt69+6t7t27a9u2bXZ7YJrVqFHD8jkzMjLUpk0bpaena/To0YqMjFRBQYG+//57q+MsXezF8vnnn6t79+4223TmWjCbO3eubr75ZkvvUUdUpnyNjIwsty7hLu6oA5Q+R85wtL7qiZgr4o48oKI8fufOnapVq5bq1aunmjVr2iyv6J5zpg7kCHvnsfQj1v/5z3/04Ycfas+ePVa9u4uLi7Vy5UqNGTPGZpvx8fFasGCBZs2aZfUGdU9wpEytKH+SXP/+5Wg8l6N8vbQOOX36dM2YMUNBQUFOf46cnBw1a9ZMP//8s9auXasLFy5o2bJlluXFxcXKyMhw6oVhjpyHS7m7THWkvi5V/nherjx2ypQpSkhIUHBwsOrXr+/0D0eAJ9GDswLXX3+9S+O5REdHKzo6WkeOHNE999zj1Lq+vr76/fffbebXqVNHvXv31l//+lenYwoPD1dcXJySk5PLTDNkyBBFR0dr5syZDm/3ueeek4+Pj+66665y05kzy2uuuabcxuGyYrj99ttVt25dPf3003bXq8zbh6+66irFxMRo6dKldo/rL7/8ottvv11XXXWVFi5caLN8/fr1OnbsWJnjqCQmJmr79u0aOXJkueMahoWFKSIiwqHGzcsRd3h4uMaPH6/HH3/c6k2bleHodSBJ48aN05EjR7R27VqX99e0aVNVr17d6ftl7ty5evPNN7V7927LvGHDhunYsWN68803bdIvXLjQcn9KF3tmf/vttzZvat6/f7+CgoJ0zTXX2N2vs/fg0KFD5ePjo1dffVUrVqxQYmJipSsZU6dO1apVq7R///5y05mvZXt5lbPS09M1efJkq1+LDx06pFtvvdXlx+DtnUNHFRUVacWKFVq4cKFNTA0bNrQZZ6+0kpISLV68WM2aNbM7nl1ZzF8iPTl+WGWOiSdjaNGihYqKimy+QJmvwbK+NJvz1op6xJU2ePBg+fn56W9/+5vzwTvJlfhcURXXq7M8XX8pT+3atbVlyxbVqlVL3bp1s/smXHt8fHz0+OOPa9q0afr99981aNAg+fv72y1Pn3/+eZ05c6bCeoCj18Kf/vQn3X333U713KlM+XrPPfdoy5Ytdh+NvnDhgtvypctdByjN0frqpaoyZmeVl8c3a9ZM1157rd3GzUvZu+ecqQO5Q3p6um677TYdOnTIKl+bNGlSmb3ofHx8lJaWpmXLlrk8FIEz3FWmuvL9y9F4Lmf5Kl2sGxYVFbk0HuN7772nw4cPa9CgQXrllVfUuHFjm/O/cOFCZWVlue17iT2eKFNdra87ezwvV35l/kEkLCyMxk1ccejB+f9OnTqlIUOGKDExUTfccINq1qypvXv36umnn9add97p0jbfe+89XbhwodxfjYqKinTixAlJFwfTXrVqlf7zn/9Y/VpZ2t/+9jd16dJFN910k2bMmKEbbrhBPj4++uijj3T06FG1b9++zH0lJSWpVatW2rt3r2666Sa7aebOnWsZ6+RSv/76q06cOKELFy7oyy+/1Msvv6y///3vSktLq/QvzxXFUKNGDf3973/XkCFDdMcdd+ihhx5SRESEfvrpJ61evVpff/21ZbBkVyxdulRdunTRn/70J6WmpuqGG25QUVGR3n33XS1btkw5OTl64YUXNGzYMD3wwAMaP368atWqpa1bt2rKlCkaPHiwhg4danfbsbGx+vHHH+2Oq1pZnoxbkpKTk/XSSy/pyy+/VFxcnNvjL0/16tU1ZswYpaSk6K677qqwAJ0xY4bOnj2rvn37qkmTJvrll1+0ePFiXbhwwelKd+vWrTVixAgtXrzYMm/YsGHKzs7WqFGjNH/+fPXs2VOnT5/W0qVLtX79emVnZ1vGrYyJiVGLFi00fPhwzZo1S2FhYdq/f7+mTZumpKSkchu6y7sHLxUcHGz5EnL69Gm7vQhOnTplyWPMQkNDy/wlODw8XAMHDtSTTz6pDRs2WOafPXvWsp0ffvhBTz31lIKCgnT77bc7FKt0cRy/SyvZv/76q/bv369XXnnFZhyg4cOHKzU1VbNmzaqwonYpe+fQURs2bFB+fr5Gjx5t07th0KBBSk9PV2xsrKT/Ht+zZ8/qyJEjevbZZ7Vnzx5t3LixzPM8ePBgdenSRZ07d1ZYWJi+/PJLJScnKzIy0qPj4FXmmHgyhujoaN1+++1KTEzUwoUL1bx5c+Xm5mrixImKi4tTo0aN7G5rzJgxGjJkiFM9M0wmkx566CHNmDFDDz74oKpXr17Zj1QmR+Nz9h69lKevV3fxRP3l5MmTOnfunFXaOnXq2AxBEBoaqnfffVcxMTHq1q2btm/fbtML054hQ4ZoypQpWrp0qR555BE9/fTTmjx5soKCgnTvvffK399fb7zxhh5//HFNnjy5zJdBuHKtzp49W9HR0U7nfY64tHydOHGiNm7cqJ49e+qpp57SLbfcYqkHz5s3T+np6Wrbtq0k+/l4nTp1FB4e7tEYzXWAwsJCm/vFz89PdevWtfx9ueqrjsZ8ubgzj7/0nnOmDlRZFy5c0D/+8Q+lpqaqVatWVsvuv/9+PfPMM/rkk0/sjqXYr18/dezYUS+88ILq16/vlnjK4s7j7Uzdz5l4PF2+msuUoqIiHT58WM8995y6d+9e4Xce831cXFysH374QZs2bVJaWpr69++vkSNHqn379ho8eLDN+Q8PD1dycrI2bdrk8CPRjpYTZp4oUx2pr5fenrPH0xWVyWOBKxk9OP9fcHCwZcyZ2267Ta1atdL06dM1ZsyYCruil6VGjRoVFgyffPKJGjRooAYNGqht27ZavXq1li1bVuaAwtdee60OHDigXr16KTk5WW3atNFNN92kJUuW6JFHHtFTTz1V5r6uv/563X777XryySfLTNOjRw/16NFDRUVFNsuefPJJNWjQQBEREbr33ntVUFCgrVu3ltkY66qyYrjzzju1a9cu+fv765577rE8PlFQUKBZs2ZVap/NmzfX/v371b17d02ePFmtWrVS7969tXXrVsujEYMHD9a2bdv09ddf69Zbb1WLFi20aNEiPfHEE1q5cmWZFVmTyaS6det65K2anoxbuthL9LHHHrOpGJSUlHjkS9elxo8fr5ycHIdeNtO1a1d98cUXGjlypKKiotSnTx+dOHFCmzdvtnmxhSNSU1NVUlJi+dtkMmn16tV6/PHHtWjRIrVo0UK33nqrvvrqK23fvt2qV4ifn582b96sa665RsOHD1erVq2UkpKipKSkcu9Rqfx70J7Ro0crPz9fMTExdr+w9+rVy5LHmKeKhh14+OGHtXHjRu3Zs8cy76WXXrKs3717d/3000966623nDq227dv14033mg1ZWRk6Prrr7fbsDdw4ECdPHnSofGS7Ln0HDoqPT3dMlzEpQYNGqS9e/fq9OnTkv57fFu3bq2pU6eqZcuW+vjjj60eVb30fomJidGbb76pAQMGKDIyUqNGjVJUVJQ2b97s8fvK1WPiznveXgyrVq1S165d9eCDDyo6OloPPfSQ7rzzTpsx2UozV7idjWvUqFG6cOGCy2W7oxyNz5V7tDRPX6/u4on6S4sWLWyOXVlvgw8JCdHmzZtVt25dde3aVd99912FMfv5+Wn8+PF6+umndebMGU2cOFFr167Vzp07ddNNN6lVq1Z69dVXtWzZMi1YsKDc7Th7rUZGRioxMdGm/HWX0uVrYGCg3n33XT366KN64YUX1KlTJ3Xo0EGLFy/WQw89ZNXIYC8fr2zPM0diNNu0aZPNOb/lllus1rtc9VVHY76cXM3jL3XpPedMHaiyduzYoVOnTmngwIE2y1q2bKmWLVuWOxbivHnzPHbfXMpdx9vZup8z8XiyfDWXKU2bNtUDDzygvn37atWqVRXGab6PmzZtqtjYWG3btk2LFy/WG2+8YektaW/845CQEPXs2dOpMdqdKSck95epZhXV10tvz9nj6SpX81jgSmYyLtfI4gD+MGJjYxUREeHxBgLgj2Du3Ll6+eWXdeTIkaoOxWVRUVG6//779cgjj1R1KPCwP8L1CgAAgP89PKIOwGH5+fn64IMPtH37doffWgr8rzp79qyOHj2qzMxM9enTp6rDccnJkyf19ttvKzc3Vz179qzqcOBBf4TrFQAAAP+7aOAE4LDExER99NFHmjx5sstj0wL/K1588UWlpqaqV69e5Q4NciWLjY1Vfn6+Fi9e7NGX0KDq/RGuVwAAAPzv4hF1AAAAAAAAAF6LlwwBAAAAAAAA8Fo0cAIAAAAAAADwWjRwAgAAAAAAAPBaNHACAAAAAAAA8Fo0cAIAAAAAAADwWjRwAgAAAAAAAPBaNHACAAAAAAAA8Fo0cAIAAAAAAADwWjRwAgAAAAAAAPBa/wcbBLn6H3adqAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "P(tag|start) = P(tag|'.')"
      ],
      "metadata": {
        "id": "Mn0VivHxmZiy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "len(train_tagged_words)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "czeC3fiTmXtg",
        "outputId": "01e1c5f7-ae5d-41dd-9607-0d544535a2cf"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "53751"
            ]
          },
          "metadata": {},
          "execution_count": 127
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def Viterbi(words, train_bag = train_tagged_words):\n",
        "    state = []\n",
        "    T = list(set([pair[1] for pair in train_bag]))\n",
        "\n",
        "    for key, word in enumerate(words):\n",
        "\n",
        "        p = []\n",
        "        for tag in T:\n",
        "            if key == 0:\n",
        "                transition_p = tags_df.loc[:, tag]\n",
        "            else:\n",
        "                transition_p = tags_df.loc[state[-1], tag]\n",
        "\n",
        "\n",
        "            emission_p = word_given_tag(words[key], tag)[0]/word_given_tag(words[key], tag)[1]\n",
        "            state_probability = emission_p * transition_p\n",
        "            p.append(state_probability)\n",
        "        p = sum([j.tolist() for j in p][0])\n",
        "        pmax = max(p)\n",
        "\n",
        "        state_max = T[p.index(pmax)]\n",
        "        state.append(state_max)\n",
        "    return list(zip(words, state))"
      ],
      "metadata": {
        "id": "uCuHyHyumY66"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "random.seed(1234)\n",
        "\n",
        "rndom = [random.randint(1,len(test_set)) for x in range(5)]\n",
        "\n",
        "test_run = [test_set[i] for i in rndom]\n",
        "\n",
        "test_run_base = [tup for sent in test_run for tup in sent]\n",
        "\n",
        "test_tagged_words = [tup[0] for sent in test_run for tup in sent]\n",
        "test_run"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1NoEeWGImfdZ",
        "outputId": "3ca2c6b9-7954-4ac8-eab9-746af5f7829e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('Вот', 'PRCL'),\n",
              "  ('здесь', 'ADVB'),\n",
              "  (',', 'PNCT'),\n",
              "  ('в', 'PREP'),\n",
              "  ('этом', 'ADJF'),\n",
              "  ('самооправдании', 'NOUN'),\n",
              "  (',', 'PNCT'),\n",
              "  ('—', 'PNCT'),\n",
              "  ('корни', 'NOUN'),\n",
              "  ('невероятной', 'ADJF'),\n",
              "  ('долговечности', 'NOUN'),\n",
              "  ('трифоновской', 'ADJF'),\n",
              "  ('прозы', 'NOUN'),\n",
              "  ('.', 'PNCT')],\n",
              " [('XII', 'ROMN')],\n",
              " [('Не', 'PRCL'), ('можешь', 'VERB'), ('!', 'PNCT')],\n",
              " [('Литература', 'NOUN')],\n",
              " [('Речь', 'NOUN'),\n",
              "  ('идёт', 'VERB'),\n",
              "  ('об', 'PREP'),\n",
              "  ('антимонопольном', 'ADJF'),\n",
              "  ('законодательстве', 'NOUN'),\n",
              "  ('.', 'PNCT')]]"
            ]
          },
          "metadata": {},
          "execution_count": 178
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "start = time.time()\n",
        "tagged_seq = Viterbi(test_tagged_words)\n",
        "end = time.time()\n",
        "difference = end-start"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 381
        },
        "id": "RF8tnK2zmjIw",
        "outputId": "ad686221-b7e9-41e7-de50-92ba3673563d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.0 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-179-03375a79cc52>\u001b[0m in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mstart\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mtagged_seq\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mViterbi\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_tagged_words\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0mend\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mdifference\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mend\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0mstart\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-177-00d00700ed6c>\u001b[0m in \u001b[0;36mViterbi\u001b[0;34m(words, train_bag)\u001b[0m\n\u001b[1;32m     17\u001b[0m             \u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstate_probability\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     18\u001b[0m         \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mj\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtolist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mj\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtolist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 19\u001b[0;31m         \u001b[0mpmax\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmax\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m         \u001b[0mstate_max\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mT\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpmax\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.9/dist-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36m__nonzero__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1525\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1526\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__nonzero__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1527\u001b[0;31m         raise ValueError(\n\u001b[0m\u001b[1;32m   1528\u001b[0m             \u001b[0;34mf\"The truth value of a {type(self).__name__} is ambiguous. \"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1529\u001b[0m             \u001b[0;34m\"Use a.empty, a.bool(), a.item(), a.any() or a.all().\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all()."
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"Time taken in seconds: \", difference)\n",
        "print(tagged_seq)\n",
        "print(test_run_base)"
      ],
      "metadata": {
        "id": "qBh6Z88rmkdF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "check = [i for i, j in zip(tagged_seq, test_run_base) if i == j]"
      ],
      "metadata": {
        "id": "xTEAyaNGmlyT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "accuracy = len(check)/len(tagged_seq)"
      ],
      "metadata": {
        "id": "n4aafvD7moZN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "accuracy"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zpxMMNQxmo_w",
        "outputId": "d3b2b1ea-5ea9-4dbe-8d0d-86ab857f18b2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8867924528301887"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "incorrect_tagged_cases = [[test_run_base[i-1],j] for i, j in enumerate(zip(tagged_seq, test_run_base)) if j[0]!=j[1]]"
      ],
      "metadata": {
        "id": "RvwT4tlImuuX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "incorrect_tagged_cases"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KMCshinUmvV7",
        "outputId": "2cc963db-758c-4ebe-f1a8-74832f7bc109"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('Peter', 'NNP'), (('Walters', 'VB'), ('Walters', 'NNP'))],\n",
              " [(',', ','), (('58-year-old', 'VB'), ('58-year-old', 'JJ'))],\n",
              " [('of', 'IN'), (('British', 'JJ'), ('British', 'NNP'))],\n",
              " [('this', 'DT'), (('cement', 'VB'), ('cement', 'NN'))],\n",
              " [('The', 'DT'), (('Sacramento-based', 'VB'), ('Sacramento-based', 'JJ'))],\n",
              " [('which', 'WDT'), (('*T*-110', 'VB'), ('*T*-110', '-NONE-'))],\n",
              " [('*T*-110', '-NONE-'), (('has', 'VB'), ('has', 'VBZ'))],\n",
              " [('in', 'IN'), (('north', 'VB'), ('north', 'JJ'))],\n",
              " [('the', 'DT'), (('purchasing', 'NN'), ('purchasing', 'VBG'))],\n",
              " [('being', 'VBG'), (('examined', 'VBD'), ('examined', 'VBN'))],\n",
              " [('Robert', 'NNP'), (('Black', 'VB'), ('Black', 'NNP'))],\n",
              " [('the', 'DT'), (('Richmond', 'VB'), ('Richmond', 'NNP'))]]"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk\n",
        "nltk.download('punkt')\n",
        "\n",
        "sentence_test = 'Я пошел гулять. Прогулка была хорошей. Я увидел лису и зайца.'\n",
        "words = word_tokenize(sentence_test)\n",
        "\n",
        "start = time.time()\n",
        "tagged_seq = Viterbi(words)\n",
        "end = time.time()\n",
        "difference = end-start"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jtuOAzbnmzLy",
        "outputId": "e8d91a01-4bc0-49c2-da98-328ce1390b88"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(tagged_seq)\n",
        "print(difference)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q4bEsqtDmz1l",
        "outputId": "88b5b4cc-0785-4835-b837-226cde03687a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[('Twitter', 'VB'), ('is', 'VB'), ('the', 'DT'), ('best', 'JJS'), ('networking', 'NN'), ('social', 'JJ'), ('site', 'VB'), ('.', '.'), ('Man', 'NNP'), ('is', 'VBZ'), ('a', 'DT'), ('social', 'JJ'), ('animal', 'VB'), ('.', '.'), ('Data', 'NNP'), ('science', 'NN'), ('is', 'VBZ'), ('an', 'DT'), ('emerging', 'VB'), ('field', 'NN'), ('.', '.'), ('Data', 'NNP'), ('science', 'NN'), ('jobs', 'NNS'), ('are', 'VBP'), ('high', 'JJ'), ('in', 'IN'), ('demand', 'NN'), ('.', '.')]\n",
            "22.862608909606934\n"
          ]
        }
      ]
    }
  ]
}