nlp_for_sec_filings.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "nlp-for-sec-filings.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/allen44/nlp-sec-filings/blob/main/nlp_for_sec_filings.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UuUOZFOW5_NP"
      },
      "source": [
        "# NLP on Financial Statements\n",
        "\n",
        "## Packages\n",
        "The `project_helper` module contains utility functions and graph functions. The `project_tests` contains the unit tests for all the problems.\n",
        "\n",
        "### Install Packages"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fa4RgA0wx_mH"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R_kploq_8bgn"
      },
      "source": [
        "### Mount gDrive"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xSdi4Vh05_NV",
        "outputId": "14e5b250-6179-4580-c972-2ee8eb3c67e8"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "% cd '/content/drive/MyDrive/Colab Notebooks/nlp-for-sec-filings'\n",
        "% ls"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "/content/drive/MyDrive/Colab Notebooks/nlp-for-sec-filings\n",
            " Bag_of_Word_Exercises.ipynb                  \u001b[0m\u001b[01;34m__pycache__\u001b[0m/\n",
            " config.py                                    Readability_Exercises.ipynb\n",
            " \u001b[01;34mData\u001b[0m/                                        README.md\n",
            " LoughranMcDonald_MasterDictionary_2018.csv  'request library.ipynb'\n",
            " nlp-for-sec-filings.ipynb                    requirements.txt\n",
            " project_5_starter-Copy1.ipynb                STNG.ipynb\n",
            " project_helper.py                            tests.py\n",
            " project_tests.py\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1hsGSTuh5_NY",
        "outputId": "227a6fd7-d735-424a-88b1-201ee1ecedf4"
      },
      "source": [
        "!pip install -U -r requirements.txt\n",
        "!pip install retrying"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already up-to-date: alphalens>=0.3.2 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 1)) (0.4.0)\n",
            "Requirement already up-to-date: nltk>=3.3.0 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 2)) (3.6.2)\n",
            "Requirement already up-to-date: numpy>=1.15.4 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 3)) (1.20.2)\n",
            "Requirement already up-to-date: ratelimit>=2.2.0 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 4)) (2.2.1)\n",
            "Requirement already up-to-date: requests>=2.18.4 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 5)) (2.25.1)\n",
            "Requirement already up-to-date: scikit-learn==0.24 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 6)) (0.24.0)\n",
            "Requirement already up-to-date: six>=1.11.0 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 7)) (1.15.0)\n",
            "Requirement already up-to-date: tqdm>=4.19.5 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 8)) (4.60.0)\n",
            "Requirement already up-to-date: pandas-datareader>=0.1 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 9)) (0.9.0)\n",
            "Requirement already up-to-date: yfinance>=0.1 in /usr/local/lib/python3.7/dist-packages (from -r requirements.txt (line 10)) (0.1.59)\n",
            "Requirement already satisfied, skipping upgrade: seaborn>=0.6.0 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (0.11.1)\n",
            "Requirement already satisfied, skipping upgrade: pandas>=0.18.0 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (1.2.4)\n",
            "Requirement already satisfied, skipping upgrade: empyrical>=0.5.0 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (0.5.5)\n",
            "Requirement already satisfied, skipping upgrade: IPython>=3.2.3 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (5.5.0)\n",
            "Requirement already satisfied, skipping upgrade: statsmodels>=0.6.1 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (0.10.2)\n",
            "Requirement already satisfied, skipping upgrade: matplotlib>=1.4.0 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (3.2.2)\n",
            "Requirement already satisfied, skipping upgrade: scipy>=0.14.0 in /usr/local/lib/python3.7/dist-packages (from alphalens>=0.3.2->-r requirements.txt (line 1)) (1.4.1)\n",
            "Requirement already satisfied, skipping upgrade: regex in /usr/local/lib/python3.7/dist-packages (from nltk>=3.3.0->-r requirements.txt (line 2)) (2019.12.20)\n",
            "Requirement already satisfied, skipping upgrade: click in /usr/local/lib/python3.7/dist-packages (from nltk>=3.3.0->-r requirements.txt (line 2)) (7.1.2)\n",
            "Requirement already satisfied, skipping upgrade: joblib in /usr/local/lib/python3.7/dist-packages (from nltk>=3.3.0->-r requirements.txt (line 2)) (1.0.1)\n",
            "Requirement already satisfied, skipping upgrade: idna<3,>=2.5 in /usr/local/lib/python3.7/dist-packages (from requests>=2.18.4->-r requirements.txt (line 5)) (2.10)\n",
            "Requirement already satisfied, skipping upgrade: certifi>=2017.4.17 in /usr/local/lib/python3.7/dist-packages (from requests>=2.18.4->-r requirements.txt (line 5)) (2020.12.5)\n",
            "Requirement already satisfied, skipping upgrade: chardet<5,>=3.0.2 in /usr/local/lib/python3.7/dist-packages (from requests>=2.18.4->-r requirements.txt (line 5)) (3.0.4)\n",
            "Requirement already satisfied, skipping upgrade: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.7/dist-packages (from requests>=2.18.4->-r requirements.txt (line 5)) (1.24.3)\n",
            "Requirement already satisfied, skipping upgrade: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn==0.24->-r requirements.txt (line 6)) (2.1.0)\n",
            "Requirement already satisfied, skipping upgrade: lxml in /usr/local/lib/python3.7/dist-packages (from pandas-datareader>=0.1->-r requirements.txt (line 9)) (4.2.6)\n",
            "Requirement already satisfied, skipping upgrade: multitasking>=0.0.7 in /usr/local/lib/python3.7/dist-packages (from yfinance>=0.1->-r requirements.txt (line 10)) (0.0.9)\n",
            "Requirement already satisfied, skipping upgrade: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.18.0->alphalens>=0.3.2->-r requirements.txt (line 1)) (2018.9)\n",
            "Requirement already satisfied, skipping upgrade: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.18.0->alphalens>=0.3.2->-r requirements.txt (line 1)) (2.8.1)\n",
            "Requirement already satisfied, skipping upgrade: setuptools>=18.5 in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (56.0.0)\n",
            "Requirement already satisfied, skipping upgrade: decorator in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (4.4.2)\n",
            "Requirement already satisfied, skipping upgrade: prompt-toolkit<2.0.0,>=1.0.4 in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (1.0.18)\n",
            "Requirement already satisfied, skipping upgrade: pickleshare in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.7.5)\n",
            "Requirement already satisfied, skipping upgrade: traitlets>=4.2 in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (5.0.5)\n",
            "Requirement already satisfied, skipping upgrade: simplegeneric>0.8 in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.8.1)\n",
            "Requirement already satisfied, skipping upgrade: pygments in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (2.6.1)\n",
            "Requirement already satisfied, skipping upgrade: pexpect; sys_platform != \"win32\" in /usr/local/lib/python3.7/dist-packages (from IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (4.8.0)\n",
            "Requirement already satisfied, skipping upgrade: patsy>=0.4.0 in /usr/local/lib/python3.7/dist-packages (from statsmodels>=0.6.1->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.5.1)\n",
            "Requirement already satisfied, skipping upgrade: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.4.0->alphalens>=0.3.2->-r requirements.txt (line 1)) (2.4.7)\n",
            "Requirement already satisfied, skipping upgrade: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.4.0->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.10.0)\n",
            "Requirement already satisfied, skipping upgrade: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib>=1.4.0->alphalens>=0.3.2->-r requirements.txt (line 1)) (1.3.1)\n",
            "Requirement already satisfied, skipping upgrade: wcwidth in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.2.5)\n",
            "Requirement already satisfied, skipping upgrade: ipython-genutils in /usr/local/lib/python3.7/dist-packages (from traitlets>=4.2->IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.2.0)\n",
            "Requirement already satisfied, skipping upgrade: ptyprocess>=0.5 in /usr/local/lib/python3.7/dist-packages (from pexpect; sys_platform != \"win32\"->IPython>=3.2.3->alphalens>=0.3.2->-r requirements.txt (line 1)) (0.7.0)\n",
            "Requirement already satisfied: retrying in /usr/local/lib/python3.7/dist-packages (1.3.3)\n",
            "Requirement already satisfied: six>=1.7.0 in /usr/local/lib/python3.7/dist-packages (from retrying) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bJtyEx93vv63",
        "outputId": "e5faef32-150f-42ab-fb3a-f51223f5bd22"
      },
      "source": [
        "!pip install -U  pandas"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already up-to-date: pandas in /usr/local/lib/python3.7/dist-packages (1.2.4)\n",
            "Requirement already satisfied, skipping upgrade: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas) (2.8.1)\n",
            "Requirement already satisfied, skipping upgrade: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas) (2018.9)\n",
            "Requirement already satisfied, skipping upgrade: numpy>=1.16.5 in /usr/local/lib/python3.7/dist-packages (from pandas) (1.20.2)\n",
            "Requirement already satisfied, skipping upgrade: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "79wPiv445_NZ"
      },
      "source": [
        "### Load Packages"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Swz1Od-P5_Na"
      },
      "source": [
        "import nltk\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import pickle\n",
        "#import pprintcvxpy \n",
        "import pprint\n",
        "import project_helper\n",
        "import project_tests\n",
        "import retrying\n",
        "\n",
        "from tqdm import tqdm"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iabyxt5A5_Nb"
      },
      "source": [
        "### Download NLP Corpora\n",
        "You'll need two corpora to run this project: the stopwords corpus for removing stopwords and wordnet for lemmatizing."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JekNHhch5_Nb",
        "outputId": "3b85fa6a-b83d-46f9-e168-3457ef77fb36"
      },
      "source": [
        "nltk.download('stopwords')\n",
        "nltk.download('wordnet')"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n",
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data]   Package wordnet is already up-to-date!\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T0_rB2xU5_Nc"
      },
      "source": [
        "## Get 10ks\n",
        "We'll be running NLP analysis on 10-k documents. To do that, we first need to download the documents. For this project, we'll download 10-ks for a few companies. To lookup documents for these companies, we'll use their CIK. If you would like to run this against other stocks, we've provided the dict `additional_cik` for more stocks. However, the more stocks you try, the long it will take to run."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A7LBB-3d5_Nc"
      },
      "source": [
        "cik_lookup = { 'AMZN': '0001018724', 'STNG':'0001483934' }\n",
        "\n",
        "cik_lookup1 = {\n",
        "    'AMZN': '0001018724',\n",
        "    'BMY': '0000014272',   \n",
        "    'CNP': '0001130310',\n",
        "    'CVX': '0000093410',\n",
        "    'FL': '0000850209',\n",
        "    'STNG':'0001483934',\n",
        "    'HON': '0000773840'}\n",
        "\n",
        "additional_cik = {\n",
        "    'AEP': '0000004904',\n",
        "    'AXP': '0000004962',\n",
        "    'BA': '0000012927', \n",
        "    'BK': '0001390777',\n",
        "    'CAT': '0000018230',\n",
        "    'DE': '0000315189',\n",
        "    'DIS': '0001001039',\n",
        "    'DTE': '0000936340',\n",
        "    'ED': '0001047862',\n",
        "    'EMR': '0000032604',\n",
        "    'ETN': '0001551182',\n",
        "    'GE': '0000040545',\n",
        "    'IBM': '0000051143',\n",
        "    'IP': '0000051434',\n",
        "    'JNJ': '0000200406',\n",
        "    # 'KO': '0000021344',\n",
        "    # 'LLY': '0000059478',\n",
        "    # 'MCD': '0000063908',\n",
        "    # 'MO': '0000764180',\n",
        "    # 'MRK': '0000310158',\n",
        "    # 'MRO': '0000101778',\n",
        "    # 'PCG': '0001004980',\n",
        "    # 'PEP': '0000077476',\n",
        "    # 'PFE': '0000078003',\n",
        "    # 'PG': '0000080424',\n",
        "    # 'PNR': '0000077360',\n",
        "    # 'STNG':'0001483934',\n",
        "    # 'SYY': '0000096021',\n",
        "    # 'TXN': '0000097476',\n",
        "    # 'UTX': '0000101829',\n",
        "    # 'WFC': '0000072971',\n",
        "    # 'WMT': '0000104169',\n",
        "    # 'WY': '0000106535',\n",
        "    # 'XOM': '0000034088'\n",
        "    }"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zR8DsScg5_Nc"
      },
      "source": [
        "### Get list of 10-ks\n",
        "The SEC has a limit on the number of calls you can make to the website per second. In order to avoid hiding that limit, we've created the `SecAPI` class. This will cache data from the SEC and prevent you from going over the limit."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "euz-R-C35_Nd"
      },
      "source": [
        "sec_api = project_helper.SecAPI()"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3280g5zU5_Nd"
      },
      "source": [
        "With the class constructed, let's pull a list of filled 10-ks from the SEC for each company."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZSNp634q5_Nd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "08da06d1-96cd-4c1d-91eb-6230b9f000fd"
      },
      "source": [
        "from bs4 import BeautifulSoup\n",
        "\n",
        "@retrying.retry\n",
        "def get_sec_data(cik, doc_type, start=0, count=60, sec_api=sec_api):\n",
        "    newest_pricing_data = pd.to_datetime('2021-04-26')\n",
        "    \n",
        "    rss_url = 'https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany' \\\n",
        "        f'&CIK={cik}&type={doc_type}&start={start}&count={count}' \\\n",
        "        '&owner=exclude&output=atom'\n",
        "     \n",
        "    print(rss_url)\n",
        "    sec_data = sec_api.get(rss_url)\n",
        "\n",
        "    # if request doesn't work, wait and try again shortly, \n",
        "    # via retyring.retry decorator\n",
        "    \n",
        "    try:\n",
        "      # print('using encoding with ascii')\n",
        "      feed = BeautifulSoup(sec_data.encode('utf-8'), 'xml').feed\n",
        "      if type(feed)!=None:\n",
        "        entries = [\n",
        "            (\n",
        "                entry.content.find('filing-href').getText(),\n",
        "                entry.content.find('filing-type').getText(),\n",
        "                entry.content.find('filing-date').getText())\n",
        "            for entry in feed.find_all('entry', recursive=False)\n",
        "            \n",
        "            if pd.to_datetime(entry.content.find('filing-date').getText()) <= newest_pricing_data]\n",
        "    except UnicodeError:\n",
        "      print('using encoding with utf-8')\n",
        "      feed = BeautifulSoup(sec_data.encode('utf-8'), 'xml').feed\n",
        "      if type(feed)!=None:\n",
        "        entries = [\n",
        "            (\n",
        "                entry.content.find('filing-href').getText(),\n",
        "                entry.content.find('filing-type').getText(),\n",
        "                entry.content.find('filing-date').getText())\n",
        "            for entry in feed.find_all('entry', recursive=False)\n",
        "            \n",
        "            if pd.to_datetime(entry.content.find('filing-date').getText()) <= newest_pricing_data]\n",
        "    return entries\n",
        "\n",
        "def test_get_sec_data(cik='0001483934', doc_type='10-k', start=0, count=60):\n",
        "  sec_api = project_helper.SecAPI()\n",
        "  get_sec_data(cik=list(additional_cik.keys())[0], doc_type='10-k')\n",
        "  get_sec_data(cik=list(additional_cik.keys())[0], doc_type='20-f')\n",
        "  get_sec_data(cik=list(additional_cik.keys())[1], doc_type='10-k')\n",
        "  get_sec_data(cik=list(additional_cik.keys())[1], doc_type='20-f')\n",
        "  return\n",
        "\n",
        "test_get_sec_data()\n",
        "\n"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AEP&type=10-k&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AEP&type=10-k&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AEP&type=20-f&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AXP&type=10-k&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AXP&type=10-k&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=AXP&type=20-f&start=0&count=60&owner=exclude&output=atom\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OBFeAOot5_Nd"
      },
      "source": [
        "Let's pull the list using the `get_sec_data` function, then display some of the results. For displaying some of the data, we'll use Amazon as an example. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eSdWAP6Z5_Nd",
        "outputId": "435e74e8-f9d5-4dbe-fe66-b9a53392193b"
      },
      "source": [
        "example_ticker = list(additional_cik.keys())[0]\n",
        "sec_data = {}\n",
        "\n",
        "for ticker, cik in additional_cik.items():\n",
        "    if ticker != \"STNG\":\n",
        "        print(ticker)\n",
        "        sec_data[ticker] = get_sec_data(cik, '10-K')\n",
        "        \n",
        "    else :\n",
        "        print(ticker)\n",
        "        sec_data[ticker] = get_sec_data(cik, '20-F')\n",
        "    \n",
        "\n",
        "pprint.pprint(sec_data[example_ticker][:1])"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "AEP\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000004904&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000004904&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000004904&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "AXP\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000004962&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "BA\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000012927&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "BK\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001390777&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "CAT\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000018230&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "DE\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000315189&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "DIS\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001001039&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "DTE\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000936340&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "ED\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001047862&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "EMR\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000032604&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "ETN\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001551182&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "GE\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000040545&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "IBM\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000051143&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "IP\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000051434&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "JNJ\n",
            "https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000200406&type=10-K&start=0&count=60&owner=exclude&output=atom\n",
            "[('https://www.sec.gov/Archives/edgar/data/4904/000000490421000010/0000004904-21-000010-index.htm',\n",
            "  '10-K',\n",
            "  '2021-02-25')]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NLOXhHNS5_Ne"
      },
      "source": [
        "### Download 10-ks\n",
        "As you see, this is a list of urls. These urls point to a file that contains metadata related to each filling. Since we don't care about the metadata, we'll pull the filling by replacing the url with the filling url."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E_0mqfut5_Ne",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3369a0c2-af55-4083-e645-c9b0c78bdeaa"
      },
      "source": [
        "raw_fillings_by_ticker = {}\n",
        "\n",
        "for ticker, data in sec_data.items():\n",
        "    raw_fillings_by_ticker[ticker] = {}\n",
        "    for index_url, file_type, file_date in tqdm(data, desc='Downloading {} Fillings'.format(ticker), unit='filling'):\n",
        "\n",
        "            if ((file_type == '10-K') or (file_type == '20-F')):\n",
        "                file_url = index_url.replace('-index.htm', '.txt').replace('.txtl', '.txt')            \n",
        "            \n",
        "                raw_fillings_by_ticker[ticker][file_date] = sec_api.get(file_url)\n",
        "\n",
        "\n",
        "print('Example Document:\\n\\n{}...'.format(next(iter(raw_fillings_by_ticker[example_ticker].values()))[:1000]))"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading AEP Fillings: 100%|██████████| 30/30 [00:21<00:00,  1.39filling/s]\n",
            "Downloading AXP Fillings: 100%|██████████| 32/32 [00:02<00:00, 12.32filling/s]\n",
            "Downloading BA Fillings: 100%|██████████| 30/30 [00:02<00:00, 10.49filling/s]\n",
            "Downloading BK Fillings: 100%|██████████| 16/16 [00:04<00:00,  3.77filling/s]\n",
            "Downloading CAT Fillings: 100%|██████████| 40/40 [00:05<00:00,  6.90filling/s]\n",
            "Downloading DE Fillings: 100%|██████████| 33/33 [00:05<00:00,  5.88filling/s]\n",
            "Downloading DIS Fillings: 100%|██████████| 38/38 [00:02<00:00, 17.82filling/s]\n",
            "Downloading DTE Fillings: 100%|██████████| 28/28 [00:03<00:00,  8.43filling/s]\n",
            "Downloading ED Fillings: 100%|██████████| 24/24 [00:03<00:00,  6.98filling/s]\n",
            "Downloading EMR Fillings: 100%|██████████| 28/28 [00:02<00:00, 11.37filling/s]\n",
            "Downloading ETN Fillings: 100%|██████████| 9/9 [00:02<00:00,  3.31filling/s]\n",
            "Downloading GE Fillings: 100%|██████████| 30/30 [00:04<00:00,  6.89filling/s]\n",
            "Downloading IBM Fillings: 100%|██████████| 28/28 [00:02<00:00,  9.71filling/s]\n",
            "Downloading IP Fillings: 100%|██████████| 33/33 [00:03<00:00, 10.90filling/s]\n",
            "Downloading JNJ Fillings: 100%|██████████| 38/38 [00:02<00:00, 16.09filling/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Example Document:\n",
            "\n",
            "<SEC-DOCUMENT>0000004904-21-000010.txt : 20210225\n",
            "<SEC-HEADER>0000004904-21-000010.hdr.sgml : 20210225\n",
            "<ACCEPTANCE-DATETIME>20210225082736\n",
            "ACCESSION NUMBER:\t\t0000004904-21-000010\n",
            "CONFORMED SUBMISSION TYPE:\t10-K\n",
            "PUBLIC DOCUMENT COUNT:\t\t169\n",
            "CONFORMED PERIOD OF REPORT:\t20201231\n",
            "FILED AS OF DATE:\t\t20210225\n",
            "DATE AS OF CHANGE:\t\t20210225\n",
            "\n",
            "FILER:\n",
            "\n",
            "\tCOMPANY DATA:\t\n",
            "\t\tCOMPANY CONFORMED NAME:\t\t\tAMERICAN ELECTRIC POWER CO INC\n",
            "\t\tCENTRAL INDEX KEY:\t\t\t0000004904\n",
            "\t\tSTANDARD INDUSTRIAL CLASSIFICATION:\tELECTRIC SERVICES [4911]\n",
            "\t\tIRS NUMBER:\t\t\t\t134922640\n",
            "\t\tSTATE OF INCORPORATION:\t\t\tNY\n",
            "\t\tFISCAL YEAR END:\t\t\t1231\n",
            "\n",
            "\tFILING VALUES:\n",
            "\t\tFORM TYPE:\t\t10-K\n",
            "\t\tSEC ACT:\t\t1934 Act\n",
            "\t\tSEC FILE NUMBER:\t001-03525\n",
            "\t\tFILM NUMBER:\t\t21676556\n",
            "\n",
            "\tBUSINESS ADDRESS:\t\n",
            "\t\tSTREET 1:\t\t1 RIVERSIDE PLAZA\n",
            "\t\tCITY:\t\t\tCOLUMBUS\n",
            "\t\tSTATE:\t\t\tOH\n",
            "\t\tZIP:\t\t\t43215\n",
            "\t\tBUSINESS PHONE:\t\t614-716-1000\n",
            "\n",
            "\tMAIL ADDRESS:\t\n",
            "\t\tSTREET 1:\t\t1 RIVERSIDE PLAZA\n",
            "\t\tCITY:\t\t\tCOLUMBUS\n",
            "\t\tSTATE:\t\t\tOH\n",
            "\t\tZIP:\t\t\t43215\n",
            "\n",
            "\tFORMER COMPANY:\t\n",
            "\t\tFORMER CONFORMED NAME:\tKINGSPORT UTILITIES...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6ynavVpL5_Ne"
      },
      "source": [
        "### Get Documents\n",
        "With theses fillings downloaded, we want to break them into their associated documents. These documents are sectioned off in the fillings with the tags `<DOCUMENT>` for the start of each document and `</DOCUMENT>` for the end of each document. There's no overlap with these documents, so each `</DOCUMENT>` tag should come after the `<DOCUMENT>` with no `<DOCUMENT>` tag in between.\n",
        "\n",
        "Implement `get_documents` to return a list of these documents from a filling. Make sure not to include the tag in the returned document text."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GEW2DQiA5_Ne",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "78ae8394-3621-44fa-cfe0-20b67e842903"
      },
      "source": [
        "import re\n",
        "\n",
        "\n",
        "def get_documents(text):\n",
        "    \"\"\"\n",
        "    Extract the documents from the text\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    text : str\n",
        "        The text with the document strings inside\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    extracted_docs : list of str\n",
        "        The document strings found in `text`\n",
        "    \"\"\"\n",
        "    \n",
        "    # Write regexes\n",
        "    doc_start_pattern = re.compile(r'<DOCUMENT>')\n",
        "    doc_end_pattern = re.compile(r'</DOCUMENT>')\n",
        "    \n",
        "    # Create 3 lists with the span indices for each regex\n",
        "    doc_start_is = [ w.end() for w in doc_start_pattern.finditer(text) ]\n",
        "    doc_end_is = [ w.start() for w in doc_end_pattern.finditer(text) ]\n",
        "\n",
        "    # Create a loop to go through each section of the text section in the list\n",
        "    extracted_docs = []\n",
        "    \n",
        "    for ind, section in enumerate(doc_end_is):             \n",
        "        extracted_docs.append(text[doc_start_is[ind]:doc_end_is[ind]])\n",
        " \n",
        "    return extracted_docs\n",
        "\n",
        "\n",
        "project_tests.test_get_documents(get_documents)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tests Passed\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pm66-9P95_Nf"
      },
      "source": [
        "With the `get_documents` function implemented, let's extract all the documents."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": false,
        "id": "wUk85djG5_Nf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "eeb261db-a0f2-48d0-b68f-6bb4ec091be9"
      },
      "source": [
        "filling_documents_by_ticker = {}\n",
        "\n",
        "for ticker, raw_fillings in raw_fillings_by_ticker.items():\n",
        "    filling_documents_by_ticker[ticker] = {}\n",
        "    for file_date, filling in tqdm(raw_fillings.items(), desc='Getting Documents from {} Fillings'.format(ticker), unit='filling'):\n",
        "        filling_documents_by_ticker[ticker][file_date] = get_documents(filling)\n",
        "\n",
        "\n",
        "print('\\n\\n'.join([\n",
        "    f'Document {doc_i} Filed on {file_date}:\\n{doc[:200]}...'\n",
        "    for file_date, docs in filling_documents_by_ticker[example_ticker].items()\n",
        "    for doc_i, doc in enumerate(docs)][:3]))"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Getting Documents from AEP Fillings: 100%|██████████| 24/24 [00:04<00:00,  5.73filling/s]\n",
            "Getting Documents from AXP Fillings: 100%|██████████| 25/25 [00:00<00:00, 41.69filling/s]\n",
            "Getting Documents from BA Fillings: 100%|██████████| 28/28 [00:00<00:00, 46.31filling/s]\n",
            "Getting Documents from BK Fillings: 100%|██████████| 14/14 [00:01<00:00, 12.36filling/s]\n",
            "Getting Documents from CAT Fillings: 100%|██████████| 20/20 [00:01<00:00, 10.96filling/s]\n",
            "Getting Documents from DE Fillings: 100%|██████████| 20/20 [00:01<00:00, 12.73filling/s]\n",
            "Getting Documents from DIS Fillings: 100%|██████████| 22/22 [00:00<00:00, 37.74filling/s]\n",
            "Getting Documents from DTE Fillings: 100%|██████████| 20/20 [00:01<00:00, 18.27filling/s]\n",
            "Getting Documents from ED Fillings: 100%|██████████| 22/22 [00:01<00:00, 17.77filling/s]\n",
            "Getting Documents from EMR Fillings: 100%|██████████| 27/27 [00:00<00:00, 37.48filling/s]\n",
            "Getting Documents from ETN Fillings: 100%|██████████| 9/9 [00:00<00:00,  9.36filling/s]\n",
            "Getting Documents from GE Fillings: 100%|██████████| 21/21 [00:02<00:00,  9.97filling/s]\n",
            "Getting Documents from IBM Fillings: 100%|██████████| 24/24 [00:01<00:00, 18.02filling/s]\n",
            "Getting Documents from IP Fillings: 100%|██████████| 24/24 [00:01<00:00, 18.65filling/s]\n",
            "Getting Documents from JNJ Fillings: 100%|██████████| 22/22 [00:00<00:00, 30.15filling/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Document 0 Filed on 2021-02-25:\n",
            "\n",
            "<TYPE>10-K\n",
            "<SEQUENCE>1\n",
            "<FILENAME>aep-20201231.htm\n",
            "<DESCRIPTION>10-K\n",
            "<TEXT>\n",
            "<XBRL>\n",
            "<?xml version=\"1.0\" ?><!--XBRL Document Created with Wdesk from Workiva--><!--Copyright 2021 Workiva--><!--r:df29644c...\n",
            "\n",
            "Document 1 Filed on 2021-02-25:\n",
            "\n",
            "<TYPE>EX-4.C\n",
            "<SEQUENCE>2\n",
            "<FILENAME>ex4c20204q.htm\n",
            "<DESCRIPTION>EX-4.C\n",
            "<TEXT>\n",
            "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\"><html><head>\n",
            "<!-- Do...\n",
            "\n",
            "Document 2 Filed on 2021-02-25:\n",
            "\n",
            "<TYPE>EX-4.D\n",
            "<SEQUENCE>3\n",
            "<FILENAME>ex4d20204q.htm\n",
            "<DESCRIPTION>EX-4.D\n",
            "<TEXT>\n",
            "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.01 Transitional//EN\" \"http://www.w3.org/TR/html4/loose.dtd\"><html><head>\n",
            "<!-- Do...\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TMRi4FJx5_Nf"
      },
      "source": [
        "### Get Document Types\n",
        "Now that we have all the documents, we want to find the 10-k form in this 10-k filing. Implement the `get_document_type` function to return the type of document given. The document type is located on a line with the `<TYPE>` tag. For example, a form of type \"TEST\" would have the line `<TYPE>TEST`. Make sure to return the type as lowercase, so this example would be returned as \"test\"."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l_4XEWZ55_Nf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6bd28aab-d23e-4563-eb9b-d00c1a475bb9"
      },
      "source": [
        "def get_document_type(doc):\n",
        "    \"\"\"\n",
        "    Return the document type lowercased\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    doc : str\n",
        "        The document string\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    doc_type : str\n",
        "        The document type lowercased\n",
        "    \"\"\"\n",
        "    # Write regexes\n",
        "    type_pattern = re.compile(r'<TYPE>[^\\n]+')\n",
        "    \n",
        "    # Create list with the span idices for each regex\n",
        "    doc_types = [ w for w in type_pattern.finditer(doc) ]  \n",
        "     \n",
        "    extracted_docs = doc[doc_types[0].start()+6:doc_types[0].end()].lower()\n",
        " \n",
        "    return extracted_docs \n",
        "\n",
        "\n",
        "project_tests.test_get_document_type(get_document_type)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tests Passed\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Sp_0b-2H5_Nf"
      },
      "source": [
        "With the `get_document_type` function, we'll filter out all non 10-k documents."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DT05yl1k5_Nf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a60ffa92-fc1d-4e73-9d82-947beaa82f8d"
      },
      "source": [
        "ten_ks_by_ticker = {}\n",
        "\n",
        "for ticker, filling_documents in filling_documents_by_ticker.items():\n",
        "    ten_ks_by_ticker[ticker] = []\n",
        "    print(ticker, end='')\n",
        "    counter = 0\n",
        "    for file_date, documents in filling_documents.items():\n",
        "         \n",
        "        for document in documents:\n",
        "            # print(document[:200])\n",
        "            if ((get_document_type(document) == '10-k') or (get_document_type(document) == '20-f') ):\n",
        "                 \n",
        "                ten_ks_by_ticker[ticker].append({\n",
        "                    'cik': additional_cik[ticker],\n",
        "                    'file': document,\n",
        "                    'file_date': file_date})\n",
        "                counter += 1\n",
        "    print(f\":\\t{counter} 10-k's\")\n",
        "\n",
        "# project_helper.print_ten_k_data(ten_ks_by_ticker[example_ticker][:5], ['cik', 'file', 'file_date'])"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "AEP:\t27 10-k's\n",
            "AXP:\t18 10-k's\n",
            "BA:\t20 10-k's\n",
            "BK:\t9 10-k's\n",
            "CAT:\t31 10-k's\n",
            "DE:\t18 10-k's\n",
            "DIS:\t16 10-k's\n",
            "DTE:\t12 10-k's\n",
            "ED:\t18 10-k's\n",
            "EMR:\t21 10-k's\n",
            "ETN:\t7 10-k's\n",
            "GE:\t18 10-k's\n",
            "IBM:\t18 10-k's\n",
            "IP:\t18 10-k's\n",
            "JNJ:\t17 10-k's\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XZIEZnAd5_Ng"
      },
      "source": [
        "## Preprocess the Data\n",
        "### Clean Up\n",
        "As you can see, the text for the documents are very messy. To clean this up, we'll remove the html and lowercase all the text."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i-3jhnVZ5_Ng"
      },
      "source": [
        "def remove_html_tags(text):\n",
        "    text = BeautifulSoup(text, 'html.parser').get_text()\n",
        "    \n",
        "    return text\n",
        "\n",
        "\n",
        "def clean_text(text):\n",
        "    text = text.lower()\n",
        "    text = remove_html_tags(text)\n",
        "    \n",
        "    return text"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oUtTFTMQ5_Ng"
      },
      "source": [
        "Using the `clean_text` function, we'll clean up all the documents."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eUii9C9N5_Ng",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7f681b8a-ae81-4a33-d031-2ce0d778f535"
      },
      "source": [
        "for ticker, ten_ks in ten_ks_by_ticker.items():\n",
        "    for ten_k in tqdm(ten_ks, desc=f'Cleaning {ticker} 10-Ks', unit='10-K'):\n",
        "        ten_k['file_clean'] = clean_text(ten_k['file'])\n",
        "\n",
        "\n",
        "project_helper.print_ten_k_data(ten_ks_by_ticker[example_ticker][:5], ['file_clean'])"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cleaning AEP 10-Ks: 100%|██████████| 27/27 [00:39<00:00,  1.46s/10-K]\n",
            "Cleaning AXP 10-Ks: 100%|██████████| 18/18 [00:26<00:00,  1.45s/10-K]\n",
            "Cleaning BA 10-Ks: 100%|██████████| 20/20 [00:35<00:00,  1.77s/10-K]\n",
            "Cleaning BK 10-Ks: 100%|██████████| 9/9 [00:03<00:00,  2.5110-K/s]\n",
            "Cleaning CAT 10-Ks: 100%|██████████| 31/31 [00:55<00:00,  1.78s/10-K]\n",
            "Cleaning DE 10-Ks: 100%|██████████| 18/18 [01:01<00:00,  3.44s/10-K]\n",
            "Cleaning DIS 10-Ks: 100%|██████████| 16/16 [00:32<00:00,  2.01s/10-K]\n",
            "Cleaning DTE 10-Ks: 100%|██████████| 12/12 [01:00<00:00,  5.04s/10-K]\n",
            "Cleaning ED 10-Ks: 100%|██████████| 18/18 [01:19<00:00,  4.44s/10-K]\n",
            "Cleaning EMR 10-Ks: 100%|██████████| 21/21 [00:06<00:00,  3.0310-K/s]\n",
            "Cleaning ETN 10-Ks: 100%|██████████| 7/7 [00:27<00:00,  3.99s/10-K]\n",
            "Cleaning GE 10-Ks: 100%|██████████| 18/18 [01:15<00:00,  4.22s/10-K]\n",
            "Cleaning IBM 10-Ks: 100%|██████████| 18/18 [00:04<00:00,  4.1610-K/s]\n",
            "Cleaning IP 10-Ks: 100%|██████████| 18/18 [00:43<00:00,  2.42s/10-K]\n",
            "Cleaning JNJ 10-Ks: 100%|██████████| 17/17 [00:17<00:00,  1.01s/10-K]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[\n",
            "  {\n",
            "    file_clean: '\\n10-k\\n1\\naep-20201231.htm\\n10-k\\n\\n\\n\\naep-2020...},\n",
            "  {\n",
            "    file_clean: '\\n10-k\\n1\\naep10klegal20194q.htm\\namerican electr...},\n",
            "  {\n",
            "    file_clean: '\\n10-k\\n1\\naep10klegal20184q.htm\\namerican electr...},\n",
            "  {\n",
            "    file_clean: '\\n10-k\\n1\\naep10klegal20174q.htm\\namerican electr...},\n",
            "  {\n",
            "    file_clean: '\\n10-k\\n1\\naep10klegal20164q.htm\\namerican electr...},\n",
            "]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9PcBZ8-p5_Ng"
      },
      "source": [
        "### Lemmatize\n",
        "With the text cleaned up, it's time to distill the verbs down. Implement the `lemmatize_words` function to lemmatize verbs in the list of words provided."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e9EzuFMJ5_Ng",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d3cf0f6a-9bca-4e8b-edcd-5669e1563350"
      },
      "source": [
        "from nltk.stem import WordNetLemmatizer\n",
        "from nltk.corpus import wordnet\n",
        "\n",
        "\n",
        "def lemmatize_words(words):\n",
        "    \"\"\"\n",
        "    Lemmatize words \n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    words : list of str\n",
        "        List of words\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    lemmatized_words : list of str\n",
        "        List of lemmatized words\n",
        "    \"\"\"\n",
        "    wnl = WordNetLemmatizer()\n",
        "    lemmatized_word = []\n",
        "    for word in words:\n",
        "        lemmatized_word.append(wnl.lemmatize(word, pos='v'))\n",
        "     \n",
        "    return lemmatized_word\n",
        "\n",
        "\n",
        "project_tests.test_lemmatize_words(lemmatize_words)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tests Passed\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wbwLG2Fu5_Nh"
      },
      "source": [
        "With the `lemmatize_words` function implemented, let's lemmatize all the data."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ph9lthL25_Nh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "113d61f4-4599-4961-8669-a853e8800699"
      },
      "source": [
        "word_pattern = re.compile('\\w+')\n",
        "\n",
        "for ticker, ten_ks in ten_ks_by_ticker.items():\n",
        "    for ten_k in tqdm(ten_ks, desc='Lemmatize {ticker} 10-Ks', unit='10-K'):\n",
        "        ten_k['file_lemma'] = lemmatize_words(word_pattern.findall(ten_k['file_clean']))\n",
        "\n",
        "\n",
        "project_helper.print_ten_k_data(ten_ks_by_ticker[example_ticker][:5], ['file_lemma'])"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 27/27 [00:14<00:00,  1.8310-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:04<00:00,  3.6710-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 20/20 [00:06<00:00,  3.3010-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 9/9 [00:00<00:00, 10.4210-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 31/31 [00:44<00:00,  1.44s/10-K]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:04<00:00,  4.1410-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 16/16 [00:03<00:00,  4.3010-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 12/12 [00:04<00:00,  2.9410-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:06<00:00,  2.6610-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 21/21 [00:01<00:00, 16.2910-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 7/7 [00:01<00:00,  5.0410-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:06<00:00,  2.9610-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:01<00:00, 16.4910-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 18/18 [00:04<00:00,  3.9510-K/s]\n",
            "Lemmatize {ticker} 10-Ks: 100%|██████████| 17/17 [00:01<00:00,  8.6210-K/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "[\n",
            "  {\n",
            "    file_lemma: '['10', 'k', '1', 'aep', '20201231', 'htm', '10', ...},\n",
            "  {\n",
            "    file_lemma: '['10', 'k', '1', 'aep10klegal20194q', 'htm', 'ame...},\n",
            "  {\n",
            "    file_lemma: '['10', 'k', '1', 'aep10klegal20184q', 'htm', 'ame...},\n",
            "  {\n",
            "    file_lemma: '['10', 'k', '1', 'aep10klegal20174q', 'htm', 'ame...},\n",
            "  {\n",
            "    file_lemma: '['10', 'k', '1', 'aep10klegal20164q', 'htm', 'ame...},\n",
            "]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l2pA6Pm75_Nh"
      },
      "source": [
        "### Remove Stopwords"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jDDjria35_Nh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "232fb68d-eadb-4612-eb77-4fc7da1ea96b"
      },
      "source": [
        "from nltk.corpus import stopwords\n",
        "\n",
        "\n",
        "lemma_english_stopwords = lemmatize_words(stopwords.words('english'))\n",
        "\n",
        "for ticker, ten_ks in ten_ks_by_ticker.items():\n",
        "    for ten_k in tqdm(ten_ks, desc='Remove Stop Words for {ticker} 10-Ks', unit='10-K'):\n",
        "        ten_k['file_lemma'] = [word for word in ten_k['file_lemma'] if word not in lemma_english_stopwords]\n",
        "\n",
        "\n",
        "print('\\nStop Words Removed')"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 27/27 [00:09<00:00,  2.7610-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:02<00:00,  7.4610-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 20/20 [00:02<00:00,  6.7010-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 9/9 [00:00<00:00, 19.2410-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 31/31 [00:32<00:00,  1.05s/10-K]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:02<00:00,  7.7310-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 16/16 [00:01<00:00,  8.2310-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 12/12 [00:02<00:00,  5.5610-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:03<00:00,  4.9610-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 21/21 [00:00<00:00, 30.7810-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 7/7 [00:00<00:00,  9.6310-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:03<00:00,  5.6910-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:00<00:00, 31.5810-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 18/18 [00:02<00:00,  7.4710-K/s]\n",
            "Remove Stop Words for {ticker} 10-Ks: 100%|██████████| 17/17 [00:01<00:00, 16.0210-K/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Stop Words Removed\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WsamuaHX5_Ni"
      },
      "source": [
        "## Analysis on 10ks\n",
        "### Loughran McDonald Sentiment Word Lists\n",
        "We'll be using the Loughran and McDonald sentiment word lists. These word lists cover the following sentiment:\n",
        "- Negative \n",
        "- Positive\n",
        "- Uncertainty\n",
        "- Litigious\n",
        "- Constraining\n",
        "- Superfluous\n",
        "- Modal\n",
        "\n",
        "This will allow us to do the sentiment analysis on the 10-ks. Let's first load these word lists. We'll be looking into a few of these sentiments."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v3OFOKZf5_Ni",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "875fad7d-03b3-4014-9135-c376ce4c3bda"
      },
      "source": [
        "import os\n",
        "\n",
        "\n",
        "sentiments = ['negative', 'positive', 'uncertainty', 'litigious', 'constraining', 'interesting']\n",
        "\n",
        "sentiment_df = pd.read_csv(os.path.join( 'LoughranMcDonald_MasterDictionary_2018.csv'))\n",
        "sentiment_df.columns = [column.lower() for column in sentiment_df.columns] # Lowercase the columns for ease of use\n",
        "\n",
        "# Remove unused information\n",
        "sentiment_df = sentiment_df[sentiments + ['word']]\n",
        "sentiment_df[sentiments] = sentiment_df[sentiments].astype(bool)\n",
        "sentiment_df = sentiment_df[(sentiment_df[sentiments]).any(1)]\n",
        "\n",
        "# # Apply the same preprocessing to these words as the 10-k words\n",
        "sentiment_df['word'] = lemmatize_words(sentiment_df['word'].str.lower())\n",
        "sentiment_df = sentiment_df.drop_duplicates('word')\n",
        "\n",
        "sentiment_df"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>negative</th>\n",
              "      <th>positive</th>\n",
              "      <th>uncertainty</th>\n",
              "      <th>litigious</th>\n",
              "      <th>constraining</th>\n",
              "      <th>interesting</th>\n",
              "      <th>word</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>abandon</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>abandonment</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>abandonments</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>abdicate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>54</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>abdication</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86035</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>wrongdoing</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86036</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>wrongdoings</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86040</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>wrongful</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86041</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>wrongfully</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86049</th>\n",
              "      <td>True</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>False</td>\n",
              "      <td>wrongly</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>2720 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       negative  positive  uncertainty  ...  constraining  interesting          word\n",
              "9          True     False        False  ...         False        False       abandon\n",
              "12         True     False        False  ...         False        False   abandonment\n",
              "13         True     False        False  ...         False        False  abandonments\n",
              "51         True     False        False  ...         False        False      abdicate\n",
              "54         True     False        False  ...         False        False    abdication\n",
              "...         ...       ...          ...  ...           ...          ...           ...\n",
              "86035      True     False        False  ...         False        False    wrongdoing\n",
              "86036      True     False        False  ...         False        False   wrongdoings\n",
              "86040      True     False        False  ...         False        False      wrongful\n",
              "86041      True     False        False  ...         False        False    wrongfully\n",
              "86049      True     False        False  ...         False        False       wrongly\n",
              "\n",
              "[2720 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DLl7w9025_Ni"
      },
      "source": [
        "### Bag of Words\n",
        "using the sentiment word lists, let's generate sentiment bag of words from the 10-k documents. Implement `get_bag_of_words` to generate a bag of words that counts the number of sentiment words in each doc. You can ignore words that are not in `sentiment_words`."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M6y9W7j95_Ni"
      },
      "source": [
        "from collections import defaultdict, Counter\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "\n",
        "\n",
        "def get_bag_of_words(sentiment_words, docs):\n",
        "    \"\"\"\n",
        "    Generate a bag of words from documents for a certain sentiment\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    sentiment_words: Pandas Series\n",
        "        Words that signify a certain sentiment\n",
        "    docs : list of str\n",
        "        List of documents used to generate bag of words\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    bag_of_words : 2-d Numpy Ndarray of int\n",
        "        Bag of words sentiment for each document\n",
        "        The first dimension is the document.\n",
        "        The second dimension is the word.\n",
        "    \"\"\"\n",
        "    \n",
        "    # TODO: Implement\n",
        "\n",
        "    bag_of_words = np.zeros((len(docs), len(sentiment_words)))\n",
        "    for ind, word  in enumerate(docs):\n",
        "        for sen_id, sent in enumerate(sentiment_words):\n",
        "             #bag_of_words[ind, sen_id] = word.count(sent) \n",
        "            if sent in word:\n",
        "                bag_of_words[ind, sen_id] = 1 \n",
        "    print(bag_of_words)         \n",
        "    return bag_of_words\n",
        "\n",
        "\n",
        "# project_tests.test_get_bag_of_words(get_bag_of_words)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iZfxwZnx5_Ni"
      },
      "source": [
        "Using the `get_bag_of_words` function, we'll generate a bag of words for all the documents."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "scrolled": true,
        "id": "ZqJoT6Sa5_Nj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d0ab5061-3cd6-48a0-dc86-0d7b897c21c1"
      },
      "source": [
        "sentiment_bow_ten_ks = {}\n",
        "\n",
        "for ticker, ten_ks in ten_ks_by_ticker.items():\n",
        "    lemma_docs = [' '.join(ten_k['file_lemma']) for ten_k in ten_ks]\n",
        "    \n",
        "    sentiment_bow_ten_ks[ticker] = {\n",
        "        sentiment: get_bag_of_words(sentiment_df[sentiment_df[sentiment]]['word'], lemma_docs)\n",
        "        for sentiment in sentiments}\n",
        "\n",
        "\n",
        "# project_helper.print_ten_k_data([sentiment_bow_ten_ks[example_ticker]], sentiments)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[1. 1. 1. ... 0. 0. 0.]\n",
            " [1. 1. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 1. ... 0. 0. 1.]\n",
            " [1. 0. 1. ... 0. 0. 1.]\n",
            " [1. 0. 1. ... 0. 0. 1.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[1. 1. 1. ... 0. 1. 1.]\n",
            " [0. 0. 1. ... 0. 1. 1.]\n",
            " [0. 0. 1. ... 0. 1. 1.]\n",
            " ...\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 1.]\n",
            " [0. 0. 0. ... 1. 1. 1.]\n",
            " [0. 0. 0. ... 1. 1. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 1. ... 1. 0. 0.]\n",
            " [0. 1. 0. ... 1. 0. 0.]\n",
            " [0. 1. 0. ... 1. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 1. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 1. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 1. 1. 0.]]\n",
            "[[0. 1. 0. ... 0. 0. 1.]\n",
            " [0. 1. 0. ... 0. 0. 1.]\n",
            " [0. 1. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 1. 0. ... 0. 0. 0.]\n",
            " [1. 1. 0. ... 0. 0. 0.]\n",
            " [1. 1. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 1. 1. 0.]\n",
            " [0. 0. 0. ... 1. 1. 0.]]\n",
            "[[0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 0. 0. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 1. 1. 0. 1.\n",
            "  1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1.\n",
            "  1. 1. 0. 1. 0. 0. 1. 1. 1. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 0. 1. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 1.\n",
            "  0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 1. 1. 0. 1. 0. 0. 0. 0. 0. 1.\n",
            "  1. 0. 0. 1. 0. 0. 0. 1. 1. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0.]]\n",
            "[[0. 0. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 1. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 1. 1. 1.]\n",
            " ...\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 1. 1. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 1. ... 1. 1. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 1. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 1. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 1. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 1. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 1. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 1. 0. 1. 0. 0. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " [0. 0. 0. ... 1. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [1. 0. 0. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 0. ... 0. 1. 1.]\n",
            " [0. 1. 0. ... 0. 1. 1.]\n",
            " [0. 1. 0. ... 0. 1. 1.]\n",
            " ...\n",
            " [0. 1. 0. ... 0. 1. 0.]\n",
            " [0. 1. 0. ... 0. 1. 1.]\n",
            " [0. 1. 0. ... 0. 1. 0.]]\n",
            "[[1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 0. 1. 1. 1. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0.\n",
            "  0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [1. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 1. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]\n",
            " [1. 1. 0. 0. 1. 1. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 1. ... 1. 0. 1.]\n",
            " [0. 0. 1. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 1. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 1. ... 1. 1. 1.]\n",
            " [0. 0. 1. ... 1. 1. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[1. 0. 1. 1. 1. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 0. 0. 1. 1. 1. 0. 0. 1. 0. 0. 1. 1. 1. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 1. 1. 1. 1. 0. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [1. 0. 1. 1. 1. 1. 0. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 1. 1. 1. 1. 0. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  0. 1. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]\n",
            " [0. 1. 1. 1. 1. 1. 0. 1. 0. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1.\n",
            "  1. 1. 0. 0. 1. 0. 0. 1. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 1. 1. 0. 0. 0.\n",
            "  0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.]]\n",
            "[[0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0.\n",
            "  0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]\n",
            " [0. 0. 0. 1. 1. 1. 0. 1. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
            "  0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1. 0. 1. 0.\n",
            "  0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 1. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 1. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 0. ... 0. 1. 1.]\n",
            " [0. 1. 0. ... 0. 0. 1.]\n",
            " [0. 1. 1. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 1. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]\n",
            " [0. 0. 0. ... 0. 0. 1.]]\n",
            "[[1. 1. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 1. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " [0. 0. 1. ... 0. 0. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " [0. 1. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " ...\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]\n",
            " [1. 0. 0. ... 0. 0. 0.]]\n",
            "[[0. 0. 1. ... 0. 1. 1.]\n",
            " [0. 0. 1. ... 0. 1. 1.]\n",
            " [0. 0. 1. ... 0. 1. 1.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 1. ... 0. 0. 0.]]\n",
            "[[0. 0. 0. ... 1. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]\n",
            " [0. 0. 0. ... 0. 1. 0.]]\n",
            "[[0. 1. 0. ... 1. 1. 0.]\n",
            " [0. 1. 0. ... 1. 1. 0.]\n",
            " [0. 1. 0. ... 1. 1. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n",
            "[[1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " [1. 0. 0. ... 1. 0. 0.]\n",
            " ...\n",
            " [0. 0. 0. ... 0. 0. 0.]\n",
            " [0. 0. 0. ... 1. 0. 0.]\n",
            " [0. 0. 0. ... 0. 0. 0.]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wrvzvsSd5_Nj"
      },
      "source": [
        "### Jaccard Similarity\n",
        "Using the bag of words, let's calculate the jaccard similarity on the bag of words and plot it over time. Implement `get_jaccard_similarity` to return the jaccard similarities between each tick in time. Since the input, `bag_of_words_matrix`, is a bag of words for each time period in order, you just need to compute the jaccard similarities for each neighboring bag of words. Make sure to turn the bag of words into a boolean array when calculating the jaccard similarity."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5i9bKjh25_Nj"
      },
      "source": [
        "\n",
        "\n",
        ">Jaccard Similarity\n",
        "\n",
        ">Now that we have the bag of words, we can convert it into a boolean array and calculate the jaccard similarity. The jaccard similarity is defined as the size of intersection divided by the size of union of two sets. For example, the jaccard similarity between two sentences is the number of common words between the two sentences divided by the total number of unique words in both sentences combined. The closer the jaccard similarity value is to 1, the more similar the sets are. To more easily understand our calculations, we plot the jaccard similarities.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EJroyGZd5_Nj"
      },
      "source": [
        "from sklearn.metrics import jaccard_score\n",
        "\n",
        "\n",
        "def get_jaccard_similarity(bag_of_words_matrix):\n",
        "    \"\"\"\n",
        "    Get jaccard similarities for neighboring documents\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    bag_of_words : 2-d Numpy Ndarray of int\n",
        "        Bag of words sentiment for each document\n",
        "        The first dimension is the document.\n",
        "        The second dimension is the word.\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    jaccard_similarities : list of float\n",
        "        Jaccard similarities for neighboring documents\n",
        "    \"\"\"\n",
        "    \n",
        "    # TODO: Implement\n",
        "    bag_of_words = bag_of_words_matrix.astype(bool)\n",
        "    \n",
        "    jaccard_similarities = []\n",
        "    \n",
        "    len_ = len(bag_of_words)\n",
        "\n",
        "    for in_, v in enumerate(bag_of_words[0:len_-1]):\n",
        "    \n",
        "        u=bag_of_words[in_+1]\n",
        "  \n",
        "        jaccard_similarities.append(jaccard_score(v,u))\n",
        "\n",
        "    return jaccard_similarities\n",
        "\n",
        "   \n",
        "\n",
        "\n",
        "# project_tests.test_get_jaccard_similarity(get_jaccard_similarity)"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bYFuzVvK5_Nj"
      },
      "source": [
        "Using the `get_jaccard_similarity` function, let's plot the similarities over time."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nOPJ4j8M5_Nj",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 492
        },
        "outputId": "425da263-4163-40ce-d1cb-25109dd6c2c8"
      },
      "source": [
        "# Get dates for the universe\n",
        "file_dates = {\n",
        "    ticker: [ten_k['file_date'] for ten_k in ten_ks]\n",
        "    for ticker, ten_ks in ten_ks_by_ticker.items()}  \n",
        "\n",
        "jaccard_similarities = {\n",
        "    ticker: {\n",
        "        sentiment_name: get_jaccard_similarity(sentiment_values)\n",
        "        for sentiment_name, sentiment_values in ten_k_sentiments.items()}\n",
        "    for ticker, ten_k_sentiments in sentiment_bow_ten_ks.items()}\n",
        "\n",
        "\n",
        "project_helper.plot_similarities(\n",
        "    [jaccard_similarities[example_ticker][sentiment] for sentiment in sentiments],\n",
        "    file_dates[example_ticker][1:],\n",
        "    'Jaccard Similarities for {} Sentiment'.format(example_ticker),\n",
        "    sentiments)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HxpGawqW5_Nk"
      },
      "source": [
        "### TFIDF\n",
        "using the sentiment word lists, let's generate sentiment TFIDF from the 10-k documents. Implement `get_tfidf` to generate TFIDF from each document, using sentiment words as the terms. You can ignore words that are not in `sentiment_words`.\n",
        "\n",
        ">TFIDF\n",
        "\n",
        ">From the sentiment word lists, let’s generate sentiment term frequency–inverse document frequency (TFIDF) from the 10-k documents. TFIDF is an information retrieval technique used to reveal how often a word/term appears in the chosen collection of text. Each term is assigned a term frequency (TF) and inverse document frequency (IDF) score. The product of these scores is referred to as the TFIDF weight of the term. Higher TFIDF weights indicate rarer terms and lower TFIDF scores indicate more common terms."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VcnmBBM35_Nm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "984151d1-62e6-49b3-e204-3a4522992459"
      },
      "source": [
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "\n",
        "\n",
        "def get_tfidf(sentiment_words, docs):\n",
        "    \"\"\"\n",
        "    Generate TFIDF values from documents for a certain sentiment\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    sentiment_words: Pandas Series\n",
        "        Words that signify a certain sentiment\n",
        "    docs : list of str\n",
        "        List of documents used to generate bag of words\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    tfidf : 2-d Numpy Ndarray of float\n",
        "        TFIDF sentiment for each document\n",
        "        The first dimension is the document.\n",
        "        The second dimension is the word.\n",
        "    \"\"\"\n",
        "    \n",
        "    # TODO: Implement\n",
        " \n",
        "    vectorizer = TfidfVectorizer( vocabulary = sentiment_words)\n",
        " \n",
        "    tfidf  = vectorizer.fit_transform (docs).toarray()\n",
        " \n",
        "    return tfidf\n",
        "\n",
        "\n",
        "project_tests.test_get_tfidf(get_tfidf)"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tests Passed\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JHNhYlWt5_Nm"
      },
      "source": [
        "Using the `get_tfidf` function, let's generate the TFIDF values for all the documents."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "41-nPlGs5_Nm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dca3f492-1167-4b3d-d97e-6d6021b9f53a"
      },
      "source": [
        "sentiment_tfidf_ten_ks = {}\n",
        "\n",
        "for ticker, ten_ks in ten_ks_by_ticker.items():\n",
        "    lemma_docs = [' '.join(ten_k['file_lemma']) for ten_k in ten_ks]\n",
        "    \n",
        "    sentiment_tfidf_ten_ks[ticker] = {\n",
        "        sentiment: get_tfidf(sentiment_df[sentiment_df[sentiment]]['word'], lemma_docs)\n",
        "        for sentiment in sentiments}\n",
        "\n",
        "    \n",
        "project_helper.print_ten_k_data([sentiment_tfidf_ten_ks[example_ticker]], sentiments)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[\n",
            "  {\n",
            "    negative: '[[0.01096464 0.01233952 0.00616976 ... 0.        ...\n",
            "    positive: '[[0.02451587 0.         0.00468565 ... 0.        ...\n",
            "    uncertainty: '[[0.         0.         0.00285177 ... 0.00445166...\n",
            "    litigious: '[[0.         0.         0.         ... 0.        ...\n",
            "    constraining: '[[0.0089781  0.         0.         ... 0.        ...\n",
            "    interesting: '[[0.         0.         0.         ... 0.        ...},\n",
            "]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I4Vlddmz5_Nn"
      },
      "source": [
        "### Cosine Similarity\n",
        "Using the TFIDF values, we'll calculate the cosine similarity and plot it over time. Implement `get_cosine_similarity` to return the cosine similarities between each tick in time. Since the input, `tfidf_matrix`, is a TFIDF vector for each time period in order, you just need to computer the cosine similarities for each neighboring vector."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z-21LZ065_Nn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0c8de7a8-c7ac-4fae-bfda-d5745a2c6b54"
      },
      "source": [
        "from sklearn.metrics.pairwise import cosine_similarity\n",
        "\n",
        "\n",
        "def get_cosine_similarity(tfidf_matrix):\n",
        "    \"\"\"\n",
        "    Get cosine similarities for each neighboring TFIDF vector/document\n",
        "\n",
        "    Parameters\n",
        "    ----------\n",
        "    tfidf : 2-d Numpy Ndarray of float\n",
        "        TFIDF sentiment for each document\n",
        "        The first dimension is the document.\n",
        "        The second dimension is the word.\n",
        "\n",
        "    Returns\n",
        "    -------\n",
        "    cosine_similarities : list of float\n",
        "        Cosine similarities for neighboring documents\n",
        "    \"\"\"\n",
        "\n",
        "    cosine_similarities= cosine_similarity(tfidf_matrix[0:],tfidf_matrix[1:])[0].tolist()\n",
        "       \n",
        "    return cosine_similarities \n",
        "    \n",
        "project_tests.test_get_cosine_similarity(get_cosine_similarity)"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Tests Passed\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DgoxY7z85_Nn"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "def plot_similarities1(similarities_list, dates, title, labels):\n",
        "    assert len(similarities_list) == len(labels)\n",
        "\n",
        "    plt.figure(1, figsize=(10, 7))\n",
        "    for similarities, label in zip(similarities_list, labels):\n",
        "        plt.title(title)\n",
        "        plt.plot(dates, similarities, label=label)\n",
        "        plt.legend()\n",
        "        plt.xticks(rotation=90)\n",
        "\n",
        "    plt.show()"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QsUsjQkh5_Nn"
      },
      "source": [
        "Let's plot the cosine similarities over time."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FGzkgxI95_Nn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 492
        },
        "outputId": "76ab32c7-08d5-4ce7-a01b-a47b0047d8d7"
      },
      "source": [
        "cosine_similarities = {\n",
        "    ticker: {\n",
        "        sentiment_name: get_cosine_similarity(sentiment_values)\n",
        "        for sentiment_name, sentiment_values in ten_k_sentiments.items()}\n",
        "    for ticker, ten_k_sentiments in sentiment_tfidf_ten_ks.items()}\n",
        "\n",
        "\n",
        "plot_similarities1(\n",
        "    [cosine_similarities[example_ticker][sentiment] for sentiment in sentiments],\n",
        "    file_dates[example_ticker][1:],\n",
        "    f'Cosine Similarities for {example_ticker} Sentiment',\n",
        "    sentiments)"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlMAAAHbCAYAAADvSh/AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOydd3yb1fWHn6thS7a894oT21l2trMTssPeEFaYbYFSKB1Ax49CKS2lpUBpKS0tM9CwGlpmAgkZJGGExNl2BonjbM94yZZsjfv745Vsx3Gm/Uq2cx9/9JH0rnP0Stb71TnnniuklCgUCoVCoVAozgxDsB1QKBQKhUKh6MkoMaVQKBQKhULRCZSYUigUCoVCoegESkwpFAqFQqFQdAIlphQKhUKhUCg6gRJTCoVCoVAoFJ1AiSmFIoAIIQqFENMCbLOPEMIuhDCe4f52IUSW7/GrQojfdcKXRUKIW06w/nkhxENnevzjHNMqhPhQCFErhPhPVx67NyGEOEcIsSPYfigUPRElphSK4yCEuEEIsc4nJg77hMDkzhxTSpknpVzRRS62IIRIF0K8K4So9ImGrUKIW30290kpbVJKz5kc27dvcVf4KaW8QEo5z+fzrUKI1e3Wf19K+duusNWGq4EkIE5KOaerDiqE6CeE8Aoh/tHBOimEaPB9dvy3n/nWPSKEcPmW1QghvhRCTDiOjWghxMtCiFIhRL0QYqcQ4hdd5L8UQuT4n0spV0kpB3bFsU/Tj74+X0yBtq1QdBVKTCkUHSCE+CnwDPB7tAtxH+DvwGXB9OsEvA7sBzKBOOAmoCyoHrVBaATr+yYT2CmldJ/ujie5wN8MVAPXCiFCO1g/3CdE/bcn2qx7W0ppAxKA1cB/hRCig2P8GbABg4Eo4FJg1+m+DoVCoS9KTCkU7RBCRAGPAndLKf8rpWyQUrqklB9KKR/wbRMqhHhGCHHId3vGf0EVQsQLIT7yRR2OCCFW+YWEEKJECDHL9/gRIcQ7QojXfFGHQiHE6DZ+pPqiTRVCiD1CiHtP4PYY4FWfr24p5QYp5SLfcY765S+EWCGE+J0vImL3pcDihBDzhRB1Qoi1Qoi+bfw4KoLRZnmM73VWCCGqfY/T26xfIYR4TAjxBdAIZPmWfU8IMRh4Hpjgj9D49jkqjSiEuFgIsbFNBGdYm3U/F0Ic9J27HUKImR34+BvgYTTBYxdCfFcIYRBC/EoIsVcIUe47/1HtztV3hRD7gGUdnWyf8LkZ+BXgAi45wXtzXKSULmAekIwmgtszBnhDSlktpfRKKbdLKRe08WOQEGKJ73O2QwhxTZt1rwohnhNCfOw7R2uEENm+dSt9m23ynZdrhRDThBAH2uxfIoR4QAixWWhRtpeEEElCi9DWCyE+E0LEtNl+vO89qhFCbBJt0tm+9/23QogvfPsuFkLE+1b7fanx+dJhlE6h6M4oMaVQHMsEwAL87wTbPAiMB0YAw4GxaBdWgPuAA2hRhyTg/4Djzdt0KfAWEA18APwNwCe+PgQ2AWnATODHQojzjnOcr4HnhBDXCSH6nPwlch1a9CoNyAa+Al4BYoFtwK9P4RgG3z6ZaJE7h9//NtwE3AFEAHv9C6WU24DvA1/5ojbR7Q8uhBgJvAzciSY0/gl84BOyA4F7gDFSygjgPKCk/TGklL9Giy6+7bPzEnCr7zYdyEKL/LT3eypaNOh453sykI723r0DHLcO7ET4BPitwH4pZWUHm3wNPCaEuE0I0b/dvuHAEuANIBHtPf27ECK3zWbXAb8BYtAiWo8BSCmn+Nb7o2dvH8fFq4DZwAA0wbgI7fOcgPb+3+vzJQ34GPgd2mfofuBdIURCm2PdANzm8zXEtw2A35dony9fHccXhaLbosSUQnEscUDlSdJCc4FHpZTlUsoKtAvWTb51LiAFyPRFtFbJ40+CuVpKudBXz/Q6mjADLSKRIKV8VErZ7KtZegHt4tgRc4BVwEPAHl80Z8wJ/H9FSrlbSlmLdoHcLaX8zPea/wOMPMG+AEgpq6SU70opG6WU9WgX6qntNntVSlnoi5a5TnbMdtwB/FNKuUZK6fHVWjWhiVgPEArkCiHMUsoSKeXuUzzuXOBpKWWxlNIO/BK4Thyd0nvEF+VzHOcYtwCLpJTVaGLmfCFEYrtt1vuiNP5bW2F2jS8atx/IB644jp0fAvPRhGOREGKXEOIC37qLgRIp5Sv+aCTwLtpnwc//pJTf+N7X+Wji/3R4VkpZJqU8iPb5WuOLejrRfmz4Pyc3Agt9n2WvlHIJsA64sM2xXpFS7vSd03fOwBeFotuixJRCcSxVQLw4cb1MKm0iLb7Hqb7Hf0KLAiwWQhSLExcMl7Z53AhYfHYzgdS2F2O0iEBSRwfxpYF+IaXM822zEXjPl47qiLb1VI4OnttO4DMAQogwIcQ/femyOrR0TbQ4etTg/pMd5wRkAve1OwcZQKqUchfwY+ARoFwI8ZYQIvUEx2pLR++diaPP7XH9FkJY0QTLfABfJGUfWuSlLaOklNFtbp+2WfeOb1milHKGlLKgI1tSSoeU8vdSynw0kf8O8B8hRCza+RnX7vzMRUsZ+mn/+Trp+9qOU/2cZAJz2vkyGe1HRVf5olB0W5SYUiiO5Su0CMjlJ9jmENoFxE8f3zKklPVSyvuklFloabyfdlTPcxL2A3vaXYwjpJQXnmxHX7roSTTREHuadk+H+4CBwDgpZSSt6Zq2Au54EbmTrQPtHDzW7hyESSnfBJBSviGlnIz2Pkjgj6fod0fvnZujhcKJfLsCiERLqZUKIUrR0qVnlOo7VaSUdWgpy3CgH9r5+bzd+bFJKe/S04/jsB94vZ0v4VLKP5zCvif7HCgU3R4lphSKdvhSXw+j1SBd7ovAmIUQFwgh/COy3gR+JYRI8BXSPgz8G1qKpnN8UaFatJSU9zTd+AaoF1qRtVUIYRRCDDle6k4I8UffepMQIgK4C9glpaw67RNw6kSgRSdqfJGSU6mzaksZkC6ECDnO+heA7wshxgmNcCHERUKICCHEQCHEDF/NkdPnx6me4zeBnwittYGN1pqqUx3tdwtaLddQtFTVCGASMFwIMfQUj3FKCCEeEkKMEUKECCEswI+AGmAH8BEwQAhxk+/zafZtO/gUD1+GVjPWFfwbuEQIcZ7vs2oRWkF7+kn3hAq0966rfFEoAo4SUwpFB0gpnwJ+ilZUXoH2y/se4D3fJr9DqwnZDGwB1vuWAfQHPgPsaFGuv0spl5+mfQ9aTcwIYA9QCbyINjy+I8LQalhqgGK0yMulp2PzDHgGsPp8+xr45DT3XwYUAqVCiGOKr6WU64Db0YrDq9FSp7f6VocCf/DZLkUrav7lKdp9Ga0+bSXauXWi1SadFF+h9UzgGSllaZtbAdrrbxud8o+U89+eOUX/2iLRivwr0SJqs4GLpJR2X53auWh1dIfQzsMf0c7NqfAIMM+XlrvmZBuf0Ekp96O1Dfk/Wv9fHuAUrjFSyka0ersvfL6M74wvCkUwEMevi1UoFAqFQqFQnAwVmVIoFAqFQqHoBEpMKRQKhUKhUHQCJaYUCoVCoVAoOoESUwqFQqFQKBSdIGizdMfHx8u+ffsGy7xCoVAoFArFKVNQUFAppUzoaF3QxFTfvn1Zt25dsMwrFAqFQqFQnDJCiL3HW6fSfAqFQqFQKBSdQIkphUKhUCgUik6gxJRCoVAoFApFJ1BiSqFQKBQKhaITKDGlUCgUCoVC0QmUmFIoFAqFQqHoBEpMKRQKhUKhUHQCJaYUCoVCoVAoOoESUwqFQqFQKBSdQIkphUKhUCgUik6gxJRCoVAoFApFJ1BiSqFQKBQKhaITKDGlUCgUCoVC0QmUmFIoFAqFQqHoBEpMKYKO1+uhsa422G4oFAqFQnFGmILtgOLspv5IJR8+/TiHv91BfEYm2aPHkZ0/juTs/giD0voKhUKh6P4oMaUIGge2F/Lh04/jcjoZe9nVHP52B9+8v4A1/3uHsKhoskaNJXv0ODKHDsccagm2uwqFQqFQdIgSU4qAI6Vk4+KPWTHvBaISk5jz0GPEZ2QC4LDXU7KxgN3r1rDz69VsXb4YkzmEPsNGkJ0/juz8sYRHxwT5FSgUCoVC0YqQUgbF8OjRo+W6deuCYlsRPFzNTSx98e8Ufr6UrFFjuOCe+7CE2zrc1uN2caCokN0Fa9hdsIa6inIAknMGaMJq9DjiMzIRQgTyJSgUCoXiLEQIUSClHN3hOiWmFIGirrKcD576PWXFu5hw9fVMuOr6U66LklJSuX8vu9dpwqp0104AIhOSyM4fS3b+ONJz8zCazHq+BEUvxOPxsndLFVkjEoLtikKh6MYoMdUJvF6Ju8mDq9mDq8mDu9mDu9mLq6n1uavJQ4jFROqAaMKjQoPtcrdk39bNfPTMH/C43Vxwz33kjB7XqePZq49QvH4tuwvWsG/zRtyuZkKsYfQbkU92/lj6jhyN1RbRRd4rejO7Csr59IWt3PDIOGKSw4PtjkKh6KacSEz12pqpxrpmKvbXtwghd5MHV5O39XGbe1eTt0UU+e+19V48bu9p2Y1OCiNtYAxpA6JJGxBDWGSITq+wZyClpODj91g5/xViUtK47P4HiU1N7/RxbTGxDJt5HsNmnoerycneLZsoLljD7oJv2PHVKoTBQPqgvJbRgdHJKV3wahS9kca6JgC8nuD8sFQoFD2fXiumSotrWfT8lg7XGc0GzCFGTKHavTnUiCnESFhkCKYQI+ZQ/3ptudm3zBRqbFneup8Bc6iRxrpmDu6o4eDOanauKaVw5UEAYlPDSRsYQ/qAGFIHRGMJP3vSUK4mJ4v/+Szbv/ic/mMncv4PfkyINazL7ZhDLeSMHkfO6HFIr5fS3d9qdVbr1rDitRdZ8dqLxKZltAirlP4DMBiMXe6HomfisLuC7YJCoejh9No0n7PBRU1ZY4s4MrURTQaDvgXLXo+X8n31HNxRzcEd1RzeXYu72QsC4tNtpA2IIW1gDKn9owm19k49W1NWygdP/o6K/XuZfO1NjL18TlAKxWvLS9ld8A27163hwLateD0eskeP47L7f6UK188AKSXSK0EIhKBXnMPP39zB1s8Pct1DY4lL63gwhOL4SK/k24IycvKTdP9uVSiCyVmZ5rOEm0nOigqKbYPRQHK/KJL7RZF/fl88bi9lJXWauNpZzdbPD7Jp6X6EgIQ+EVpacGAMKdlRhFh6/ltSsmk9H//lCSSSK3/xCP1G5AfNl6jEZEZdcCmjLrgUZ4Odgo/f5+t332Tj4o8Zed7FQfOrp+D1eKnYb+fwrhoO76rl8O4aHPXtIjkCBLQILHwiSwAYfPf+Zf71vp205z5hBgj/xViA0Whg6g0DyRgcq+trdPbSyFTJlkqik8KITuz6aHBbDu+uZclLRYRHhpI2ULUtUZyd9Pwrdw/AaDKQmhNNak40Yy7qh9vloay4jgM7tcjVpqX72bB4HwaDILFvREvkKjk7CnNIz0lHSSn55v0FrH7rNeIzMrnsvge7Va2SJdzGxDk3ULp7Jytff5k+Q4YTl5YRbLe6Fa4mD2V7ajm0q5bDu2oo3VOHu8kDQGS8hT55cUQlWAGQUnvP8QW3pVdqD6X0rWt9jARJm8e+/aT073P0Mq9XsnNNGWUldbqLqd6a5lv8UiG5E1OZfE1/Xe001jUDquZMcXajxFQQMJmNLdEoLtEuYKW7a1vE1frF+yj4ZC8GkyCpbyTpvm2T+0VhNHfPKVaaHY18+o+/sHPNFwycOIXz7rwXs6X7dS0XQnDe93/EvAfuYeGzT3LD7548q9spOOzNWsRpVw2HdtVSua8er1eCgLg0G4MnpJCSE0VKdjS2mMCNVPW4vexcUxYQW70xMuVu9uByevAGoIzDaW/W3YZC0d1RYqobYA41kpEbS0au9gu82enm8K5aDvhqrtYtLGHtxyUYzQaSs6JI6BNBdKKVqMQwohOthEeFtqZHgkD14YO8/+RjHDl4gKk3fof8i6/o1rU0tphYzr39Hj54+vd8teBNJl93c7BdCghSSuqrnC3C6fCuGqpLGwEteprYN4KR5/YhJSea5OyoXlvP155AioENi/dRVlLH+XcM0dVOIKNtgbT17doyCj7dy7hLs+g3LD5gdhWKk3F2fFv2MEIsJjKHxJE5JA6ApkYXh76t0UYLflvNluUHjmrZYDIbiGojrqISwohKtBKdGEZYVIiuwmZ3wTcs+ttTCKORqx58lMyhI3Sz1ZX0HzeRvGmz+Oa9BfQbMZq0QbnBdqnLkV5J1aEGn3jSap4aarQ2AKFhJpKzoxg4PpnUnGgSMyO7bdRTT6SUOBoCJwYO7aqh6qBddzuBjLYF0tbBndVUHbCz8O+byRqRwORr+hMR2/0i4IqzDyWmegChYWb6DU+g33CtQ7PXK7FXO6ktd1Bb3kiN7/7IoQZKNlceVbtgCjUSlWA9KpIVlRhGVIKVsMgzF1rS6+Xr/77Nl/+ZT2LfbC67/0EiExK75PUGihm33sGBoi0seu4pbvrjs4SG6Vuoqzcel5eyvXVtisVraXa4AQiPDiU1J4qUnGhS+0cTmxIe1Ghmd8HV5MHrDlytT6CiYI4ARtsCGZly2l1EJVrJnZTK2o/28OZvjjD2kn4Mm56OwXj2/RhQdB+UmOqBGAyCyDgrkXHWY4pzvV6J/YiTmvJGassdLfeVB+zs2Vip1cP4MFv8Qqs1kuUXXBab+bhCq6mxgUXPPc3udWvInTKDWbffjTmk53V+D7GGccHd9/H2I79g+av/4vwf/DjYLp0S7mYPtZUOTUxXaLcjh+yUl9S3RCxjUsLJGZ1Iak40KdlRRMRZunXqNVgEul4qUMIjsJGpwAq38KhQRp2XSU5+Iivf2skXC3axY00p024YRFK/yID5olC0RYmpXobBIIiMtxIZb4V2mSuvx0tdlVO7ALeJaJXvq2f3hgqtf5CPEKuJqAQrtphQwqN9t6hQvK5KvvzPX7EfKWf6rXcy8vyLe/RFOm1QLmMvn8Oa/71Ndv5Y+o+bGGyXAK1uTnufHNRWNLZ57GhJ1fkJDTcRkxTG0OnppGRHkZIThdV2dnfeP1UCPZLPaXcREoBatN5aM+Wwu4hN1iLIkfFWLrp7GMUbKlj19k4WPLGOIeekMf7yLELDzt5BJYrgoMTUWYTBaCA60dd3Ji/uqHUej5f6ytaIVm15Y0vU49C3NTQ1uvE0f4ur4RMQZkLCr2LtogiKvvy6VWxFhxIeFUJ4dCi2NgKsu9fiTLj6eko2FbD4X8+S0n8gtti4k+/UBTgbXL5z3NgaZfKJp/a9nKyRIUQnWMkYFENUopXIBF9tXIL1rOqq39UEMoLj8XhpanQHREz11popp70Zi621f6AQguxRiWQMjmXNh8VsWX6A3RsrmDwnh/6jk3r0Dz1Fz0KJKQWgNUiMTgojOunYuiGv18PqN19n7QcfEpeezaiL7sLrDcNe00SD71a2p5aG2mY8rmPnMrSEm31iK+SoKJetjQizhGsfRX/PoQ77EuFfTmv/Ilq3xd/biHbb+w7adpS48HWSFD6dN/3We1nwu/tZ9NwzXPyThzAYhPZF7O/y7WswKRBg8DWYbLu+A6SUOOpdmjBtm5bzCdWmRvdR29tiQolKsNJvWHxLXZsmmqy9oplrdySQKapAio5ARYuklIGz5ZU4G9xYI46NuoZYTZxzzQAGjU9hxfztLHmpiO1fHmbK9QN1b1qqUIASU4qT4LTb+fjZP1GysYChM89jxm3fx2TuOBIipaSp0d0isOw1TTTWNmGvaW5ZVrnfTmN9c0ujx26FcTL7ti7jX/c8i8ky8vT2bd/hWwit+WSb4mYhICLOQlSClf6jk7QRmL4IU2S8BVMPatDaWwh08XTgbAVGJLqbvR3+gNKDJocb6ZUnjMQm9Ingqp+PpnDlQb5+bzdvPfoN+RdkMurczG4fIVf0bJSYUhyXir17eP+px6ivrGT27fcwbNb5J9xeCIEl3Iwl3HzCOc68Hi+Ndc00+EVWbRNO3/B0f8SoJRIkWqcXOWoKEtqt80eaWta1HKx1WdvRa/7u2226dHu9/dmwsIzqQ6sZefE5hEUnI71to17HRspau353vF6LNmlRpog4C0aT+kLvTvTGaBFw7JQ/etkJZPF5vWbLGnHitLbBIBg6LZ2sEQmsXvAt33y4h53flDH1hoGkq+luFDqhxJTiGKpLD7Hhkw/ZsnQxlvBwrn3kcVIHDO6y4xuMBmwxFmwx3a8/TM6oXzDv/rvZvXY+Nzx2dndHPxsIZI+pgEamAvS6glGbZbGd2v9keHQo531vCIMmVLHyzR28/+cNDByXzMSrcgiLVAM0FF2LElMKQIukHCjaQsHC99ld8A0Gg5FBE89hyo3fITz67Pk1Fx4dw+w7f8gHTz7Gl/95g3OuvyXYLil0JKCRqfre1/spGKMGT3ekamZeHNc/PI51i0rYsHgfJVsqmXBFNrmTUlWvNUWXocTUWY7b5WLHlyspWPg+FSXFWCMiGX/FNQyffWHARrV1N/qPmcCQ6bP55v0F9BuRT/pgfaf+UASP3hgtkl4ZsNfVnSNTbTGFGBl/WTYDxibz+Rs7WDF/B9u/KmXa3IEnLElQKE4VJabOUhpra9i0ZBEbF39MY20Ncel9mH3HDxl8zrQe2YCzq5l+y+3sL9rCouee5uYn/tbju6MrOsZhd2lDMwMwICJQURx/oXYgCGzNma9m6gzElJ/YlHAu/+lIdnxdyhcLdvHOY2sZPiuDMRf1wxyqBoAozhwlps4yKvbuYf2iD9i2egUel4t+I0cz6sLLyBw6QvVkaUOINYwL77mPtx7+Octf/Sfn/+AnwXZJoQNOezOWMHNAokbOAKX5enPq0hRq7PSoVyEEgyak0HdoPF/+bxcbFu9j17pyplw3gL5q8uRTxj96u/6Ik/oq3+1I631dlQMkWn1sbCgRvntbjIWIWO1xeHQoxl4yDZASU2cB0uuleMM61i98j31bN2MKDWXItNmMuvBSYlPTg+1etyV1wGDGXXkNX7/7FlmjxjBg/ORgu6ToQvx9i6ISrAERU72yjimQBfz1rk5FpdpjsZmZcdNgrTfVGzv42Dd58jnX9u+Wg2MCjb9P3rFiydGyrNnpOWofU6iRiFgLkXEWkvpGggB7dRP2aielxbU0NRzdW08ICIsKJcInsmyxFmwxoUTE+gRXTOgJpzbrTigx1Ytpdjoo/HwpGxZ9QPXhQ9ji4jnnhlsZOvM8rLaIYLvXIxh/5XWUbCxgyQvPkTpg8FlbR9Yb8afDrDYzNWX62wvcvHy9txFpV4opP6n9o7n2wTFs/Gwf6z4u4Y1Hzo7Jk6VX0ljXTF2Vk/ojjmOjS1VO3O16iIVYTUTEWYiIs5I2IEZ7HGvxLbNgCT+x8HE1ebBXazbsR5qor3ZiP+Kk/kgTFfvr2bOpsmV+UT9Gs6FFWB0lttpEurpDirbXiqnSXTtZ/fbrxKSkEZOSRmxKKjGp6UTEx2MwBP/E60ldZTkbPvmILUs/pamxgeScAVx07wP0HzcJo6nXvuW6YDSZuOCe+3j95z/ik388w1W//A3C0Hu/YM8mOlPQ3Bl7ehPoRqQGgzhqAnX9bDVj0WnOSaPJQP75fek/OumoyZNnfyeP2JRwXWzqgdfjxdngxmFvxml34bS7cNhdOO3NOOwuHPUuHPXNmliqdh7VVBi0/4WIWAuxKeH0GRJHZFuxFGvp9JyH5lAjMcnhxCR3fE6l1AZPdCS27NVO9hdW0VB3bNPn0HAT4y7JYui04GVaeu2VtdnppKnBTtHKZTQ7GluWG81mopNSNJGVmkZMSqomtlLTsUZE9ohwYkdIKTn87XYKPn6fb7/5EoD+4yaRf+FlpA4YFGTvejaxqelMu/m7fPbi39nw6UeMuuDSYLuk6AL8qb1AiClt2pXA1kwZTPp/lznqm7FEmGms1f+1Oeyu416Eu4q2kycvnbeNgkUlzP5Onq42j4eUEpfTg8MnhJw+MeS0u3A2NGvCqEUwaeKp/RRVbQmxGLHYzFgjQkjIjCB7VIKvdslCZJwVW2xo0KetEkJgjQjBGhFCYmbH23jcXm2GjepWkVV/pImoBGtgnW1HrxVTfYYMY+7v/4yUksbaGqoPH6T68CHf/UGOHDpA8fq1eD2tH77Q8PCWSJZfZGlRrTTMlu6ZQ/e43exc8wXrF75P6a6dhIaHk3/R5Yw8/2Ii4xOD7V6vYdisCyhev5ZV818lc+gI4tL7nHD7PbV7+Lj4Y+Kt8QyNH8qAmAGYjaoBaHfiTPsWnQmuJs8xUQC9cNhdmMyGgEyf4mxwERYZEjAxZTlJ9/OuwD958poP9xyTctIDr1fyzYfF1JY7jooiOe0uvJ6OPzMGo8BqM2OxhWCNMJPQJwJruBlLRIhvubl1ve95b5l9wWgyEBlvJTI+uOKpPb1WTPkRQhAeHUN4dMwx/YK8Xg91FRUtAksTWQc5sG0r21YtP2pbW2zcMSIrJiWNqMSkoKTOHPZ6Nn/2CRs//Qj7kSpiUlKZ+Z27yJ06gxBL9/qQ9QaEEJx7573Mu/9uPn72SeY+9tQx3dGllKwrW8drha+x4sCKo9aFGEIYFDeIofFDGRI/hKHxQ+kT0afHRkJ7A/7aokBEplqmdwnA262lw8y4mjwn37gT+PtZxabqnwZzN3twN3l0qZkKNtWlDRQs2ttSCxQZbyWxbyRWW0gbUWQ+6rnZYlTfHd2MXi+mToTBYCQ6KZnopGT6jcg/ap2ruYma0sOayDqkRbWOHD7AzjVf4qyvaz2G0UhUYjK22DgMRiNGkwmD0YjBaDr2uUm7NxqNGNosb33e8T5GY+tzkOz8ejWFny/D3dxEnyHDmX37PfQbka9qeXQmPDqGc++8l/ef/B1fvDOfKTfcCoDL62JJyRLmFc2jqKqImNAY7hp+F9cOvJYmTxNbKrewpWILWyq38N9v/8v8bfMBiAyJPEpcDYkfQpxVFbgHCn9kyhKu/9dgS31WJ2tOTgWH3RUQMdXU6EZKsITrH9kLZBQx0PiF9sxbc9XcgT2Ys1pMnQhzSCgJffqS0KfvMesc9XVtUobavb36CK4mJ163B6/Hjcej3TW3+3wAACAASURBVB/13O3G61/uOfMvOqPZzODJ0xl14aUd+qfQj5wx4xk641zWfvAuyUNy+dJQxPzt8yltKKVvZF8envAwl2RdgsXUmhZOtaVyXt/zAHB73eyu2c3Wyq2ayKrcwgtbXsArtXRCmi3tKHE1OHYwYWbVMFQPnHYXRpOB3Q27AChvLCcOfbphO1qiYCY8Oqf7nHYX1ogQ6qucutppaaIZgNRboAcLBJJTncBZ0b1RYuoMsEZEYo2I7FRht5TyKGHlaSu03B48HjfS4+lQlCX2zSIsMqoLX5HidBh41cVsXr+St57+Ne9NPsiIjNH8atyvOCf9HAzixNFBk8HEwNiBDIwdyFUDrgKg0dXItiPb2Fq5lc0Vm9lSsYVPSz4FwCiM5ETnHCWwcqJzMPbyEamBwOmL4FQ4SoEYGt0OXW0BHKESl0PfOhyH3RWQepJACpyu6H7eXfFHpnpj1O1sQompICGEwGgyqVYFPYitlVuZVziPJXuXEJ8bwvlfJfJA7eVc/71HO3XcMHMY+Un55Ce1pporHZUUVha2RK8W713Mu9++C4DVZCU3LveoFGFKeIqqoThN/Omw0uZazOibXvFfMGtEJRavvj+EnPXNAREdrak3FZnqDI76ZhC987WdTagruUJxArzSy+f7P2de0TwKygqwmW3cOPhG5g6eS3HKMr5+9012fLWagRO6tjt6vDWeqRlTmZoxFdAimfvq97GlcouWIqzYwvxt83F5tYtMuDkck8GE8FU4t9y3EVgC0fJc+P58T064vv0xUmwpTEufxvSM6WREZnTp6w4kTl8TyPrmOmL1ttXQjMEkqJFHSEY/MeVxe2l2enpd6q0lehPR+6I3Dp/4NRjUj6GejBJTCkUHONwOPtz9Ia8XvU5JXQkp4SncP/p+rup/FbYQra4m8cprKdlUwGcv/I3UgYOIiNVvXi8hBJmRmWRGZnJx1sUAuDwudlbvZEvlFvbU7mmpu5JtOtpJKVuW+Zf7l7Usb7ONf/2JjrGzeid/Wvcn/rTuT+RE5zA9YzrTMqYxJH7ISdOc3QmHvZmEPhHUNtXqLqYc9S5Cw40t4lcvWntnBaIo3Jd6C0ABurPBhRAQau19lyyHr8ZN0bPpfZ9MhaITVDoqeWv7W7y9421qmmrIi8vjiSlPMDtzNibD0f8uRpOJC++5j9d+fi+f/P0Zrv6/RwM6otJsNJMXn0defOCbCu6v38/n+z9n+f7lvLz1ZV7Y8gIJ1gSmZkxlesZ0xqWMI9QYGnC/Tgen3YU13Ex9Xb3uthx2F8LiBZ3bFvkjOJbwwESmTCEGTCH6f+Yd9Vq7B9ELozeOAKVlFfqixJRCAeyu2c1rRa/x0e6PcHldTMuYxs25N5OflH/CWqSYlDSm3fQ9PnvxOTZ88iGjLrwsgF4Hj4yIDG7MvZEbc2+ktqmWVQdXsXzfchYWL2TBzgVYTVYmpk5kesZ0pqRPIcbSvYZ8ez1emhrdWGxmaivrTr5DJ3Ham3GFOEHfAXYtvbMCleYL5FQ8gYi2BQNHvYv4dH1GkSoChxJTig6pb65n2b5lZERkkBef1+2jDGeClJI1pWuYVziP1QdXYzFauKL/Fdw4+Eb6RvU95eMMm3U+xeu/YeUbr9JnyHDiz7J2FVGhUVycdTEXZ11Ms6eZtaVrWb5/Ocv3L2fpvqUYhIERCSOY0WcG0zOm0yfyxN3jA4HTN3u9KcxAo6vxJFt3HofdRWN4YCJgEKgRdq6AjUDTa5Lj7oCjvlml+XoBSkwpjsLldfGfHf/h+U3PU91UDYDZYGZo/FBGJo5kVNIohicMJyq057ZmcHlcLCpZxGuFr7GjegexlljuHnE31w689owiKC3d0R+4h4V/e4obHnsak7l3fvGfjBBjCJPSJjEpbRIPjnuQoiNFLN+nCasn1z3Jk+ueJCsqq6XOaljCsKDUWfmLp10h+rVDaG+v1lYVEDsQmGH2gRQ4DruL2OTe12/N49YipKrHVM9HiSkFoEVplu1bxp/X/5m9dXsZmzyWu4bfRV1zHRvKN7C+fD3zCufx0taXAMiJziE/KV8TWImjSLGlBPkVHB+H28Ge2j3srtnNtzXf8vHujyl3lJMdlc1vJv6Gi7Iu6nTkLTw6hvO+fy/vPfFbvnzn30yZe1sXed9zEUKQF5dHXlwe94y8h4P2g6zYv4Ll+5bzauGrvLT1JeIscUzLmMa0jGmMTxl/VLNTPXE2aOkwh8muuy2PL6VYLksxou/rC2xX92aiEwMzdZU2RU7P/QF3PFrEb2+MTDnr0EYNRATbk4CgxJSCzRWbeWrdU6wvX09WVBbPzXyOc9LOaakVmtFnBqCJkq2VW1lftp4N5Rv4qPgj3t7xNgDJ4cmMTBxJfmI+I5NGkhOdE/CIQ6OrkT11mmhqeztoP9gyOs0kTIxOHs0jEx9hctrkLu3NlJ0/jmEzz2fth/+l38jRZOQO7bJj9wbSbGnMHTyXuYPnUttUy+qDq1m+fzmflHzCu9++i9VkZULKBKZlTGNyyjm6+uIXHfXGGl3tQOsFs4oyEsnU3VZomAmDMQBF4QGqmfLPAWiNCKG4pph1ZeuYM2BOr+irFsgu8gHn7blQvh1ufg+SAj9IJtAoMXUWc6D+AH9Z/xc+KfmEOEscD41/iCv7X3nMqDU/VpOVMcljGJM8BgCP18O3Nd9SUFbAhvINrCtdx6I9iwCICIlgRMIIRiWNYmTiSIbED+myuqtGVyPFtcWtgqm2VTT5MRlM9I3sS158HpfmXEp2VDbZ0dn0ieyD2aDfF9fUm7/LvsJNLHruaW75098IDdN/EtieSFRoFBdlXcRFWRfh8rhYW7qWZfuXsWL/CpbtX4bRa+J2nqLKUQn07XL7foFTS+BSbw6zHZPQ9yvX4ZvkGLQeaS5Psy52PC4vLqcHqy0Eu6sBgHpXPejQZKLJ4Z8D0Mzr215mwc4FmAwmrux/ZZfbCjSOul7a/bzuEOxZCQh49SK48b+QNirYXumKElNnIbVNtfxr8794c/ubGIWRO4fdyW1DbiPcfHoXfqPByKDYQQyKHcTcwXORUnLQfrAlLbihbAOrDq4CtLqrIfFDWtKCIxJHnLTuqsHV0CKYimuL2VWzi+KaYg41HGrZxmww0y+qH8Pih3F5zuXkROeQFZ1FRkSGrqLpeIRYrFx4z/28+fADLH35eS68576A+9DTMBvNTEybyMS0iTw47kG2HdnG0j3LYA3sqd0DjO5ym46WaFF5lx+7PX4x5TTZiQjRd9SWs01ReIOrgaKqCmYypOvtNLQWuhdWFQIm9tXtI1eHyFvbuet2V+0G4PE1jzM8YTjZ0dldbi+QNPbWefm2faTd3/AOLLwP5l0Kc/8DmROC65eO9Fox5ZVeXF4XXulFSolHeo56LJF4pbfjG168Xu2+ZXupbX+8fWMtsQyIGdCtQ8/Nnmbe2v4W/9z8T+qb67k853LuHnE3SeFJXXJ8IQTpEemkR6RzSfYlANQ4a9hQvqFFYL1W9Bovb30Z0OquRiWOYmTSSNJsaZTUlrCrZldLpKm0obTl2CGGEPpF9WNE4giuir6qJdKUHpF+3EhasEjpP5DxV17HVwveoKmxgbypM8kaNfasLUo/HYQQ5MblkmXL5iW+0M2O0+7CFGrkcLP+Ysov3FxmJ2Gn+YPlTGxFxFpo8jS1fAfqZQe0qWQOlh0EHdOXfjEaGm5m1+5dzMiYwcaKjdz/+f28edGbAauz04NeWzNV9D4kDIYB50LSJ/DapfD6FXD9G5A9I9je6cIpXYWEEOcDfwGMwItSyj+0W98HmAdE+7b5hZRyYRf7elos37+cHy//cUBtptnSmNVnFrMyZwVtlFJHSCn5dO+n/KXgLxywH2BS6iR+kv8TBsYO1N12tCWa6X2mM73PdACcbidbK7eyoXwDBeUFLNyzkHd2vtOyfagxlKyoLPKT8rUoU1SWJpps6T1qct/xV16L1+Nm6/IlFBd8gyXcxoAJk8mdMpPUAYO6teg+G/A37KxwVATAlhZ9iI2NwlCv73eCs76ZxD4RHHEc0dWOv9bHYjNzcPcB0nQUU37h5jDVU99cz/jU8cwZOIe7PruLJ9Y+wcMTHtbNtt401jdjMAhCw7rXD8JOYS+HfV/ClAe051FpcNsiTUy9cS3MeRUGXRRUF/XgpO+gEMIIPAfMBg4Aa4UQH0gpi9ps9ivgHSnlP4QQucBC9Ch0OA2yo7L50agfYRAGDBgQQmiP/TcMGAy++zbLhRDHXdd2XyEERmFsOW5JbQmf7fuM+dvnM69oHonWRGb0mcGszFnkJ+UHLXqyoXwDT657ks0VmxkQM4B/zvonE9MmBsUXAIvJwujk0YxOHs3t3I7H62FXzS7KGsvoF9mPVFtqjxJNx8NgNDL5upuZOGcu+7ZspGjVcopWLmfzZ58QnZTC4HOmkztlBtFJycF29azEXzytiSl9h9z7xUB6fArs1c+OlBJHgwtrhJkqp761YG1bMBywHyAtALZKPVpNZE50DmOSx3DbkNt4ZesrjEsZx3l9z9PRA/1w1DdjiTD3rh9X2z8C6YXcNg2MbYlwy4cw/2p4+ya48l8w9Org+agDp3KFHwvsklIWAwgh3gIuA9qKKQlE+h5HAYcIMn2j+vK9od8LmL38pHyuGnAV9c31fH7gc5buXcp7u97jrR1vER0azfSM6czKnMX4lPGEGPUP6e6t28szBc/w2b7PSLQm8ujER7k0+9JuJ1SMBiMDYwcGJEoWDAxGI31H5NN3RD7NjkZ2rvmSbauW8dW7b/LVgjdIHZhL3pQZDBg/GYtNdUEOFE67NoVHZWMl6eZcXW011DfRZGwkK6YfTTracTV58LollvAQqhwHT75DJ/ALHGOYpKyhTFdb/ijYXpdWL+Wvk/rhyB9SUFrAI18+Ql5cHukR6br6oQeO+l44L1/RBxCbDYnt/q/CYuHm97Xo1LvfA1cjjLo5OD7qwKmIqTRgf5vnB4Bx7bZ5BFgshPghEA7M6hLveiARIREt3aAbXY18eehLluxdwpK9S/jfrv9hM9uYkj6FWZmzmJQ6iTBz1/4qrnZW8/ym53lnxzuYjWbuHnE3N+fe3OV2FKdPiDWMIdNmMWTaLOoqK9i2egVFK5ex5IW/seyV58nOH8fgKTPoN2IURpOqr9ITZ4OLyEQL1U3V5Ibo2wfnSHUtDrOdYVH92I5+U9e0zMtnM1PiPIKeJbH+aNsh14GWCbb1tGUKNVLcsJtYSyyxFm3EoNlg5ompTzDngzn8bOXPmHf+PMzGnvV/0+vm5Ws8oo3im/QjrcdUe0IjYO4CeOcm+OCH0NwA4+8KvJ860FX/bdcDr0opnxJCTABeF0IMkfLo/zIhxB3AHQB9+gR/Sgm9CTOHMStTq6Fq9jTz9eGvWbpvKcv2LWPhnoVYjBYmpU1iVuYspqZPJaITX+pNnibmb5vPC5tfoNHdyFX9r+IHI35AvDW+C1+RoquIjE9g3OVzGHvZ1ZTv2U3hyqVs/2IlO9d8gTUikoETp5A3ZQZJ2f17Vwqgm+Cwu5AWN7ghMjTy5Dt0gtpaO05TA1lRQ9nORt3stBYz+9N8XTOwpENb9c2EhpnYU1+sm41WW1qn9V01u44ZvZdmS+ORiY9w3+f38eyGZ/np6J/q7k9X4qhvJjK+FzUj3bEQpAdyLz3+NiFhcN0bsOA78MkvNEE15f7A+agTpyKmDgIZbZ6n+5a15bvA+QBSyq+EEBYgHo4edyyl/BfwL4DRo0fLM/S5RxJiDGFK+hSmpE/hofEPsb5sPUv2LmHZvmUs3bcUk8HE+JTxzOozi+l9prf8+joZXull4Z6F/HX9XznccJip6VP5Sf5PevyQ4bMFIQRJWTkkZeUw9cbvUrJpPUWrlrNl2ads/PQjYlLTyZsyg8GTpxGZkBhsd3sF/h5JrhCnJqZC9L2YOeqbcZjt9IvqBzqKqbZF4VUVVZh0FFNabVYIu2p26WajxZavvq24ppiLsy4+Zv25fc/lmsPX8ErhK4xJHsM56fo2fO1KHHYXYb0pzVf0PkT3gZQRJ97OFApz5sF7d8Gy32qCaubDHUezeginIqbWAv2FEP3QRNR1wA3tttkHzAReFUIMBiyA/sNkeigmg4mxKWMZmzKWX477JZsrNrN031KW7F3CI189wqNfP0p+Uj6z+sxiZp+Zx21dsLZ0LU+ue5KiqiIGxw7mt5N+y7iU9hlYRU/BaDKRnT+W7PyxOBvs7Pz6C4pWLmP1W6+x+q3XyMgdyuAp0xkwbjKhYSpte6b4eyQ1mbQJjiND9I1MuRslItqje6rd2aZdQdX+Kh2llGbLEm4OiJhy2psxWCV2l52c6JwOt3lgzANsqNjAg6sfZMGlC0gM6/4/PNzNHlxOD5be0mPKWQu7l8O4O09NFBlNcMU/tUjV6qc1QXX+H8DQPUbBny4nFVNSSrcQ4h7gU7S2By9LKQuFEI8C66SUHwD3AS8IIX6CVox+q5TyrIo8nSkGYWBE4ghGJI7gp/k/ZUf1DpbsXcLSvUt5/JvHefybxxmWMKyl5UJGRAbFNcX8ueDPrDiwguTwZH4/+fdclHVRt2nFoOg8lnAbw2aex7CZ51FbXkrRquVsW7Wcxc//lWUvPU/2mPHkTZlB5rCRGIzda1BBd8df79Ng0uqXokIj0WuGPikloskckOiDo80IuyOOI7qKKX8/q901u4kOjdPRkq9mKlKLuh0v4m4xWXhyypNc9/F1/HLVL/nX7H91u8E27fG/X70mMrXjE/C6IPfyU9/HYICLn4EQG3z1N01QXfpX6ObvXUecUs2Ur2fUwnbLHm7zuAiY1LWunX0IIVo6iv9w5A8pri1m6V4tYvV0wdM8XfA0/aL6sa9uH1aTlR+N+hE3Dr6xRzetU5ycqMRkJlx1PeOvvI7D3+6gaNVydny5kh1friQsKprBk6eSN3UWCZn9gu1qj8Df96lWHMEgDISbw3UTU81ONwavkcgo/UdqOu0uDEaB2WLUvzVCfTNxGWHsr9/PjIjhutpy2F3gm0PxeJEpgKzoLH459pc8/OXDvLjlRe4cfqeufnWWtp3daTyijXbryWz7ACJSIS3/9PYTAs79HYSEw+d/1Eb5Xfkv6GGDCXpRp7DeR1ZUFlnDsrh92O0ctB/ks72fsergKiamTuSOYXeccl2VoncghCB1wCBSBwxi+i3fo3jDOoo+X8aGTz6m4OP3yZ0ygylzbyM8OibYrnZr/BGBaiqJt8TrGtHdX3EYgPjoaN1s+PHPyyeEoMqhn5jy97NymhuQLkmaTb8uU+5mD+4mD3VUEG+NJ9py4vN4ec7lfH34a/6+6e+MTh5NftJpXtgDiH/0pdVQD0+Ohcv+BsOvC7JXZ0iTHXZ9Bvm3nlmaTgiY/n+aoFryMLgcWnNPc88JFCgx1UNIs6VxS94t3JJ3S7BdUXQDjCYz/cdMoP+YCTjq61j30f8o+Oh/7Fr7FRPnzGXEeRdjNKl/747w1xZVcJj4MH1Hu+4p2wdAaoKeSTcN/7x8bq+bmqYa3ez4+1nVCa3LerotDb36rfuFb5n34CkNqhFC8PCEh9lauZWfr/w5Cy5ZcFIBFixaIlPuQ1p67JsXeq6Y+nYxuJ0w+ASj+E6FST8CcxgsvB/euAauf1MTWD0AVWSjUPRwrBGRnHP9Ldzy5HOkDcxlxWsv8vrP72XvFv1GjvVk/AXoZe5DJFr1LVQ+WKHNL9knUc8e4RpO36i3mqYaJPqVrPojKhWUYTaYu2xuz47wC9/9rpITpvjaEm4O54mpT1DlrOKhLx+iu5bvtkSmPFr0koProHxbED3qBNs+gPBE6DO+88caeztc/g8oWQWvX6kVtvcAlJjqIbgrKqhfupTyv/yFqhdfpOGrr/DU9owPmSIwxKSkccUvHuHynz2E29XMgt/9ig///AfqKvWfzLcn4bC7CA0zUd5UpntkqvyIlm5LitW/35vDrvVj0jPFB60C57BnP32j+mIUejYH1aI3dYYjWmTKWac1haw9CCcQSXlxedyXfx8r9q/gje1v6OZfZ3DUN2M0GTA7DmgLDCZY/3pwnToTXA7YuRgGX9x1heMjboCrX4GDBTDvEmjQ9zPdFag8QDfE63TiLNqGY/MmHJs24dy0Gdch3ww9RiN4PC3bmvv0wZKXi3XIECx5eVhyczFG6jvUW9F9EUKQnT+OzKEjWffhf1nz3n8oXr+WcVdcw+iLr8AU0ktGDnUCp91FaLiJI84jJFgT0LEpOdU1dYQTmBFbDt8UOXoXn7dM79JcTE7UqUWLzhS/cHOYG7TI1Jd/hZV/0laGRkHCQEgcrN0SBmn3tiQQgrmD57Lm8BqeWvcUIxNHkhun77RBp4ujvhlrhBlRdxAsUZA1DTa9CbN+rfVh6insWgquhqPn4usK8i7XUn5v3wivXgQ3vwcR3XcuUyWmgoyUEtfevTg2bcKxaTOOzZtxbt8ObjcA5tRULMOHEXPzTViHDceSOxivw4GzqAjn1kKchYU4N2+hftEnLcc0Z/bBmpeHJc8nsPJyMUboO2WGonthCglh/FXXkTtlBitef5Ev3n6drSuWMP2WO8jOHxts906I1+th3+YNuBo+Q9T37fLjO+3NmMK0PjgJYQldfvy22OucSIMXs0Xfod5ej5emRjcWm5kyvSNTDf7U2x5mxExAzwkH/akwp8muRaZq9mnppKk/g4rtWlps24ewfl7rTpZoSMxFJA7it7E5XGXayAPLf8o7l71LuLn71N847L55+eoOQWQ6jLxZa3q5YyHkXRFs906dovfBGguZk7v+2APOhRsXwBvXwSsXaHP7RXfP2VN6rZhybC2k6p/PY0pKxpSUiDk5GVNSEuakJExJSRgswRkl4KmpwbFliyacNm3CuXlzS7rOEBaGZehQ4r7zHazDh2EdNgxTwrFf9gaLBdukSdgmtXajcFdX4ywswrl1K87CQho3bqRu4aKW9SGZmZqwGtJGYKmJdXs9kQmJXPrT/2Pv5o0se+V53nviUbJGjWHaLbcTk5wabPeOora8lK0rPqNwxVLqq7Sev+7yrhchDrsLwrToboI1gSO4utwGQG1TLdJhxGD16j4lUFOjGyRYbCEccfrKwXWy2SpwGjSBo+M8x84GF1JIoiJtWnNVeznEZGp1NX6khIYKTViVb4OKbVC+Hba+S7SzlidCQ/lOSiK/fWk0j1uyEAmDIXEQJPgiWkFqSeCPTFF3ACJTIXu6JqrWv95zxJS7CXZ+okWljDrJiX5TtKjUv6+Gly+AWz6AuO43w0evFVPeulqa9uyh4auv8dqP7SJjjIrClOwTWknJmJL9QqtVfBkiIjr1JShdLpw7duLYtBHn5s04Nm2muaREWykEoTk5RJw7G8uwYViHDyc0Oxtxhg0YTTEx2CZPwja5jcA6ckQTWIU+gbVhA3ULW9uFhfTt20Zg5WopQiWweiWZw0Zw85+eZcOiD/nq3TeZd98PGH3JlYy7/BrMQfphAeBububbtV+xddli9m3dBELQd9hIxs+Zy5Lnn9HFptPuwhijpaoSwhI4wiFd7Oyp3YPFHU5ouP5fsy3FzL40X4ghBL3km9PuAoOk2egkJzqHkjL9avIc9c24zE5yYnzpRHs5xLbrpyYE2BK1W9bU1uVSQn0p+eVF3LV9Ps9VrWW8u4bLN70FzfWt29mSWlOEbe91prG+mZiUcKg4pE2/YjDCyLnw+RNaBK6bRmCOongFNNV1fYqvPRlj4dYP4fUr4OXztQhVUvdK2/ZaMRU+cSLZH30EgMfegLu8DHdZGa7SMtxlpbjKynCXasuchUV4qo4NjQurtSWSZU4+WmiZErVlxrg4hMGAlBL34cNHp+sKC5FNWgzcGB+Pdfhwoq64Auvw4ViGDMFo0zfkbIqNxXbOZGzntIZf3VVVWmqwsBBHYSGNBQXUffyx7wWLVoGVl4d16BCsw4cjzD2reZqiY4wmM6MvuZJBk6exav4rrPnfOxSuXMa0m77HgPGTAjqhcnlJMVuWLWb76hU4G+xEJiQxcc5c8qbNJDI+kYYGO0ue18e20+7CbNamkkmwJrBDRzFlddmwxVl1OX5bnA2+efkizFRVVhFr1S/a4rQ347W4CDWFkm5LpwQdxZS9mQZTXWtbBHsZ9DnFKbOEgMgUiEzh9qxprF1yB7+v3MKwH6wkC3NrmtB/v/41rWGknyN/12qZGNrVLwsppTaBc7gBSiogKl1bMcInpja+AdN+0eV2u5yiD7TatX5TT75tZ0kZDrctgnmXwqsXwk3/g9SR+ts9RXqtmGqL0RaO0ZZFaFbWcbeRzc24yis00VVaiqusXLsv10RXw9q1uMsrWmqZWjCZMCUmIF0uPBWVAIjQUCy5ucRcdx3WEcO1dF1qakAvVsfDFBeHbcoUbFOmtCxzV1a2iCtnYRGNa9dS5xOihogIbFOmEDFzBuFTpqjIVS/AFhPLBffcx7BZF7D0lef56Jk/kJE3jBm33Ul8RqZudp0Ndrav/pwtyxdTvmc3RrOZ/mMnMmT6bPrkDUMEYE4uV7MHt8uLy1SPQOja+HZP7R6s7jSio/X/n3G0nZfvQBVxFv2meHHYXTSZGsmKytJ9ypbamgYcxnpyo3PA44LGKi2SdJoYDUYeP+dxrv7gah5Y+QDzL5yPJToD+s9u3cjrhdp9WoqwYhu87dUiYTrgatI+h1azQ1sQ6Uu5x2Rq0bUN/4YpD3TvaVU8Ltj+EQy8AEwBGtiSMBC+swjmXaaJqhvegcwJgbF9Es4KMXUqiJAQQtLTCEk/fj8Y6fXiqarShFa76BaAZdhQrUh84ABEDxo1ZYqPxzZ1Kraprb8u3BUVNG7ciH35CuzLl2vRK7OZ8LFjsc2cQcSMGZiTu+/ICsXJSRuUy42P/5nNn33KF2+9xms/+yGjLriECVffQGhY10RNpdfL/qItbFm2mF3fdUWASwAAIABJREFUfIXb1UxC3yxm3HYngyZPw2oL7MAI/+gwu6GWOGscJoN+X4HFtcXkuAdhtek/Msuf5rOEa/Py6VlY77S7sBtrTqmJZmex1ztwmO1kR4+DhkpAQviZvbbEsEQem/wYP1j6A55c9yS/Gv+rozcwGCCmr3YbeD78503QqZ6utfu5byhpZJvrzqibYcF3tBRazkxd7HcJJavAWaN/iq89sVmaoHrtMvj3lXDdG1q9WZBRYuo0EAYDpoQErSh8SF6w3dEVU0ICkbNnEzl7NtLjwbFxI/VLl2FfupSyR39L2aO/xZKXpwmrmTMJHTCgW0TeFKeHwWBkxLkXMmD8JL5463UKFn7AttWfM2XubeSeM/2Mo0X1VZUUrviMrZ9/Rm1ZKaFh4eRNn83Q6bNJytJ3OP2J8IupalGhtUXQkT3VJQx2W7QiY51x2o+umRoYO1A3Ww31TdQajjDkFJtodoamBg/OyAayo7Khcpe28AwiU37OST+HW/Nu5dXCVxmXMo7ZmbNPvpMO+NtLWH1d5I8SU4MuBmsMbHi9e4upove1CYqzZwTedlS6lvJ7/QqtU/qceTDowsD70QYlphQnRRiNhOXnE5afT+ID99NcXNwirCqf/RuVf30Wc1qaL2I1k7DR+Qg1lUmPIiwyitl33MOwWeez9OV/8Mnf/8ymzxYx87bvn7L48bhd7C74hq3LFlOyaQNSesnIG8akOXPJGTcRc0jwe+f4L2JVskLX6E2Tp4mqmmoALOGBEVNmixGDSXDEeYQ4SxxenWw12ptwRjSQEz1GJwsa0iuRTgOmZLCF2FpTbp0QUwD3jryXgrICfv3Fr8mNy9V1bsHj0dr9XOuQ35LmA63H1LBrYd3LWrPKcP1StmeM1wPbPoIB5wVv/jxbItzyIcy/WutFddULMOSq4PiCElOK00QIQWh2NqHZ2cTfcbvWmX35cuxLl1Hz1ttUv/Y6hqio1jqryefoXmiv6DqSsnK4/tE/UbhyGaveeJV//99PGDbzPCZfdzPWiI6bwVbu38vW5UsoWrUcR10tttg4xl0xh7xps4lO6l6pYGfLXG+HyLcO0c3O3rq9hLi0z701gA0765rrcHvdxFnjqNDBjtcrcTd6ccY2tI6wkz7Z1sXTtjQ1uhFSEBEZpi2w+3ow2Do3BZDZaOaPU/7INR9ew89W/oxXz38VsyGwg2xa5uVz+Rp2hrarqxt5E6x5Hja/DRN+8P/snXd4W+XZh++jPW15r8SxE5wEQqYDJARICJBQaKFQZiGM0gGFUmi/UChtgUIHo5QCoWzKLqtlQxiBUqBQsqAQMuyYLI94STra63x/vJIsyc6wfeQRdF+XL9tH0nlfeej89IzfM6R72yu2fAi+jsHP4hsslkJY8jw8cy5Yhld05sRUjkGhKymh4LTTKDjtNGJeL54PPsDz9go8776L+6WXkPR6LHPmYD9qIbYjF6Ivy+4stByDR9JoOHDB0dQdPJcPn3mCNa+/xMaPPmDe6UuYdvRiNBotIb+P9R++x+cr3qSlYQMarY4Jsw9m6pGLGDd9JpoRWjibiAi0RrdTbF6QtXU2uzZjjospk21oIlMmmyHpfl5oKqQ9C/P5gt4wIBE1BqmwVoiDXZuBMnBuUXUtrxwAoLigQBxQSUwBjLWP5ZpDr2Hpv5aybM0yLqu/bNDn7A9JMRX8SnhLZVJ+IFTOEqm+ORdlzTNswHz5IujM6QX8w4UpD87+x7D/jHJiKodqaKxW8hYtIm/RIpRIBN/q1XjeXoG8YgWt114H116HaepUIawWLsRYV5ersxrBGC1Wjjz3B0w98hhW/O1e3n7gLj57+3VKx9Wy4aP3iQSDFI2pZv6SCzjgiIVY8vKHe8t7JOANgwRBnY9SS/aEvejkE9EG8xCIKb8njCXPkJzLV2QuAjqysg5Afp4NjRSvpwu4gDKR+lGRLW07AKgojqdjve2iDV+vjtXEsTXH8nHLxzzw+QMcXH4wh1Ydqsp59wa/HEZn1KL3bk1P8aUyawm8fDnsWA1j6odsb3skFhOWCHVHg2GEZB1GwHUkJ6ZyZAVJp8N68MFYDz6Y0it/QaihATkurNpv+wvtt/0F/dix2BcuxHbUQiyzZuXqrEYoxdU1nPrr37Hxo/d599EH2Pjxh+w/bz4HHrmIirpJo0oQ+z1h9GYJRVIoNmdv+HCTs4kyjajFGYrIlN8TorDSSmegBSBujaC+mEqkSUuLUlIqgewMXN/avgOQGFcaj9x42lSJSqVyxUFXsHbnWq56/yqeO+G5rP5NpOKXQ1jsejGweVdeSQeeAsuvFqNyRpKY2v4JeFrhgG8P905GFLmrV46sI0kSxro6jHV1FF/4I8JtO/G88w7yirfpfuIJuh5+GF1JCYUXfI+C009HY86+yWGO/iFJEpPmHk7dIYeixGJodaPTyDXgCSGZRform5Gpza7NTNEIq5GhSvOZbXq2+EV3mIhMqU9HlyiqrypOqYULZmdSdEvHTqCMCeU14oBn56CLzzMx68zcMv8Wznj5DK7691Xcc8w9PRG3LOL3hDHbdKLuKG8XBfCmPCFYPv8HHPuHkRMFWvcCaA1Qt2i4dzKiyP5fTY4cGejLSik443Sq772Xuv/8h6rbbsMwYQI7/3gjDccsovOhvxHz+4d7mzn6QKPRjlohBUJ0KCYRXclWFCIai/KV+ysKKcFo0aHVZvdlNhyKEgnFMMVtETSShnxDdlKu29tF5Ku2bGzPwaC8i3sPjs5uEfEqdMQbH7IQmQKY4JjAVYdcxUctH/Hg5w+qfv6+8MshzKZ4WnRXYgpEqi8kwxfPD8m+9oiiiHqpCUcJsZcjSU5M5RhWtDYreccuZtzfHmLcY49irNuPnTfeSMPRx9D54EPEfL49nyRHjr3E7wkT1geQkLIWvWnxthCMBrFFHUMWlQIw20TNVIGxIGvO5K1dInU4sSLFsDNLkSm37CWmjaAzxJ+LZ2dWxBTASfudxDdqvsGda+5k7c61WVkjFb87hNkgRo3tsmYKoHouFO0nRt2MBJpXg2sbHDDMXXwjkJyYyjFisMyezbiHHmLc449hmjSRnTfdJETVAw/kRFUOVQh4wgT0HgpMBVlrh9/s2gyAMWwZkuLzhJhKRKayJRIBup1uwtoglfkVPQezEJmKxCKEPFEkczx6E/IJ0ZYlMSVJEr+Z+xsqrBVc8d4VuILZqQMDMZfP7wlj1nrEgfw+uvl6NiZsErZ9BO0bs7anvWbdi6DRiREyOdLIiakcIw5LfT3VDz7IuCcexzR5MjtvvkWIqvvvJ+b1Dvf2coxSFEWJj0JxZ9X9vMnVBIAmoMdkGxqPKRBdg12BrqzOG/S6A0SNofSmg4D6kamt8laMYQsGW7ys16uOYefusBls3DL/Ftr97Vzz4TVZWyfkjxCLKpglUX+228gUwPQzQdIKm4ThRFFEvVTtfOHQniONnJjKMWKxzJpF9YMPMO7JJzAdcAA7b/kTDUcfQ8d99+VEVY5+EwpEicUUXFJ2Z9c1uZooNBUS8sWGxhZBTolM+bMXmVIUhaAvitac4l8V9osPlWnobsAUtmHPizejqOR+viemFE/h8lmX8/bWt3FL6vt0QYr7udIOJseeC8vtZTDxWPj0STFceLho+xy6m4Z+Ft8oISemcox4LDNnUn3/fdT8/UlMBx5I+59upeGoo+m4516inpyoyrF3BOIRnG6yO5dvs2sztXm1+D2hoa2ZshuSo2SyQWegE13QiDk12pYw0lSZRmcjpoiVgtTic8hami+VJQcs4YgxR9AhKchEVD+/L2HYGWneffF5KrPOET5bG19XfT97zboXQNLA5OOHbw8jmJyYyjFqMM+YQfV991Lz1N8xTZ9G+5//TONRR9Fx9z1EPZ7h3l6OEU7CcLJdactaJ5+iKGx2bWa8tY5YREkXHlki4A0jSRDVhfBH/FlL8yUETl5+SiRFzo6YanA2YInYsSUjUwkxld3IFIj6qRvm3YAEfBULqn7+QCIyFdoK+XsppvY7GmzlsHoYU33rXoSaw8A6NF5co42cmMox6jBPn071PfdQ8/RTmGfMoP2222g86mg67r47J6py7JJEBMevk7PmMdUd7MYVdDFOXwsMkWGnLCJgXaHsekw1OBswh209410ga5Gpps6v0EUNPWlSTzsggWVoLuQFpgJ0kJVh0cnIlL9xz/VSCbQ6mPFdaHgT3M1Z2NUe2LkeOjYM/yy+EUxOTOUYtZinTWPsPXdT88wzmGfOpP22v9Bw1NG033UXUTk73jc5Ri8JMRXQe7KW5tvsFJ18FVrhw2S2Z4ipgLtnMLBKJOfyJUbJZCnN19CxGX3M2JN6g6yIqXA0zM4uIQyTkT1Pm4iIaEe/z/Qe5/Ltiplni7+dtY9nZ2O748sXAQn2/9bQrz1KyImpHKMe89QDGXv3X6l59lks9fV03H6HEFXLluVEVY4kiTRfQOfNWgF6whahGJGO6hWZCvtEV5SK+OPu54khx9mKTG1rF7Py0orq5VbV19ni3oIubARSfn5ZcD8fLvyeMEaTBq0U2fvIFEDRBKg5HNY8JubjDSXrXoDqOWAv3/N9v6bkxFSOfQbzgVMYe9cyap57FstBB9Fxx51CVN25jKg7O8aCOUYPAU8YNAohbSBrkakmVxNmnRlTcshxRs2UysOAQVycTXFbBCArNVOKotDS0Q5kPCeP+mKqwSXSiWKthJhqA2v2mgaGEr8cwmSOi6H+iCkQnlPdX8GW91Xf1y7pbBSdfLkU327Jiakc+xzmKVMYu+xOav/5D6yHHEzHnXFRdfsdRF3ZM+PLMbIJeEJgioKUvVEyTa4mavJqCHpEF1gva4QsiKmAJyQiU1lM8+307STmF95SptTUZcKyQEUauhswR+xirX0xMiWHsBjjFge7M+zsiwNOAGP+0Dqir3tBfM6l+HZLTkzl2Gcx7b8/Y+64g9rn/4l1zhw67rqLhqOOpuWaa3G99BLhHTuGe4s5hhC/J0zEEKTAWIBemz338/GO8QS8ITQ6Cb0pZaxLNAKKumJKiSkEvJGkx1SeIS8rzy3RyQd9pPn06g4mb3Q2UpmoObMZRFo0S3P5hgO/HMasj1u69DcypTfDtFNFZ52/W/3N9cWXL0JVPTjG7vm+X2NyYirHPo9p8mTG3HE7tS88j+2Iw3G/8grNS6+g4aij2XTkQnb8/P/oeuIJAhs2oETVjxzkGBkEvGGCeh/FWeoI84V9tHhbhMeUHMZs1ac7hXt3AurWSwX9EZSYsGDIpvv5JucmTGEhptLqwDxtqg+8bXA2UKqpQJLAaNFBwAXR4D4VmTJJrr0z7OyLmUvEz+N/z6q/uUy6t0DzmpxR514w+lsjcuTYS0yTJlF1660o0SjBTZvwrVqFf9VqfCtX4n7lFQA0djvmmTOwzKrHUj8L09SpaEymYd55DjUIeMLCFsGcnQjHV+6vABjvGC/qmOwZ9VJyi+prps3l686e+3mjs5ECpQQkMFriYioWFUaShXbV1glFQ2yTt7FYKcFk0yNppCFzPx8KlJgYaWQp6Nx7w85MKmdA+TRY/TAc/AN1N5jJly+Jz7l6qT2SE1M5vnZIWi2myZMxTZ4MZ52FoihEmpvxrVqFb9Vq/KtX0X7bbeLOej3mKVMw18/CUl+PeeZMdAW5uVSjEb8njDu/O2v1UolOvvH541nr6e5dL5WFzrdEh6LZpqezuZOJBRNVXwOEmKrTHI7JqkejiUfbvB2iVd+oXmSqydVEVIlii+ZjSLVFgH0izRfwhVEUMEdb996wsy9mnQOv/h80rxXiKluse0EIt8La7K2xj5ATUzm+9kiShL6qivyqKvJPEO/Aok4nvjVr8K9ejW/VarofeZSuBx4EwDBhApZZs7DMrsdcX4++qio9nZNjxBGLKQS9YZyOTqotFek3unYA1eDrGtQaTa4mtJKWans1//HspLQ6I6KZhchUwrPIlMUhx4qi0OBs4CDl2+kCMdHJZ1QvMtXobARAHzb3rDUEQ46HCr87Ln7D2/tfL5XK1FNg+dVi+HG2xJS7Gbb/Fxb+Kjvn38fIiakcOfpA63BgP/JI7EceCUAsGCTw+ecicrVqFe7ly3E+8wwAutJSEbmKpwaNkyYhabW7O736hP0QDYnUixITH7GoKHhOfp3yWYlm3J56LPVrpfd9tQYRjTDl93wYbKAZuSWYIV8ERQGfTqbEPC39Rl8HUA2hwc15bHI1MdY+Fr1WnzTSTCMLkamAV1yc9RYJd8idlTRfi7cFX8SHJWpPr5dKjJJRMTLV4GxAK2nBr8NckdLJB/tEZCpp2BneDnn1Az+RuUDUMX32DCy6QfUmAKAnxXfAt9U/9z5ITkzlyLEXaIxGLPX1WOrrgR+gxGIENzXgXy1Sg77Vq5BfE0NINVYr5hkzMM+aibG2Fl15BfrKCnQlJdkRWR/eCW/+WnVn7f4hiULkhLgy5qeLLVOG+MoUY8a8rLpb++NDjvs07AyqMyx7s3Mztfm1RKMxgr5Ib/dzuQVQ1/QwUTPl1Qlz2jQxpaBKx1eDswEAXdCEqbCPyJQpf9BrJGh0NlKdV03QG06xRWgDjV4IiFFOMi2rcQ0uMgUwawn872nR2Tf9dBV2l8G6F6FkfyiuU//c+yA5MZUjxwCQNBpMkyZimjSRgjPPBCDc3Ixv9ZqkwOq4c1m627VOh760FF1lBfqKSvQVcZFVXi6+r6xAa+9nyqThbSGkJiyE8UeCRguSVkSJJE38a20fX2tS7tvHsbTbM84VCULQLbqsAi4xIiX5tavnNueWntuDe+HvZbD1iCttPqAT0TYV2O0omdDg5zlGYhG2yFtYMHYBQa/wmDJZ+6qZUldM+eUQOr0GV0yIpqTHVCQIKBAJDHqNhJhSAhrMqUX1iVomFdN8Dc4GJhVMIuAJ96yV8JjaB1LpyciUxjW4mimAcYdBQY1I9aktpjw7YcsHMP8X6p53HyYnpnLkUAl9ZSX5lZXkf/N4AGJeL+HmZsItLYSbW8TnlmYiLa3416zB/dprEImknUNjs6GvqIgLrgr08aiWvqICXUUl+rJSJH38It29BZ67AEomw2mPDKzNeqiIRSEop4utXYoxF8hO8bhBpt4S7HaUjApiaru8nUgsQm1+bc8Fs1c3XxbSfJ4wJrueLn+G+7mKw3AbnY2UmkoJeaK903wmh4gaqUAgEmCbvI3jKr+FoqSIUU8b2PYN93Mx5FjBpJEH3s2XQKMRNgkrrhcu5UUTVNkjAOtfBhRhEppjr8iJqRw5soTGasVYV4exru8wuRKNEunoJNKSIbhaW4g0txD47H9EuzPSNJKErrQUfXkZOt969AYt+mNOx/TpOiwHHTQEz2qAaLRgdoiPvcHVDW8uUW35RG1RQO/t3c2ngmBL7eQLdPXYFaSRjQJ0bxizzUBnQBjQJtN8ru2qrdHgbKDOOplYTOldgK5iUXiTqwkFhWp9LVtJGRLtaRu88BghBOQwJmMEjRQbfJoPYMZ34Z3fiXl9R18z+PMlWPcCFO0HpQeod859nJyYypFjmJC0WvRlpejLSjHP6LsjJ+b3E25pjUe0WpJfhz99h+BOH56ABeWLe4B7mPD6axhqaob0OYwWEmk+g1WLUWtMuy3i9RDyvEw0fOSAz58QU7X5tbRu9QEZTuGRIPg6B3z+XeGXRW1Re+YoGdd2wDbo88eUGJudmzmt4lAg07BzJ9jVE1OJdGKFdgxbaUsfJVM5U7V1hhO/HMKsDwzcsDOTvErY7xhY+wQcebU6dYe+Lmj6N8z76T6RWh0qRm77TY4cOdCYzRjH12KbNw/HKadQ8pNLqPx2LeMOWseEmy9g0mefMf6lFwFwv/HmMO925OL3hIlpIxTaMiJjATd+d4hYeCP+QcxtbHI1UWouxWawiRmA9OEUngUCnhAmq7BFMOvMWPQWcYNbncjUDnkHgWiAsboaIGPIsdwKNvVqwBqdjeg0OvJjhT1rJYxB9wFbBBBpPrNW7v9Mvt0x6xwRJWxQ6f9//Suiczfnet4vcmIqR47RxNaP4PUrYeKxcMQVSJKEsa4O0/RpyMuXD/fuRiwBT4iwPtC7+FxuzSxbGxBNriZqHcLY0O/pI82XsBGQ1O3mFIXaejoDnekeUyql+TY5NwFQphUpqeRzSszLUzEy1ehspCavhpA31rOWr1N0qe4jYirgCWOWutRJ8SWYuBispbD6UXXO9+WL4KiGiunqnO9rQk5M5cgxWpBb4elzIH8snHRPmq9T3qLFBL74gtB29Wpl9iUCnjB+vad38bncPGgxpSiKGHCcPx4QYspo0aHVpry8JuqlNOpVVkTDMUKBqHA/92eMklFJTCVMNB0xUWeWTF0GXKJTUEWR0+BsYIJjQrK+zWzT71MeUxCPTMXa1a0B0+ph+hmw8fUe0T5Q/E5ofEdEpQaZ4ovGoiiKurMoRzI5MZUjx2ggEoKnzxUdcWc80auQ2754EQDy8jeGY3cjHr8njEfr6l187m4hOkgx1e5vxxv2UpsvIlMBObQLWwREIb5KJESHqa8hxyqJqQZnAxXWCpSAJr5WppGmOmk+X9jHDs8OJjgm4JfD6IxadAZtyiiZ0R+ZikZjBL0RzLFW9QvqZ50jUnOfPjG482xcDrEw7N//FF8oGmJ122ru/exefvTmj5j75FxOeekUtstfjzd4OTGVI8doYPkvYdtHcOKdUNa7w8YwZgymKVNwv5FL9fWFVw7i13ootWREOOTmQYup1E4+EMKtT8NOjU7VNF/aXD5/Z0/xuaKoKqYmOCYQ8ITR6jTojfH9Jww7VUrzJTr56hx1IhXWS7SN/shUognCoobHVCbFdVA9V3T1DSYatO4FIfSq9uzO7o/4+bjlY+5aexffW/49Dn3yUM59/VzuWHMH7f52jh9/PK3eVs569SzW7lw78D2NEnLdfDlyjHTWPgGf3AdzL4EDv7PLu9kXL6b91lsJNzejr1SxJmMfIOAJEXB4ekem5FYi4cGde7Ozt5iyF2bO5YsXazsHt1YqCVd3g0VLd7C7J80XcKlmRNrkamJe5TwCzaJrMDmDUk6NGA3ePyvRyTfBMYHPPe4UMRVfxzr6xZRfjkcSNW51a6YSzFwCL/wYtv4Hxh3a/8cHZWh4C2af3+doKG/Yy9qda1nZtpJVbav4X8f/iMQiaCQNkwomcdqk06gvq6e+tB6HSUTOzzngHC5++2IuWH4B18+7nuPGHzfYZzliyYmpHDlGMs1r4eXLoeZwOPq63d41b9ExtN96K/Kbb1J47rlDtMGRTzQaIxJQCOi8vSJT4e4dxAY5hafJ1YRNb0sKtYAnTGl1hiu43AJ2dcVUItIRNgSIKbEMW4TBs03eRjgWFqk3TzijOzEuoFRKvzU6G9Fr9Iy1j2WlZ03PXEPPTtBbwTh4m4fhJmHmatG4suObNeXb8NovYPUjAxNTm96AaDDZxecOuVndtppVbatY2bqSL7u+JKpE0UpaphRNYckBS5hdNpuZpTOxG/p2wa/Nr+Xx4x7nsncu4xf//gVb5C1cOO3CfXIwfE5M5cgxUvF1wVNLwFIEpzy0Rw8ZQ00NxsmTcS9/IyemUugZJdPbsNPT0QYMbkhsk6uJ8fnjkSQJRVHwe0J9GHa2qutQTc/z8unFXL5Cc7xmSsUUH8B+jv1Y7/FnGHa2gc6k2ly+BmcDtfm16DQ6/J4wBeXWnnX2gRQf9EQSVZnL1xcGK0z9Dnz6FHzjxn7/brq/+AerCstZ2fouqz77Mxu6NqCgoNfomVo8lQumXkB9WT0zSmb0WHDsBQWmAu5bdB/X/ec67lp7F1vdW7nu0OswaA17fvAoIiemcuQYicSi8Oz3RATg/Nf3epxG3uJFtP/ldsJtbejLRn/RrhokxZSu91w+T3c3gxVTm12bObRSRALCwSixiJLuxwQiMlVzGETDEFVpRE480iFrMubyubapcv4GZwMSErX5tazxfEppdUrqUm5TdV5eo7OR6aWiFd8fH5EDxMXUvvF37HfHa9zMZG/008xzYNXf4PPnYPb3dnvXdl97MmW3suW/NAaaIN+AadM/mV4ynYtmXMTsstlMLZ6KSWfa7bn2hEFr4IZ5NzAubxx3rLmDZk8ztx15GwWm0T+8OkFOTOXIMRJZcT1sfgdOuAPG7LkYNIF98WLa/3I78ptvUXj2WVnc4OghIaYks5J+UYhF8bgHV1skh2Ta/e3JTr5kXUxqFCfsh4AT7OVI7ZvA+xlw3qDWBfG8jBYdXUExly9tlIwK7/obnY1U2aqw6C1iBmCqQFRxlIwv7KPZ28x3HN8hEooSCUbTC9BLJqmyznDjl0NIxDAWFO75zgOlahaUThGpvgwx1eJpYWXbyqSA2uLeAoBFZ2GmdQzf7HIy+8jrmDJ1CXqtOvMWU5EkiR9O+yHVedVc/e+r+e4r32XZ0cuStYajnZyYypFjpLHuBXj/z1B/nmh57gfG8eMx1u2HvHx5TkzFSXS9WezpY2TwtiOHB3fRaHI1AT3F5wnhltbNl7BFsFdAaC3EVHAJJT6Xzy5sESBjlExe5aDrwhu6G9jPsZ9o6fdFepuQlkwc3AJxEl5WidosSHFa97RB7RGqrDPc+D1hzDovUn5F9haRJJi1BF6/kh1N7/BJ1MXKViGgdnjE/Ea7wU59aT2nTjyV2WWzmVQ4Cd0/L4SQDqaeo85Imt1wbM2xVFgruHTFpZz96tncuuBW5lTMyeqaQ0FOTOXIMZJo3wDP/1i0Jn/jpgGdwr5oMR133UWkowNdcfGeH7CPk/Bjsudl1HnILXjCg4vgJG0RHIlOvj5GySTFVLmY0Yc69ggBTxiTVc82fyc6jY48Q564wb1DGLsOgnA0zBb3Fo6sPpKgV4i/XjVTKomcRG1WnaOOgDMlshcJiojevpLmk0Pxein1i88VRWG7vJ1P2j5hZaCBlWOraHnvUgAcRgc5qdnjAAAgAElEQVT1ZfWcvf/ZzC6fTZ2jDm2q31kkCBteFwXsWRZSCaaXTOeJ45/gkrcv4aI3L+JXc37FdybuulN5NJATUzlyjBQCbvj7WaKw97RHQWfc82P6wL54ER3LliG/9RYFZ5yh8iZHH4lZeQX5GR1H7hY8kYH9jBM0uZrQa/RU2aria2VEViDpfq7Y4mJK2vvi3d3hl8PYi0zJUTLJDinXdlGfNQi2uLcQUSLxaFGGQAwH4mlLdUROg7MBo9ZIla2KHdudPWt528Ud9pUCdHcQM52qiClFUdji3iLEUzzytNMnPLkKTYXUG4s4r6uF2Wc+z37FB6CRdmMp2fgOhGQ44NuD3ld/qLJV8cg3HmHpv5Zy7X+uZYt7C5fVX7b7vY5gcmIqR46RQCwGz18EXZvh3BcHZepnrKvDUFuLe/nynJhCiI6QNkCJLdNjqlmVyNS4vHHo4mNiEjVTvaI4gB8rkhJDUakrPOAJUVpjTzfsjEbA3TzoQbqpnXyB9oznpLIreaOzkfH549FqtCliVL9PuZ8D+N1+SjUuyO9/elRRFJpcTXzS+kmy7qnD3wFAsbmY2WWzxUf5bNFZuvkdePQkaF0PJQfu/uRfvgjG/GFJp9oNdu486k7++N8/8tAXD7FV3srvD/t9v7oFRwo5MZUjx0jg/Vth/cuw+A+DjipIkoR98SI677ufSFcXusIsFryOAtxuHwGdhzGZc/ncLcgRI5JWixKNDujcTa4mJhb0XBwD3hAanYTelJJGkVtAa0T2DdIdNAVFUUTNlE0vRskkbBE8rWKsiApiSiNpqM2vZVuTCyDF+ykhctQZJdPgbOCg8oOAVDFqgB293c8VRYFwGMkw+trq/XIEs27vDDtjSoxGZ2NSPK1qW5WsjSu1lHJw+cHMLhcCqiavprdvU+0CyK+GNY/C1FN2vVA0DOtfgcnHgW54fqY6jY6rD7ma2vxabvrkJs5ffj53LLyj97SCEU5OTOXIMdw0vAUrboADT4E5F6lyyrzFi+m8+x7kt9+m4NRTVTnnaEV2+wjovJSY0wt/Y64WvBEDOpuZsKf/XX2haIjt8naOrTk2eczvCWO26tMvbnIr2MuQu7oG/BwyCQeEBYPJaqDT3ckER9zDKuExlTcGJdrGQGflNDobqbZXY9QaewYP2zMiUyqk+eSQTJuvLbn/gDeMJIHRouszMtX18MN03v8Atc89O6qsP6LhGKEQmA0uyOstdGNKjE3dm9LEkzMoUp7l1nLmVc5Liqex9rF7Nr3UaGDmWfDuH6D7Kyio6ft+Te+JlO3+JwzuCQ4SSZI4a/+zGGMbw9L3lopOv6OWMalw9HRy5sRUjhzDSfdX8OwFUHoAnHC7ar49xsmT0VdXIy9/42svpnyeIAG9l5KMyJSvsxkFCa3JNCAxtdW9lagSTdoigIismOx9eEzZK5A72we0/75IdL2ZbDo62zrTbRFARKZc94Ju7/zJMmlwik4+6Kk5Sw5vThTUqxCZSnTyJdbyy8LwVNJIPXP5rD3PIfD5F0Q7Omi56peMvf8+pD7GnoxEdmfY6Qq6OPmFk9npF8+3ylbF/DHzk+KpylY1MMfwGWfBu3+ENY/Dwqv7vs+6F8BggwkL+3/+LDB/7Hwe/cajXPz2xSx5bQk3HXETC8YuGO5t7RU5MTVCCYSj7HD62dHtT/vc6gpwzAFlnD+vj9BujtFFyAdPnQ0ocMZjqhr5SZJE3uJFdD70N6JOJ1qHQ7VzjzZC3ih+Qx+GnZ0dgAOdeWCGhJkDjoH0Ib0J5FYo3T++njokaoswRwnHwr1GyQQNBRDzDciGIRgNslXeyuKaxYAQiAaTFq0uLlw8bSBpwNpTg6bEBmZEmmqLAKT7WXnawFyQln4KNzcjmUx4P/yQ7sceo/Cc/lmHDBfJ9KUpCob0eqDPOz5np38nl868lG+O/yYVNpWsExxjhUha+zgsuBI0GV2k0YhI8U1cDPrBmXKqyaTCSTx5/JNcsuISLl1xKUsPWsrZ+5894q93OTE1DCiKgssfZnu3n2ZnulhKfN3pDaU9RiNBeZ4Ji1HHb19ex+c7XPz+5KmY9OpNoc8xhCgKvHwZtH4OZz0Dheob19kXLabzvvuRV7yD4+STVD//aCHqg4C19ygZ2ekCHGhNA7uQJDymavJrksf8nlAfc/laYcJC5Ab1xFQi0hHQCRFTaEoZJWNyIMuBAZ/7K9dXxJRYT7TIkxFtk1tFtCh+cfZ1dRN0PYK/66h+r9XgbMCsMye7If2pYrQP9/NwczN5ixcTdbnYecufsMyZg2miOn5X2cQXd6s35/XuHl3ftR6A0yadRr5RnfE8SWYtgWfOg8YVUHdM+m1bPwRfR3IW30iixFLCQ4sf4pfv/5KbPrmJLe4tXHnwlclGj5HIyN3ZKCYWU9gpB9nh9LE9LpKaM6JM3lB6watRp6GqwEyVw8wBFXlUOczJ76sKzJTlmdBrNSiKwh0rGrj1zY1s7vBy75J6SvNGzruKHHvJf++Fz56CI6/u/SKnEqYDp6CvrERevvxrK6YioShENMSM4fQOobAfT7wgfDCRqUprJWZdzziaXk7hQQ8E3WAvR+5qG9A6fZGcy6cTc/nS0nz5Y5E7Bp5S3OTcBPSk3gLejGibZ2eayPF3i3E20WD6G8C9ocHZwPj88cl2eL8nTGG5JWWdlOLzUIjIzp3oq6oovWIpm084kealV1DzzNNoRnhBeiAhphy9BzZv6NpApbVSfSEFMOk4Mdtz9SO9X2fWvQg6M+x3tPrrqoBFb+HWBbdy2+rbeOjzh9gmb+OW+bfscqjycJMTUwPAG4zQ4vLT7AzQ7PTT7ArEhZKPZmeAFpefcFRJe4zDoqfKYWZckZVDJxQzJkUoVTrMFFkNexXGlCSJS4+qo67Uxs+e/pQT7vyA+86ZzdQxWfhHzJEdtnwIy38JE78Bh/9f1pYRXX2L6XrsMaKyjNY+Ml+EskmieFpvyaitcTcjR4xoNBq0A7wQN7maqHX01EslnMLT3M+TxdoVyJ1fDGidvkjUTMlaUaSclubLr0IeREqx0dmITtIxLm8cIISbJT9jlIy9p14q5PUNaq25lXOT3wc8IUy2+GuZpw3GHJS8LdzWBrEY+qpKdEVFVPzuBrZfeBHtf76Nsl9cMeA9DAW+RJqvsHe6fX33+uwVWuuMMO0M8ebN094z4zMWgy9fEgIrW3MCVUAjafhZ/c+oyavh+v9cz5JXl7Ds6GXJSOZIIiemMohEY7TJQSGSnP6kOBLpOPG1M6PFWSNBqd1EVYGZGWMdHD+tQgilFLFkM6r7o/7G1Aqqiyz84OGVnHrPh9xy6nS+OS0Lk8hzqIu7BZ4+Fxzj4OR7RNdNFslbvIiuhx7C88475J8wvB07w4G/r/EukHQ/t+XZBlT0H1NiNLmamF0+O3ks4RSeLNSOrwOg2MpUrpkKodFKdEc7gZTIlHs7VB8yqGL3BmcD4/LGJeez+eUQRZUpF1y5DcqnJr8NeQZWL+UKumj3tycjYEpMETVniZSipz0tAhbe0QyAvlK8ztkXLMBx5hl0PfQQtiMOxzp3LiMVv8uHhjCGovS0pS/s4yvXV2kdoaozawl8tAw++zsc+hNxbPt/hSgegSm+vji57mSqbFVc/u7lfPeV73L7wtuZXjJ9uLeVxtdKTCmKgtMXTqbdWlw9kaWEeGpzB4ilB5XIN+updJipzDcxe1wBFQ4TVQ4hkiryTckU3FAzpTKfFy45jAsfW8UlT6xhY6vMZUdPRKMZ2YV6X1siIXj6HAh5hTGnKfvRRNO0aejKy3Evf+NrKaYS6TCbPSOVJ7fiiRixlRcyEPOAVm8rgWggo5MvnsrJrC8C/JKdaESdmXzQU1vUFexEQsJhdIiUor8b8sfg/moQYqq7gSnFU5LfB1JrpmJR8O5M6+QLedUpPg/6IihKXIwGPRD2pqX5ws3pYgqg7Ior8H30Mc1XXsX4F19Amz8yI/T+LhcWjQspw4y3wdmAgpJdC4DS/UWEb/WjMPcS8eZh3QugNULdouytqzKHVBzCY8c9xiVvX8L3Xv8evzvsdxxbm0UR2k/2WTHVsNPDy581J0VTQkAFwrG0+xl0GirzTVQ6zBw6oZgqh4mKuFCqcpioyDdjVTmqpCYldiNP/OAQfvXPz7l9RQMb2zzcevp0LIaRu+evLa9fKd4Rnvo38QI3BEgaDfZFx+D8+1NEPV60tpEb0s8GCYGT32uUTDOeiIGS4jKcSv9Fzq46+SBzLp+ITMkh8f+oZHZUDZBEbdZWfycOo0MU5rrFIFvyxyJ3rh7Qef0RPzs8OzhhPyG8w6EokXCsp2bK1wlKLCPNNzAxleqyDin2Afa+3c8TYkpX0dPtpjGbqbz5Zr464wxarr2WqltvHZFdX36XD5PGBXnpxfIbujcAMKkgy35KM5fAS5fC9k+EsFr3ouj0M+Vld12VGZ8/nsePe5zL3rmMpe8tZYt7Cz+c9sMR8TvfqyuuJEnHAn9BTOi8X1GUP/Zxn9OAawEF+FRRlO+quM9+s7XLy1/e3kSJzUilw8zkcjsLJ5WKCJPDFP+897VKIxmjTstNp0xjUrmd37/6Jaf81cd9586mymHe84NzDA1rHoOVD8Chl8KUoS0Gz1u8mO5HHsXzr3fJP/74IV17uHG6hX9UUUF6xEJxNyOHjYwvLsfZvr3f593s7C2m/J4+RsnIraC3IMt+8b3OxIBCYRn45TCmuPt5T/H5NvE5fwxy5/IBnXezazMKSprvE6QIxKTHVI/IGWiar9HZiEVnocIqxJE/VYx6erufh5ub0ZWU9Co2Nx84hZKf/IT2P/8Z94IF5J848lJXfjmEpY8hxxu6NmDT27JfA3TgyfD6VbD6YZC0Ih288FfZXTNLFJgKuG/RfVz74bXcufZOtri3cO2h12LQDm8Twh7FlCRJWmAZcAywHfhEkqQXFUVZl3KfOuAqYJ6iKN2SJA27D/zhdSVsuP4bGHSjw9RtsEiSxPcPH8+EUhuXPrGGE+98n3uW1FM/7us9SmRE0LwGXv6ZmH111DVDvrx55kx0JSXIy9/42omprm4XCjFKHEVpx4NdzUQULbaiYhiAmGpyN+EwOigwFSSPBTIHAkPcsLMcuUvUNgkxFez/E8kg4A1TPNaWPpcv7jGl2CsHXIDe0J0eLUpG26x9z+ULBfxEQ/3v4gMhpiY4JiTfzKYNiZbj61jTxZS+qm/RUfT9C/C89x6tv70ec/1sDGNGVoGy3xujoA/DzvVdovg862/ojXY48CT4/J8ivafRwaSRkyLrLwatgd8d9jvG5Y3jzrV3ssOzg9uOvC3t/3Go2RulcTDQoCjKZkVRQsDfgUzp/wNgmaIo3QCKouxUd5v9R6/VfG2EVCpHTirlnxcfis2o48x7P+aZlduGe0tfb7yd8NQS4ctzykOgHfr0q6TRYD/mGDzvvUfMN/DOq9GI2+0lqPNTassw7GwXF2tbYVFfD9sjm52b06JSkBFZSSC3ga0cubMDjVYHuoxC+AHi94QwW/V0BjpTPKZ2gKTBr1iIhgc2B7DR2Yheo2esfSyQInASNVOJyFR8lIzcMfCi+k3OTT1jcMiIgiUjU+lpvtR6qVQkrZbKG28EoPkXvxjwrMVs4Q9oMRuCaYad0ViUjd0bmVw4eWg2MfMcUYe28kGonS8MUUcxkiTxo+k/4uYjbubzjs95/MvHh3U/e6M2qoDUK/L2+LFUJgITJUn6QJKkj+JpwX2TsB92rocNr8FHf4VXr4DHT4N//Ai6twz37gDYr9TO8xfP46DaApY++xm/e2Ud0cyq+hzZJxqBZ88XF4bTH01zjB5q7IsXowQCeN7797DtYTjwygECuj7cz51iTt5AxVSTqymt+ByEmDJadGhTm1HikSlPZwf2oiJg8BGIWDRG0BvpI823HeyVyE7ngM+9ybmJ2vzapDlir9RlxpDjgXYNdge66Qp0JSNg0GNjYbbFa6YkLViEUFRiMcItLeirdt2xbBhTRflvfo1/1So6739gQPvKBuFglEhUh9mafrndJm/DH/Fnv14qwdiDoXgioIyaLr694djaY3ni+Cf40fQfDes+1HqbrAPqgAXAGOA9SZKmKoqS9l8tSdIPgR8CVFdXq7S0yigKeNvFzLSuJvG5+yvojn8dLyhNYrCLIZJfvS86JI74uaiL0fV2uh1KHBYDfzv/YG54eR33/buJTTs93H7mTPJM6rwzzrEXrPgtNP0LTlwGVbOGdSuW2fVoCwuR31hO3rGLh3UvQ0nAE+49l09RkF2izsde2H+B2x3opjvY3UtMBeRQui2CoohIzsRjkdd1YCssxtk8+GHHgbgFg84i4e3yptdM5Y/B3THwxECjs5GZpTN71sqMtnnaRBdqfPyIe4DmoJnF5yDqwHRGLTqDVqyT4rIeaW+HcHiXkakEeSecgPzuu7TfcQfWefMwHzhlt/cfCpJdnhljhtZ3C+fzIYtMSRIc/EN46zqYvG+l+0fCQOS9EVM7gLEp34+JH0tlO/CxoihhoEmSpI0IcfVJ6p0URbkXuBdg9uzZwxcqiYTEC09XU49IShVP4dSCSknkuQtqYcJRUFgjvi6If7YUij9S13ZhxLjiBvj073DczcM+PFKv1XDdiQcysdzONS98wUnLPuD+cw+itvjr1dE15Pi64LUr4H/PQP35MPPs4d4RklaL/ZhjcL30ErFAAM0AR6iMNsI+hbA+gFWf8jfv68ITEhdpW2H/awoTY2T6SvOl+VkFZfFaYi9H7vyMiv0mgRpiKi5wwkYxMiatZqqqfsD1Up6QhxZvC6c6egZj+z0hJI2E0Ry/VMitaam3gUamMm0RIGOuYYb7eabH1K6QJImKa6/Fv2YtzUuXUvuP59CYh7cRJzmXLz99Hxu6NqCTdGk/g6xz0Pdh+plg7O3EnmNw7I2Y+gSokySpFiGizgAyO/WeB84EHpIkqRiR9tus5kb7TdADHRvTo0pdTSIV594u2nsT6MxxcVQD4+f3CKWCGnBU790QyPwxcNoj0PCWSP09ehIc8G1Y/HvIH95iyLMOGcf4Yhs/fnwV3172AXedNYt5+w1fymmfZv2rYuaer5PAvCswLRw5zsx5ixfhfOopvO+/j/3okTlCQm0UvwSF6XYoyMIWwWI1oR1ADVPSFsHRW0zZC1NeK+L1RYqtHE/nCuyHzAM29nu9TBIWAkGdqH8rNBUKR2v3DjjgBOS2DnR6A/21tWp09S1wTFYdUsK7LmNe3kBrphqcDdj0NsosKWNpPKFdzuXry2NqV2jz86n84x/Yet75tN10ExXXDH3TRyp+p+goNRemd5Su71rPeMf4oe1Ck6SckMoSexRTiqJEJEm6BFiOsEZ4UFGULyRJ+i2wUlGUF+O3LZIkaR0QBZYqitKZzY3vkYY3xYDHBNZSIY7GzU0XS4W14p9WrW6K/Y6GH/8HPrgd/n0LbHoTFvwC5vwYtMOXYps7oYgXLj6M7z/yCec8+F+u+dYBLJkzbtTbQowY/N3w2pXw2d+JlU7hwXE38bsVei5RNvPzRcMfggawHHQQWocD9/I3vhZiSlEUNEEDOkvGDe4W5LARm6P3aI+9ocnVhElrSrb0Jwh4wulDjuMlAX5NHtFIBHuROm9geubyuYG4+7mvA6IhyB+L+4t27MXFdLf27yU4ES2qc9SlrZU2azBjxIvcuVO8dir9SzRkdvL1WsvbDmUHJm/rj5gCsM6ZQ+H55wt39PnzsS9Y0K/9qYlvp4jemYvTf/8bujYwp2LOcGwpRxbYq5opRVFeBV7NOPablK8V4Gfxj5FB9aFwxhPx6NK4oVXjOiPMXwrTThNGjW/+BtY+AcfdArWHD90+MqgusvDcRYdy+VNr+c0LX7C+Vea6E6YMi3v7PsWG1+Gln4K3ndaZP+XcTYezYVWIyeV27ljRwH6lNk6cMfyt2pJej+3oo5Bfe51YKDTih8MOlnAwiiamxWTNeJmTW/BEDNgLS/p+4B7Y7NpMTX5NcjgvCOHm94QyOvlEZEoOi/pJm0piKlEU7taIAcNFpqIMj6kN2ItK+i2mGpwNmLQmquw9f6v+1NSbosS7E1PTfB3ozWbC/ewSbXQ2srA6vQzCL4cpKLeKKFtmmq95B1qHA41170sUSi6/DO8HH9By9a8wv/gCuqKBNRsMlkCX+D2ZS3t+bp3+Ttr97SOi1ieHOuy7V1F7mSiyK5syfGHNgnFw5pNw5lMQ9sHD34TnftDjoTIM2E167lkym4sWTOCJj7dy9v0f0+UdmE/M1x6/E/55ETx5OjFLIY9OfYh5H8/BFdLw8PcO5sVLDuPg2kKuePYzPt028A4rNclbvJiY14v3gw+GeytZJ5EOs9gzmkHkFuSIEXvpwGZZ9tXJFw5GiUUU4ZGUsg6ALEqbyCsamHjLJOFn1S2JFFuhuTDpMSXEVAf2AazV0N1AbX5tmkhME1NBN0T8STGlKArujnYM/XTV7/R30h3s7lUr5PeGMdn1EHBCLNxrLt/eRqUSaAwGKm+5mZgs03L1r1D6GT1TC5/Tg04KoC8ekzyWcD4fsuLzHFln3xVTI4lJx8LF/4UjroB1z8Ods4WtQlS9WV39QauR+MWxk/nz6dNZs83JicveZ2ObPCx7GbVsehPumgufPUXHzJ/w7dDv+PV/dZw4o5Lllx/B/IklGHQa7j67nhK7kR88spJWV2C4d431kEPQ5OUhL39juLeSdbqdIg2Wl5f+ZirSvZ1AVI9tAILDH/HT7GnubYsg9+Ux1QoGO7IsGloGasPQaw+eMHqTlu5wFza9DaPWmBRTMVsF3q4u7MX9j4I1OhupK6hLOxZIjbYl3gTGR8n43S6i4TAGa2Yedff01ckXCUWJBKM9tggAKd5gwrCz/+LXNHEipf/3czzvvovzqaf7/Xg18Lv8mDMMOzd0DdEYma8JkQEax6pJTkwNFXozLLwafvyRqDl4/Uq4dwFs/XjYtnTSzDE89cM5BMIxTlr2AW+tG76I2agh4IIXLoHHT0Ex5vH0zL8x97+H0uyJcu+Sem49bQb55p4LaqHVwAPnHoQ3GOGHj64kEB5eM0HJYMC+cCHyihUoI+AFKJvsjLuOFzjS5/J52kX6bSBpty3uLSgovTr5eswt+3A/7+xAq9NhyVNnCG+i663T35nuMaW34vFHUZRYvyNTrqCLnf6dadEiJaYQiPtZAb3czxNdg4Z+pN5gF7YIqU7rGesoirJbw849UXD22VgPPZS2P/6R4OamAZ1jMAS8Ucw6b5ph5/qu9ZRby3GYBla3l6MHZ2sLDy+9mM/ffWtY95ETU0NN0QQ4+zk47VFRtPzgInj+YvAO3El4MMysLuDFS+YxvsTGDx5dyV/fbRy2cPiIp+FtEY1a+zhdsy7hFG7kig+1LDqgnDcun8+iKeV9PmxSuZ3bzpjJ/3a4WPrsZ8P+87UvXkTM7cb78fAJ+aGgo1ukVosL0p2ePV2iIHggkaK+ZvJBT0qxV2QqbthpKyxC0qjzcuuPF2qnu59vj6f4hIDM66dQTBSfpwqcoD+CEktJXXrSI1MJP6v+pvkanY3kGfIoNvfsMc1pPcP9POp0ovj9AxZTkkZDxR/+gMZopPmKK1AG6A4/UHx+CbMhfc0NXRuYXJBL8Q2Wts0NPPmbpQQ8Hgorh7cuNSemhgNJggNOgEv+C/Mug8/+DnfUwycPQGzoIxcV+Wae/tFcjp9awY2vr+dnT3867BGUEUXADS9eCo+djGKw8kL9w8z972E0doW448yZLDtrFoXW3RdzH3NAGUsXT+KlT5u5693GIdp431jnzUNjteJePrBhuKMFp0ukrsszhIUcT//ZByCmmtxNaCQN4/LGpR1PmyuXXKgF7BUDrmHaFYnIVJe/K91jKn8M7rjvk724f+v1FS3qZdiZHHIsCsMHGplqdDayn2O/tE6+NDGajEyJdRIeU7oBiikAfVkp5dddR+Dzz2m/664Bn2cg+IMGzJaeN1CBSIAmd1Ou+HyQfPXpap667ip0BgNn/PYmKifuP6z7yYmp4cRghWOug4s+hPKp8MrP4P6jYMeqId+K2aDljjNn8vNjJvLPNTs4496P2Oke/hqfYafxHfjrobDmUdyzLuZs3S389H0th+1XzBuXHcG3pu/9C/xF8yfw7RmV3Lx8A8u/aM3ipnePxmDAtnAhnjffGvJ36UOJx+UnRpSKgp5CZiJBPF4xaNg2APfzzc7NjLGN6eUN1GvsiqIIUWAvR+7qUK1eSqwl/Jg6AxlpvvwxyHFH8v7aMDQ6G7HoLGl2D71HybSKIbnx1JS7ox2d3oDOuPfTHhRFocHZ0Kv4PODJGCWjM4ExDxCdfACGXQw53lvyjl1M/kkn0XnPvfhWDc1rrKIo+MMWLDZt8liDs4GYEssVnw+Cdf9+h3/eeB2OsnLO/O3NFFWN3fODskxOTI0ESibBuS/Bdx4AdwvcdxS8dJlw0h5CJEniJ0fVcffZ9Wxsk/nWne/zu1fW8er/Wmhx+Yd0L8NOUIaXL4dHv42iM/H6IQ8zZ+URfNoS4KZTpnH/ubMpzeufi7gkSfzxO9OYPtbB5U+t5csWd5Y2v2fyFi8i6nLh++STPd95lOLzhAjqfdiNKTVTnjY8YSN6gw6jpX+F0yBsETJTfCAKtTU6Cb0pftEMOCESQLGVibl8/YwU7Y6AHMZg1eIMOkVkKhwA707IH4vc2YHRasVg7n9ReKbvU9rgYRDpN3uPJ5/cIfys+uPR1+HvwB1y9+7kk1MiewlbhPh5++sxtTvKrr4afVUVzVf8gqjHM+jz7YmQ7CGGHlPKa8X6LjFGJheZ6j+KovDfF57ltTv/xJj9p3D6tX9U9Y3KYMiJqZGCJMHUU+CST4TB5+pHROpv9SPCd2UIOfbAcp698FDGFVp5+MMt/Pjx1cz9wwoO+f1bXPjoKu7+VyMfb+7EFxqebp0/t+EAACAASURBVMSs0/SeiEatfAhP/YVcYLqVC9/VMrPawfLLj+C02WMHbHZq0mu5b0k9eSY93394JR2eoMqb3zushx2GZLHg3oe7+kK+KBFDMP135RYeU7Y8+64fuAuisShb3FuoddT2us3vCWO26nvWiqfE/FqHMOxU6QU/HIoSCcdQTOJ/r9BUKJzPAfKrkDvbB2aL4GxIS/FBz+DhtDSfracucCBr9ZVOTKwlSWC06OJiKt0WQWOxoMkffAG/1mal8sYbCbe00HbD7wZ9vj3hbxG/G0tBXvLYhq4NWPVWqmzD7z2XLRRFwfXSS3Q9/jgxvzpvxJVYjHcevpd/P/E3Js09nJOuvA6jZeSMRlNr0HEOtTDlwbG/hxnfhVf/D178Cax+FI7/E1RMG7JtHFCZx9MXziUYifJli8yard2s3eZkzVYnr8dTVFqNxKQyOzOrHcwY62BmdQHji61oNKPUVT3ogbeuhU/uQymcwL/mPcJPPjASifq4/sQpnHXIOFWeW2meiXvPqefUu//DRY+t4vHvz8GgG9r3NRqTCfuC+chvvkn5b36NpNXu+UGjjJhPSoqOJHKz8Jiq6L+42eHZQTgWpjavDzElhzHZ+/CYioh5bGrVTCXSYZHEXD5zUbrHVMfH5PUzCtYV6KIr0LXr1Js9pQC9qEcEyR3tjJs2i/68peprJh+IKJjJphdjazw7xWSKOAlbBLWmNVhmzaT4wh/RcddfsS2YT96xx6py3r7w7xT1X+ainiaIDd0bmFQwKc3Pa18i5vXScu11uF96CYCOu/5K0fnn4TjjTLT9bFZIEAmHeW3ZrWz8z7+pP/5E5p99gWoNHWqRE1MjlfID4fzX4NMn4Y1fw73z4aAfCHsFkzot1nuDUadlxlghlhJ0eoKs3eZMiqsX1zbz+MdbAcgz6ZgeF1Yz448r2ENx9ojgq/fh+R+Dcyu++h+xtPMEXnnbxUE1dm4+ZTo1Kg+HnjbGwS2nTucnT67h189/zh+/M3XIR/vYFy3G/epr+FauwnrIwUO69pAQ0KEtyKgJk1vxhA2MLS7r+zG7YVcz+SBjSG98HQA5PlBZrVEyidRbQCe8q4rMRdD8pbgxfwxyZzsVdRP7dc6+OvlARNt0eg16Q1xoy61QcxgA0UgEj7Mbe3EJ3YHuvV6rwdlAgbGgp3A+TtooGU8bVB+SvC3c3Dyo4vO+KL7oIjzvf0DLNddinjEDfXnfnbiDxd/ZCTgwl4hi+pgSY0PXBk7c78SsrDfcBDZuZMdllxP66itKfnoploMOouPue9h5y5/ovO9+Cs87l4Kzz0Zr3/vIcNDn5YWbb2Dbuv8x/+zvMftbJ2fxGQycnJgayUiSiFBNOg5W3ACf3Adf/BPmXAiTvwnFE9WbKdgPimxGjtq/jKP2FxekWEyhsd3Dmq1O1mxzsmZrN3eu2EQs3sBSW2yNR66EuJpcnjfkkZhdEvLC27+Fj++Gglo+POJRLn7fiDckc/Vx+/O9w2rRZinS9q3plWxsk7ljRQOTyu1877DeEY9sYjvicCSTCXn58n1STOlCBnTW9K7UmHMHngG6nyfEVKZhJ4ii8L7m8sk+kaJXey6fVyvq7QpNPe7nYWMxftmtXupNTjHsjARFHVg8zefp6gRFIa+4hO7t/RNTmbVZkOK0Hg2Dr7PXkGPzjOn9ek57QtLrqbrpRjafdDLNV11F9QMPZCXS4e9yAQ7MZaKwf7u8HV/Et08Wnzuf+wet11+Pxm6j+sEHsc4Rgrj6/tn4P/uMjr/eTftfbqfzwYcoXHI2heecg3YP8zHlrg7+8Ydr6dqxneMu+Tn7H37kUDyVAZETU6MBswOOvwVmng3Lfyku/m//FgonwOTjhNgaewhohidVo9FI1JXZqSuzc9pBoqvCG4zw2XZXPHrVzfsNHfxzjagfMOo0HFiVz8x4BGtGtYPKfNPQD13e8qGIRnU3EZj1A66WT+a5N7qZNsbCn06dTl1Z/+tq+svlR09kY5vMDa+sY0KpjfkT1StU3hMaiwXbEUfgfvMNyn519YgLmw8Gb9CLMWzBYEsXU76OHShIA+rka3I1UWwuJs+Q1+u2XgOB5VYw5SO73Gh1Osz23o8ZCIkOO1kjPLSSc/lsZchxK4j+Frs3Ohux6+2UWkrTjvu94d6GnfaEYWdK1+D2jXu1jqIoNDobOX788X0+r8JyS9xvTwGreA5Rj5eYy6VK8Xkmhpoayq66ktbfXEPXI49QdN55qq/hjzfumAtFNmFfLD6P+Xy0/vZ6XM8/j2XOHKpuvgldSfrfoHnaNMb+9S4CX35Jx1/vpuOuv9L1t4cpOOu7FJ53Xp9zEzu3b+O5P/yGgMfDyVdey7hpM4bqKQ2InJgaTVTOgPNfBdcO2PCq+PjobvjwDrAUwcRjhbCacKSwXRhGrEYdcycUMXeC+CdRFIVmV0DUXsUjWI98tIX73xeOxEVWA+NLrNQUWaktsVJbZKWmWHxvNqgsEkM+WHG9GOlTMI5VRz7Ghe+b6fY6+dkxE7lowYQhG/6s0UjcetoMvvPXD7nkidU8f/E8JpQM3SxJ++JFyG+8gX/NGiz19UO2brZp7m5DgxZbnjntuKdjJyANzLBzF5180WiMoC/Sh/t5BXJHu6qGnYnIVLfUgVFrxKq3ishUXlW6wOkHm7o39RktCnjCPfVScob7eUf//azafG14wp5eETCxVgiTLb+X+7latgi7wnHqqXje/Rftf7oV69xDMU3qX4p0T/jlMAZNAK1e/P7Xd61HK2n7/BmMRoKNjWz/6U8JNW6m+OKLKf7xRbutvzTtvz9jbv8LwU2b6Lj7HjofeJCuRx+j4PTTKPzeBejLhKDfseFLnr/xOjQ6Hadf8wfKxo/8n1dOTI1G8qvg4B+Ij4AbGt4Swmr9y7D2ceHRMn6BEFaTvpE2fX24kCSJKoeZKoeZb04T7zJDkRjrW92s2epkXbObpk4v725s55lV29MeW5FvorZYiKvxxT2Ca2yBpf/pwi3/gRcuhq5GQrMu4Lf+U3nstS4mlxt46LyDOLBq6OrREliNOu4/dzYn3vkB3394Jc//eB75Fv2eH6gCtvkLkAwG3MuX71NiqjUuLPLz04Wp3N0NFPZbcCiKQpOziePGH9frtqBXlGCbrH24n2/uVNWw0+8JIWkkOmM7KTQVCgHk2g4lk1M8pvZ+PUVRaHQ1csy4Y/pYK0xecVyMZogcd3yt/gxv3lXxuRJTeoSbZ2vaOmraIvSFJElU3HA9m084kealS6l55mk0/fDN2hN+n4LZ0NOxu6F7A7X5tWKe4ijH9eKLtFxzLRqzmbH334dt3ry9fqyxro6qP91C8SUX03nPvXQ99jjdT/4dxynfwXnIbJY/ej/24mJOvuq3OMqyU8+mNjkxNdox5cGBJ4uPaFikrja8CutfhY2vw0uSmAU46Rsw+fhhq7PqC4NOw7QxDqaNSc+be4IRvurw0hT/+KrDy+YOL6/+rwWnr6egWCPB2EKLEFfF1jTBVekwp9c6dW8RnXpf/AMc1Xx+9GP88N8WWt1dXHzkBC49qg6jbvg62sYUWLhnST1n3vcRlzy5mofOOwjdEETHtDYr1sMPR37jTcquvHKfSfV1dHcDhvRRMoqCx+0BCvsdmeoMdCKH5b7rpeJF4WZ7Rpqv5jDkrg4q/p+9846Oqz7T/+dOH82MerMkd8uWAYPpYEwLYFNSCCSklw2kNxIIJdmwYdNIgITNbja72ZD8kk3OkkYCCcUONUASiAlgA5ZsWba6RqM2vdy59/7++N5p0sxoVGxJ9n3O8TmJpLnfOyOhefS+z/t5181fSycWknG4LIzFdfq5polK9bpLBJFcmlnVbTQ2ij/uz18tCiZygZ2QXiUTHB3B4fZgdZTOWiuUzYpHkmja5L18Kfq5qEzNdwA9W5bqapq+/jV6P/oxfN/5Lg233jJv147GLDidmVZz+1g7ZzQu7XyiGovh/frXmfj1byg77TSa7r47XVGaqeyrV9N0xzep/eQnGP3h//DKnx7l1Vefp8bp4q0fvZ6KJWKkwDBTR5fMVlhzvvh36R3gfVWYqo6H4PHbxb9FkrMqJrfdwgnNFXmrROPhBAdHwzlm6+BImF2HxggnMr+0bGYTK2rKaKuCdyd+w5ne+0AyETrjBr4Xu4x7/zjMmlozv/34Fk5eUTXlnIXQaauq+fqVm7jpt7v52kN7+cqbjz8i55Zv30bo8ceJ7d6Nc/PiziWUqrEJP1BHXVV15oMxP6G4hMkkzXjpcKGdfJBn7YqqQnBIB3a+guesrbN6DvmUvZevoaxB7PeUw2KSr9uHq6ISi7X0qub+8f3A1GqRklRJxJRM6zLoBcmUzjIFR32zymZVO6qpcuT+95ZaJeP05DFTAwNIViuW2vkJ8BeS+/zzqXr3uxn76U9xn38eri1b5n5ROUo0WUa5S/yBMh4bZzgyzIaqpZuXih88SP/1nyPe0UHNRz9K3ac/hWSZu42wtrTQtWkDr+77B03uSjbtepWBt11D+I1vpOajH8W+5sgO58xGhpk6WiVJYkVN4ya44OYlkbMqRVUuG1UuG6dMMkCapuELxrOqWX5aun7DG3t+QqXm57fKudwpX8PQn2uAYT50zmq+sH3D/Oex5qhrTl9OhzfIvc8eZEOjh3edseKwn+m+8EKwWgns2HnUmKlgIIIVqK3KqnoGhwglbbjcZTOuwKWxCHnM1JS1K9ExUGWilpp5BXZCBsEwGh3luJrjchhTgZG/T21fTrNTuxAWIQPsTOEKhqCsNv3HV3DEh6duZtWIfGBQyLx+DrcVRobBXgFW0V4UWIRlR6RiWv+FGwn/7W8M3HIrax58YO4XDAwQUSto0CuWHeMdwNINn/sfeoihL9+GZLOx/H9+iPvcc+fluqqi8NiPvs+eJ3Zy/AUXc8mHP4U6NsbYvT9m/Je/xP/gg5Rfdik1H/sYjvXzm2mbTxlm6lhRTs7KD52PC2O1d/HmrGYiSZKoL3dQX+7gTPUl+Ps/w8heWLEFddvXOdPVxl0jEQ6Ohjm+qXyKGVtMuvWyNjqHQ3z596+yutbFWWsO77oEs8eDe8sWgjt2UH/TF478VOVhUDgYp5LJrbcBQrIdd23xcex8Oug/iMvqmjLxBiI8DVMXAgcVsdIlJ8OkTeNuplE0JFPV6GQsNpaDRRD080eoacnaUaaBI1x8YXnnRCeV9sop3KfUepd0Diy1SkZXYNRH88bSK6epSb4r11055XM5S6LDwzm/e+SBgcOWl5osk9NJ053f5tA73sngv3wFpNPndD3N30dMLcdZKd5mO8aWpplS43G83/wmE/f9Eucpp9D8nbvnjcslx2P88Z5v0fWPv3PmW9/BOe94L5IkYa6vp+HWW6j5yIcZ+38/ZfwXvyDw8CN4LrmY2o9/HMdxx83L+fOpoyMgYWhmclSIjNXVP4KbDsD7H4BTPgDe1+APn4G71sOPLoFnvwsTvQt9t6VruB1+/jb4+dWQjMI1/wv/9DCmllNoqSpja2st7ztr5aI2UgAWs4l/f/fJrKwp4+M/f5HeschhP9OzfTvywACxV1877GcdCcVCMqopidWeVXkMDBJM2ma1J6/L38Xq8tV5jWZ0cpsvDewURi67WmSPqpjm4KdioQQmJyiaomMRhJnSylsIjo6kjZsyIdAJ0jSbqApxn1IG0ZlnlUwiGiEeDs8o6D4YHiSSjExpJ8KkHYCTV8kcQTMF4Dz+eOo+8xmCO3bgmOOmp/jwIBpmnNXCvLePtVNfVi9M8BJRorubQ+96FxP3/ZKa665l5U//37wZqUjAz6+/+iW6XtrFRdd+gq3vfN+Un0NLTQ31N3yedU88Tu0nPkH4b89z8Kqr6f3ox4i+8sq83Md8yTBTx7rMVlGRuvzbcP0e+NizcMGtoMRFYPueTfCzt8DuX4O8SJcdh0fgj58X+/R6X4BtX4dPvgDHvXnRhO1nqnKHlR994HRUDa776S5C8cO7B9HzhgvBYiG4c8dhPedIKRnRUO1yzi9nLaBXpupnPmbf5e/KSz4HYabsZRbMqYGBFLAzJlxTtpkyFS8UFVVq6k21C/NR46yBQB+Y7cQ1J3Ismj4r0dtX7FLienq1qGjrLZWZCnmzGFMj4nnNwJQWCp9DpqXodOuZKb0ypcbjKL4RrIcJi1BINdd+iLLTTsMTBkti9s434hOvk7NWVP3ax9qXFKwz8OgODl79NuT+AVp+8J/U33gj0gzyeMXkH/Zy3203MXyoizd//lY2b5s6JZstc2UldZ/5NOueeJy66z9L9OWXOfSOd9LzoWuJ7No1L/c0VxlmylBGqZzVBTfDR/8Mn3kZzr8JRrvg/uvgrg3wh+uhb9ec2xXzomQcnvs3+N7J8OL/g9Ovhc+8BFs+BZalP3q8utbF9999Cp2+ENff9zKqevhec3NlJa6zziKwYyfaYvjezlFazIzkzC3LJMb6kTUznhmukgnLYYYjw3kn+UAnhU/GIgDBiDwF2GlSZv/axqNi6k22ib186TZfRTPB8VEg01KU+6avKBfnPmW13lQlp2I0n1gEEC1Fi92MxWbWzxFmKjkoTOmRrEwBSGYzTd+6A4DK0dn/ERMbE2R4Z5WbuBLnoP9g3vC5Egpz8G1vp/cTn2T8//6PRO/CdgPURIKhr36N/uuvx752LWt+dz+eC+ePPD58qIv/+/KNRAITvO2fv0brGaWH/c0eD7Uf+xjrnnic+i/cSKyjg+73vo/u976PyIsvzts9zkaGmTJUWNWr4cIvwmdfEa3A9dvFrsAfXQT/eZYwMimY35GUpom1Ov9xOvzpNlhxNnzib3D5neA6vPmiI62trbXc9sbjeGyvl7t2dhzWszzbtyH39BBvbz+s5xxuRZNRrAk7lrLcqmRwWLw5zxSLcNAvwLKFzFQ0JE8FdjqrCY6N466pTYen1UgEaQ4+NdUOi1tF2zfd5qtoSVeLUkuOEz3TvyGnqkV5DY5upuwuC0TGQFPSbb40HHQGE3adE53UOeuosE+dokzvNUxEIB5Im6mEjkXINlPxSITRvp6Sz52trM3NqCYwTdMmLabIhNif6HTb6JzoRNGUvJWpePteYq++SvTFFxm6/V85cMk2Dmy/lKGvfo3gU0+hRg5/mz+lRF8f3e9+D+O/+AXVH/gAK//3Z/NqZntefYVffuVmJLOZd97+bVraZjexbHK5qLn2WtY99icavngriZ4e4gcOzNt9zkZGAN3Q9DKZRCtwzQUQu0sYmZd+IYzMY7dD6yWw+T1iMtBymJca970oVur0/g3qj4f3/Q7WvuHwnrnAev/ZK+nwBvnPpw6wvsHDlScfnraH5+KLGfrK7QR27MCxceNhOeNIaCQygkN25VaLgNDYCOCcsZkqNskHwnh4qrN4S8Eh8CwjNDyKJ2ttTSmtt2JK7+Wz+AG9zefvg9Xnp6tFqTZfKZWpQpN8qbPSrcs0YypDP5ckE+6q0l/HAxMH8po2EGgEp9sqwueQB9iZ+Xl/4YFf849HHuSTP/o/LLbFvUA9GhChK6fHSsew+EMon5lKdAtzuOpXv0RTVcLPPEvouWeZ+O1vGf/FL5CsVpynnYp761ZcW8/Fvr71sAyJBB97jIFbvwhAy3/8O56LL57X67c/9zSPfP+7VDc1c9Wtt8/LvkqT00n1+99P5TveseCMPMNMGZqZHBVw6gfFP98+MQn4yn0CEFpWA5uugZPfI9qF8yl/nzBue34lWDdv+jc4+X2LkpM135IkidvffDwHhkPc9NvdrNIXR8+3LFVVlJ1xOsFHd1D32c8u2am+4egwjqQblycXKBn0BwDnjH+JH/QfxCJZaPG05P18LCRPXXLsaSS410fT+owpLcXgFFOqWhQwjWOWzFRYXOKsihaCXh8ms5kyfXFsKcZt//h+ahw1U7hP4jllLTkOTqWfu6qrMRVZG5ItVVPp8ndxdevVeT+f3msYymOmTKYcIORoXy/JeBzvwQM0b1jchj8aFmUtp9tKe3s7ZZayvD9Die5usFiwNjcjWSzYV6+m+v3vQ43Hib74IqFnniX87LMM33kX3HkXlvp6XFu34t56DmVnn42lam4DNVoiwfDd32Hspz/FccIJNN/zXWwt+X/WZ6tdf/wdT//vvbRsPIG3fOGfcbjmd2XWfFLrZyvDTBmaverWwyW3wxu+DAeegJd/Dn//ETz/A2g8USxm3vR2KJvD9Eo8BM/dI7hYmgbn3gBbPwf2w7+EeDHJajbxg/eeylu+/ywf+dkuHvzUVhorSqdPl6ry7dsZ+srtxPftn/c9ZUdKw2Ef9qST8oosbpqSJBQSWSPXDCoqIICdK8pXYDVNDd9qmkY023iAAHbWbSQ0NoA7y7iV0norplRlaoJRqhxVmEJe0FRhpl4fwV1di0n/40Lu6YG64oa7UPgc9NblZPq5OxNAn0leqj/UTzQZLXxWUKaq0ZW1SkbPTA0MYGloyAk9TwyJatXgvr2L20wlIkQTNuy2JCaziY6xDtZXrcckTa2eJHp6sOlGKlsmux3Xli0CIHrzTchDQ4Sfe47QM88SfPxx/PffD5KE48RNuM/ZiuvcrTg3bZoRRFPu76fv858n9spuqt77Xupv+gKmeaz4aarK0z//MS8+9Htaz9zC5Z+6cdFXFGcrIzNlaO4yW2D9NrjmZ3DjPrjs2+Ljj9wEd2+AX70f9u0EZQZhTlWBf/wM/v0U+POdsPFN8OldcNFtx5yRSqnaZePeD5xOOJ7kI/+7i2hiDqNhBeS5+GKQJII7lu5U3/DYKBImqiszwW/Cw4RkG06nbUaEcCi84BhAjiuoSU0EtUEPa3uJWGoFsDPLTMlzDBanSOEjmjcHi0BFi4Bo6mdpiQTy0FDRa6maygH/AdZVFTZTGWCnXplKrZIZmUQ/j4n7kuIy+VQsfA4QDctianDykuP+XCyCpmn4h8XXDOxf5Lm+4CBRpYKyMvFad4x3FORLJbq7sa5aOe0lrY2NVF59NS33fJf1f3mOVff9H7Wf+AQAIz/4Ad3vejf7tpxD3/WfY+I3v5n2ZyD45JN0XXU1iQNdNN9zD43//KV5NVJKUubh/7ibFx/6PZu3X8Ebr7/5qDVSYFSmDM23yqrhzI+Kf0N7RLZqz6/g9QfAswxOfIeoWNW2Fr5G19Ow40vg3QMtZ8A7fgHL5wbQO1q0vsHDv73zZD78v7u46be7+d47N89rO85SW0vZaacR2LmDus98et6ueyQ1NjGBjWVUVWSZqeAgoaQNd/bHSpCsyvQF+/IuAoYsuGWqihMeAU0hhDD8uZmpuVemLDYTI8nhTF4KoGI5wVEfy1pFHkceGBArbYpoIDRANBktaHBiIZm6VOsy6E1TyTVVJTg2wrozzk5/rWloDABJD1xPVrGgezKhkIwrOhbBB0iCtK4/D+dpmeXb4fExkok4JrOZwX3taJq2eFvR/j6iajlOj4X+UD9hOZw3L6VpGnJ3N2WnnTajy0tmM87Nm3Fu3kzdpz9FcnycyF//SujZ5wg/8wzBRx8FwN66DtfWc3FtPYey007DZLejyTLD99zD2L0/xr5xIy33fBfbyunN3Ey187//nfbnnmbrO9/PGVe+/bB+r5KKSlLVcFgXLvZhmClDh0+Nm+CyO+CSfxWZqpd/Idp1z90jTNLJ74HjrxLLmgFG9sPOL8O+R6BiBbztx+Lzi/UX5gLp4uMauGl7G996tJ0NDW4+9YYixnQW8mzfjvdrXyPe2Yl9Xf7KxWKW3x+mjkn088AgQdmOp3pmLefeQC9JLVkYi5BCCKR32ImJwYAsWrBTKlPu2QMPRbZIrJJZWb4S/MKcaZ5lBEdH2XC2PslXQl6qWPhc07TMhB2INp/eeosE/CiynJ4aBMBffNrswMQBGsoa8NimVpTTPCuXFSa84KoFswUtmUT2einPqkxNeMVru/rk0ziw63mCIz7KZ7jS5ogpMEBUraC6vCxNPs9nppSREdRIZM5mxlJVRfnll1N++eVomkZ8337Czz5D6NlnGf/5zxn7yU+QHA7Kzjgdxe8n9spuKt/1ThpuueWw5I2CoyPsffYpTr3iLZz51mvm/foAMVnhz/t87HjNy+PtXm7ctoH3njX/prBUGWbK0OGXxSYAmse9WUw67f6lqFj94bPwyC3i43aPYEVZnHDxV+DMj4N1/jNBR4s+dv4a9nmD3LVzH+vqPVx6wvxtV/dccgner32NwM6d1C1BMxUMRKiDSTmmQUJJO8vqls3oWulJvoLAzvyrZEIJ8RdyuvWmKGLUf8Psv0/RkIzTZWUsNibafN4+cFYRjsqoSnJGjKli1SI5rqAk1dwAelaLTzyvjJmSgtObqUJ5qYwZtUFfhmWV9HpBUXLafBNDwkxt3HohB3Y9z8C+vYvYTPUTVdfhrPKwZ2wXJsmU9zVI9IicmG3l/O3glCQJx4b1ODasp+baa1EjEcIvvEBYD7InJyZouvsuKq64Yt7OnKzdjz+KpmmcfOmb5vW6/ojMEx1edrzq5el9PqKygsdh4aK2etY3LGz8wzBTho6sPI1wzmdhy2eg/0V46efw6v2QCIqVNhd+ccntBVwISZLEN6/axMGRMJ//1cusqN7CcU0za2EVkrWhHucppxDcsZM6PZOxlJQBTmbMVHK8j6hinTH9PM2YKp+mMpXKF6Xo55FkDrAz6fWCnD9TVKqiIRmr20xMien087/rjKlc7lOipxdpmmpD50Qn9WX1lNum/sxMef1CXmgW7bZ89HOpSGVKURW6/F2c3pi/TZ9jRrPo5/mwCBPeISSTiTWnno7Fbmdgfztt55xf9HkulNSJAWLaZpwVTjrGOlhdvhqHZeofh4lD3QCHpc2WkqmsDM8FF+C54AKAw94eVZIyex7fwZqTT6Oifu5/5HkDMXa+NsSO17z8rWuUpKpR77Fz9anNbD++kTNX12CzLHz82zBThhZGkgQtjtGd8gAAIABJREFUp4l/l35TTO25Z74z7ViWw2rmh+87lTf/x3N8+Ge7ePgz51JRNj/rHsq3b8P7zTuIHzyIfXV+I7FYlYiIYH42Zyo8LN6c3TPEInT5u2h0NVJmLcv7+Wg+44FEMBjLAXamJvm0OfzOj4USeKrEr+w0/bxyRcbgZFWmrNOMth+YOEBrZf72cDTbIGqavkpGvCmm6efZbb5AGAr4+P5QP3ElPn1lKrWXr0Z8XcZM5bb5yuvqsdrsNK5tZXDf4g2hx8bGABNOj432kXZOqT8l79clenoEFuEIUt4Pd86s8+9/IzwxzknTrIgppi5fiB2vednx2hAv94o9k6trXVx77mq2H9/I5pZKTKbFFf9YeDtnyJDVaRipWaq+3MF/v+9UvIEY//Lgq/N2Xc8lInAd3PmnebvmkVAsGUOKWcGiivUkuoKjYgosOxBeiopN8oG+eNgiYXXoZwUHwVVLaHws56xU602dwxtANCSj2oX5EAH03vQkH2SqRYnePmzLlxe8TqpaVCx8Dnq1KB4EOZKuGAVHfVhsdhxu0VJR43EIxwqeVaydCJkAv9NVqDKVacv6vYNUNoj/39TaxvChLuTEHLcRHyZFxwIAaI4kQ+Ghgjv5Et3debEIS1kv73iIivoGVp906vRfrEvTNHb3TXDnjnYu/s7TvOHup/nWo+0oqsaN29az83Pn8cQN53PrZRs5ZUXVojNSYFSmDBla8jppeSWffkMr331sH5cc18gVJ84sF5RP1qYmHCedSHDHDmo/+pF5uMsjo5HoCI6kC/OkQlJobByomBH9XNVUDvoPFoRNQibHlP5rPzgkgJ09ucDORE8vWCyiMjWLFSWKrCLHFGRbEmSoMTkg5ofyZoJdPix2Ow6XW0yH9fRQdvrpMD6c91p9ob6i1aLc1ptY6ZJeJaNjEVLPV+7rK7oiZzozFQvLSBLYTWGxXD0L2GmuqcHkyLTGJrxDbDh7KwBNGzaiPvAbvF2ds15JcjgVDYil8MOaaPvm28kHOhZhHvNSC62RnkP07X2V897zT9MSyZOKygsHx9j5upedrw0x4I9hkuCM1dW858zj2HZ8I82VziN053OXYaYMGToK9IkL1/JEu5d//v0eTl9VRX353MP75du2M3znnSR6e4tWOhaTfFEfDtmFzZU7Ih0MhoGKGdHPhyPDRJPRgpN8ICorjuypweAgmruR0NhoLrCztwdr8+xbObGwqODErGGQoVrWKzIVLQRH9lNeIwxOcmxMTIctb4Hd+Re/pgxOSa234RRjSqefj/pyXsPUKpRC6pzopMnVhMvqyvv5aFAAT6WIqK7lMKaaM3mpWChELBSkQq9MpTAQAx17F5+ZSkSIRoWR6E2KzN366qkA3NliERazXv7TI5itVo6/IP8qmskTeBMRGbvFxLmtdXzukvVctLGBatfSZFEZZsqQoaNAVrOJu6/ZzBXfe4Zb7t/DvR84bc7ZCM/2bQzfeSfBnTupufbaebrTwytfxIcj6aLMnfULOREmFNOwWs3YnPmzT/nUNSEm+YqZqRyEAEBwiEjlCSjJnklYhD5sLcshOrvNualqUdgs9vJVR0PiExXLCYz+Jd3iS4FBrUXM77QQzZCMySRhc1rS04mZJccjrN6cad8kerqL3nexnXyQvUomBezU23z9/diz9kOmsAiVjcJMlZVXUNm4jMHFCO8MDhJRxULnrvg+6px11Dqnmvg0FmHF0VGZSkQjvP7nJ9hw9rmUlWcWWuebwCt3WLhoYwPbjmvg/A11lNmWvhVZ+s/AkCFDAKyrd3PLZW3c/ofX+eXfe3nnGXP7JW1racFx/PEEdiwhMxX14ZDduMuzTFNgkJBsw+0pm5HBnG7BMQiTk97LpyQhNExIqgR6pgA7yzedAB3FzUfhczJ7+SrsFViDIlMkpvlGqF0upsFSQfdilcTOcVEtKhSqT/GsJEnKMTlKUiY8MZ5rEousyEmqSQ76D3JO0zlFn5dYcqyzsdwNaKqKPDiI+6KL0l+XNlMNmRZ2U2sbh3a/tPjgnf4+Ymo5kgSvh/cUJp+nsAgl0M+Xgl7/85PIsSibtwvkwrP7R/ivpw/kncA7a00NVvPRFdk+up6NIUPHuD5w9iq2rK3hq398nZ7R4vyfUuTZvp3Y7t3I/f3zcHeHX76ID2fSRXm2mQoOEEza8cxwIexB/0HKbeVicq6AYtlrV8LDgEZAEWenTIfi96P6/aIyNUulWm/j0khmlYxkRnHW6AYnlzFVbJqv099ZcI0MZFpvgKhMme3grCI0NgqaloNFSBmCfOoN9iKrctHKVNpMpZcc16OMjqIlEjkTbn6vqJBVZo3aL1u/kYh/goDPW/D6C6LAABG1AofLRJe/q3D4PIVFOAoqU5qm8fLOh2hYs47GtesZCcX5+C9epMsX4rpz13D/J7bwt1sv4mtXbuLc1rqjzkiBYaYMGTqqZDJJ3Pn2kzBJEjf++hUUtUg6uASVbxNTfYE/LY2pPl9oBJvipCwnxzQkKlM1M+OXpSb5ClU9FEUlHklmGQ9RPQnJ4v+nzFSKSG5dMXszlZp6G9G8GSxCeROhiQnd4GTOstTX5wS3syWrMof8h4q33sLZ9HMdpClJGSxC1uuY6OlBc+dnWhWjrKfPSi2JDnnBZAVnVUEsgquyCmvW82par+emFhsiIdBHVK3AVCaR1JLFK1MWS042bKmqf+9rjPb1cNK2y5Ekibt37iOaUPjZtWdyy2Vti3YCbz5lmClDho4yNVc6+cqbj+eFQ2Pc+2zXnK5lW7UKe1sbwR075+nuDq/GJ8RIuiMrM6X5+wknbbgbZvam1eXvKkg+B4iHxeJu5yT6eTCqCWCnnhtJVYvmEuKP6ZkpnzKY2ctX0UJwZBJjqqenaF6qNyCqRcUNTvaS46F0+DwD7NSp7rIsKpbl+cPlqaB7ocyZpuprazy2HNOWNlPNufTzVF4qpdrlK7E6nIvQTA0QowbZIYYE2qqOfizCyzsfwuFy07blPF4b8HPf33t4/9mrWFfvXuhbO2IyzJQhQ0ehrjqlme3HN3DXjn10DAXndK3y7duIvvQSsneRtVPyKBgQrc1sYGdkuBcVE+4ZrJLxx/2MxcYKks9BtMOAKZWpYFgWwE69opXKMVnn2OazuyyMxPU2X6Avh36egmgm+oozpqab5IOs1huIVTL6hF1mlUxmETGKgubJn73qnOik2d1cMJsVjyTRNP17FfKmWXOFKlPZeSkAk9nMsnWtDOzbW/C5LIj8/USoImz247Q4We7J//1I9BwdWITQ+Bj7X/gLx19wMRabnX/9w+tUOq189qL53Rm62GWYKUOGjkJJksQ33rqJcqeFz/3yZRLJ2U2RgchNwdIAeEZCohqQPWEXGhEVo5kwplJrZIpVpnL2yoGoTEkmgv7glAXH5upqzO78FZxSFA3JOFwWgnKQansV+PuhoiXdevPU1KLG4yS9XqzLi+SlJjqRkApWi1RVIxaWMwYxNJQ2U4GRYRyecqx20WpL5aW08vxmqRhlXTwnYUadnpSZymARTOXlmD0i2C8n4oTGRqlomLqaZFnrRnzdB5FjhcGhR1yBAaJJD2OSj9aqVswm85Qv0TQN+VA3tpWrjvz9zbP2PLEDVVE46ZLLeOTVIZ4/OMYN2zbM2zaGpSLDTBkydJSqxm3nG2/dxOuDAb73+P5ZX8e+Zg321nUEd+yYx7ubfyWUBIqeuc9echwcHQVmRj9PTfIVZUxNXiUTHAR3A6Gx0SmTfHPldEVDMpYyUemqMdlAldOTfA63B6vdIVpumjZtZarF04LTkh+GGI/IoOmvXzIO0fHMkuPREcqzFhynGVMVU02irMocChTPZkWzSeuh4RwsQnZVKqCzriZXpkDkpjRVZahr9j/f8y3FP0hCsTOk9BVs8Smjo0cFFkFVFHY/9igrTzwZZ20j33h4L22NHt41x0nipSjDTBkydBRr2/GNvP3UFv7zqU7+0TM+6+t4tm0n8uKLJH2+eby7+ZWgn4uMRraZCvlFjmqmlSmbyUaTqzBoMxaa3ObzorkbCY6OTqlMFcsxlaJYSAanqC7WKGL3IOWizTdTxtR0eSlIVYtSE3aZNl/uJF83UlkZmnMqZLEn0ENSTU7LmAJwuswQ9uXQz7PN1PjQVCxCSstaRbh70eSmEhGiYfF9mjCPFA6fd+uTfEsci3DgxecJjY2yedsV/OiZLvrGo9z2puMwH+Vh83wyzJQhQ0e5bnvTcSyrcHLDr14hkkjO6hqe7dtA0wg+9tg83938aTgyjEMWVZK0wVFVQqEYJkmirKKiyKNz1eXvYlXFqrwtmpRyKisggJ3WelQlmaafa4kE8uAgtjlM8omzEqh2Yd6qE2JVSWovX3pqcBrGVEJJ0B3oLp6XSu/KywZpZgLokxlThSorJWWzUpkzcwg0VTCmNG2KmfJPAnZmy+kpp2pZ8+KBdwYGiOrAzqglXGQnn86YWuKVqZd3PISnto6ydZv4/pMHuPT4Rrasndn+y6NFhpkyZOgol8dh5a63n8TBkTB3PDK7Nx17ayu21asJLOKpvtRePotDwpzi2ERGCcoWXG47piLGaLK6JoovOAZhpuxllsxZwUGCJsGkSk/XDQ6Cqs4pfK5pYupNtgoTVRPTBwr0Nl82Y0oqK8Nck78CdyhwCEVTSqoWOdzWDP3c00A8EiYeCaeD7iAyU4WM24GJA5gkU3F6vL4ix6mNiQ+461EDAdRweEr43F7mSi9Xnqym9W0M7GtH0+aGAZkXBfqJquUAxG1hWqvyZ8YS3d1LHosw2t9Lz6uvcNLFl3HXzv0omsYXL984/QOPUhlmypChY0Bnr63h2q2r+dlfu3lm/8xbdZIk4dm+jcgLL5AcGzsMdzh3icqUO6fFR3BAMKYqyku+TlyJ0x/qL2oEAGLBRGZqMJmAyAhBTbzhe/SWYrpaNIfKlBxTUBWNqDUMQE14AmweEtiIhUOZSb7ePmwtLQW5WKVwn6aEwgHcjVMm+TRFQe7txVZgGq1zopMWdwsOS+EdkdGgjMVuxhLP7OXLP8k3REVDY8Hn1bR+I9GAPw32XFAF+omqlQBUV5UXzKYlerqxNjctaSzC7j89gslsQVt/Bve/1M91W1ezoqb0dU1HmwwzZcjQMaIvbN/Auno3X/j1bvwRecaPL7/0UlDVRdvqG4mO4Ey6KcuGSAYGBf18BuHzQ/5DaGglVaacntTUmzAeQVmcnc4x9U2fY5pO2Xv5yixlOAMDelVKD9anUAW9xRlT+8f3Y5bMJVWL0rgCJHDVZTGmREg8OTSEJstYC7SpptvJB1l7DbPo5xnGVKZi48+DRcjWsjS8cxEgEgL96TbfyvrCU5WJ7m5sK5duXkqOxXjt6cdpPXML33yqn3qPnU9cWNikHwsyzJQhQ8eIHFYz371mMyOhOF/5w2szfrx9wwasK1cw+qN7Cf/lL4fhDucmX9SHW62YRD8fJJS04a6bOlZfSCkswnSVqWg23FJviYXiZgHs9IhKWKKnF8lux1JXV+gy0yp7L1+afl7RnGZMeWpq0TQtXZkqpAMTB1juWY7NPDUwnj4rVS2ymcVzctWC2ZKhn6eqYKm9ciumGoJSslnieSV0M6VXwFz1yP25wE5VUfAPe6nMg0VIqaZlOTbnIoF3+vuZMDeiSElaG/KbyTQWIc9rt1S097mniEfCBFafwSu9E9x8aRtu+9Ktss2HDDNlyNAxpE0tFXzqDev43Uv9PLxncEaPlSSJZV/5CiST9HzoWnqu+zCx9kXwBqbLF/FNafPFR3uRVQvuIlWCyTroP4iExKqKVUW/LpYDt9SBndFkDrBT7uvF2tKCZJr9r9qcvXzZ9HO9WlReW48yMoIWixVdWXPAf6BghifnObmyqm3uDBZBMplw6fsN0wHqPG2+VDZrOjOVJq2HhsHmBrsbub8fyeHArJ8THPWhKgqVjYWnKk0mM43rNjCwGELogQG85jqi1hBtNfkn+dJYhCVamRJ7+B6mZvkq7tktc9LySt568tLNfs2XDDNlyNAxpk9euI4TWyr40u/2MBycGezQdfbZrHn0Eepvvpnonj0cfOtVDNx8S7o9s5DyRX3YZEfOKpmQV+zFc9c1lHydLn8Xze5m7Ob8O+cgEwp3TF4lE4jmTLwleuaBMZXeyzdEjb0SIiMZYKck4aqqJtFbfJIvlozRE+iZtvU2pXWZWiUzMoy7uiYd4k/09CDZbFgapr6uqWzWtGcF5UxlypWhn1ubmtJmdGJIX3BcpDIFIoQ+0n2IRCxa9OsOuwIDTKgVxCyhIpN8OhZhidLPB/e34zvUhW/5qQyHEtz2xuOO+r17pcgwU4YMHWOymk1855qTiCQUbv3tnhlPQZlsNmr+6YOs27mDmms/ROCRRzhw6WV477wTxe8/THc9vcaCE5gUSw79PN0KqyqdMTXdTj4AOa6gJFWc6TbfIJgshPz+dD5L0zTBmJojFiFVmfKqA1SbdINXsZzgqA93ZRVmiyXDmCowNXjQfxANrYQcUyKHm5WqTAVGfempQdDzWSuW5624dU50TpvNAoimSOvZ9PNJWIQJHYtQUSQzBdDU2oamqQx1LjC8M9BHJOkiaY9T68yf08tU9ZZmZerlnQ9jdTj5ibeaKzc3cerKqoW+pUUhw0wZMnQMal29h5svbePx9mF+tat3VtcwV1RQf+ONrH30Ecovv5yxH/+Ezm3bGf3xT1Dj8Xm+4+KSFTkTns4Gdo5PAKS5T9NJURW6A93Ths9jeRhTmisX2KmMjYl2zhywCACxcAKTRcIne6nR9F/ZKSxC1iQfkoS1JX+7JcV9KrbeBVI5MCuoKoQzVPLg6EguFqG7p2Dmp5RsVjKhkIwrGThoin6ex0yZrdb0dGQhLWsVVaAF5U0lIhAdR5Gd2D2FMRyJ7m4wm3Oe51JRxD/Bvr8+g2/ZiWhmOzdflr/6dizKMFOGDB2j+uCWVZy9poZ//cPr9I5FZn0da1MTTXd8k9W//x3OE09k+Nvfpuuyy/E/+CCaOvudgDPRaGw0A+zMWnIcCoQAcFdVl3SdgfAAcSU+fVUlmEUKBwgOErE35gA709WiuQI7gzJ2lxlN0rLo5806sDNFP+/B0tiIyZbfwHROdGIxWVhRXry1JDJTNoiMgpoETyOaqhIcyRg3TdOmZUxNHz6fNDXobkCNRFDGx3PN1NAgFfWN02bOHG431c3LF3aiLzCADFgSZZSXF97DmOjpxtrSjGRdervr9jz5J5Rkkj8mVvOJC9ayrCI/+uFYlGGmDBk6RmUySdx1zUmYJIkbfvUKijo36KFjwwZW/M8PWfGTH2OurGTgpps5+La3EXruuXm648IajgzjSIo3sHSbT44Riqo4HBYsBUzGZKUXHE+LRZi8SmaIoFmYqJTBmS7HVKqiIRmz/p4l6OcSmmdZDpG8lEm+VeWrsJoKv4EnZQU5ruDIYUw1EAn4UZVk+qzksE+E3fXMT0IRhjmZVIkrcXqC02ez0qtkyiSITQjG1KBo6U3FImTyUoqqEYzlx3o0rW9jYH/HwsE7A/3st7iwqnbqagq3vpYqFkFVFV750yOMla+grL6JD59X/L+RY02GmTJk6BhWc6WTf3nz8bxwaIwfP3twXq7pOvtsVv3m1zTdeSeqP0DvtdfRc+11xPYevqqBLyom+SDL4ISGCCZteMrdJV+na2L6BceQZQZSmanQEEEErDEN7EznmEqfJCx4VmovXywA7gZisQTJRDzdepsum9U50VnSdB3oZjSUop83EhgRHKj0WT16gFpv8/mj4nGxpMoh/yFUTS0JiwDgMAsQKe56saiZDBZB0zQmvEM5jKn/evoAW7/1JIP+qUHzZa1txIIBxgcXaBgi0M9es/jeNxUYeNA0DblIi3Qx6+BLLxIcGeZvjja+ePlGHNbSNwocCzLMlCFDx7iuPqWZbcc1cOeODjqGgvNyTclkouJNb2TNIw9Tf8vNxF59lYNXXc3AzTen3zTnU76IL6sypRucwCAh2Y67srLk6xwMHKTaUU2Fvfgev2i28ZBjEB0nqIjyURrY2dOLpaEBk73wVGApyt7LVxMaz0zyIapgajRK0ucrWAGLyBH6Q/0lTfJBapVMqjJVn0U/n8SY0itT2fse90+IAPi0ZirVJpX05dt56OcR/wRyPJYTPn+pZwJ/VObLv39tSgWqaf0C56YC/XSZhJlqqc0/faiMjqKGw0uyMrXr0T8SsbioPe4ULt9UOrftWJFhpgwZOsYlSRLfuGoTHoeFz//qZRLJ+cs5mWw2aj74Qdb+aSc1111L4JFHOXDZ5Xi/Pb+Tf76oD2fSjSSBrUyHBwYHCCVteGpngEUoYScfiKk3k0XC6jCnqzihhAWz1ZoBdvbNHYsgzsrayxfw5jCmPDW1yH0C/1Bokq/LL6pts6pMuRuz6Oe6meruEXvllgmTE41nfl4OTBzAIllYWV7cLKTP0kb1c3Rgp8WSBpxODKUWHGfeuDuHg7hsZh7b6+WRV3PXx9Q0L8de5lq43JS/n36zuHdXRf41OpON6FLRxNAgvbtfYo9nI19+86aCq32OZRlmypAhQ9S67Xzzqk28NhDg35+Y//Fyc3k59TfcwNodj1J+xRWM/eQndF6yjdF7fzwvk38j0REqtRrsLmuaeZMc7yei2HA3lA4U7PKXZqaiOtxSkqQMYyqq4anOAnb29M5pjQyAqqjEI0mi1hBWkxWPv183U3q1qLZOTPJReP9fapKvVDOVBmnay8FWRmDEh8Vux+ES7dJETw+25ub0XrlwVmWqc6KTleUrsZqLh6ujoQSSBPZkJpslDwxgbWxEMov2UQqLUNkgKlUxWaF7LMKHtq7mhOZybnvgtZy1SJLJxLLWDQwuEAld8/czoq+SyaHwZylxKMWYWlqVqacffAANaD33Yk5oLl61PVZlmClDhgwBsO34Rt52agvff7KTf/SMH5YzrMuW0fTNb4jJv80nMXznnRy47DL8Dzwwp8m/4cgwHrUqhzEV9orMkruuuJkyTYhJxnBgjEAiMG1eCkSbyuHJYkwBwXACd41o86ixGMnh4TktOAaIhYVRCZsDVNsrkZJRYaZGfJgtFsrKK5B7RbWjkHHrHO/EZrKx3FP8XtJLjt1WYRB19lNw1Ed5TV2WSezJhM+TKjFZSV+jlJ18QBp4KoX1vXyuujxYhCEkyURFvcAmHPCF0DRoayznjqtOZDyS4BsP51ahlrW24evtJh6Z/XTqbDUU6kPSW805y7azlOhZelgEORGn/c+P0+NZy+fecvpC386ilWGmDBkylNZtbzqOZRVObvjVK0QTyvQPmKUcGzaw4odi8s9SWcXAzbdw8Oq3EXp2dpN/I9ERXElPzptYcESYHM80jCnLqAhBj42JKklpbb7sVTJ6ZcofzKAKpmm9laqUwfGbxqix6EF6vc3nrqlFMplI9PZhcrkwF8iGdfo7WV2xGrOpeGA4GpJBArsrF6QZHPHlwSIIM3VoNIyGyC4pqPQF+6atgEH2KhkvOKvAYptipvzeQTy1tZgt4nXuHBaYi9YGNyc0V/Dhc9fwy129/KVzJP2YpvVtoGkMde6b9h7mW+2xYRyyB5MVrPb8r3Wie+lhER64/2EscpTj3nApte655f+OZhlmypAhQ2mVO6zc+fYTOTgS5o5HDn/2JD35d9ddqIEAvdddR8+HriXe0TGj6wxHhrHLZbmMqbExANzTAB9N+kqd0YTIcE1HPwdhchxZe/k0k43Q+ERmkq9HxyLME/18HB/VJr0Sprf5ymuyJ/lWFMyxHJg4wLqqEg1Omd4mzV4lMzqSNonK+DhqKJTO/Oz3htKPj6ixkijroLdJs+jnWiJBcng4B4swMZSLRdjvDWE2SayqEdWf6y9uZWVNGbf+bk+6OrasdQNIEgP7j3BuKhGhnQTOpJsyj63g9yLRvbQWHMuKyis7HyLoqOZDb79koW9nUcswU4YMGcrRlrW1fOic1fz0r908u39k+gfMUZLJRMUbr8hM/r32Gv0fvE58rgT2VVJNMh4bx5Kw57T5Qv4AMD393BwQZmosHsBpcdJQNn1gXVSmUm2+ISKOJp3FlKpM6ViEedrL59O81KC/QVcsJzDiy2JM9RZkTKloDIWHZlAtyl0lk5RlwhPjaSxCaq+cdYVupoYz059hVbyOpZyVMVOCfi4PDYGmTaGfp/JSAPu8QVbVlGGziLcth9XMN9+6ie7RCP/2uMj52ctc1DQvP/K5qcAAHTYb1UoVZZ781Zs0FmEJ5aV+/Ps/Uxkeou3CS7EbKISiMsyUIUOGpuimSzewts7FF37zSpojdLiVPflX8Z53A2ALTH/2aHRUjMnHLJklx5pGKBjDYjFhLytMo1ZCIUwR0Uob1fNS000qKXooPLsyFbSICkrKuCV6ekXrrWpue8tSK3KG1QFqkkmwOFAdlYTGRvHU1qGpKnJf3xTGlKyDNBUdH7C2opRqUUIYnHgQ5DB4GghlTQ2CyEtBhjG1fzhTmQorUSwmC8vLpzeQ6R2AemUqjUXQGVPxSIRoMEBFQ/YkX4j1DZ6c62xZV8s1p7Xwwz938dqAqCw2rW9jcH/HEaPvAxDoo91mo1ypxlkgfJ7GIqxYGpN8o6E4L+14CMVs5cq3v3mhb2fRyzBThgwZmiKH1cx337GZ4WCc2x987YiebS4vp/qjqcrU9F/vi/qwKnZQpYzBiY4TTJjxuB1FzVF8XyZbE0pGSspLxfVQeHZmKigJ0+TJWiVjXb58ziPkMT0zFTL7qU7EoLyZsH8CTVXx1NSRHB5GSySmIBgOjYgcmKrnmUpu8+UwphpypgZBxyKYTOkdgJ3eEFY9ixXRYqyuWF2Usg6gqZqo7Hn0qUF3g8AikGFMpSf5GgV+IZ5UODQaprV+KoD1i5dvpKrMxi2/3UNSUWlav5FYOMTY4PzzzAopMN5Fv9WCPel3yjSXAAAgAElEQVTOrBiapDQWYdXSqEx9548vszqwj7VnnZee5DRUWIaZMmTIUF6d2FLJp9+wjvtf6ueRPYMLfTsFJYCd4pd9tsEJJW24K4uPcU+mspc2yZdnlYwmzs9pvc0DYyoalLHYJVSTQk00kJ7kA/DU1mb2/00Kund4RftNA5wWJ83u6fEQOTkmAHdDGg6abvP19GBdtgyTzUZSUekaCVFmE2ZK0VTWVUxv2uKRJJoGDrsKckS0+QYGxKJmnSnlT2MRhJk6OBJG1WDdpMoUQGWZjdvffDx7+v385LlDLNPhnUeSN7Vv5HXQQIrZMu3fSUpjEZZAZer1gQB7n34Mi6Zw3luuXOjbWRIqyUxJknSpJEkdkiR1SpJ0S5Gvu1qSJE2SpNPm7xYNGTK0UPrkhevY1FzBF3+3h2E9qL3Y5Iv6cMqTRtKDA4SS9mnD5/H23GxNqZN8gKisxEMQDxCUbWlgZ7r1Nh9mKiRjLhP/uyY8DhXLs4CdhRlT+7JI9qsrVmOSiv+q1zQta8Ius0omdVa6fdnbkw6fd49FkBUNZ9bkWmnhcx3BYNXxBXqbz1JXh6TvUBxPATv1Nt8+PeierzIFcPmmRi7e2MDdf+ogZKvC4XIf0dxUR+AgVsWOpkgF23xpLEJz6dyzhZCmafzrH17lxNDrNLRupG7l9H9gGCrBTEmSZAa+D1wGHAe8S5Kk4/J8nQf4LPD8fN+kIUOGFkZWs4nvvuMkwgmFL96/Z+GWyBaRL5qpTKXMlOYfICTbcNctK/ZQYntnbqZyVsnoVZxQzJQGdqZbb3Oc5AM9M+UUk2o1QV9OZaq8to5Er952W5b7PNuzzFQpgfBETEFVtKltvhEfzvIKrDYRqpa7ezLhc93glNksMzorvbZGCujn1OfFIpRVVGJzCifZ6Q1ikmBNXf78myRJfPXK47GYTHzpgddobN3AwJE0U1EvjbKomjnLC7T5lggW4dFXhxh47RU8CT+nXvbGhb6dJaNSKlNnAJ2apnVpmpYA7gPekufrvgp8C1icf74aMmRoVlpX7+HmS9t4bO8wv97Vt9C3M0W+iI9aSUzgpdp8keFeVEx4Ggu3VLRkkvj+/ahO8RgT0rRgS8jkmBwpuCUQjCTTwM5CrbfZKBpMoNrEedVKEiqaCYz6sDmd2MtcyL19WJctm/IGvc87MzMVywZ2hobAbANnFYHRzNSg4vejTEykGVOd+iSf0zqzylS6sicJdAUu3UxlYxG8Qznh8/3DIVbVuLBbCk+ULatwcvNlbTzbOcK4u4nR/l7ikfC09zMfaleCrFfE61SozbcUFhzHZIWvP7yXsxMdOMsraD1jy0Lf0pJRKWaqGejN+v99+sfSkiTpFGC5pmkPzeO9GTJkaJHon7as4qw11dz+h9foHTvydOli8kV9VCMyPalpvtCwMH3uInv5EocOocXjqOVij1qlrXzaNSgweSGwTj8PhLMWARdmTM20rhcLySRsMUxIVKmqXpnKcJ8EYyr3nGhCrF1JqVTuU/o56aFwJIngiC8rL6U/L73Nt88boqXKiVl/FzFJpZnRdOZMES1EzVmLPDiYB4uQqbbtHw6xrkCLL1vvOWMFp62s4jc9ZtA0BvfPjFc2G8mqTKeksApx//kC6JqmCcbUIsci3PvsQfzDw9RPHODEiy7FssiraItJcw6gS5JkAr4D3FDC135EkqRdkiTt8vl8cz3akCFDR0gmk8Rdbz8JSZJ4773P85sX+9Lj9wstX8RHhVqDySRhc4jKRTqkXV2YMZVq8aluYaZqbOUlnRcNydjLLJjNJggOoWmCaZUGdvb1imzMsqktxpl2SWMhmaglRKWlDDOkM1O5jKlcA7N/OJhzTmtl6/TnBFOtS9uUVTLpSb6eyYypUE6GySU5pqWsQwb34FQGQTKTDCuQTKaxCElZJjg6ks5LJZIqh0bCtDZMb6ZMJok7rt5Er6UODemItPoOjryOLEk0mgXrK19mShkbW/RYBG8gxvef7ORNjkNISJx48faFvqUlpVLMVD+Q/V9ri/6xlDzACcBTkiQdAs4CHswXQtc07Yeapp2madppdfpmcEOGDC0NtVSV8cP3n0qZzcKNv36FC+58ip/99VDObraFkC/qw61UiF1vOoogNCF2CxYLoMc72pGsVhSXePOrsZe2wDUWTGRI68FBInhQFSVTLerpzdt6m2neTI4rJGWVsNlPtaRfq7w5bXCUUBhlbGxK0L0jKy8lAY2uRqZTbmXKC55G4pEwiWg063npo/3Ll6OoGgd8IVqzputcZmdJz0tMKJqxRAfFTr4hkdFKVab8wwLgmapMHRoNk1S1KYypQlpX7+GjFx/HqK2aV1/eXdJj5qKOgb8DUG0S34d8lakU7HQxYxG+9Wg7ajJJ09Bu1px6BuW19Qt9S0tKpZipvwOtkiStliTJBrwTeDD1SU3T/Jqm1WqatkrTtFXA34A3a5q267DcsSFDhhZMW9bW8vBntvLjD55GQ7md2x54ja3fepL/evoAwdiRgXtmS1EVxmJjOJKunL18oWAESYKyAvvqQFSmbK3rwCR+DVbbSjNT0ZCcecMMDhG0iTf99MRbXy/W5VOJ5N5AvKTrZ87J2suHCcpqSEpWIv4JPDW1yP0FJvm8Qew6JdyEVBLrKpYdqg8Ogbt+KhahuwdLQwMmp5PesQiJpJrTenOZSjNTscn080mMKb9X5NBSjKlU0L2UNl9KHzt/LdHKFka79hOMJkp+3GzUPrIbu6pil+qxOsxY8pDCE90p2OnirEy91DPO/f/o50PNAeKhAJu3Xb7Qt7TkNK2Z0jQtCXwK2AHsBX6ladprkiT9qyRJBhbVkKFjTJIk8Ya2Bn778S3c95Gz2LjMwx2PtHPOHU/wnZ0djIcP75tXtsZiY6iaijXhzDCmFJlQJImrzIapSNsp1t6OY0MbJt1r1NlLo5XHwnKGtB4cImgSJipDCe9Nh7SztXcwUOKz0s9J7+UbEfTz8uYsInldGgI5OejePhRMt8RMlAYNjYUTmCwSVosC0TFwN2ZapWmqe0/aDKTI5631bkz620ipZipNWp9MP58M7NQrU/uHg0gSrK0r3UzZLCYuufBMbGqc7/76mZIfNxt1+LtolWXiSVdhLEL3oUWLRVBVjdv/8Dp1HjtNg/+galkTKzdtXujbWnIqKTOladrDmqat1zRtraZpX9c/dpumaQ/m+doLjKqUIUNHvyRJ4qw1NfzvtWfywCfP4ey1NXzviU7O+dYTfO2PrzPkP/yDvcPRYQBMcUumMhXyEkza8JQXfvNN+nwoo6M4NrZRZRVf1+CoLunMaFDOgoMOEkRUtDw1tSihEMr4eN7K1N6hmZmpVOvNxyDViWh6Jx+IapGcYkxNOmufN5huidm00mKx0aCM02VFiuhZVk8e+nlvD9aVuTv51tW7qbSIs6otpWXO0qT1sC9tpsyVlZjKBAZhwjuI1eHEWS5e1/3eECuqy3DMcDfc2WedCsDzz/+DF7vHZ/TYUqVpGh2RITYkZKIxS85uyGwluruxNi9OLMIDr/Tzcu8E128uw7u/nZMuuRzJZPC8ZyrjFTNkyNCcddLySv77faex83Pnsf34Rn7yl0Oc9+0nufX+PXSPHr7x9JGIPhEWNeVUi0KyHXeRvXgxHdZpb2ub0S/BDNzSKtLkwSGCSWca2JnCIuSvTAWnfKyYUpUpvzSWoZ+PZqpFcl8vpooKzBWZ9uREJIE3EGeDbqZKfW7RFLAzmEs/N5nNuCqrUMNhFN9IBovgDbGswoHHkTEHUolVsJRxS7f5JmMRhgapbGhMtyf3DwdprS8tL5WtqmVNONweVqsj3PLb3SSS8z8w4Y14mVDjbNCsRENKwcrUYl1wHI4nueORdk5sqaCu/0UsNjvHn3/xQt/WkpRhpgwZMjRvWt/g4bvv2MyTN1zA209r4bcv9nHhXU9x/X0v5QSj50vD0WEkTUKOapmqQGBArJIpEqBNTfI5NmyY0XlyXEFJqmLqTV8IHEqY08DORIoxla8yNcs2X8wSpiYe0c1Uhkie6OnF1jK5KiXab+sbZ2Y+YqkcWIp+7m4gODqCu7oGk8mcfl62lZk230wyTNmKhmUcThVUWd/L1z8JizCUzkvJisrBEif5JkuSJJrWt7HRPMb+4RA/eOrArO63mDrGBHqhzV5DNJSgrBgWYRHmpf7r6QN4A3G+dPEq9j77FG3nnIfDbezhm40MM2XIkKF514qaMr7+1k08c/OFXHfuGna+7mX7PX/mwz/bxcu9E/N2zkhkBFvSCRrpCbv4SA8J1YK7fqqhSSne3o61qSmnqlOKYjmMKR3YGVVzFhzD1KBxTFbo8oVmdFY0lAATxC1RahQlTT9PEclTy5Sz1aG3EttmaqbCcmaSD8QqmRFfhp2VFaBWVY3O4dCsqkXJhEIyruC0ihaw5qrLoZ+rqkJgeCidl+oeFStrCq2RmU5N6zcS8w1y5cYq/uPJ/ez3zq+hbx8Tpry1rIVYUMZRDIuwyCpTvWMR/vvPXbxlcxPWg/8gGY+zedsVC31bS1aGmTJkyNBhU0O5gy9evpHnbn4Dn72olRcOjnHl95/jvT96nr8cGJnzehpf1EejWbSIUhN2oSHxxl+Mfh5rb8e+ceOMz4umeEyeLGBnKJ61u64Xc2UlZk+u0djvDaHO8KlGQzJmByBp1CiqzpgSRHJNUUgMDExZptzhDeJxWGjUQaSlP6+EaL0FvYAErjoCI76s8HmGMdU/ESUqK7OqFqURDGZhLBWlDC0WSzOmQmOjKMlkJnyum5/ZGDeAZa1i6fEH10u47BZuuX8P6ky/EUXUMd7BiqSCxbUWVdUoy2Om0liElYurMvXNR/ZiliRuvnQDL+98mGXrNtCwZnpavqH8MsyUIUOGDruqXDY+d8l6nrvlDXzx8jY6vEHe/T/Pc9UP/sJjr3tnbap8ER/1knjjTQXQQyOiupIyOJOlRqMkDh2acYsPMriCVGVK0yAUCOVM8uVbcDzT8DnoVTBnEkCvTDXrwM46kkNDIMtT2on7hkJsaPCUhENISVVU4pGkXpkagrIaVMlEaGw0jUWQe3ow19RgdrvpHC6+dHja5wQ4Jb+4bkh839OTfEOi2pdaJbN/OIQkzQyLkK3Gda1Ikolw7wG+fMVxvNg9zi+e757VtfKpfXQvG+JxorYUsDMfY0qv6i2iytTfukZ5eM8QHzt/Lcm+/YwP9HGSgUOYkwwzZciQoSMmt93CR85byzM3XchXrzwBXzDOdT/bxWX/9gwPvNyPMsOqgS/qo8aU2sunr5IZHwUK08/j+/eDqmLf2Dbj+8/wmGwQGiKiWAWwszrFmOqbUi0CkZdyznAaLRaSUez6Xj5NSofCxYLj1CRf5ixN02gfCrBhxi0+YdicHj2A7mkkMjGBqiTx6LmzRHc2FiEzyTdTpc2oJr5H8oRgb2WwCAKTkMEiiJU1TtvMXruUbA4ntStXMbCvnatOaebc1lq+9WgHAxPRWV0vW6FEiN5QH23xBFFTapVMvsrUoUWFRVB0FEJzpZOPnLeGl3c+hMPtYcPZ5y70rS1pGWbKkCFDR1wOq5n3nbWSJ2+8gO9ccxJJVeOz973MRXc/xX0v9BBXSqOq+yI+qlRhZFKVqaBfvNm7qvOjDtLh89m0+SbBLYOIiUFPbS1aMinyPwXM1ExNTjSYQLZG8WDGVt5EPBYnEY3gqakl0aszprKmBr2BOIFYcubnZFfbdPZTMM2zylpbkzJT3hB1HjuVZfkn14o/J/31U71gcSD7RH4uY6aGMJkteGpr9bNmN8mXrabWNoY6O9A0lW+8dROKqvHl37865xbzvvF9ALQlEkQlQdrPV5mSe3oWFRbhl3/vZe9ggFsvbyMZHKfz73/jhAsvwWKb+ffTUEaGmTJkyNCCyWo2cdUpLey8/jz+672n4HFYueX+PbztB3+Z9rGKqjAaG8WjCsp5us0XiuOwm7Ha7HkfF2vfi8ntnlWlIBbS4ZYOs2BMmUUbzF1dizw0BMnkFCK5qBgF2bisNA5T+qyw2MtXo0npvBToWITePrBYsDZmFjl36PmiUteuZJ7T1FUy2TwrNRYjOTiYxZgKzToQnq7syQNp+rnJ5cKkDwL4hwapqG/AZDKTVFS6fOFZn5VS0/o2EtEoo329LK8u44Zt63m8fZiH9gzO6bqp8PmGhExUE/eftzJ1aPEsOPZHZe7e2cEZq6q5YtMydj++A03TOOkSo8U3VxlmypAhQwsuk0ni0hOW8eCnzuFnHzqD5ioBcEyqhdlA4/FxFE2hTPFgsZqw2swQDxKMS3jchWnc8fYO7G0bZpQrSika0uGWkiQqU5KofnlqatO76yYTyYcCMSYiMsctK93kaKrgWYXMfqqTyfQkH4Cnth65rxdrUxOSxZJ+TGqSb8MszZSzzKJXpuoJjgzrz6sOuS/VUlyBpqUm+WaJRQglkCSwx3vTwE5rU1P6ezHhHUovOO4Zi5BQ1Jz9f7NR03pRgRzUlx5/cMsqTmyp4CsPvsZEZPa0/n3j+6g02alXFKKyCyBnpRHoWIQscvxCqn8iyvvvfZ6xSILb3nQcqqKw5/FHWb351PRrbmj2MsyUIUOGFo0kSeK89XV8750nA4KLWUg+ndZtS5Rl3sQCg4IxVZkfeaCpKrGODhxtM2/xgWhTZeCggwRVVxrYmc4xTapMteuwzrYZVKbikSSapu/lS0T1BceZ1luip3fqJN9QiHqPnSrXzNo16dalJQxqUqySGR3B6nBid7nSa2v+P3tvHh/JXd75v6u7q++WWurWLc19aHyMx9jGDjaQDcFgYvCy4UrMLtmQBRaTsGzChmTDb0mWLPzAP4eNuXKQsCQmmJgQE0JYsonzS4AxeDwz+BppNJeO0dWSutVV3V3dVV21f3yrqrslzYyk7jlk6vN68cKSWv3t6tGonnmez/P+BLdvYzavoZYN9myywHGAp1Id/dwZ8VmWRW5umvY6vxRszuher/aeXiJt7UyfPAFAwO/j4//mINmizv/41olNP+/I0gj7fVGkaJpSwSQUDeD3N95Sq0tLmKp61TtT/zKW4d7f/xdOZwp8/u23cMNAO6eeepJCLuvhEFokr5jy5MnTllSmJIopfyVYK6aUGUE/70yt+T36xARWsUh4eOObfFAHt7Tp56ouu8BOfXICSZYJdDfCQl+wYZ0b8TI5PqasNU9n1XDp55LkI97RaTOmVsfIbNQvJa6p0RROosfFIkiS1MCYckKHN9+ZsotRuwMm6OeimCopeSqlEh02sNPZGtzdZDHlwDun7c4UwHX9bbz7Fbv46pEpvndqYcPPaZgGY9kxhqsStPXb4ddrmc9rhejVkGlafPofx/h3f/JDuhNhvvG+O3nN9aILdfw736Stq4cdh15yVV7bi01eMeXJk6ctKaczheZ36efV3BTFapB4V/+a36ONCGJ1aLOdKbUiCjctB4aGokl1Ju0p5MFBJH/j5tnIrMJgR4S28PoNyM7obcm3YGMRRC5fvDOFVShQXV5u6ExVTYuxeWXDfilxTTpy2I+/5ETJiM6Ug0WoTIyL2JpksulukabqRGJ+KC5S9Xdg5vN1WAThYXI6UyfnFAaSEeKhwAWfb73q2ztMduY8JaWGqPiVV+1lRyrKb379WUqV9S08ODq3fI6KWWF/SZDpS0rlAlgEhzF15TtTyyWdd/3ZER78zknecFM/X3/gZeyyw6IXJseZeuE5bnr1PRcNA/e0fnnFlCdPnraknM5UtYg7eivMnAMg3rd2J0AbOQF+P6G9m4MTaqousAgO/byo1wE7Jy4YI7NR83mpPkqmarpRMmKTz4msqRVTE0tFNN3cZGfKDm526OfxbgEHdRhTdViEU/MKnbEgqfja5v5LXpdSIRKxAAu9JP7MnGJqeU4UUzVgp7opMOha6t8nMBgzY6Pu58Kyn//xb25kfLHIp/7h5IaebyQrulzDygK09VNULtSZOndVsAgvTOd5/cPf5Z9GM/z2G67nU289RDRYK0p/9Pffwi/L3PCvXn1FX9eLWV4x5cmTpy2phdICyVCScsGoYREyglOU6F775lU+MUJo1058oY0XA9V6uKUyI4CdikAVWJaFPjG5KuDYiZE5sImsPABNLtSiZGz6ub4GY8rJPdyo+dw5K1xXIBqhTorLuTr6+UQDFmGzAE2wY2uCYqyoq+L24xQaublZkCTau3uomhanM5s3uq9U7669SD5fw6gP4GW707zttiH++F/O8tz55XU/3+jSKEFfkB3qErQNoKmVNYupq4FF+NrTU7zxs9+jbFR59N0/wTtetqNh2aJSKvLCP/8j+++4i2jbxuKUPF1YXjHlyZOnLan54jxdoW7KRcMd86k2PuBCniltdHTTI76yA7e0GVMC2GmS6ExTzeUwVXVVZ8qJkdl4Z0oUHFpAJeULY4USojOV7kKfWt2ZOjmnIElsOuJFdKbmIZhAUQXQsi3djVWpoE9PE9wuNvmawSI4G4qRgHh+fVm8n7Ux3zTxzhSBYJCpbJGyYTbNmHIkh8N079jFzNhqw/lv3HOAzliQX//aMxjVC2+P1mtkaYQ9iSFkwIwP2J6pNcZ8565cwHHZqPJbf/0sv/qXP+LmbUm++csv55btHase98K//BOVUombPON5S+UVU548edqSWigt0BcQXQ0n5FjNCgjkWsWUkc1izM4SHt44+RzEiApwO1OKLrpbiXT6ggHHJ2zz+WbGfFLAwvDrpGI9lPLLVHWdRKqLysQk/o4O/PFaUTM6q7CtM9owyln/WZValEyip4ZgSKXRp6fBNJGHtpFRyyyX9E0XU86GYtgn3hM9W0QKBvGnxJ9Vbm6WDtcvJbxZe1o05gPhm5oZO4m5AgjbHpX5nTdcz/PTeb7w3bOXfB7LshhdGmU4LMagWrAfrBqBv/5xlYmJK+KXms6VeMsfPMmfPznBu1+xiz9/5+10JVZ3Xy3L4kff+Vu6d+6mb+/mljA8rS2vmPLkydOW1HxxnrQkNpPcMZ9SJOCXCMdW34TLI2LEE2pikw+c2JVZFEn8qz/eWedjGmzsTJ2YzRMN+tnWGd34WRFx0+9M1GER0mnBmFoj4Hgz5nPnLNFtW0E/T3c1YBFOOZt8mzzH6bZFpCwA+kIeua8PySduQ7m5mTosghNw3Lpiqn/fMHpZY2FydTbfa2/o5e7renjo709ybqFw0efJlDJky1n2+cVrK/lEUbWyM1XDIlzeztR3xxa49+Hvcnpe5fNvfwm/8boDBPxr39qnXniWhclxDt39M5virHm6sLxiypMnT1tOpmWyWFqkE/tGFpfBNFGLBvFYcM0bhbPJt7IzpVcFzOpSuYCNUTIzKD7hKRI+JrsztaLIcWJkfL6N3bg0VccIlQlbFtH2beRtiGZbSuTy1Z9TNqqcXShsyi+lV6oYFbPWmYr3uGclOtONWIQmN/mc9y9sLkCoHX12zsUiVLQSxeWcC488NafS1x4msYENyEvJMaGv9E2BwCf8zn03EPT7+K9//exFo2Yc8vmwKbbgSqYoqqMrPFOXO+DYNC0+88Qp/t2f/IB0PMjj77uT197Qd8HHl4sFvvMHDxPvTDH8sldcltf04yyvmPLkydOWU66cw7AM2kxBIA/Hg1DICO5T+9pFRXnkBIHubgKpxhFgrihu8suli9OwtfoMO2UOxYoTkIM2sHMSf1caX6RGXndiZIZ7NzbiAzuXL1AkZVSRkkNutyjenrTz/2odsNPzBaqmxb5NbvKBE9w8DwmBRYi2JwkEg1QmJvBFo/hTKcbmFdrCgTXHRxs6y6wxpgIrsAjJ3hqwsxmj+1pq6+oh2p5k5uTaoM7e9jAfet0w3zu1yGNPT13weUaXRFG+TytBNE3JzkwOr+hMXU4sgsAePM0n//co9x7s5+vvvZPdXRd+vyzL4tuf/RTLmTnuff+vI4fDLX9NP+7yiilPnjxtOTmMqXhVFCqiWzSNYoSIJ1ebbkEEHK814itWjHWdWarPsFNmUYwI8VRKADvX2OTbTIyM+1oLOiX/Mp3VKrQJLEJADhIoFKFabehMnbQz+YabKKbCoSpUVDHmW8iQSNUYU/L27UiSZKMKEpseD7mes8p5zHAX1YWFOiyC2CRM9vRhmk5kTWvM545ceOfY6s6Uo5+7bRsv3dHJR//2wmT0kaURBuODJNR5Aey0w5tXdaYmxi8LFuHETJ43fPq7/NPoPP/t9dfxP992iNglWFxHv/U4p546zCvu//cMDF/X0tfjScgrpjx58rTl5DCmwob413g4JmPlpikYQeLp7lWPNysVymfOrBkjU1wnsLGk6iIyxCcJ0nrZT6LTxgdMTRFcscnnmM83EiNTf5biy7pYhPxChkQ6jWHn5NV7pkbnFGS/xI5UbBPn2D4mvw2zjNfo57CSMdUcqkAr2J2p8jl0Q6zkB10sQq0zdT5XoqRX2ddC87mjvr3D5GZnKObXxiD4fBIf+9kbLwrxHM2OMtw5DMvnXWCnJEEo1tiZ0sfHW45F+KujAntQqlT5yrvu4N/fufOSxe350RP88yN/yp7bfoJbfuZft+y1eGqUV0x58uRpy8npTAUqIUHvln2U5sepWj7iPUOrHl85dQoMg/CBRr/U7LJGxbi4V8qRpupia7C4BKaOUqqSSKUxKxWM2VnkoZWbfJvrGFV1E12rkvdlSZlmA2OqMrHamzU6q7ArHScY2Pivc6erEraWALDi3S793KpWqZw/T3DbNhbVMouFSlOjt5KiEwj5CRSn0TVhyHexCHMzhBNthKKxmvn8MhRT/ftFMb2Wb8rR7q44v/IqAXVdaZ0q6AUm8hPs79wP+fN2Z0psQ670xVXGWxdw7GAP/vNXf8RNg0m++St3ceuOzkt+XzG/zDf/5/9LIt3Fa/7j+z3T+WWUV0x58uRpy8npTPk02WVMKfOi0Ej0rvaoaCfsTb79jWJR4FQAACAASURBVMXU8cnsus90I0McYKeqEU+l0afOg2Wt2Zka6oxs2ETtjBOz/pw95uu36eeCMbUy/290dnOZfFDvYxKerLK/HV0rCZ7VzCzoutjkm29uk885KxILQDmPXhTm7VoxNeuaz538vz1drR3zAfTs2oPP77+gb8rRu16xGwCLxmpqLDuGhcVw2w4RKdQ2sGYun2VZVMbHW+KXmlku8VYbe/CuV+zikV+6ne7EpT1PplnlWw8/SCm/zOs/8Btrbrh6ap28YsqTJ09bTplihrZgG5VitcaYygjfTbxr9ZhPGx1BikRWrakfm8it+0ytUCOFFwwZ07REgTMptrZWdqY2az7XCmL0VgwopAJxTHwUlpYEqmBF/p+i6ZzPlTZfTBV0MaIy7HicivDeJFJd6BPj7nU1u8kHYqTo+PN1xQK/n0BPDyAM6Mk6xlR3IkR7tPXUcDkYonvHrov6poALdvncTb6ATQ5vG1gzl6+azbYEi/D9Uwvc+/vfZWxO4bP3v4TfvAj2YKV+8FdfZfyZY/zUv383PTt3N/U6PF1aXjHlyZOnLadMKUNXpKsWhQKoWTGqWgvYWT4xQnjfvlUhxMcm119MlRTdxSKohg3sTKWouPEutc6UGyOzGb+UUhclE+5EzS5iWaady9eY/+cUOZtlTJVUnXBcRlJnwSeTV8oAtK1kTM2rxIJ++to3vwWmqTrhkDD767kygZ5upECAqqGjLGTcTb5TmwxsXq/69g0ze3psFbxzPRrNjtIWbKOnbK/wtQ/YPxcrzOfnmtvksyyLz/7TKd7+hR/QEQvy+Pvu4nU3Xhh7sFLjzxzn+499mete/q+48ades6nX4Glj8oopT548bTllShm6ol21KBRAXVaQgFh74zafZVloIyOEVviljKrJs1PLBAOX9pFYlmUXbvYmn0M/tztTUiSCP512H+/GyDSxYacFVFKxHpQFG4vQmRZbg4OrM/k2s8kHoCkV0dlTVwA7U4IxJYVCBLq7GZtX2NPEJh/YxWhQA0BfUt0RXz4zj2WZJHv63MiaVmMR6tW/dxijXCYzfmna+UqNLgnzuaQIw7zjmVo55nOwCPImPFN5TWAPPvHtUV53Yx+PP3Dnht4PZWmBv334k6QGhvjpX3rA80ldIXnFlCdPnracMsX6zpTtmVI1YtEAvhXdJ2N6GlNRVsE6R+cUSnqVaLDx8WtJL1epGqboQCgzdfRz0ZkKDg423LQ2GyMDNc9USVbpTAyQt/MGY8GwyP/b1lhMRYN+BpKRNZ9rPWcJYOeciJJZzODzB4glO+yA4yEkn09gEZoscEoFnbBf0MX1+WyDXwqgvaeX6WWNYqV6Wcznjvr32Sb0S4z6VsowDU5mT9bM50A10iuyIVcypmwsQnCDWISR2TxvePi7PDEyz4fvvY6Hf+7mS2IP6lU1DL75qU9glMu8/gO/4fGkrqC8YsqTJ09bSpZlsVBaIB3sQi9XRTGgl1DLEvH46qJCs2NkVhZTjl9qPXl22krGFEkX2KlPTqzqQLwws7kYGXBwBRblQIlUxy43Ky9cEl2dlYypvT0bJ6y711WwR1TKHMR7bSxCCsnnQ58YR962neWizrxSbqqYMipVjHKViE/BMiWMzILLX3KxCD19LjOr1YypeiXSXcQ7Opm5yEbfWprIT1CulgUWIX8eomm0sijEV3am9PFx5P5+pGBwradaU3997Dz/+jPfo1Cp8hfvuoN33nVp7MFKffcrX2J69AXufvcvkxpcvdXq6fLJK6Y8efK0pbRcXkY3ddII87LrY9KDJJLtqx6vnRgBSSK0b1/D549N5EjFgsjrQAo4PiZnm08xY8RtkrrTmarXyOzmYmRAFG5WUMdPlbaO3SiLC4SiMaR5EfMirxjz7W+ii1NSdUHuVmfBxiIkUl1YpkllYpLgtm2cyjSPKnCBpyyik4ZqtdaZmp0hEAoRS3bU8v8u45hPkiT69g0zfYmNvpVyzOf7O/bbjKkBijaIdFVnanz9AceWZfGRbzzPf3r0OAcHkvztL9/FbevAHqzUqaee5Mjf/BU33f0zDN/5yg1/v6fm5BVTnjx52lKaL4mioh1RzITjMuRnBP08lV71+PLoCMHt2/FFG7tExyez3Lwtua4zSw1RMrMitqYzjZHJYGlaw+jNsixOzGxukw9EMVWVS3RUq/jahwRjKt1Vx5gShduCzX7av8lzLFP4wCIRHxQX7SgZcZaRyWCVywS3b3NRBc10i2oIhgx6VRQK9YypZE+foKzPK6TjITpi6+/obEb9e4dZnp+jkFs/GmMkO4Lsk9nVvgvy09A2gKbUhV/bcrEI6/RLHZ/M8cXvn+P+27fxyH+4ne62jY/mcnOzfPuzv0fPrr385L/7pQ1/v6fm5RVTnjx52lJaKNom6aoohCJxmcrCBBUzQLyrf9XjtROrzefLRZ3TmQKHhtZXTLnFQDQA6hxKyRIBxzaRvP7GOZvXWC5tLkYGROFWCaikqia0D6EsLAhD+NQk/nTaLQpP2ubzzQQcA5RLBpZpEQ6KDT4z6nSm0jUD9dAQY/MqYdm3aV+Wc00AYWMavSK6TvVRMi5jqknK+nrVtwnf1OjSKHuSe5D9MuSnoK2uMxWvdaZcLMKO9XWmDp9ZBOA//fQ+5HViD+plVCr8zUMfAwle/4FfJ9BC4rqn9csrpjx58rSl5AA7o4YoIsKxIMrsOWA1sLOqKOhTU6tiZI5PCb/UzdvWzvFbKWdMFfHnscwqalEnnkq7+AC5bszXjPkcROFWCuRJmUC006Wf6yvGiSN2MbWvd3PFh1sg+kXnqSAlsEyTtnQXuotF2O5u123WlwV1Y9LKFHpJdHHk/n4s02R5bpZ2e5Pv1Jx6Wc3njnp27sbnD6zbN2VZFiNLI8J8XlZBW3Y3+aCxM+VgEda7yXf49CJ7u+ObDpB+4n/9IfPnTnPPA/+Z9u7eTT2Hp+blFVOePHnaUnKKKVkX45BIQkadnwZYFSVTHh0FILwi4Pj4RA5JgoODqz1Wa0lTK/j8EnJlbgWwcwokqWFry4mR2SxIs6TqqP48KX8YvVKmpORpS3cLxtS2RvN5ZyxIV3xzN2HXxySJnDqlIjoaiXQXlfEJkGXk3l5OzSlNG8Ldwq10Fl2VRIctFELNLmHoFZI9fczmNZSy0RRlfb0KBIP07Nx90ViZei2UFljSloRfKi9+1mgfpKTq+HwSoWhtiaEysX7GlF41OXIuy0/sXs1GW49O/MsTPPN/vs1t972J3bfcvqnn8NQaecWUJ0+etpQyxQwJOUG1CEgQigZQl0SB5ZjCHbkxMis6U8cms+ztjq876sXhWUnq3Apg5wSBvt6Gra3NxshAjWeVC+TpDCZQFsUIKN6exJiZbWRMzSns64lvmiOkOSHHljhD0ST7ugSwMzgwgGpYTC9rTXOfSqodBmwtYeSNBr8UiIDjsStgPq9X//5h5k6PUTWMSz52NCuK8nosgpvLl5Ab/gwq4+vHIjwzlaOkV7lj18aLqcWpCb7zR59m8MAN3PXWf7vh7/fUWnnFlCdPnraUMqUM6WgaTdUJRQP4/D7UnOiurKSfayMn8Hd2Eujucj9nWRbHJ3PcPLS+ER+IMVXYYUw1ADunCK4KOM5v2nxe0aqYVYuCTT9XbMZUxAIsC9nGIliWxclZZdN+KajrTFXnAIm8XVw59HN5+zZOZwQXqtkCR1N1wlEfkmShZ0vIAyuKqZ6+lkTWbER9ew9g6BUy585c8rHuJl9DMWXTz1dhESbWjUU4fFoUshstpipaiW889DGC4Qg/8/7/soqt5unKyyumPHnytKWUKWbojnRTKtRiPJR8kXBQQg41bkKVR0YJD+9v6BycWyySK+ocWucmH9ghvQmbMWV3pgSwc7Ih3kXTq5xdKDThlxIFjaCf99YxpoRJPGiP+aayJQqV6qY3+ZxrAojo5yGaQllaIhiJEIxE0cfHCW7bzpjDfWpy9FZSdcJhE8sCPbPcYD73+f20pbsYs8eWqU2OLTeq/n1iKWE9JvTRpVEG4gO0BdtqYz67MxVdhUVYf8Dxk2eWGO5N0LmB7UXLsvj7P/w02enz/Myv/BfiHRvHKHhqvbxiypMnT1tK9Z2pcEwGy0ItGsRjjTdhS9cpj42tHvFNiHX49WIRQIypwjbPygF2hvwBqgsLDZ2pk3PKpmNkoDGXr7Nt0I13kbPCMO8wphy45f5Nms/FNen4ZR+B0rQdJZMhkeqiurSEWSwKxtS8SjDgY6hj85t8YBejIR1D82EZRgNjqi3djc/vv2KbfI4SqTTxVHpdvqmRpRHhlwJYnoJYFwRCYswXX4FFmJhYFxahbFQ5Mr604a7UM//n7xj53v/Py95yP9tuOLih7/V0+eQVU548edoysizLjZJxo1CKS6h6gHh7YwFTPnsWq1IhvAKLcHwyRyzo35CpWlMdUvgsipUgnkqhT4lxT33A8YhtPm9mkw/szlRyF8pihmh7EnN6RuTkdQmO1mgLOkaaWhE+sIKIkskvZGrmc0TA8di8yq50jMAmVvbrVVIqRGQNoyDGUfWeqfaeXpHJN6dckU2+evXvHWbmEp2pol5kPD8uyOdgM6bE6y8pOtG6MV81m8VUlHVhEZ6ZWkbTzQ0VU3NnTvHEF/+QnYdu4fZ//eZ1f5+nyy+vmPLkydOWUb6Sp2JW3Fy+Gv08RLyj0QPlbPKF9jdu8h2byHFwMIl/nav+1apJuWjUOlNGWPilpgREU67rTDUTIwP1uXwFUqn95BcywsM0OYE8NIjkE7+yR2cV+tvDtG3C5O6oFtwsomSUxQXaUl3uNpo8NOTG1TQrraAT9ivoJdE9lPtrUTLJ3n4ySpm8ZlzWGJm11L9vmHxmHjW7dMHHjOXGsLCEXwrsYmoQo1JFL1eJtNX+DDaCRTh8ehFJgjt2rW9Mp6kq33joY0TbO7jnfb/q/ix4ujbk/Wl48uRpyyhTFB6idF3IcTU3RaEqk0g3Mna0EyNIwSChnTtrn9OrnJjJb2jEVy6Iba+IQz8v+0h0plYRyUGYzzcbIwM1uKUmqyTTw268i2BMrYiR2eQosXaWXYyqc+iRLkr5ZRIOY8rnw+juZSpbanr05pLWpWV0XaAo5IF+SqpCuVAg2dPLySu8yefICT2+GG9qdEkU5bXO1JTwSzmes7ox30awCIdPLzLc20Yyemm/lGVZfPtzv4e6tMi9/+nXiSQ275XzdHnkFVOePHnaMnIYUyl/mqphEo7LFGbOAhLxnsZ8vPLICUJ79yLVEaGfO7+MYVrrJp8DLpgxHPVjKvOopaooOiYn8LW14U+K57Isi5FZZdMjPrBz+SSDGBqBYBxlcYF4Ko0+OekypvSqyZlMgX0tKKbCYQtMHdUSr1nQz8U22pmsuO5mC5xy0cCyIGwuoJcj4j2Lx1merd/ka43RfaPq3rkLvyxz/iI5fSNLIySCCfpifRcAdtZ1psbHwee7JBZB06scncjyE+sc8R35m7/i9JEf8Mp/+4uucd7TtSWvmPLkydOW0UJJGLLbLDHSi8Rl1DkR6ZLo2+U+zrIstJFRQsOrR3zAhjf5ACIBlaIRwDQt4p3pVQHHM8siRmaz5nPnrGqgSEoKUC4U0LUS8WhMGMLtztS5hQKVqtkUFsE5KxwUBUHeEFuQDhYhuG1bXYHTPGMKIGLOUin4ke1CI+tiEXoZm1dJRmXS8cubybdS/oBMz849F+9MZUfZ32FvhNYDO9fI5dPHJ5AHBi6JRTg+maNsmOuCdU6deI5/+Yv/xb477uLm175+HVfl6WrIK6Y8efK0ZTRfFCHHsaropITjQZSFWQDiXT3u44z5DNWlpdUxMpM5BjsidCfWHybrjnPIuViERCqNPjHR4I0ZmW0uRsY5qywrpPzhGmOqagG4CIbRueYI6yB8YJWSQSRQBEApC2O4A+yUbfN5wCexPRXb9DlQx7PSpzEUs4ZFsDtT7T29Ikame/MA0mbUt2+YubOnqBr6ml8fy47VjfgagZ2wujO1nk0+xy/10h0X90sVclm++T8/QbKnl7vf/StX5f3xtD55xZQnT562jBZKC8TkGJYmbv6RuIy6JMzD9cDO8ogY26zc5Ds2kd3QiA9q7KewlUG1gZ3xZAeV6ekVfqnmixxNrVAKKHQG21wsQrikAbUw5ZOzCj4JdndtvmPkdtt8ogBUSqb42B/AXF62GVMqO9OxTYXvrnVWuDyJnivXbfLNEu/oJBAMcXJeYc8VNp876t83TFXXmT+7Gt5ZBUpGqc58XgN2Flfk8rlYhHX4pZ48s8j1/W20Ry+8QGCaVb718Ccpqyqv/8BvEIpubqnB05WRV0x58uRpy2i+OO9u8gGEYzJKXiHgg3C8djPWRlZv8s3lNaaXtXWHGztyOyvGjNuZilRN0HWXSA7Nxci4Zyll1ECBVCRF3gZ2BpdFweOMx0bnFHakY4TlzVOv3fcPUYjmFY1YsgNzVnT5gtu3cWpeYV8LPExOBydkLGOWjRVYhD4W1Aq5on7FzeeO+vfa8M41Rn0moivYgEUAaOtHUwSnSw6JPwcXi7D94p0pTa9ybCJ3Sb/U4cf+gonnnuFV7/yPdG3fedHHerr68oopT548bRktlBboitYVU3EZtVAhHg00jEC0kRPIg4P4E7ViwPFLbWSTD0QxFYoG8BdmUYwwATmIb1GAP4MriqkDTRDJQRQeSrBAKiYgmj6/H/9chkB3N76wGE2OzioMt8B8DsIUTjCOks255nMAq2+AiaVi05l8ILAIAP6SGCnW6OczDebzVhRum1G8M0VbV/eaJHRLgoAvwO723eITK4CdkXgtl68ybmMRLtGZOjqepVK9OF/q7PGnefKvHuX6n/xpbvhXr97klXm6kvKKKU+ePG0ZZUoZ0pG0CM71SYRkE7VkEW9rHIGUT4ysHvFNZpH9Etdt0NPkktaVGRSrzQZ2NjKmnBiZ4Sb8UtWqSUWz0AIFOhNDYpOvM4UxVdvkK1WqjC8Vmy483DGfMSPo5w6w017tnwp3YFrNm89BgC0DsoVVFKNEub8fvayhZpdI9vRyysnku8LAznr17R1m+gIbfbvbdyP77W5jHbCzuCKXzymmLjXme/LMIj4Jbtu5tl8qv5DhW5/+/0gPbedVv/iejV6Kp6skr5jy5MnTllA9/VwUOAGk4hyqESTeXus2mcUilfFxQvtXkM8nclzX377h8VhJqdRy+aoRl/tEIIDcK0zvTozMdX2bL3IcnpUmq6Q6dokCJ5VuYEydmlexLJre5HNxD5VJrHgveTtKRh+fINDby9iyKLZaAdEUUTImukM/H+hneU6ME5O9fZycU0iEA3Qnrkwm31rq3zeMurjg+tTq5fqlQHim2oRPTlMrjZt8Np/rUliEw2cWuXGgfU3gatXQ+eanPo5p6Lz+A7+xKmvS07Urr5jy5MnTlpCqq2hVje5otw3sDGLlZ1CNEPFUl/u48smTYFkNnSmjavLM1DI3b9B8Dja9246SUSuyAHZOTiAP9CMFAoAY8QEMNzHmcwqc+iiZREcKY27O3eRzNgabZUw5o7ewdg4tlMYolxuwCKfmVfw+iR3p5k3PJbVCJFhGL/iRwmH8HR3knGKqp4+xq7jJ5+hivinXLwV2MeV0phpDjivnxi+JRShVqhyfzF1wxPfPj3yRmbFR7n73++nsv3hR5unakldMefLkaUuoRj9Pu/Tu0uwZqpaPRHef+zhtRNwQw8O1m+DonEJJr27YLwViTBWJy5j5GVTNsoGdjUTyEzNKUzEyUBu9leQCHeFOlMVForK4MbubfHMKwYCPHS3AFQQjtg/MZnYlnGJq+zbG5lS2p6KEAps3uTvSVJ1woIiuhUSxIUnkbMZUe28fp+bVq+aXctS1YycBOcjM2OpRnxtw7AA72wewLIuSohOuH/OtI+D46fEsetXijjX4Uid/8D2Ofutxbr7n9ez/ibuauyBPV1xeMeXJk6ctIYd+3h3ttrtFMuqs8KnEe3e4j9NGRvC1tRGwjc4g+FIANw9tbJPPskQUSjjqo5hfxrSwgZ01HxM0HyMDNVO4FigQKQcwqwYRsUyGPOgwpkQXZ725gheSyDX0Q0VBsURhFovEqS4uItvAzlZt15UUnYgvj14M1jb5ZmcIxWIUCbJYqLTE6N6M/AGZnt171uxMNWTyAbQNoJerVHXTZUxZliUYU5fwSx0+s4DfJ3HbCr5UduY8//tzn6Jvz35e+fZfbP6CPF1xecWUJ0+etoScYsrpTIXjMmpGdDjifbWbWPnECOH9+xvGRscmcnTGggx1RjZ0pl6uUjVMIsGyi0WIRSKCxWR3ploRIwO10ZtPLlKxcQhhTYz+nK3B0dl8034pECPFcFhUavlK0D5L8Kx8g0OcWyy2LHS4VNAJk0UvSA1YhGRPP2Ou+fzqdqZAmNDnz57GqFTcz/ksifaQyBMkL0j7tA249POo3Zmq5nLrwiIcPi38UvFQwP1c1dD5m9/7OD5/gHs/8Ov4A5tHa3i6evKKKU+ePG0JuWO+cFp0VmKySwlP2J4pq1pFO3mS0IpNvuOTOW4eSm7Yl+MiGHyqC+wMl4VR3OlMuTEyTRZTpXwZgFjYco3QweU8UjSKP5UiV6wwly83HXAMonCLhMR1KCXw+QMEbNxDpq2Lqmm1ZLvOqFQxylXCxjzVUh39fG7WjZGBKx9wvJb69x+gahjMnT3tfi5Q/4A6xpRr4I+Lwqdy7hxwcSxCoWzwzNTyqgiZyeeeITN+llf94ntoS3c3fR2ero68YsqTJ09bQplShkgggmyEsExLdKZyeSQsYkkxvquMT2CVSoTrNvmWSzqn5tVN+6UAIlLW7UyFVFEAON0ix3zeTCYfgJbNUfWV6AjFUBxg5/wCwcFBJEni5Jw4t1nzOdg+Jll0opSiQSItwpQBTgfE+9SK0ZszupQLc+L/+/sxq1XyC/Mke/s4NacQDwXoa7/6W2uOCX3GRiT4LYiZdbfI5booGfu6om2iM+ViEbZduJg6Mp7FMK1VsM7TR58iEAyx+7Y7WnIdnq6OvGLKkydPW0IuFsFGCETiMqpSIBb24/MLo3R51Daf13WmfmT7pQ5t0C8FtZDesLWAoocIyDLSnCh0ZHvMNzLbfIwMQCmniFy+SAplMUMgFEKaPu9S1kftTb5mx3yWZQkDv18UZ/nlAm0pwZjyp9OcVKpITcbVONLcYkp02uSBfvILGcxqlXa7M7XnKm/yOYolO2jv7nHhnRIgUfe68ucbgJ1Qi5JxsQiDF97Ae/LMIgGfxC3baz+HlmVx5uhTbLvhIHLw6qEhPDUvr5jy5MnTllCmlGmkn8dklGKVeLx2E9JOjEAgQHDPHvdzxydzSBIcHGrf8Jku3LI6h2KESaS6MKam8Hd24o8L4/YLLYiRAdDyGqVAgVSsF2VhwWZMna/5pWweU7NdHMc8HZZy4Aug5AT9XB8X22hj8yrbOqNNxdU4copRv1YARCROblaMy5I9fZy0DfXXivr2DjNzcgTLslZ/MX8e2kSx5BZT7phvHLm//6JYhMOnF7lpKEmszi+1dH6SfGaOXS95aQuvwtPVkFdMefLkaUtoobTQmMsXqqDqfuJt9Zl8Jwjt2oWv7qZ2bCLLnq74mpDES8kZ50T0KRQzTjwlNvnqY2RGWhAj45ylygU62wbJL2aIJ9qxNM1lTJ2cVdnfk2i6i+O+f+YiZqwHdWmRRLq7xphqYYHjjEkDxQIE/AS6ulzGlNSWZkEtX1Xy+Ur17xtGzS65XrwG5afriikdOeQnEBQF56UCjtWywbPnl1eP+J7+IQC7XnJbi67A09WSV0x58uRpS2i+OO9u8gFErCVUPUS8s7ZmXh4ZbRjxWZYlzOeb8EuBoFz7/BJycRLFCNrdoknkba2LkXFULAnGVCrej7K4QCwoOlDBoSEsy2J0TmmNX6pQ67apci+WaRJvb8eYmyMwNMSZBZU9Ldrkcwo3n1pE7koh+Xzk5mYIyEGmK6K4vRY2+Rz17zsArA3vZPk8tNc6U6uxCBfe5Hvq3BJV01oF6zxz9Cm6tu8kkUq36Ao8XS15xZQnT56ueRX0AiWj5NLPAfzaecpmgHi6FwBjcRFjfp7Q8AH3+8YXi2SLOjdv27hfCnDhoJYyS6HsI57sQJ+ZIWh3i0Znm4+RAZtnVQmiBVQ65CSFXJaIPWqSh4aYy5dZLuktwiI4uXzTKD5xE4+K2Dxynb3oVat1nSm1giRZoFQaAo7be3o5lRHBx9fSmC+9bQeBUIiZlcVUWYHysks/F8XUSizChTtTT55eRPY3+qVKqsL0yRPeiO9FIq+Y8uTJ0zWv+eI8IBhTWqGCLyBRXjgLQKLHzkpzyee1LLVjk2Ld/9AmYmRAFB7heJDiUgbTgqhfBtN0A46deJdmsQhGxcQ0A2hygVg5IOJwNB0kCXlggNG51pjcQXTbAMLlcRRLFDKhYgmAqYjonLRq9KapOuGggVH0ue9Zbm6W9p5eTs4Janx/+8bYX5dT/kCA3t17V4ce5wXPzB3zqbWQ4/VgEQ6fWeTQUJJIsOZDO/ejo1im6Y34XiTyiilPnjxd81ooiW2w7mi3uJHFZNQ5scofH9gNiBEfQKguRub4RI5o0L/puBLBs/KjqKLYCFfEJqHTmToxoxAL+hnqaC7DzjFql2SVkPBqE1IUAr29+IJBTtobg62IXXHGpOHyOIohCplgdhmAkwFRFLZik885KyxrGCUf8oAYVwpgp4iR2dMdb4oafznUv3eY+XNnsCyj9sk6YCdAKV8b8+kTE8CFsQh5Tee5NfxSZ48+RSTRRu+evS2+Ak9XQ14x5cmTp2teDrCzK9LldovUBcEuineJ0Ys2MkKgt5dAR22Ucmwyx8HB9k3Hr5TUCuGwgWIDO0NFMZpyuiytiJEB0JZEMaMFCvgdnMD8gmt0H5lV6EqE6IxdeFts3WepOpIPQlKBfNlP5IsJ2AAAIABJREFUKBqD6Wn87e2cUGEgGWnYOGv2rJClAKLDVshlMcplkr19dmTNteOXctS37wBmtYphZWufrAN2umiJeB1j6iJYhKfOLmFaNOTxmdUqZ48/zc6bb8Xna35r0tPVV2v+xrRIuq4zNTWFZscaeLryCofDDA4OIstepIGna0dulEw0jaZmiSRklKy42SU6xU2qPHKC8P7aiE/Tq7wwnec/vGLXps/VVJ2IDKoN7AwuZjFCIQJdaSzL4sRMnntv6r/Es1xapYwoDKtyCX1ZdKEC09PIL38FIAKOh1sw4gO7WxSRkCRQSpYw1U9MIG/fztic2tLtupJSoc0QnC+5v5+sHXAsJ7uYyy9cU5t8jvr3ip8hw1wA21NWD+yslAzMquV2pirjExfFIhw+vUjQ7+Mldb696bERNFXxRnwvIl1TxdTU1BSJRIIdO3ZcExC3HzdZlsXi4iJTU1Ps3Lnzar8cT55cZYoZwv4wCTmBVtBJd8ZRJ1VCsowcDmOWy5TPnCX+qle53/P89DKGaXHzJv1S1apJuWgQ9ms2sDOANDODPDSI5PMxnSuR14ym/VIA2sICIBGKgrKwQDgWR5o/TXBoiKppMTavcP/tFw/RXfdZqk4kXAVAUTQS3QNUjn2X8KFDnM6o3LkndYln2MBZBZ1UeQkQwM7lMycByPoTwMI1ZT53FG1Pkuzpo5ipK6bqgZ2LojvpeqYuEXD85NlFbt6WbOB2nT36FD6/n+0Hb758F+LpiuqaGvNpmkYqlfIKqaskSZJIpVJeZ9DTNaf5ksAiSJJESa0Iz1ShTCIqugPlsVNQrRKu2+Q7NmGTzzeJRSg7pHXfMooRItGZQp+ccgOOWxUjA1BaEq813hZBWcwQS4gCTR4aYmKpiKabLdnkA3t0KQuPVj6XJ97RiT4zQ6m7n7Jhtmz0ZpkWmqojl7IggdzTQ252GknyMaUL7MO1OOYDwZvSzYUavLMO2Fl06efyJbEIy0Wd56fzq/L4zhx9ioHh6wnHrr1i0tPmtK5iSpKk10qSNCpJ0ilJkj60xtf/syRJL0iS9IwkSf8gSdKm/wnlFVJXV9777+la1EJpge5oN6bTLYoFUEoW8YQwfpdHxPZV4yZfjoFkhO7E5ojhDVEyRph4qttmTDUWU61gTGnLBSyqtLcnVzGmRlsUV+OepepE5CK66UMrFIgFgmCazCdEWPSeFo3eykUDy4JAUSGQjCEFg+TmZmnr6uLUokZY9jHYce1s8tWrb98BLDSq2P+wrAN2ag5aIhG8JBbhB2cXsSwa+FL5zDwLk+PsuvnWy3sRnq6oLllMSZLkBz4D3ANcB/ycJEnXrXjYMeBWy7IOAo8Bn2j1C/XkydOPrzLFDOlImnLRAAvCsoZqBIknRUSMNjKKFI26ME0Qm3ybhXVC3U2zOodiRIgn2jCLxVpnalZhW2eUeAvM2iWlTDlQoDOaRlnIELUz4eShIU7aWIRW+YtKqk7Yl0cJ9AAQqQrI1Lmw8PS0IuBYnCOK0UBJRe4SYNXc3AztPX1uJt+1tsnnqH+f2AitmGIxoB7Y6XSmoolgDYuwbe3O1JNnlggFfA0/h2eOPgXArls8vtSLSevpTL0UOGVZ1hnLsirAV4D76h9gWdYTlmUV7Q+fBAZb+zJfPMrlcnz2s591P56enuZNb3rTVXxFnjxd+3Jy+Zy1/hBZCkaQeKobEDEy4f37kXziV9p8XuN8rrRpvhTUEAKh8hQF3U/E3rqq70y1zBReMCnKBTp9SbSCSrhcwReP408mGZ0TRVs02HzRZpoW5YJOxMqi+MV7Fy4I7MPzUhu9beFNxe6sJef98xWKyH2icMvNzZLs6WVs7trc5HOUHtoOBChbuTWBnQDhuFzDImzfsebzHD6zyC3bOwgFan6pM0d/SLK3j46+C4cie9p6Wk8xNQBM1n08ZX/uQnon8HfNvKgXs1YWU/39/Tz22GNX8RV58nRtq6gXKeiFhlw+StOARKJbrKqXR0YJrRjxAZsmn0MNbmmpk5iWRKQq/DPBoSFKlSrnFgotMZ8DqGUfmqyStL1EQUVFHhpCkiRGZ5WW8KUAKvboLWzOk0cUmsFsDl8sxnOq1NLtOqez51OLyINDaAUVTckTSXUzs6y1rAN2OeTz+/FLUUyrUodFED2CkqoTigbwB3wXxSJkCxVOzOQbRny6pjHx/DPsuvk2z1LxIlNLt/kkSXo7cCvwygt8/V3AuwC2XaAt6ui3/+Z5XpjOt/LlcV1/G//t9ddf9DHnzp3jnnvu4a677uL73/8+AwMDPP7440xPT/PAAw+QyWSIRqP80R/9EcPDw5w+fZr777+fQqHAfffdx6c+9SlUVUVVVe677z6y2Sy6rvPRj36U++67jw996EOcPn2aQ4cO8epXv5oHHniAe++9l+eee4477riDL3zhC1x/vXiNP/mTP8mDDz7IgQMH+OVf/mWee+45dF3nIx/5CPfdd99Fr8OTpxeLHCxCV7QLLS9u0FVFQBTjfTvQz5/HVNVV5nPZL3F9/+aLHaezoiszwC5CBZsxNTjIs3MiRuZAkzEyAJgmRT1MKZwhVoyyBAQziwR37aVsiOy/117f2/w51EZvkeosC1IMJJXAzCy+bds4lSnw1ts6L/EMGz9LrhSQt+1i2Q44LoSSgN6yAvGyK1/DIkBjlMzFsAg/OCu2GOvN5xPP/4iqrnsRMi9CraczdR4Yqvt40P5cgyRJ+mngvwJvsCyrvNYTWZb1h5Zl3WpZ1q1dXV2beb1XRGNjYzzwwAM8//zzJJNJvva1r/Gud72Lhx9+mKeffpoHH3yQ9773vQC8//3v5/3vfz/PPvssg4O16WY4HObrX/86R48e5YknnuBXf/VXsSyLj3/84+zevZvjx4/zyU9+suHct771rXz1q18FYGZmhpmZGW699VZ+93d/l5/6qZ/ihz/8IU888QQf/OAHKRQKV+4N8eTpKqoB2GnfoA1FdAvi/bvQTtjm87qA4+OTWa7ra2tYR9+oSqpOKOKnUBQm5GB2mUBPD75QqGUxMgAUFyhXE2iySqhk5/FNCwTDmUyBqmm1JOAY6ujnlSkUPUgs2UF1YpJq3wDFSrWlozfnLFkvIO/YS85mTGUkcca1iEVYUw5jqj7kOO4wpsYJXtAvtUhY9nHTYJ1f6umnkMMRBq+7+D/qPW09racz9RSwV5KknYgi6m3Az9c/QJKkm4E/AF5rWdZ8K17YpTpIl1M7d+7k0KFDANxyyy2cO3eO73//+7z5zW92H1Mui3rx8OHD/PVf/zUAP//zP8+v/dqvAYLZ9Ju/+Zv88z//Mz6fj/PnzzM3N3fRc9/ylrdw991389u//dt89atfdb1U3/nOd/jGN77Bgw8+CAiExMTEBAcOHLjY03ny9KKQEyXTFeliuSBu0GXFpp+nuij8zd+Bz0dor4jlMKomz0wt85Zbh9Z+wnVKU3XCUQlVEcDOwNy8SyRvVYwMgJmdwDDjaIECPqUDJIlQUWvc5GtRF8cZk0ZYIq8NkuhMU/nuMdTb7gKjdSZ356yApOM3deTBQXLPPA3AeCVMMOBjqLP59+6KyBnzJfoAkdeY7I66WIT219+75rc9eWaRW7d3EgyInoVlWZw59hQ7Dt6MP+BBkV9sumQxZVmWIUnS+4D/DfiBP7Es63lJkn4HOGJZ1jeATwJx4C/tOfCEZVlvuIyv+7IqFAq5/+33+5mbmyOZTHL8+PF1P8cjjzxCJpPh6aefRpZlduzYcUl+08DAAKlUimeeeYZHH32Uz3/+84D4S/i1r32N/XV0Z0+eflzkhBx3RbuYVecJhPwU8zn8PotIoo3F0VGCO3bgi4g1+5NzKsVKtSnzOdgdiHCV/JIN7Jw4j3znnQC80KIYGYDK/HkggSYXsPJForE4PsQm3+icguyX2JmONX0O1IqpsC+PUjBIDcTBMJiJpiEPe1qUyeecFbREB13u6yP397NE25M8m9XZ3RXfdMTPFVd+CmLdEBD3hZJSoW93u4tFWGuTb1EtMzKr8MHX1Oj4mfGzqEuLHvX8Rap1caYsy/qWZVn7LMvabVnW79qf+3/sQgrLsn7asqwey7IO2f/bsoXUWmpra2Pnzp385V/+JSCKmx/96EcA3HHHHXzta18D4Ctf+Yr7PcvLy3R3dyPLMk888QTj4+MAJBIJFEW54Flvfetb+cQnPsHy8jIHDx4E4DWveQ0PP/ywC5A7duxY6y/Sk6drVAulBYK+IG3BNjt4WEZVSsQjfiRJonziBOH6cGPXfN5cMaUVdMJyGdUIEW9vpzo/T3CbCOsdmcm3zHxeyohiUQsUKGfzDYypk7MKu9Jxt7vR9Fn2mDQk5VHyBaJ+8e/pMTlJOh6iowXZf/VnBY08/qgfXzRKbm6aZE8fY3Mq+67BGJkLKj/t+qUcEGkkEUS3f6evxZhy/FL15vMzT/8QgJ0eX+pFqWuKgH4t65FHHuELX/gCN910E9dffz2PP/44AJ/61Kd46KGHOHjwIKdOnaK9XXBv7r//fo4cOcKNN97Il770JYbtX/apVIo777yTG264gQ9+8IOrznnTm97EV77yFd7ylre4n/vwhz+MruscPHiQ66+/ng9/+MNX4Io9ebo2NF+apyvahSRJYvQWl1GLBvFYiOryMvr0NKG6YurYRJbOWJBtTY6RSopOJFBA0YPE4+LvtTw4xPSyRl4zWgLrBCgt2nErMQllaZGIJIHfj9zXx8is0jK/FAgfUyBgYZgmhm4Q0UWszI+sRMs9TJqqI5cVZHsUmpubJd7Vy/lcaev4pcBmTAk/rFbUsSxBP6+4xdSOVd/y5JlFokE/Bwfb3c+dOfYUvbv3EktufsPU07Wrayqb71rQjh07eO6559yPHQ8UwLe//e1Vjx8YGODJJ59EkiS+8pWvMDo6CkA6nebw4cNrnvHlL3+54eP683p6ejAMo+HrkUiEP/iDP9j4xXjy9CLQQnGBrohYWCmpOpGoj7mKn77+NjT771uj+TzHoaFkU6vnliU6EOGuZRQjTK/TLdo2xIhNPr+uFZt8gJYVzxeOBVAWMnQHo8h9fahVOJ8r8XMvbc771XCWqhMO6Sh2cHOoUEQKhzmm+nnj/tYWOCWlQkLLI3e1Y1QqqEuLGDGxLbjnGmZMrVJ+Gna+HIBSvkY/rzw7cUEswuHTi9y6oxPZL/oVxfwyM6dO8rI3/fyqx3p6ccgrpprU008/zfve9z4syyKZTPInf/InV/slefL0olKmlGF3cjcg2E/tbT4xeuvspDwyAuCO+ZZLOmPzKm+4qf+Cz7ce6eUqVcMkZC6i6kEipr1lNzTEieOLAOzvbVFnKi+gmYlQEKNSJlSxkIcGGZtXW3oOOFEyGoopOibBpSzSwCBKpdryblFJ1UmWCsi7u1ienwPLYlkW19JKo/tlV3nZjZJxgJ2RuOhMrYVFyChlxuZV/s1LatvdZ48dAcvy/FIvYnnFVJN6+ctf7vqnPHny1Hplihlu77sdsDfErCJVy0eiqxfthRH8qRQBG7XyzFRz4caOHKO2rzyNhdiu88Vi+Ds6ODF7rmUxMvVnJX1iw0teWCL4shtavskHtqner6JIwssTmJ6l1CUKz1Z2i4xKFaNiCsZUfz9ZG4swa8UI+i22b5FNPgk76HhVyHGQ/MTEmliEH5wVxfYdu2rMrjNHnyLW0Un3jl2X+RV7ulryPFOePHm6ZmUBiq7QHe2maphUtCo+3cYi9GxDGx1pNJ9P5JAkuKnpTT5R4FglsRYfWs67RPJWxshglClpPnRfmXZDdDhCSznkbQKLEA36WxoGrBV0wuTIm234ZRnfxCSL7SJWppXdohpjSkXetoPcrCimzmhhdnXFCPi3yq3HKaZEweluQ9qdqeCO1ebzw6cXiYcC3Dggun9Vw+Dcj46y6+Zb3bgjTy8+eX+ynjx5umZl2TezdCSNZjOmrIoopmLd26iMnWrwSx2bzLGnK950vpwLB7WBofL8wmWJkSF/HtVqR5NVYprodEV0Q2zyzSns7WkNfsFRSdUJW4soRph4exIqFcbDnXREZVIt3ORzio6grrrAzmAkyolc9ZqOkVkp952vDzmWQDYKmPn8mliEw2cWuW1Hh1swnh95gUqpyE5vxPeilldMefLk6ZqViQlAd6TbvUFX7XiZYMXC0nVC+0UxZVkWxyayTfOloFYM6BXBSQpMTSPbBY6IkWlRMbU8hWK1UQoUCBUtfD4fQaOKPCjO2t/CblFVN9G1KhFzDqXiJxYR7KpRfxt7uxMtzYpzo2T0AvLu61iem6Gtu5epnHZNBxxfUHXAznBMpjrpBBw3dqbm8xpnMoVGJMLRH+IPBNh+46Er93o9XXF5xZQnT56uWVWdzlQ07Y6OjOICYBE4L0ZHTmdqYqlItqg3FW7syDmrXK0SCPgJaBrBbUOcmHFiZFpUECxPUTTbbPq5TiwUQQKUVDcLaqWl+XVulIyZIV80ifpE1M4RI86eFhvCnTFpyMrjSw+Qm5vFn0xjWWwtxhQ0ADs1O5evcgHG1OEzwi9Vn8d35uhTDF53I8Fw68a1nq49ecXUZdDnP/95vvSlLwHwxS9+kenpafdrv/RLv8QLL7xwtV6aJ09bSqZdTHVFutxuUaW0TCwkURk9iRQKEdyxAxDhxtA8rBPE1qDPB0XDTywSRkIwpkZmWxcjA8DyFGWzDU1WsfIlIpIPX3s7pwqiSzTcyk2+Qg3YWShUiOgGyDJnpdhlYUwBRENFLCyW5+fQIqLI3VKbfOD6pUCM+cQmn8AiyHV5rCD4UolQgOv7hV8qO3Oe7Mx5L9j4x0DeNt9l0Hve8x73v7/4xS9yww030N8v/kL+8R//8dV6WZ48bTmZQMAXIBlKcl4VgbNaqUA8GkMbGSG0dy9SQPwaOz6ZIxr0t6SbU1J1IpEqSjFINFpjTL3w/PmWxcgAsDyJYd6AJheozCkkKjrBwUFG58Qm377e1pvCLbOAZcmElCLVnj5Mydfy0VtJrYBlEk2YKAsLmFWDJX8bAZ/E9lRronGumNprBVNJ0UkNxKmMjiP39eFbgUV48swSL93Z6UblnDl6BMBDIvwY6Notpv7uQzD7bGufs/dGuOfjF33IuXPneO1rX8stt9zC0aNHuf766/nSl77E4cOH+bVf+zUMw+C2227jc5/7HKFQiA996EN84xvfIBAIcPfdd/Pggw/ykY98hHg8zo4dOzhy5Aj3338/kUiEw4cPc8899/Dggw9y5MgRTp8+zSc/+UlAFF1Hjhzh05/+NH/+53/O7//+71OpVLj99tv57Gc/i9/vb+174cnTVpAkulKSJLnFQLFskOyJUj4+QuLVP+0+9NhEloOD7S3JfCspOuFghbweoosA+P0Eens5MXOiaYZVvYzsLJYZRvMrFLNZ+os68q5hRmcVOqIyXfHQpZ9kndLs0VvVLAJJgotLKJ29QOu7RZqqIxsFgqk4ORuLMGVE2ZmOuSDLLaO6zlRJrRBNyFQmJlaN+GaXNc4uFLj/9pop/cyxp+gcGCLZ03vFXq6nq6Mt9lN9ZTQ6Osp73/teTpw4QVtbGw899BC/8Au/wKOPPsqzzz6LYRh87nOfY3Fxka9//es8//zzPPPMM/zWb/1Ww/O86U1v4tZbb+WRRx7h+PHjRCK1mfnP/uzP8vWvf939+NFHH+Vtb3sbJ06c4NFHH+V73/sex48fx+/388gjj1yxa/fk6VpTV1QwpDRVJxj2oeoysWiMai7nxshoepUXZvIcGmpNVIem6oT9KqoRIlwxkPv6mClWUVoYIwOgLWUB8PsrWKZJcClHcFAEHO/rabUp3B6T2gkLgZlZZhNpEuEA3YnWFW0AxWyJYEVF7u5keW4WgLFisKUesCsmmzFVrZqUC4aLRZC3N27yPXnG4UsJv1S5WGTqhee8rtSPia7dztQlOkiXU0NDQ9xpp8O//e1v57//9//Ozp072bdvHwDveMc7+MxnPsP73vc+wuEw73znO7n33nu59957131GV1cXu3bt4sknn2Tv3r2MjIxw55138pnPfIann36a224TfwFLpRLd3d2tv0hPnraI6qNkQiGdvCkTsgT6IHzgAADPT+fRq1ZL/FLirApJM4OFRFBRhPl8urUxMlgWpZygnEd9IiMvolWQBwc5+bzCm24ZvNh3b1iavWGniQVJwnmFM8EO9nbHW1q0AZSWVMGY6uslNzeDLxDgpOrj7i2ERXBlF1OODywUMDDz+dXm89OLtIUD7qbn+LPHMKsGuz2/1I+Frt1i6ipq5S+WZDLJ4uLiqscFAgF++MMf8g//8A889thjfPrTn+Yf//Ef133O2972Nr761a8yPDzMG9/4RiRJwrIs3vGOd/Cxj32s6evw5OnFoHQkDQjgpOwXJvOQJr4W2rcfECM+gJtbgEUAceP0+QXPKji/iPzyGxmZFcVUy+JdSlm0iigKY5Iw2od1g1xHN4VKoaUBxyCuKSRXUGknGAoTMC2eI3FZUAWlvCYYU4OD5M7PEOnsoopv65nPwWVMORuKwZL4GVxrk+/2Xak6v9RThGIx+vcfuIIv1tPVkjfmW0MTExNuSPGXv/xlbr31Vs6dO8epU6cA+LM/+zNe+cpXoqoqy8vLvO51r+P3fu/31oyVSSQSKIqy5jlvfOMbefzxx/mLv/gL3va2twHwqle9iscee4z5+XkAlpaWGLfXcD15+nFUd1R0ZjVVx28tARDIqsjbtuGPCzPz8ckcA8kI3W3hps8zqyblogEVm2eVzdlYBKWlMTIsT1EyRWEWNcUNOFIxGA+JgrCVMTJgm+oDBfJmjHhEbCOO+JOXpcDRSqZNP99Fbm4GKyEK4i3JmLI9U04unz8vfjfXF1PncyUmlor8hD3is0yTs8eOsOOmW/B5ftcfC3nF1Brav38/n/nMZzhw4ADZbJYPfOAD/Omf/ilvfvObufHGG/H5fLznPe9BURTuvfdeDh48yF133cVDDz206rl+4Rd+gfe85z0cOnSIUqnU8LWOjg4OHDjA+Pg4L32paAVfd911fPSjH+Xu/9vemYfZUVV7+11n6tNz0kMGMg8kIUgGggQIaBAZ1AgKweBEAgICynC9fIJXLoZBBOEiKiAXBXK5oIwKiKjMeiOBjJ0RCCSEJGQgSaeH09OZ1vdHVXc6nU7SkKrdfSr7fZ56uk5Vdf32qqqzz6q91177lFMYN24cJ598Mps3bzZit8XSE9nVzZdEMo4zFdqwZbdpZJasrzng+fhaaW5wYoqyye2A01oUHTiIt7bUeZdfCqDuQ5pdZyqezhIJhYmEQryVdmIrvW6ZakqkiEs99akY+RJCQ2E+KujteUZyzSotmQixdD2hASOo2bqFhrxehEPCsIocG8kHUNzBmdr24R5pEd5Ys3u81Ja179JYW8MIGy910GC7+TohEonw8MMP77btpJNOYsmSJbtt69+/P/Pnz9/j/2fPnt22ftZZZ3HWWWe1fX7ttdd2O/a5557b4/9nzJjBjBkzPkHJLZbg0T4APT/s/GhFNmwk7/QzAPiovpkPa5o4b8pQT/TappJJNxIOhYlmsmT7D2DdPz7gy+O8G8lH7UYasiUoWfJanCSasQGH8M62Rg4pjR/wlDgdaU6kKKaaLU0wMJomWd6HTCjMoR63gLU0pgEhLvU0U0iquYmPKGRoeQGxSG69vytAxEl/0NrNF9qyZ1qEeWt30Ksg2jZn49rFCxEJMXTCJNNFtnQTufVkWyyWg47K/EpSyQzpZJZssoZYKEskkyU+xolFqfIwWSfsSiGQzKQpjEQQYF1eL2+nkQGo3UCNlNEcaSTckCKeShNzE4N63SoFTgB6LPMRzcksefUJqnv3oTAW5pDSA+8abU+rM5ofqqXGbeVbn8rPyS4+3TU7H031SUIhQde/t0e81BtrdzB5WFlb/rG1i+fTf9QY8os9fF4sPRrrTHVg6NChrFixoruLYbFYXCryK9pGUqWTdRS4tVbrNDJLNtQQDUtb1ukDZddUMlnyJUS4Vy/ernOGwHnazVe7kfpQJc2RBrK1TeQlGgkPHMjabQ2M9tiZUlWaEklCbsxZdPsOPiwoZ6QfI/naZT+v2e7orW7Ky83g8w7OVLwoSqpDWoQN1Y1s3NnUFi+VqN7BR++vYfjEo4yX1tJ9WGfKYrH0XBR6x3vvmkom2UQ8C6HSUiL9nESIS9bv5LD+JcSj3gT6tqYQaEyHiKeyRAc5c/J5Oo0MOPPy0YvmSB3pRCPxRCOJsr4kM1nPg89TzRmyGdCsMxgmXpdgdbiUkT60FjVWO5NDFxalnISdItRESjzvTvQbRcjIrkiYpkSKeEFoj7QIu+bjc4Ls1y5xs55PsikRDiasM2WxWHosYYSQhNq6jppTafKa08THjEFEyGSVZRtrPUuJAM6PpmqWhnSUvEQTsUGDeGtLPWP6l3g3jQxA7Ye0ZIpI4fwY56fSbCooA/A8uWVra1E24zg68WSa1RF/RvI1fOgE7hf2DlOzdQuR4t5kJez5/H8m6NjNFw873ZaxwbucqTfW7qCsMNZm39rFCyiuqKRi0O5dgZZgY50pi8XSYwm5P2bNroPTnAkRrW1sG8m3ems9jckMEwd7k/kcHMcjFq51EnburCE6aCBvba5rCy72hEwa6jeRTheQwcmRFU+leS9SSkjwfIRdW1B9tqVNa1NRhS8OTsMWJ4atqF8BNVs3ky4sIyTk5ki+djTWp4hlnRHZsaGOo6SqvLFmB8cMd+Kl0skk65dXMXzipz3vPrX0bKwzZbFYeiytzlRTIgXagCLEm1Nt08hUbXB+uCd42DLVnEgRFae1KN6SoqGiH/XNaW+Dz+s3o9ksmowjuqu1aFmmkKEVhZ51WbbS3BYHliI/loeIsKWgzJeg8MYdCcKZFvIH9aNmy2ZqoyUMKffeJtM01yeJttTtlhZhfXUjm2qb2+KlNq5aTqofE9D8AAAgAElEQVSlmeGTbEqEgw3rTHUzTz/9NKtWrdrvcffeey8PPfTQPo+pqqri+eef96poFku3075lCnXiffJS6V3B5+t30rsgypBy72KZmuqThLK7ckxtyHNavbwOPk9qAaIhojitRkVFRSzfmfY8Xgp2OVPN2QgFEqKxtJxwPI8BvfP3858fn6baRqKpBNmK/jTV1bJFizxvaTNNOpUh2Zwhkti+W1qEjvPxrVm8gEgsj0GHj+u2slq6B+tMdSPpdLrLztTFF1/Mueeeu89jrDNlCQqtXSRht4pqTqQIh5zpXPJVyRs+HHCTdQ7q5WmXSnNDilBmlzP1dshpkfJsGhnYLft5XjZNDCE2YBAfVDf6Mhlwa8xUQyZCfkuKbSWVjKgsapv6xEuaEyli6Xpqs44DtS6Zz6icHMm3i9YcU+HqLcTajeSbt2YHFUV5jOxThKry/pIFDP7UOKIxbyeOtvR8emzSzlvn38rb1W97es4xZWO4+uir93nMunXrmDZtWlt6hNtvv51EIsFrr73G5MmTefXVV6mpqeH+++/nhBNOIJPJcPXVV/O3v/2NUCjEhRdeyGWXXcaiRYv4wQ9+QCKRoKKigjlz5tC/f3+mTp3KhAkTmDt3Ll/96ld59tln+cc//sFNN93EU089xSuvvMJ9991HMplk5MiR/O///i8FBQXMnj2boqIirrrqKqZOnbpHWSZPnsx1111HU1MTc+fO5Uc/+hHXXnstr7/+OpWVlWSzWUaNGsW8efOorKz09LpaLF4TCTlVU75bRTUlUkTE6dIrGTgIicWoa07x3rYEXx7vYSJNnB/OULqGMBALhVnWFPN2GhmAuo1t2c+j6TT56QzNffqhirexWS7NiSRCmkQyRHldgg/yhvgWEN6SFPK0nppmJ+loddif+f9M0tqyF/poPVE3OayqMm+tEy8lIuzYuJ7aj7by6dOnd2NJLd1Fj3WmeiLpdJr58+fz/PPPc/311/PSSy9x3333sW7dOqqqqohEIlRXV5NKpbjssst45plnqKys5LHHHuPHP/4xDzzwAADJZJKFC53hs++++y7Tpk1j+nTnC9irVy8uvPBCAK699lruv/9+Lrvssi6V5YYbbmDhwoXcddddALz99ts88sgjXHnllbz00kuMHz/eOlKWnEJau/kakki2hlBWKXEnjl22oRZV75J1gvMD2ZxIEs8mKABiAweyamvC2y4+gNqNNEf6AxBtSRFvbGZ7idNV5EfCzqZEijzZTlMW4vUJ3uvb27dUBS0ao0TqqEk496U2Wprz3XyN7lQy0ZqtxIZ8FoB1OxrZWtfS1sW3dvECAIbZ/FIHJT3WmdpfC1J3cOaZZwIwadIk1q1bB8BLL73ExRdfTCTiXMqysjJWrFjBihUrOPnkkwHIZDL079+/7Tz7mipmxYoVXHvttdTU1JBIJDj11FO7XJaOnH/++ZxxxhlceeWVPPDAA5x33nkfy16LpafQnEihqRryUmnyx7qZzzc4o+DGDfTOmUq1ZMiklVQmSTydITxgEO/vaOD0Cd62flG7kYY8p7so1NhMPJliQ7yMWCbEkDIPc1m5NCdSRLNbACfQfVNhBaf44OBoKkUqXEA8VE9NbR7EC0mHY4yozG1nqnVevmgq0ZYWYd6a1vxSu5ypyiHDKKmwL6wHIz3WmeouIpEI2Wy27XNzc3Pbel6e0w8eDodJp9N7PYeqcvjhhzNv3rxO9xcW7n2I8KxZs3j66acZP348c+bM2WMuv49TlkGDBtG3b19eeeUV5s+fzyOPPLJXXYulJ9OUSJFJJYin0uSNbg0+r2FknyJK872bw64tOWgmTUljC4mRfd2uN4+nBandyM7IEai2QCpDPJlmiZRwaJ8iImHvQ1mb65oIa2s+qxSbCst9aZlq2riZTDiPgkgtm3eEac7vzeCyAvJjuT2SrzVmKpasb0uL8MbaHfQpzmN4RSHNiQQfvrOKo884uzuLaelGbAB6B/r27ctHH33Ejh07aGlp6XQi4vacfPLJ/Pd//3ebQ1NdXc3o0aPZtm1bmzOVSqVYuXJlp/9fXFxMfX192+f6+nr69+9PKpX62M5Px3MBXHDBBXzrW9/i7LPPJhzO7QrNcnDidL2lSKWbiKfSxMeMRlVZsqHG02Sd4PxoqmZpySh5TS1sKXRaHcZ6mRYBoHYDtfQmk3XiwPJTaZakCnwZyQfQVNeMqJvPKplmR2klg3wYyZdYuwmAwoIkNR9tdeOlcrtVCtwRnpIlrEmiAwe2i5cqR0RYt3QRms0y/EjbxXewYp2pDkSjUa677jqOPvpoTj75ZMa4+Wz2xgUXXMDgwYMZN24c48eP5/e//z2xWIwnn3ySq6++mvHjxzNhwgRef/31Tv//nHPO4bbbbmPixImsWbOGG2+8kcmTJzNlypT9anfkxBNPZNWqVUyYMIHHHnsMgNNPP51EImG7+Cw5S6o5QyadJZVNURBx5srbUN1EdUOSCR7GS4Gb3NLNZ5WfTLMmWkphLMxArx2P5lrqs0W0hLYBEM8qb6XzfImXAmhqyJDNJggDyfwSBvQr86UFLLHxIwDiJUr99u1syhT6MmWNaZrqk+TRQsxNi7BmWwPb6lt26+LLLy6h38hR3VxSS3dhu/k64fLLL+fyyy/f6/6Kioq2OKVIJMIdd9zBHXfcsdsxEyZM4J///Oce/9ux227KlCm7pUa45JJLuOSSS/b4v9mzZ3d6jvZlKSsrY8GCBbv939KlSxk/fvzHdswslp6Ck7CzGRUoLnV+mJe48VITB3mX+Rzc2KxsAnDSIizPFHk/jYxLYzqfFGudUYO9ylEJ+dIylc1kaWlWopkm8hG2+NTFB9CwuRoogpIIqll2hkuC0TKVSDnxUm5ahNb8UscOLyebzfB+1SKGH/lpQiHb+n+wYlumAswtt9zCWWedxc9+9rPuLorF8olp7+CUHOIEgi9ZX0N+NOx5/qKmRGrXZMCpNPOa476kKgBoScbIai0opHv1AWC0D1otjWlASGVaiLekWJfX2zcHp3G7kwss5Q7IqYmW+pI3yzRNdUmijdVE3QmO563dQb+SOEPKC9i8+h2aE/UMP9JmPT+Ysc5UgLnmmmv44IMPOP7447u7KBbLJ6YpkUTVcaZ6tQafb6hh3MBSz7uqmhNJUDc5aFEJO9Ihb6eRaUemOYpm64lnMmwtrKA4L0L/0rjnOq3B0y2ZLHkNjWwq9GdOPoDGnc7cdc3q3JfaaAkj+uT2nHwAjbUtRBtriA0egqry5todHDvCiZdau3g+oXCYIeMmdncxLd2IdaYsFkuPprlhV2tR2VHH0ZzKsGpTraeTG7fSlEgRzu4grEqmoi/g8TQyrUgYbYkQzTQTb0nxfrQXo/oV+zI5bnNDEtUMzRnId9MiHOpTRnLHGc2SSAvZcIzy8jIKYrkfTdKcSBJN1RMbMoT3PkqwPZFsm49v7ZKFDBhzOPHC3O/OtHxyrDNlsVh6NG3dfKqUfmoCqzbXkcqop5Mbt2nVJxGtIZ7OsLOktevN+5apTNEAQskokVQL+ckUy7XIly4+cGPOss5kyvmpNFuLKxhS7kdrkdLSAnnZBLWJDA15pb7FZpkk1ZIhnXbTIgwZzLx28/HVbfuI7evXMdwm6jzosc6UxWLp0TjOQD15miUSi7FkvZNOwMvM521adU1kMwnizUnWx3szpNzjaWRcmgsPRVUJpVuIpzKsiZb6lxahPoVmna7LeDJN3uDBRH0YyUdzkmS4gDzq2FnbyHYpDka8VFvCzgaigwYxb80OBvTKZ1BZflvW8+GTju7OIlp6ANaZslgsPZrm+iRkaikMKwBL1u9kQK98+pZ4H1/UXN9MJuvks3pLSnwLPm+ODwVtAlXyU2m2FJT55ng0twuqz4ZiDBzcxxcdGptIRYuISy21O+uojpTk/DQysCvmrKA4CpEob75f3ZZfau2SBfTq25/e/Qd0cykt3Y11pjpQVOR8+Tdt2tQ2X15VVRXPP/982zHPPvsst9xyyz7P0/7/LRbLJ6dxWx1k6ikqcDKdV22o8aWLD6CxPkU6myQ/lWa5FvkWfN4UHdjm4EQiMVoieb518zXXNSJaDcD2SC/f8j5JYxPJaBFR2U42naY2UhKIbr7WlqnCPiWs/qie6oYkxwwvI9XczPoVSxl+5Kd9iXWz5BbWmdoLhxxyCE8++SSwpzN1+umnc80113T5/y0WyyencUc9WRopKS9hW30LG3c2+dLFpwotzY0AxJMZNhWUez+NjMvOUHmbM5XM70VlcR5lhTFftJpq6glpNdFMlq0F5f7lfXJbpkIhx3GrC8AEx7BrkuOCQ8p3m49v/cqlZFIphh9pu/gsPThp55abb6blrbc9PWfeYWPo9x//0aVj161bx7Rp01i8eDHXXXcdTU1NzJ07lx/96Ec0NTWxcOFC7rrrLtasWcM3v/lNGhoaOOOMM7jzzjtJJBJt/79ixQqam5u55JJLWLhwYVuSzxNPPJE5c+a0nQdg2rRpXHXVVZxwwgl85zvfYeHChYgI559/Pv/2b//m6bWwWHKFxtoESorSAf2p2uDES/nRMpVKSZuDE0OoySvyfhoZlx3ZQjTrZD/fEe3tW7wUOPPyka0jnkzxQWE5J/k0ko/GFlK9C9GwM59pvKyPL/Fmpmnc5sSbFQ/pzxtrdzCoLJ+BvQt4cfECovF8Bo49vJtLaOkJ2Jap/RCLxbjhhhuYMWMGVVVVzJgxY7f9V1xxBVdccQXLly9n4MCBnZ7j7rvvRkRYvnw5f/jDH5g5c+ZuEyh3pKqqig8//JAVK1awfPlyOxWM5aCm2W0tKu43gCXrdxIJCZ8aUOq5TjodaksOmikopTAv4v00Mi61qTw0W08oq6yN+JvYsimRIptJkJ9Ms6WogmEV/uR9yiQVJESaFFkJcciA/r7omCaxuZpwpoX8YYOceKlh5agqa5csZOi4iYQj3k20bcldeuxrQ1dbkLqbefPm8fTTTwPwjW98g6uuumqPY+bOnctll10GwJgxYxgyZAirV6/e6zmHDx/O2rVrueyyy/jSl77EKaec4k/hLZYcoCXbAkBR/yEseaeGsYeUEI96P21HKklby9TOWJlv08gA1DdHyGoN+akUKwrKOdWneCmA5sYs6Wwz8VQaHTCQvIg/U56kQ3kAtGRS1EWKOdSnLlLTNG5PEE3Ws6GgPzWNH3LsiHK2ffA+iR3bGX72N7u7eJYegm2Z6kYikQjZbLbtc2trVe/evVm6dClTp07l3nvv5YILLuiuIlos3UpWQ6BOy1RBv+Es2+hf8Hkq5bRMhbLK+9FSf5J1ujQ1ZtFsHfnJDFsKy32b4BigsSFNVjPkJ9MUDR/mm04m4rTiNSZTbvB57sdLATTVNhNLJXizuQBw8ku1pkQYZvNLWVysM9UFiouLqa+v73TfMcccw1NPPQXAo48+2ukxJ5xwAo888ggAq1evZv369YwePZqhQ4dSVVVFNptlw4YNzJ8/H4Dt27eTzWY566yzuOmmm1i8eLEPVlksPZ9sOtQ2lczWWF8akhlfgs9hV8xUPJVhfV6ZP8Hn7qivlsYMZBPEUyk2FfoXFK6ESKUdZ1SywqCh/XzRAUiH46gqDc0t1EaCkbAToKkxQ14oxevr6xhSXsAhvfJZu3g+/UYcSmEv77PwW3IT60x1gRNPPJFVq1YxYcIEHnvssd323Xnnndxxxx2MGzeO9957j9LSPWM5Lr30UrLZLEcccQQzZsxgzpw55OXlMWXKFIYNG8bYsWO5/PLLOfLIIwH48MMPmTp1KhMmTOBb3/qWnajYctCSzYbRbD2REKzY5uT7mTjInx+wdNJxpvKTKTYXlvuSFkFxnKl0QxayTeRllOL+fSn0KVA7K5G2rsv6SBGH9vWv6y0dyQdtJp3OUBsNRo4pgJZ0mHi+MP/9HRw7vJzGulo2v7fajuKz7EaPjZnqLhIJ5y146NChrFixAoCysjIWLFiw23GzZs0CYMCAAbzxxhuICI8++ijvvPPOHv8fj8d58MEH99ASkbYWq47Y1iiLBbISRbMJivJCLFlfQ++CKEPKC3zRSiUzkK0nnkqxpbDct7xPAJpwuvRToTijfIwtcq6fMxqtOtaLz/jo4KQj+WjWGW0pJRWUxIMRmN1CHPIj1DWnOXZEOe8vWQiqDD/y091dNEsPwjpTB8iiRYv4/ve/j6rSq1cvHnjgge4uksUSGDLhGJqtpqgwypINO5kwqJdvCRLTLYpqA/FUmryBA3wd1h9qbiYL1ISKGN3PPwenrWVKYWt+JSMqfXSmwnEk+xEA5Ycc4puOaTQUIRF1noVjhpcz//4FFPYuo8/Q4d1cMktPwjpTB8gJJ5zA0qVLu7sYFksgyYZiaCpBUXE+736U4EtH+Pcj3ZJsBhQkwqED/I2FkYwTx1Qd68VYP1umcJypvHSWlj4DyI/5M5IPIBPNJ5zdBMDgwYN80zGLEwnzoUQYXlFIRUGEdUsXM+qY45GQjZKx7MI+DRaLpceSDYVBG5CCQlT9mdy4lVSqCYD6UKFv08gAIKG2OKbN+RW+JuzMEnG7LtPEhgz2TQdaW6aqaQgX+O6MmkLFcT6XJcNMHl7Oh2+vItnUyPBJtovPsjvWmbJYLD0W1UZAqY86zs14n9IiAGSzDQBsj/X215nCyWcVzsLWor6+JdEEQBytgpYkvQ71Ly0COI5HJltPTbTUvylrDJMVp/NmTSSfY0eUs3bJAsKRCEOOmNDNJbP0NKwzZbFYeiytGck3ZEoYUVlIab6PQc1ua9GW/EoO8znhpNP1pkQHDSQW8bca1mw9sXSGwSP9bZkCSGUaqYmUcqhPkykbx43Pq4uEOWZ4GWsXL2Dg2COIxf3JjG/JXawzZbFYeiytztTShmImDva360izCUSFnSV9fZtGZpeW0/VWPtLn1qJsE5AhGYr5n/dJU2SyKTJFZZQWBGMkXyv9KwuINlSzc9NGmxLB0inWmTLIzTff/In+74ILLmDVqlX7PObee+/loYce+kTnt1h6Kq3O1Kp0ha/xUo5WPbFMiLyhQ32bRqa9ViitjBxY5rOOkxahLlzCCJ+73jRbC0BxZV9fdUyj2SSTD63gfTfruU2JYOkM60wZZG/OlKruNq1MR373u98xduzYfZ774osv5txzzz2g8lksPYaIO9BYGxCE6lAv36aRaaW1661ylM+tRZoGbSJFyNcJjmHXXIONxZX+pXroVeJqOTmm+gzofML3XCWlKY4dXsGaxQsoGzCIXn39yyJvyV16bGqE/3t8Nds3JDw9Z8WgIk742qj9HvfQQw9x++23IyKMGzeOG2+8kfPPP5/t27dTWVnJgw8+yODBg5k1axYlJSUsXLiQLVu28POf/5zp06ezefNmZsyYQV1dHel0mt/85jf85S9/oampiQkTJnD44Yfz05/+lFNPPZXJkyezaNEinn/+eW655RYWLFhAU1MT06dP5/rrrwdg6tSp3H777Rx11FEUFRVxxRVX8Nxzz5Gfn88zzzxD3759mT17NkVFRVx11VVMnTqVyZMn8+qrr1JTU8P999/PCSecQGNjI7NmzWLFihWMHj2aTZs2cffdd3PUUXZ+KUvPon0uqWgkQn4s4uuoN4cUsbQyfLjPOZLcuQYTkWJ/pqzZTcsJqtd+/sVLSSzmaGQcZ2roMP9js0zShHJk/3weXbWCI794encXx9JDsS1THVi5ciU33XQTr7zyCkuXLuWXv/wll112GTNnzmTZsmV885vf5PLLL287fvPmzcydO5fnnnuOa665BoDf//73nHrqqVRVVbF06VImTJjALbfcQn5+PlVVVW1Zz999910uvfRSVq5cyZAhQ/jpT3/KwoULWbZsGf/4xz9YtmzZHuVraGjgmGOOYenSpXzmM5/ht7/9bad2pNNp5s+fz5133tnmlN1zzz307t2bVatWceONN7Jo0SKvL5/F4jmRcIQjBpYSCftfXYVVfR/J10ptvNz32CxQRIXC4SN91nFaprKhCGOGBKubL5MfJbF2FdlM2nbxWfZKj22Z6koLkh+88sornH322VRUVADOVDLz5s3jj3/8IwDf/va3+eEPf9h2/Fe+8hVCoRBjx45l69atAHz605/m/PPPJ5VK8ZWvfIUJEzofRjtkyBCOOeaYts+PP/449913H+l0ms2bN7Nq1SrGjRu32//EYjGmTZsGwKRJk3jxxRc7PfeZZ57Zdsy6desAmDt3LldccQUAn/rUp/Y4t8XSE8mGI77HS7WSDoUY4+M0Mu3RygG+x2YBRLNh+hxmwpmqJxkrDExahFYKKotZs3g+eYWFHDLqsO4ujqWHYlumDpC8vLy2dVUF4DOf+Qz//Oc/GTBgALNmzdprYHhh4a78Mu+//z633347L7/8MsuWLeNLX/oSzc3Ne/xPNBpt6wIJh8Ok0+l9lmtfx1gsuUAqHPVtcuOOJOPFvk063JGC4Yca0YmlhSFj/I0Dc8jSHC+md2HMgJY5+vYt5P0lCxk6fhLhSI9tf7B0M9aZ6sDnPvc5nnjiCXbs2AFAdXU1xx13HI8++igAjzzyCCeccMI+z/HBBx/Qt29fLrzwQi644IK2SYuj0SipVKrT/6mrq6OwsJDS0lK2bt3KX//6Vw+tcpgyZQqPP/44AKtWrWL58uWea1gsXtMYyTfWMpXtYybeJ6wR+o410/oeVRjpd2yWi/YKVhcfQJ/8Ohpra2wXn2WfWDe7A4cffjg//vGP+exnP0s4HGbixIn8+te/5rzzzuO2225rC0DfF6+99hq33XYb0WiUoqKitpapiy66iHHjxnHkkUfy05/+dLf/GT9+PBMnTmTMmDEMGjSIKVOmeG7bpZdeysyZMxk7dixjxozh8MMPp7S01HMdi8VLkgUl9C2JG9GKH/opIzqRbIThh5oZ9SYS8jfZaTvy+w01omOCrIQIAS073kYkxNDxR3Z3kSw9GOtMdcLMmTOZOXPmbtteeeWVPY6bM2fObp8TicRe/x/g1ltv5dZbb237vGLFin2er5XXXnttDw2A6dOnM336dABmz57d6fEVFRVtMVPxeJyHH36YeDzOmjVr+PznP8+QIUM61bRYegr5/X0eXecihBkw3owzJRpmtKHWolDcjCMK0C9A9Ym4Ax62fbCc/qPGUFBiXzwte8c6UwcRjY2NnHjiiaRSKVSVe+65h1gsWPENluDRf4SZoN+wxhh1+AgjWtlQmMrivP0f6AHhsgojOgAjRww1puU34ZCQSTewc9MmDj/H5vCz7BvrTB1EFBcXs3Dhwu4uhsXysRjzKTOjTqMaYVClmZF82Wjebrm0/KRgxBgjOhBi7MhgJezUzGYAhk+yU8hY9o0NQLdYLD2WkOQxbpiZoOaIRIykKgAIGewyOmSSmcDpcCSfymJzXYpmUIorKqkYFJzuS4s/dMmZEpHTROQdEXlPRK7pZH+eiDzm7n9TRIZ6XVCLxXLwEQ7nEY+GjWjFwuYa6gsHDjCiI8QYNXHfU1F5RThWYETHNMMnftpYK6Ild9mvMyUiYeBu4AvAWODrItLx2/kdYKeqjgR+AdyKxWKxHCCRqJm4IoB4oTlnYIihhLlh4pSVmrErlh+sZJ2tDDfUsmfJbbrSMnU08J6qrlXVJPAocEaHY84A/sddfxI4Sawrb7FYPiGxsFN9FBQU7ufIA6e1EiypLPddq1VxwvHHGlGKiLkBJvm9zCRWNYmEogw63M4UYdk/XXGmBgAb2n3e6G7r9BhVTQO1gKmayVOOO+64/R5z55130tjY6Gs5nn76aVatWtX2+brrruOll17yVdNi6SmEwlEgSkWl/yPR+vYtA2DMcZN81yIUJiTFlJf5nxYhJFHiEf+dqd79KgEYOMpMRndThCP5lPQZTTRmrnXUkrtI6xQoez1AZDpwmqpe4H7+NjBZVb/f7pgV7jEb3c9r3GO2dzjXRcBFAIMHD570wQcf7Kb11ltvcdhhPX/uo6FDh7Jw4cK2+fu6QiaTIRzueuzHrFmzmDZtWlseKZPkyn2wBJvNSxdTPuJQYkX+jrDLpNOsf+FvDPviNF91AD54by3btm7lqCn+t0y9/fKrVI4YQflQ/7O6vz/vdYYdu/8X0VyievM2Ynl5FBlwfC25gYgsUtWjOt3XBWfqWGC2qp7qfv4RgKr+rN0xf3ePmSciEWALUKn7OPlRRx2lHYfpt/8Rf3XOfXz0wdoumNd1+gwZzomzLtrnMUVFRSQSCV577TVmz55NRUUFK1asYNKkSTz88MP8+te/5qqrrmL06NFUVFTw6quv8sILL/CTn/yElpYWRowYwYMPPkhRURFDhw5lxowZvPjii/zwhz+krKys0+OuueYann32WSKRCKeccgpnnnkm06ZNo7S0lNLSUp566iluvPHGNudq6NChzJw5kz//+c+kUimeeOIJxowZw7Zt2/jGN77Bpk2bOPbYY3nxxRdZtGjRx3L6wDpTFovFYrF0ZF/OVFe6+RYAh4rIMBGJAecAz3Y45lmgNeX3dOCVfTlSucKSJUu48847WbVqFWvXruVf//oXl19+OYcccgivvvoqr776Ktu3b+emm27ipZdeYvHixRx11FHccccdbecoLy9n8eLFfP7zn+/0uB07dvCnP/2JlStXsmzZMq699lqOO+44Tj/9dG677TaqqqoYMWLPRIIVFRUsXryYSy65hNtvvx2A66+/ns997nOsXLmS6dOns379emPXymKxWCyWg5X9jgVW1bSIfB/4OxAGHlDVlSJyA7BQVZ8F7gf+V0TeA6pxHK4DYn8tSCY4+uijGTjQSUI3YcIE1q1bx/HHH7/bMW+88QarVq1qm0svmUxy7LG7mvBnzJixz+NKS0uJx+N85zvfYdq0aUyb1rWuhjPPPBOASZMm8cc//hGAuXPn8qc//QmA0047jd69gxcQarFYLBZLT6NLiVVU9Xng+ai77iEAAB6aSURBVA7brmu33gyc7W3Rup+8vF2Bh+FwmHQ6vccxqsrJJ5/MH/7wh07PUVhYuN/j5s+fz8svv8yTTz7JXXfd1ek8gHsr297KZbFYLBaLxQw2A/onoLi4mPr6egCOOeYY/vWvf/Hee+8B0NDQwOrVq/f4n70dl0gkqK2t5Ytf/CK/+MUvWLp06R4aXWXKlCk8/vjjALzwwgvs3LnzE9tosVgsFoula1hn6hNw0UUXcdppp3HiiSdSWVnJnDlz+PrXv864ceM49thjefvtt/f4n70dV19fz7Rp0xg3bhzHH398W7zVOeecw2233cbEiRNZs2ZNl8r1k5/8hBdeeIFPfepTPPHEE/Tr14/iYjNzjVksFovFcrCy39F8frG/0XyWj09LSwvhcJhIJMK8efO45JJLqKqq+tjnsffBYrFYLJbd2ddoPnOTUVl8Z/369Xzta18jm80Si8X47W9/291FslgsFosl8FhnKkAceuihLFmypLuLYbFYLBbLQUWPi5kKQHqqnMZef4vFYrFYPh49ypmKx+Ps2LHD/qB3E6rKjh07iMfj3V0Ui8VisVhyhh7VzTdw4EA2btzItm3bursoBy3xeLwtUanFYrFYLJb906OcqWg0yrBhw7q7GBaLxWKxWCxdpkd181ksFovFYrHkGtaZslgsFovFYjkArDNlsVgsFovFcgB0WwZ0EdkGfOCzTAWw3WcNq5VbOkHVCqJNJrWCaFNQtYJok0mtINpkSmuIqlZ2tqPbnCkTiMjCvaV+t1o9SyuINpnUCqJNJrWCaFNQtYJok0mtINpkWqszbDefxWKxWCwWywFgnSmLxWKxWCyWAyDoztR9VitntIJok0mtINpkUiuINgVVK4g2mdQKok2mtfYg0DFTFovFYrFYLH4T9JYpi8VisVgsFl+xzpTFYrFYLBbLAWCdKYvFYrFYLJYDIJDOlIiUiUhZd5fDYrFYuhMRqejuMlgsBwOBcaZEZLCIPOpmVn8TmC8iH7nbhhosR5HH5xsjIn8Vkb+IyAgRmSMiNSIyX0QO81gr0m69SESOMuWUisjpPp+/r4gc6S59/dRy9UpEZJKI9Pbp/N15r3zX8fv6mcK9L6+KyMMiMkhEXhSRWhFZICITPdb6goi8LyJzRWSiiKwE3hSRjSJykpda+yiDsRFVuaolIhER+a6I/E1ElrnLX0XkYhGJeqXjahWJyA0istJ97raJyBsiMstLHQugqoFYgHnADCDcblsYOAd4w2A51nt8vn8CXwa+jjP9zjmAuNte9lBnFrADWA18AVgLvAxsAL7usU1ndljOAra0fvZYawLwBvAW8JK7vO1uO9JDnYeBCnf9VGC9q/UBcLbHNpm8V9e2Wx/rar4PrAMm5+j1qwZ+B5yEO6LZrwWY796jr7v3Z7q7/SRgnsdaVcBhwLHu83GMu/0wYLGHOmV7WcqBjR7bFDgt4A/Ab4BjgIHucoy77TGPbXrGrS8GAj8A/hM4FPgf4GYvtdppVgITgXFAkR8ark6k3XoRcBRQ5pfefsvTXcI+XNh3P8m+T6j1g70s/w5Ue6y1pN36ex32eVlBLseZ22gYUAeMcLf3BZZ5bFMKeA54AHjQXerdvw94rFVFJz/6buW11Mvr1279dWCou17hpU433KvF7db/AnzBXT8aeD1Hr987wPeBfwEfAr/EdTy8Xjp8f9fvbZ8P92pDh31VHupkcBz499strZ+THtsUOC1g9SfZ9wm1lnb4vMD9GwLe9lhrLM4L0HtAEqeH6H1gDlDqsdYsDL1QdnVp6yoIAItE5B4cj3uDu20QMBNY4rHWzcBtQLqTfV53nYbbrd/RYV/MQ52Mqm4HtotIQlXXAKjqVhHxUAaA44BbcL7YvwEQkamqep7XQkChqr7ZcaOqviEihR7qhESkRFXrgCxOywqqur19l5xHmLxX7TlEVf/qas0XkXwPz23y+jWo6l3AXSIyGKe19x4R6QU8qqr/4aFWs4icApQCKiJfUdWnReSzOD/eXlIjIt8FSoCdIvJvwOPA54GEhzprgZNUdX3HHSKyoZPjrdbuVIvI2cBTqpp1zx8CzgZ2eqgD0CAix6vqXDeUohpAVbPifWXxADBTVd8RkaOB76nqZBG5ELgfmO6h1r8Do4FiYCkwUVXXuCEcL+K0/hklSM7UucB3gOuBAe62jcCfcW6klywGnlbVRR13iMgFHmvdLSJFqppQ1Xva6YzEeQvwivUi8jOch/NtEfkv4I84FfFmD3VQ1QUicjJwmYi8ClwN+JU99q8i8hfgIXZ3ss8F/uahzvXAqyJyN06LxxMi8ixwosc6YPBeAcNdOwQYKCIFqtro7vMyvsPk9Wv7EXF/OH8O/FxExuCECnjJxe75szjdl5eIyBycFrGLPNaaCVzrap2C07X4d5yu0gs91LkT6I3r8Hbg5x7qBFXrHOBWHAe+1XnqBbzq7vOSi4HficgoYAVwPoCIVAJ3e6yVr6rvQNvL1r3u+m9F5Acea3XXC+VesRnQPwEiMhqnO29bJ/v6qurWbijWASEiJcD3cJyau3Aq/vNwKuKbVNXrH+lW3QHAL4CjVHW4TxpfAM5gl5P9IfCsqj7vsc5InB+tUTgvKhtxnO6/e6xj7F65LSjtWaSqCfcNcLqqelYhG7x+d6iq15W7xfKxEZFyAFXd0d1lOVBE5I84vUCv4MS/9lbV892g+hWqOtpDrWeBlTgvlGNd3dYXyuNU9VSvtLpcpiA5UyJyKk6g3Uuq+kG77eer6gPdV7IDo51dL6vqunbbc9oui+VgwG3xGgC8qaqJdttPU1VPW91M1RWuTZ29oLzllUZ3aO2jDCer6osenq8EqGxtUWm3fZyqLvNKxz2nkefP7Sb/DxznZilwi6rWi0gpcJiqvuGhVre8/O+LIKVGuBn4MXAE8IqIXNZu9/c91gq7Q1tvFJEpHfZd67HWz9hl18t+2WXYpu7SOs5nrVNF5DsiMqTD9vM91gnqvTJy/TpoDfVTS0QuxxlRdRmwQkTOaLf7Zo+12teBftYVVwOP4nSXzncXAf4gItd4pWNaaz94FioiIl/DGVH8lDgpCz7dbvccr3RcLWPPn6rWqOoPVXWaqv5YVevd7bVeOlLuOetU9WeqeosbAvOUq/u97nCkWgsViAVnhFPEXe8FPA/8wv3s9aiZ3wG/B64EFgF3tNvn2Qg7k3YZtilwWjgV0z9x4i7WAJfluk0Bvn4/M6i1HHd4ODAUWAhc4X72ul4yVVesBqKdbI/h/chpk1rP7mX5M86gBa90qoD+7vrROI7VV318Jkw9f2Hgu8CNOF1t7fdd66PWFD+1ulym7hD1xRB4q5OLfT/wBLDSY61l7dYjwH04/bV5PjygRuwybFPgtEz9kNnrl3NaKzt8LsIJqL8DD9MVuOc2VVe8DQzpZPsQ4B2PbTKptRP4EvDZDstUYKuHOss7fO6P86JyOd478yafv8C95H2cJTDdfMCa9sGyqppR1e/g5JTxNFM47VISqGpaVS/Cedt4Bedh9RJTdpm0KYhaEVVNuzo1OElVS0TkCbxNYQH2+uWS1lYRmdD6QZ2YlWk4+bOO8FjLVF1xJU434l9F5D53+RtOnp8rPNQxrfUG0Kiq/+iwvIZzDb2iXkRGtH5Qp1vqRJy4sMM91AGzz9/RqvoNVb0TmAwUicgfRSSPdiNoc1Cra3SHB+fHAuTjDM3sbN8Aj7UeBk7rZPsFQCoX7TJsU+C0cJKQfraT7TcB2Vy0KcDXz6TWQKDfXvZN8VjLZB0Ywkl8e5a7HEO72SdyVcvEAowHRnayPQp802Mtk8/fHklAgetwUp143SVrTKurS6BG83VERGar6uzuLofXBNWuXEbcBJaq2tTJvgGq+qH5UuUOJq9fd98rEblIVY3MK2eqrjBsk0mtaar6XFB0XC1frp+IPAw8rB1GCIqTe/E3qupZXjqTWl0lSN18neHr5LntEYOTbmLILpM25bqWqjZ1/HEWkdnuPt8dKXv9eqbWXrjYgEYrpupAkzaZ1LohYDrg0/VT1W91dG7c7b/z2rkxqdVVgu5Mmew7Pcqglim7TNoURC1jzjz2+uWSlsl6yZRWEG0yqRVEm3L+Je/jEHRnapJBrY8Mapmyy6RNQdQyWUHa65c7Wl82qGWqrjBpk0mt7wZMB8xevyC+5HVKoJypjon4dNckkp4n/euIqp7m17m7yy4/bTpItIw58/b69VwtERkjIieJSBGAqm50t3t+HU3UFeLwNRE5210/CfihiFwqzoS9nmFSy9U7ujWJpoiMBY4XkS/msI7R69cJQXzJ65TABKC72X+Px5mE+MvAnar6a3ffYlU90kOtMM5opoHA31T1X+32XauqN3moZcQuwzYFVcvUVB72+uWIljgZqL8HvAVMwEmY+Iy7z+t6yVRdcQ/QByeNRB1OzrFncXI0bVVVz1IWGNb6CfAFnHxqL+IMuX8VOBn4u6r+NJd0XC1j1++gpzuGEPqxYDOg55JNgdPCZkDPpetnUiuIGdCXu3+jwA4g5n6O0C7Jay5q4SQ6LcBxPErc7fleapnS6Ybr1w/4DXA3UA7MBpYBj+NmfM9Fra4uQermM5mIz2TCMFN2BTXhmimtLwOfU9UrcbqMviAiv3D35apNJrVMXj+TWiF1J5dVpwVsqqt3hw9apuqKVo0UsEBVk+7nNJD1UMe4ljqJThuBNapa52o1eaxlSgfMXr85wCpgA05LWxNOC9j/AffmsFaXCJIzZTOgHxhBzKptUstmQD8wbAb0A8dUXbGlXfxXW9yXiPQDkh7qmNZKikiBu94WQycipXjreJjSAbPXr6+q/lpVbwF6qeqtqrpBna7mITms1TW6oznMjwWbAT2XbAqcFjYDei5dP5sB3YcFKAT6+K3jlxaQt5ftFcARuabTDddvabv1mzrs87pL0ZhWV5fABKBDm2d/GjDA3fQhTkBfTfeV6sAJql1BQmwG9APC5PUzfa9ERICj2f37O199qHxN1RUm6yTDWkbuVUCfiRuAn6vbrd1u+0jgFlWdnotaXS5TUJwpETkX+AnwAs7DAs5b4cnA9ar6kMd6Y3Ampmz/gD6rqm95rGPMLlM2BVXLcKVvr18OaInIKcA9wLvs/v0dCVyqqi94qGWkrjBcJ5nUMnKvgvhMWILlTL0DTO5YGYpIb+BNVR3lodbVwNeBR4GN7uaBwDnAo+r043qlZcQuwzYFTstwpW+vX+5ovQV8QdulX3C3DwOeV1XPYpkM1hUm61qTWkbuVRCfiXbnPRX4Cru/pDyjnUz9kktaXaI7+hb9WIDVQGkn20vxfsbq1UC0k+0xn7R8t6sbbAqUFk6Qb69OtvcGVueiTQG+fia13sVNV9DJ9XvPh3tlqq4wWdea0jJyr4L4TLjnvBMnHcc5OPnOjnfXnwd+mataXV0iBIefAotF5AWc4ZIAg3HeNm/0WCsLHAJ80GF7f7wfjWHKLpM2BVFLgM6aebN4PwTeXr/c0XoAWCAij7L793cGcL/HWqbqCpN1rUktU/cqiM8EwBe1k5YuEXkMx6nzMkGoSa0uEZhuPmhrujyVPeMgdnqscxpwF84bRvsHdCTwffW4mdGEXSZtCqKWiMwErsPpOtqj0lLVOV7ouFr2+uWIlqt3GJ3HnK3yUsfVMlUHGtHpBi0j9yqgz8Qy4DuquqDD9qOB+1XVs1QgJrW6XKYgOVMdEZFpqvqcT+cOsedojAWqmvFDr4O2L3aZtCmIWoYrfXv9ckSrE+0jVXWx3zqulm91YHfodIOWkXsVhGdCRI7EyUpezK74ykFALfA9VV2Ui1pdLlPAnSlP577aj9ZFqnqfIS0jdhm2KXBahit9e/1yR8tkvWSqrgicTSa1gmSTOAlB215SVHVLELT2R5AyoHeG1/EP++Jig1qm7DJpUxC1bjCkA/b65ZKWyXrJlFYQbTKpFRibVHWLqi5yW4d8rStMau2PoDtT3zWoZfLLYMquwHzBu0kriDaZ1AqiTQDXG9QyVVeYrGtNapm6V0F8JgBOD6jWHgTKmRKRMSJytYj8SkR+BZzoBvqZ4Mt+nFREYiJyroh83v38DeBcEfmeiET90GyHLzYdRFomKy17/Xqoloh8RkRGu+tTgJEi8iUfdE4XkXj7bao63wedwa064nAe8G0RuUREfB0h7uZiGihOIlkTHO3HSUWkn9tFhYhUAiEROdwnrRIRGdH6ufWZEJFxfuh1lDeg0R1ae4oHJWbKVHLBdnrDgTNxgt4yOMMxf6/uDOAe6jwCRIACoAZnctk/Aifh3L+ZHulcDvxJVTfs9+AD15oMvKWqdeJM7XENcCTOLOA3q2qtj9rH41SQK9TDTMPuuY8GVFUXiMhYnAzbb6vq817quFpjcGIF3tR2UyqIyGlejybdRxnOU9UHPTyfyazuRTj3p/339wVV9TS1hIjcifO8RYC/43xv/wp8Fliiqv/PQ60moME9/x9wAur9GJCwAjhaVRtF5FZgBPA08DkAVT3fQ62nVfUr7voZOPmFXgOOA37m8SjPX3XcBHwbeAhAVS/3SOe7OHWeALcCs4AVOLmSfq6qnqVHEJGv4Vyzj4AoMKt1BJyJOC0RCXn9ndqHlmh3OjR+J7IytWA2keHlOMOqrwVeB+7GyeexCpjqsdYy928E2AqE3c+ChxM64oyC2AT8H3ApUOnjvVqJm7QOuA/ny348TlbqP3qsNb/d+oVAlavzL+AaD3V+ArwBLAR+BrwC/CfwT+DHPjx/7+D8gK0Dzmi3b7Ff962Tcqz38FxXu/fmGuBb7nJN6zaPy/01YD7wO2AN8L/AI8AyPJ5o1n3WBedlaCdQ4G6P4jj0XmotwUk8eiHwsltf3EsnkzofoM6qduuLgFC7z0s91lrSbv11YJi7XuGD1gacib3PBWa6y7bWdQ91lrvPQzmQwJ0I2713VR7bVAX0d9ePBt4Gvtrx2nqodyJOKpVncF76bwFG+qDzVaDMXa/EcXiXA48BA73W61KZukPUF0Och2RIJ9uHAO94rLWcXU5NAfCauz7Y6wcU540l5n7R6ts9QHGc1h2vdJbgdPuegpM4bhvwN7ciKfbYprfarS/usM/ryqR9ZbwA10nEmTV9udfPhPs81AEl7vZ8vJ8xfTlQ5K4PxXHgruhor0day/ayLAdaPNQx+TK0jF1OTQVOCw7AOOB1j7VWuH/jOM5Uvvs5TDunxCOtjt+lfjiO9zxgg4c6fwc+564/1Vrv4jgHXjs4i9utz++wz+tnvRjnxe73wCHutrVeanRi09IO+7y2aXmHz/1xHODLOz4vHmj9DHgQ50XoSeA2HMd+CXC2x1rtHfrHgH/D6YmaBbzo9T3ryhKkDOhXAi+LSKfJBX3Qi+B0D+ThdL2hqut9iGO6H8dRDAM/Bp4QkbXAMThdml6h6jTHvgC84NrxBZyu09txvH+vWNGui2ipiBylqgtFZBSQ8lAHnFiE3jiOoqjqNgBVbRCRtIc6aXW6VBpFZI263b2q2iQiXjdzh9Tt2lPVdSIyFXhSRIbgfdxAX5x8TB3zLwlOS4FXmMzqLkCTu94A9AFQ1WUiUuKx1l9E5P9wnKnfAY+LyBs43Xz/9Fhrt3uvzjDxXwG/cp8Nr7gAeEhEZuO0aFeJSBXQC/iBhzoA40WkDse2PBHpr6qbRSSGUyd6hqrWA1eKyCTgERH5C/7EFauIRFU1BbTFzrlxaF7r1YvICFVdA+Beu6k4rdpex2hNUzdZpjjZ3f+hqv9PRJ7E6fF4wkOt9vd+pKrOcNfniMiVHup0mcA4U6r6N/fH2EQiw9/hTAfwJnACTr93ayBhtZdCqvoLcVLko6qbROQh4PPAb9Xb4NKOFXEKeBZ4VkQKPNQBpzL+pYhcC2wH5onIBhwn+AKPtUpx3sQEpxJrrYyL8NbxSIpIgao2ApNaN4pIKd47A1tFZIKqVgGoakJEpuFMU+F15t/ncFrBqjruEJHXPNQx+TL0PPA3EfknTtzUEwAiUobHzqiqXi0ixzqr+oYbCPxVnDrkSS+1cN7O91aOjk7qJ0aduMrWwT2jgDk4caoL1OP4GFXdm8NUgE+DBlR1kYh8DifcYa4PEl9tp7Wx3fZy4N891rqEDg6aqtaLM7PB1zzWyopImapW47wYhV29nSLi9UveayJyA05r2Gsi8lVV/ZOInIjj4BsnMAHopnFHXhyG04z/djeVoUjbBR8f4LlGqepqL871MTRLgGE4Tv1GVd1qULsA6Kuq73t0vjxVbelkewVOzMJyL3Tccw7EaQnbI0GdiExR1X95pWUSMZvV/YvAWJxulhfb6Uc7u48e6PVl9+SCvj3rprSCaJNJraDZJCIzgJ/jdNmPBi5R1b+4jQy/VNVveKgVxempaR3sMBCnlfnPODGW673S6nKZguJMucM878N5YP4KXK3u1BAiMl9VfRni2kk5PHNwuqC1XlUHG9AxaVPgtIJok0ktv3QM/cBMwAkCL8VxDsGp+GuAS9XDKURMaRm2aSLOtCGdaV2iqks81OoJ18+kTZ5quXplwHDgPVWt8fLc+9AsxRnQtMOE3l7LESBnai5wE86IqguA84DTVXWNiCxR1YmGyuGpgyMie4tBEJxRYmVeae2jDEactqBqBdEmk1o+fKdMOgNVwHdV9c0O248B/ltVx+eaVhBtMqkVRJs6nPso2qUc8bPnxqTW/ghMzBTOiLPW/Dq3i8ginLiIbwOeeoz7cXCKvNQCbsYZFdFZsLRnwYombQqiVhBtMqll+Ds1h73/wDwIePkDU9hRB8CNnyr0UMekVhBtMqkVRJsQkc8C/4XzUjIJJ/1MbxFJAd9WD3MYmtTqKkFyphCRUnUTPqrqqyJyFs7QXa9bb4w4OC6Lgae1k1mwRcTLYG2TNgVRK4g2mdQyaZPJH7O/ijMq7CF2BdYPwsll5HVyVVNaQbTJpFYQbQInrcQpqrpNnEz1d6jqFBE5GWdU+ik5qtUlgtTN9w2cnCBvdNg+GPhPVb3QQ63Xgcv24uBsUNVBHmqNBqrVHdLfYV9fr+I8DNsUOK0g2mRSy7BNv8LJ2t3ZD8z7qurp6EER+QKdZ3b3IzO+Ea0g2mRSK6A2LVPVce56GGfwyJHu55Wq6lkqBpNaXS5TUJwpk5hycExi0qYgagXRJpNapr9TJn/MLJaDARF5ACek5hWcSYc/VNUfiDNyerGqejafokmtLpcpKM6UG9H/I+ArOEn4FGc+omeAW0yNLPCaoNplsRwMtPv+noGTANW3768prSDaZFIriDa5WlGcjOdjgaXAA6qaEWf+1T7qYa4zk1pdxY/srt3F4zhZmqeqapmqluPME7TT3ecZIlIqIreIyNsiUi0iO0TkLXdbLy+1MGSXSZuCqBVEm0xqdZNNbxn8/p7Y4ftbg8f1kkGtINpkUiuINqGqKVW9R1W/r6q/VTc/nKo2ee3cmNT6OIUKxMI+5t/b175PqPV3nIlZ+7Xb1s/d9kIu2mXYpsBpBdGmg/D6XZOr31+TWkG0yV4/T7SKgBtw5pOtxZnf9Q1glpc6prW6XKbuEvbh4r4A/BAnq3Xrtr5upfmSx1omH1AjdgX4C24ryBzQCqJN7vlM1kum6orA2WSvnydaz+BMNDwQZ47G/wQOBf4HuDlXtbq6BKmbbwbO3Eb/EJGdIlINvIaTFsHrOYg+EJEfipNBGXCCZEXkanaNDvIKU3aZtCmIWkG0yaRWEG0Cs/WSKa0g2mRSK4g2AQxV1TmqulFV78BJmv0uTgLtM3NYq2t0hwfn1wKMwZkEuKjD9tM81umNM7nx2zj90dXAW+62sly0y6RNQdQKok32+nmmZ6ReMqkVRJvs9TtgndeB493104G/t9vndYuvMa0ul6k7RH0xBC4H3gGeBtYBZ7Tbt9gHPVMPqDG7gvgFN3yvAmeTvX4HrGPy+2tEK4g22evnidY4YD7OC8pcYJS7vRK4PFe1ulym7hD1xRBY3loxAkOBhcAV7uclHmuZfECN2BXgL7itIHNAK4g2ueczWS+ZqisCZ5O9fv4uwHlB1Gq/BGk6mZC6M8ur6joRmQo8KSJDcOb38pILgUmqmhCRoa7OUFX9pQ9apuwyaVMQtYJok0mtINoEZuslU1pBtMmkVhBt2h/X48x7GTStNoLkTG0VkQmqWgXgVpTTgAeAIzzWMvmAmrIrqF9wW0HmhlYQbQKz9ZIprSDaZFIriDYhIsv2tgtnBGFOanWZ7mgO82PBGSLZby/7pnis9QowocO2CM5cX5lctMuwTYHTCqJN9vp5omWyXjJVVwTOJnv9PNHaCkwAhnRYhgKbclWrq0tgppMxiYgMBNKquqWTfVNU9V/dUKwDwqRNQdQKok0mtYJok8VyMCEi9wMPqurcTvb9XlW/kYtaXS6TdaYsFovFYrFYPjlBStppsVgsFovFYhzrTFksFovFYrEcANaZslgsFovFYjkArDNlsVgsFovFcgD8f7wE7CdRdN+FAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 720x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JToT2MoW5_Nn"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qAvgaQFT5_Nn"
      },
      "source": [
        "## Evaluate Alpha Factors\n",
        "Just like we did in project 4, let's evaluate the alpha factors. For this section, we'll just be looking at the cosine similarities, but it can be applied to the jaccard similarities as well.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6E_3Z7VehkXw"
      },
      "source": [
        "### Price Data\n",
        "Let's get yearly pricing to run the factor against, since 10-Ks are produced annually."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_v3pgJFv5_No",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "outputId": "064c32e9-3ae8-4baa-f1d4-9bbe4f3e1c3b"
      },
      "source": [
        "from pandas_datareader import data as pdr\n",
        "import yfinance as yf\n",
        "\n",
        "# use yfinance while making sure the returned data is in the same format \n",
        "# as pandas_datareader's get_data_yahoo()\n",
        "yf.pdr_override()\n",
        "\n",
        "def download_daily_data(ticker):\n",
        "  # download dataframe\n",
        "  df = pdr.get_data_yahoo(ticker, \n",
        "                              start=file_dates[example_ticker][-1], \n",
        "                              end=file_dates[example_ticker][1])\n",
        "  df['ticker'] = ticker\n",
        "  return df\n",
        "\n",
        "test_daily_data = download_daily_data('AMZN')\n",
        "test_daily_data"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\r[*********************100%***********************]  1 of 1 completed\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Open</th>\n",
              "      <th>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Close</th>\n",
              "      <th>Adj Close</th>\n",
              "      <th>Volume</th>\n",
              "      <th>ticker</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Date</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1997-05-15</th>\n",
              "      <td>2.43750000</td>\n",
              "      <td>2.50000000</td>\n",
              "      <td>1.92708302</td>\n",
              "      <td>1.95833302</td>\n",
              "      <td>1.95833302</td>\n",
              "      <td>72156000</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1997-05-16</th>\n",
              "      <td>1.96875000</td>\n",
              "      <td>1.97916698</td>\n",
              "      <td>1.70833302</td>\n",
              "      <td>1.72916698</td>\n",
              "      <td>1.72916698</td>\n",
              "      <td>14700000</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1997-05-19</th>\n",
              "      <td>1.76041698</td>\n",
              "      <td>1.77083302</td>\n",
              "      <td>1.62500000</td>\n",
              "      <td>1.70833302</td>\n",
              "      <td>1.70833302</td>\n",
              "      <td>6106800</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1997-05-20</th>\n",
              "      <td>1.72916698</td>\n",
              "      <td>1.75000000</td>\n",
              "      <td>1.63541698</td>\n",
              "      <td>1.63541698</td>\n",
              "      <td>1.63541698</td>\n",
              "      <td>5467200</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1997-05-21</th>\n",
              "      <td>1.63541698</td>\n",
              "      <td>1.64583302</td>\n",
              "      <td>1.37500000</td>\n",
              "      <td>1.42708302</td>\n",
              "      <td>1.42708302</td>\n",
              "      <td>18853200</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-12</th>\n",
              "      <td>2163.19995117</td>\n",
              "      <td>2180.25000000</td>\n",
              "      <td>2155.29003906</td>\n",
              "      <td>2160.00000000</td>\n",
              "      <td>2160.00000000</td>\n",
              "      <td>3334300</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-13</th>\n",
              "      <td>2144.98999023</td>\n",
              "      <td>2170.28002930</td>\n",
              "      <td>2142.00000000</td>\n",
              "      <td>2149.87011719</td>\n",
              "      <td>2149.87011719</td>\n",
              "      <td>3031800</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-14</th>\n",
              "      <td>2155.67993164</td>\n",
              "      <td>2159.04003906</td>\n",
              "      <td>2125.88989258</td>\n",
              "      <td>2134.87011719</td>\n",
              "      <td>2134.87011719</td>\n",
              "      <td>2606200</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-18</th>\n",
              "      <td>2125.02001953</td>\n",
              "      <td>2166.07006836</td>\n",
              "      <td>2124.11010742</td>\n",
              "      <td>2155.66992188</td>\n",
              "      <td>2155.66992188</td>\n",
              "      <td>2945600</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-19</th>\n",
              "      <td>2167.80004883</td>\n",
              "      <td>2185.10009766</td>\n",
              "      <td>2161.12011719</td>\n",
              "      <td>2170.21997070</td>\n",
              "      <td>2170.21997070</td>\n",
              "      <td>2561200</td>\n",
              "      <td>AMZN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5728 rows × 7 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                    Open          High  ...    Volume  ticker\n",
              "Date                                    ...                  \n",
              "1997-05-15    2.43750000    2.50000000  ...  72156000    AMZN\n",
              "1997-05-16    1.96875000    1.97916698  ...  14700000    AMZN\n",
              "1997-05-19    1.76041698    1.77083302  ...   6106800    AMZN\n",
              "1997-05-20    1.72916698    1.75000000  ...   5467200    AMZN\n",
              "1997-05-21    1.63541698    1.64583302  ...  18853200    AMZN\n",
              "...                  ...           ...  ...       ...     ...\n",
              "2020-02-12 2163.19995117 2180.25000000  ...   3334300    AMZN\n",
              "2020-02-13 2144.98999023 2170.28002930  ...   3031800    AMZN\n",
              "2020-02-14 2155.67993164 2159.04003906  ...   2606200    AMZN\n",
              "2020-02-18 2125.02001953 2166.07006836  ...   2945600    AMZN\n",
              "2020-02-19 2167.80004883 2185.10009766  ...   2561200    AMZN\n",
              "\n",
              "[5728 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Hkm0QjdqkNO-"
      },
      "source": [
        "# def get_price_df():\n",
        "#   for ticker in list(file_dates.keys()):\n",
        "#     price_series = get_daily_price_series(ticker)\n",
        "#   return price_df"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OzxinEbqhydG"
      },
      "source": [
        "# def pivot_to_date_index_ticker_columns_price_values(daily_data_df, ticker):\n",
        "#   daily_price = daily_data_df.copy()[['Adj Close']]\n",
        "#   daily_price['ticker'] = ticker\n",
        "#   daily_close_by_ticker = daily_price.pivot(\n",
        "#                           # index='date', \n",
        "#                           columns='ticker', \n",
        "#                           values='Adj Close')\n",
        "#   return daily_close_by_ticker\n",
        "\n",
        "# test_adj_close_by_date_and_ticker = pivot_to_date_index_ticker_columns_price_values(test_daily_data, 'AMZN')\n",
        "# test_adj_close_by_date_and_ticker"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 875
        },
        "id": "xAP1fkC8jCl5",
        "outputId": "9c8a09f5-34ca-4273-d8f8-ab9035779d9d"
      },
      "source": [
        "\n",
        "# Make dataframe with all tickers\n",
        "dates = None\n",
        "tickers = None\n",
        "\n",
        "def get_prices_for_all_tickers(ticker_list = list(additional_cik.keys()), start='2000-01-01', end='2021-04-01'):\n",
        "  index=pd.date_range(start, end)\n",
        "  prices = pd.DataFrame(index=index)\n",
        "  for ticker in ticker_list:\n",
        "    daily_price = download_daily_data(ticker).copy()[['Adj Close']]\n",
        "    daily_price[ticker] = daily_price\n",
        "    daily_price = daily_price.drop(columns='Adj Close')\n",
        "    prices=pd.merge(prices, daily_price, how='left', left_index=True, right_index=True)\n",
        "  prices.index.name = 'date'\n",
        "  prices.columns.name = 'ticker'\n",
        "  dates = prices.index\n",
        "  tickers = prices.columns.name\n",
        "  return prices\n",
        "\n",
        "prices = get_prices_for_all_tickers(list(additional_cik.keys()))\n",
        "prices"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th>ticker</th>\n",
              "      <th>AEP</th>\n",
              "      <th>AXP</th>\n",
              "      <th>BA</th>\n",
              "      <th>BK</th>\n",
              "      <th>CAT</th>\n",
              "      <th>DE</th>\n",
              "      <th>DIS</th>\n",
              "      <th>DTE</th>\n",
              "      <th>ED</th>\n",
              "      <th>EMR</th>\n",
              "      <th>ETN</th>\n",
              "      <th>GE</th>\n",
              "      <th>IBM</th>\n",
              "      <th>IP</th>\n",
              "      <th>JNJ</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2000-01-01</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-01-02</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-01-03</th>\n",
              "      <td>12.15813446</td>\n",
              "      <td>34.16057587</td>\n",
              "      <td>25.94027710</td>\n",
              "      <td>24.55235100</td>\n",
              "      <td>13.69310665</td>\n",
              "      <td>13.79204750</td>\n",
              "      <td>23.11524773</td>\n",
              "      <td>11.81341267</td>\n",
              "      <td>12.38327026</td>\n",
              "      <td>15.65466309</td>\n",
              "      <td>7.35668182</td>\n",
              "      <td>26.58582878</td>\n",
              "      <td>72.89100647</td>\n",
              "      <td>27.12947083</td>\n",
              "      <td>26.81100273</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-01-04</th>\n",
              "      <td>12.30316639</td>\n",
              "      <td>32.87073135</td>\n",
              "      <td>25.89993858</td>\n",
              "      <td>23.84729004</td>\n",
              "      <td>13.51711082</td>\n",
              "      <td>13.77194118</td>\n",
              "      <td>24.46928596</td>\n",
              "      <td>11.83797455</td>\n",
              "      <td>12.54379368</td>\n",
              "      <td>15.46770096</td>\n",
              "      <td>7.10544586</td>\n",
              "      <td>25.52239609</td>\n",
              "      <td>70.41678619</td>\n",
              "      <td>26.45734978</td>\n",
              "      <td>25.82945251</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-01-05</th>\n",
              "      <td>12.76242256</td>\n",
              "      <td>32.03582382</td>\n",
              "      <td>27.51365089</td>\n",
              "      <td>23.30813980</td>\n",
              "      <td>13.83390903</td>\n",
              "      <td>14.09361553</td>\n",
              "      <td>25.48480606</td>\n",
              "      <td>12.20637131</td>\n",
              "      <td>13.11709404</td>\n",
              "      <td>15.75665855</td>\n",
              "      <td>7.16342115</td>\n",
              "      <td>25.47809601</td>\n",
              "      <td>72.89100647</td>\n",
              "      <td>27.37390327</td>\n",
              "      <td>26.10210609</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-28</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-29</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-30</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-03-31</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-04-01</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>7762 rows × 15 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "ticker             AEP         AXP  ...          IP         JNJ\n",
              "date                                ...                        \n",
              "2000-01-01         NaN         NaN  ...         NaN         NaN\n",
              "2000-01-02         NaN         NaN  ...         NaN         NaN\n",
              "2000-01-03 12.15813446 34.16057587  ... 27.12947083 26.81100273\n",
              "2000-01-04 12.30316639 32.87073135  ... 26.45734978 25.82945251\n",
              "2000-01-05 12.76242256 32.03582382  ... 27.37390327 26.10210609\n",
              "...                ...         ...  ...         ...         ...\n",
              "2021-03-28         NaN         NaN  ...         NaN         NaN\n",
              "2021-03-29         NaN         NaN  ...         NaN         NaN\n",
              "2021-03-30         NaN         NaN  ...         NaN         NaN\n",
              "2021-03-31         NaN         NaN  ...         NaN         NaN\n",
              "2021-04-01         NaN         NaN  ...         NaN         NaN\n",
              "\n",
              "[7762 rows x 15 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 705
        },
        "id": "WiMo6r6ySCir",
        "outputId": "5b1496fe-398b-4491-ca6f-29cf4d14ee17"
      },
      "source": [
        "def get_prices_multi_indexed_for_all_tickers(ticker_list = list(additional_cik.keys())):\n",
        "  _prices = get_prices_for_all_tickers(ticker_list = ticker_list)\n",
        "  _tuples =  zip(_prices.columns.values, \n",
        "                ('price' for i in range(len(_prices.columns))))\n",
        "  _prices.columns = pd.MultiIndex.from_tuples(_tuples)\n",
        "  # _prices.columns.name. = 'price'\n",
        "  _multi_indexed_prices = _prices.stack(level=0)\n",
        "  # .swaplevel(0, 1, axis=0)\n",
        "  _multi_indexed_prices.index = _multi_indexed_prices.index.set_names(['date', 'ticker'])\n",
        "  assert _multi_indexed_prices.index.levels[0].freq =='D'\n",
        "  return _multi_indexed_prices\n",
        "\n",
        "multi_indexed_prices = get_prices_multi_indexed_for_all_tickers()\n",
        "multi_indexed_prices"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n",
            "[*********************100%***********************]  1 of 1 completed\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>price</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th>ticker</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"5\" valign=\"top\">2000-01-03</th>\n",
              "      <th>AEP</th>\n",
              "      <td>12.15813446</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AXP</th>\n",
              "      <td>34.16057587</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>BA</th>\n",
              "      <td>25.94027710</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>BK</th>\n",
              "      <td>24.55235100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>CAT</th>\n",
              "      <td>13.69310665</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"5\" valign=\"top\">2020-02-19</th>\n",
              "      <th>ETN</th>\n",
              "      <td>101.41313934</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>GE</th>\n",
              "      <td>12.53708553</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>IBM</th>\n",
              "      <td>142.92942810</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>IP</th>\n",
              "      <td>40.07544327</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>JNJ</th>\n",
              "      <td>144.09953308</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>75960 rows × 1 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                         price\n",
              "date       ticker             \n",
              "2000-01-03 AEP     12.15813446\n",
              "           AXP     34.16057587\n",
              "           BA      25.94027710\n",
              "           BK      24.55235100\n",
              "           CAT     13.69310665\n",
              "...                        ...\n",
              "2020-02-19 ETN    101.41313934\n",
              "           GE      12.53708553\n",
              "           IBM    142.92942810\n",
              "           IP      40.07544327\n",
              "           JNJ    144.09953308\n",
              "\n",
              "[75960 rows x 1 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DnuH8acV5_No"
      },
      "source": [
        "### Dict to DataFrame\n",
        "The alphalens library uses dataframes, so we we'll need to turn our dictionary into a dataframe. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0rbw92e7_dIQ"
      },
      "source": [
        "# Resample workaround for upsampling a multiindexed df.\n",
        "# Adapted from\n",
        "#  https://github.com/pandas-dev/pandas/issues/28313#issuecomment-532231669\n",
        "\n",
        "\n",
        "def upsample(df, rule='D', method='ffill',\n",
        "             group_index: str='ticker', date_index: str='date', \n",
        "             upsampled_date_index_level: int=0):\n",
        "    \"\"\"\n",
        "    Upsample a Pandas DataFrame or Series with either a DatetimeIndex or\n",
        "    MultiIndex. This can be used to upsample financial data for a single\n",
        "    company, or upsample data for multiple companies in a single DataFrame.\n",
        "\n",
        "    :param df:\n",
        "        Pandas DataFrame or Series.\n",
        "\n",
        "    :param rule:\n",
        "        Upsampling frequency e.g. 'D' for daily.\n",
        "\n",
        "        This is passed directly to the Pandas resampler which has more options:\n",
        "        https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#dateoffset-objects\n",
        "\n",
        "    :param method:\n",
        "        String for the method of filling in empty values. Valid options:\n",
        "        'ffill' is forward-fill with last known values.\n",
        "        'linear' is linear interpolation between known values.\n",
        "\n",
        "    :param group_index:\n",
        "        If `df` has a MultiIndex then group companies using this data-column.\n",
        "        By default this is TICKER but it could also be e.g. SIMFIN_ID if\n",
        "        you are using that as an index in your DataFrame.\n",
        "\n",
        "    :param date_index:\n",
        "        If `df` has a MultiIndex then use this data-column as the dates.\n",
        "        By default this is REPORT_DATE but it could also be e.g. PUBLISH_DATE\n",
        "        if you are using that as an index in your DataFrame.\n",
        "\n",
        "    :return:\n",
        "        Upsampled DataFrame or Series.\n",
        "    \"\"\"\n",
        "\n",
        "    assert isinstance(df, (pd.DataFrame, pd.Series))\n",
        "    assert isinstance(df.index, (pd.DatetimeIndex, pd.MultiIndex))\n",
        "    assert method == 'ffill' or method == 'linear' or method==None\n",
        "    assert upsampled_date_index_level in range(0, len(df.index.levels))\n",
        "\n",
        "    # Pandas 0.25.1 does not support upsampling a DataFrame with a MultiIndex\n",
        "    # using the normal resample() function, so we must handle the two cases\n",
        "    # differently.\n",
        "\n",
        "    # If the DataFrame has a DatetimeIndex.\n",
        "    if isinstance(df.index, pd.DatetimeIndex):\n",
        "        # Normal upsampling using Pandas.\n",
        "\n",
        "        if method == 'ffill':\n",
        "            # Fill with last-known value.\n",
        "            df_upsampled = df.resample(rule).ffill()\n",
        "        elif method == 'linear':\n",
        "            # Fill with linearly interpolated values.\n",
        "            df_upsampled = df.resample(rule).interpolate(method='linear')\n",
        "        elif method == None:\n",
        "            # Fill with NA.\n",
        "            df_upsampled = df.resample(rule).asfreq()\n",
        "\n",
        "    # If the DataFrame has a MultiIndex.\n",
        "    elif isinstance(df.index, pd.MultiIndex):\n",
        "        # Pandas has very complicated semantics for resampling a DataFrame\n",
        "        # with a MultiIndex. The best way is apparently to group the DataFrame\n",
        "        # by companies (e.g. using TICKER) which creates an individual\n",
        "        # DataFrame for each company, and then apply the resampling to each\n",
        "        # of those DataFrames. It is further complicated by the need to reset\n",
        "        # and set the index. Pandas is quite poorly designed in this regard\n",
        "        # and its resampling API has already been changed several times.\n",
        "\n",
        "        # Helper-function for resampling a DataFrame for a single company.\n",
        "        def _resample(df):\n",
        "            if method == 'ffill':\n",
        "                # Fill with last-known value.\n",
        "                return df.set_index(date_index).resample(rule).ffill()\n",
        "            elif method == 'linear':\n",
        "                # Fill with linearly interpolated values.\n",
        "                return df.set_index(date_index).resample(rule).interpolate(method='linear')\n",
        "            elif method == None:\n",
        "                # Fill with NA.\n",
        "                return df.set_index(date_index).resample(rule).asfreq()\n",
        "\n",
        "        # Group the original DataFrame by companies and apply the resampling to each.\n",
        "        df_upsampled = df.reset_index(level=date_index).groupby(level=group_index).apply(_resample)\n",
        "        if upsampled_date_index_level == 0:\n",
        "          df_upsampled.index = df_upsampled.index.swaplevel('date', 'ticker')\n",
        "\n",
        "    return df_upsampled\n"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kT07r8Pb5_Np",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "outputId": "51d1daad-2d24-4db5-a1ef-313156663a3f"
      },
      "source": [
        "cosine_similarities_df_dict = {'date': [], 'ticker': [], 'sentiment': [], 'value': []}\n",
        "\n",
        "\n",
        "for ticker, ten_k_sentiments in cosine_similarities.items():\n",
        "    for sentiment_name, sentiment_values in ten_k_sentiments.items():\n",
        "        for sentiment_values, sentiment_value in enumerate(sentiment_values):\n",
        "            cosine_similarities_df_dict['ticker'].append(ticker)\n",
        "            cosine_similarities_df_dict['sentiment'].append(sentiment_name)\n",
        "            cosine_similarities_df_dict['value'].append(sentiment_value)\n",
        "            cosine_similarities_df_dict['date'].append(file_dates[ticker][1:][sentiment_values])\n",
        "\n",
        "cosine_similarities_df = pd.DataFrame(cosine_similarities_df_dict)\n",
        "\n",
        "cosine_similarities_dfs = {}\n",
        "for sentiment in cosine_similarities_df['sentiment'].unique():\n",
        "  df = cosine_similarities_df\n",
        "  df = df[df['sentiment']==sentiment].loc[:, ['date', 'ticker', 'value']]\n",
        "  df['date'] = pd.to_datetime(df['date']).dt.date\n",
        "  df = df.set_index(['date', 'ticker']).sort_index()\n",
        "  # df = upsample(df, \n",
        "  #               rule='D', \n",
        "  #               method=None, \n",
        "  #               group_index='ticker', \n",
        "  #               date_index='date',\n",
        "  #               upsampled_date_index_level=0)\n",
        "  \n",
        "\n",
        "  cosine_similarities_dfs[sentiment] = df\n",
        "  \n",
        "\n",
        "cosine_similarities_dfs['negative']"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>value</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th>ticker</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1994-03-17</th>\n",
              "      <th>BA</th>\n",
              "      <td>0.45996206</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">1994-03-31</th>\n",
              "      <th>AXP</th>\n",
              "      <td>0.37257673</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>IP</th>\n",
              "      <td>0.10975695</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1994-04-01</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.11535805</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1995-03-28</th>\n",
              "      <th>IBM</th>\n",
              "      <td>0.22303764</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">2020-02-20</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.60875266</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ED</th>\n",
              "      <td>0.92879746</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-24</th>\n",
              "      <th>GE</th>\n",
              "      <td>0.97748160</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-27</th>\n",
              "      <th>BK</th>\n",
              "      <td>0.98678383</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2021-02-17</th>\n",
              "      <th>CAT</th>\n",
              "      <td>0.00196681</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>253 rows × 1 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                       value\n",
              "date       ticker           \n",
              "1994-03-17 BA     0.45996206\n",
              "1994-03-31 AXP    0.37257673\n",
              "           IP     0.10975695\n",
              "1994-04-01 JNJ    0.11535805\n",
              "1995-03-28 IBM    0.22303764\n",
              "...                      ...\n",
              "2020-02-20 AEP    0.60875266\n",
              "           ED     0.92879746\n",
              "2020-02-24 GE     0.97748160\n",
              "2020-02-27 BK     0.98678383\n",
              "2021-02-17 CAT    0.00196681\n",
              "\n",
              "[253 rows x 1 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3Xgy4xQ05_Np"
      },
      "source": [
        "### Alphalens Format\n",
        "In order to use a lot of the alphalens functions, we need to aligned the indices and convert the time to unix timestamp. In this next cell, we'll do just that."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KFr3GrXCOXoB",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "outputId": "adebe583-42a3-4e95-ad9d-4aca5b51d92c"
      },
      "source": [
        "joined = multi_indexed_prices.join(cosine_similarities_dfs[sentiment], how='outer')\n",
        "try:\n",
        "  joined = upsample(joined, rule='D', method=None)\n",
        "except: # If duplicate index\n",
        "  joined = upsample(joined[~joined.index.duplicated(keep='first')], \n",
        "                    rule='D', method=None)\n",
        "factor_df = joined['value']\n",
        "# prices_df = joined['price']\n",
        "joined"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>price</th>\n",
              "      <th>value</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th>ticker</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1996-03-28</th>\n",
              "      <th>AEP</th>\n",
              "      <td>NaN</td>\n",
              "      <td>0.57735382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1996-03-29</th>\n",
              "      <th>AEP</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1996-03-30</th>\n",
              "      <th>AEP</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1996-03-31</th>\n",
              "      <th>AEP</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1996-04-01</th>\n",
              "      <th>AEP</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-15</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-16</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-17</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-18</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>144.29301453</td>\n",
              "      <td>0.99263108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-19</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>144.09953308</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>125087 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                         price      value\n",
              "date       ticker                        \n",
              "1996-03-28 AEP             NaN 0.57735382\n",
              "1996-03-29 AEP             NaN        NaN\n",
              "1996-03-30 AEP             NaN        NaN\n",
              "1996-03-31 AEP             NaN        NaN\n",
              "1996-04-01 AEP             NaN        NaN\n",
              "...                        ...        ...\n",
              "2020-02-15 JNJ             NaN        NaN\n",
              "2020-02-16 JNJ             NaN        NaN\n",
              "2020-02-17 JNJ             NaN        NaN\n",
              "2020-02-18 JNJ    144.29301453 0.99263108\n",
              "2020-02-19 JNJ    144.09953308        NaN\n",
              "\n",
              "[125087 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CAf2FnpPOzKc",
        "outputId": "80911000-c29e-48b4-f4a6-89ae1ce6c21f"
      },
      "source": [
        "factor_df"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "date        ticker\n",
              "1996-03-28  AEP      0.57735382\n",
              "1996-03-29  AEP             NaN\n",
              "1996-03-30  AEP             NaN\n",
              "1996-03-31  AEP             NaN\n",
              "1996-04-01  AEP             NaN\n",
              "                        ...    \n",
              "2020-02-15  JNJ             NaN\n",
              "2020-02-16  JNJ             NaN\n",
              "2020-02-17  JNJ             NaN\n",
              "2020-02-18  JNJ      0.99263108\n",
              "2020-02-19  JNJ             NaN\n",
              "Name: value, Length: 125087, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cmbm4eAk5_Np",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 688
        },
        "outputId": "6f8ba0e9-4382-48ed-9efd-2cbf4a7d094c"
      },
      "source": [
        "import alphalens as al\n",
        "\n",
        "# FIx bug workaround:\n",
        "# Just stumbled upon this issue.\n",
        "# @luca-s , you are right:\n",
        "# utils.get_clean_factor_and_forward_returns(factor, prices, max_loss=0.0) (factor and prices as shown by @niti's comment) function call reveals the source of trouble: ValueError: Bin edges must be unique raised by pd.qcut inside quantile_calc, line 132.\n",
        "\n",
        "# What it does is it groups factor by date and tries to cut every group into bins (line 165), and clearly fails, since there is no way to cut one sample (one per date) in more than one bin. But since max_loss is different from zero, no_raise is True (line 570), so exception is swallowed, quantiles for every group are set to NaN (line 158), and so the whole dataset is dropped (line 578).\n",
        "# Line numbering is given for commit 32bad52.\n",
        "\n",
        "# Therefore, in case of one asset only, one should use utils.get_clean_factor_and_forward_returns(factor, prices, quantiles=None, bins=1) or utils.get_clean_factor_and_forward_returns(factor, prices, quantiles=None, bins=[-np.inf, np.inf])\n",
        "\n",
        "# Hope that someone will find it useful.\n",
        "\n",
        "factor_data = {}\n",
        "skipped_sentiments = []\n",
        "\n",
        "for sentiment in sentiments:\n",
        "    try:\n",
        "        data = al.utils.get_clean_factor_and_forward_returns(factor_df, \n",
        "                                                             prices, \n",
        "                                                             quantiles=None, \n",
        "                                                             bins=5,\n",
        "                                                            #  bins=[0, 0.25, 0.5 , 0.75, 1], \n",
        "                                                             periods=[3, 7, 14, 21, 30, 60],\n",
        "                                                             max_loss=1)\n",
        "        factor_data[sentiment] = data\n",
        "    \n",
        "    except ValueError as e:\n",
        "\n",
        "      skipped_sentiments.append(sentiment)\n",
        "      print(e)\n",
        "\n",
        "if skipped_sentiments:\n",
        "    print('\\nSkipped the following sentiments:\\n{}'.format('\\n'.join(skipped_sentiments)))\n",
        "factor_data[sentiments[0]]"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n",
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n",
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n",
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n",
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n",
            "Dropped 99.8% entries from factor data: 99.8% in forward returns computation and 0.0% in binning phase (set max_loss=0 to see potentially suppressed Exceptions).\n",
            "max_loss is 100.0%, not exceeded: OK!\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>3D</th>\n",
              "      <th>7D</th>\n",
              "      <th>14D</th>\n",
              "      <th>21D</th>\n",
              "      <th>30D</th>\n",
              "      <th>60D</th>\n",
              "      <th>factor</th>\n",
              "      <th>factor_quantile</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th>asset</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2002-03-29</th>\n",
              "      <th>AEP</th>\n",
              "      <td>-0.00911264</td>\n",
              "      <td>0.01583884</td>\n",
              "      <td>0.03905489</td>\n",
              "      <td>0.03862047</td>\n",
              "      <td>-0.00260313</td>\n",
              "      <td>-0.01084693</td>\n",
              "      <td>0.33885528</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2003-03-20</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.00043820</td>\n",
              "      <td>0.00394247</td>\n",
              "      <td>0.00306608</td>\n",
              "      <td>0.05124845</td>\n",
              "      <td>0.10249620</td>\n",
              "      <td>0.14388747</td>\n",
              "      <td>0.24424699</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2004-03-11</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.00298373</td>\n",
              "      <td>0.00954671</td>\n",
              "      <td>-0.01282786</td>\n",
              "      <td>-0.01014278</td>\n",
              "      <td>-0.04236293</td>\n",
              "      <td>-0.12464086</td>\n",
              "      <td>0.20137915</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2005-03-01</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.01601905</td>\n",
              "      <td>0.02729199</td>\n",
              "      <td>0.00563636</td>\n",
              "      <td>-0.02313818</td>\n",
              "      <td>0.01038291</td>\n",
              "      <td>0.04479391</td>\n",
              "      <td>0.52455370</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2006-03-01</th>\n",
              "      <th>AEP</th>\n",
              "      <td>-0.00218934</td>\n",
              "      <td>-0.01724132</td>\n",
              "      <td>-0.02162027</td>\n",
              "      <td>-0.04050390</td>\n",
              "      <td>-0.06896574</td>\n",
              "      <td>-0.08429156</td>\n",
              "      <td>0.62423398</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-02-21</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>-0.00447991</td>\n",
              "      <td>0.00655557</td>\n",
              "      <td>0.01966771</td>\n",
              "      <td>0.01409490</td>\n",
              "      <td>0.04818622</td>\n",
              "      <td>0.09462386</td>\n",
              "      <td>0.67321487</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-02-24</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.00781239</td>\n",
              "      <td>0.01981726</td>\n",
              "      <td>0.01791162</td>\n",
              "      <td>0.02334218</td>\n",
              "      <td>0.03191720</td>\n",
              "      <td>0.07964932</td>\n",
              "      <td>0.97402353</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2017-02-27</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.01004890</td>\n",
              "      <td>0.01070256</td>\n",
              "      <td>0.03496729</td>\n",
              "      <td>0.04632366</td>\n",
              "      <td>0.02058814</td>\n",
              "      <td>0.00874170</td>\n",
              "      <td>0.98509581</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2018-02-21</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.01624216</td>\n",
              "      <td>0.00617113</td>\n",
              "      <td>-0.00033636</td>\n",
              "      <td>0.02507359</td>\n",
              "      <td>-0.03085932</td>\n",
              "      <td>-0.01877412</td>\n",
              "      <td>0.97516889</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-02-18</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>0.99263108</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>206 rows × 8 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                          3D          7D  ...     factor  factor_quantile\n",
              "date       asset                          ...                            \n",
              "2002-03-29 AEP   -0.00911264  0.01583884  ... 0.33885528                3\n",
              "2003-03-20 AEP    0.00043820  0.00394247  ... 0.24424699                3\n",
              "2004-03-11 AEP    0.00298373  0.00954671  ... 0.20137915                1\n",
              "2005-03-01 AEP    0.01601905  0.02729199  ... 0.52455370                5\n",
              "2006-03-01 AEP   -0.00218934 -0.01724132  ... 0.62423398                3\n",
              "...                      ...         ...  ...        ...              ...\n",
              "2014-02-21 JNJ   -0.00447991  0.00655557  ... 0.67321487                3\n",
              "2016-02-24 JNJ    0.00781239  0.01981726  ... 0.97402353                5\n",
              "2017-02-27 JNJ    0.01004890  0.01070256  ... 0.98509581                3\n",
              "2018-02-21 JNJ    0.01624216  0.00617113  ... 0.97516889                3\n",
              "2020-02-18 JNJ   -0.00134089 -0.00134089  ... 0.99263108                3\n",
              "\n",
              "[206 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GKcjTpFd5_Np"
      },
      "source": [
        "### Alphalens Format with Unix Time\n",
        "Alphalen's `factor_rank_autocorrelation` and `mean_return_by_quantile` functions require unix timestamps to work, so we'll also create factor dataframes with unix time."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vibRc0e1vHQL"
      },
      "source": [
        "def datetime_to_epoch(ser):\n",
        "    \"\"\"Don't convert NaT to large negative values.\"\"\"\n",
        "    if ser.hasnans:\n",
        "        res = ser.dropna().astype('int64').astype('Int64').reindex(index=ser.index)\n",
        "    else:\n",
        "        res = ser.astype('int64')\n",
        "\n",
        "    return res // 10**9"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oo0Qj7ovvcIy"
      },
      "source": [
        "unixt_factor_data = {\n",
        "    factor: data.set_index(pd.MultiIndex.from_tuples(\n",
        "        [(x.timestamp(), y) for x, y in data.index.values],\n",
        "        names=['date', 'asset']))\n",
        "    for factor, data in factor_data.items()}"
      ],
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0LTB0IUvsnFk",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "outputId": "74e99fca-b80e-4a10-8b21-2e0e3ae71ba5"
      },
      "source": [
        "unixt_factor_data['constraining']"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th>3D</th>\n",
              "      <th>7D</th>\n",
              "      <th>14D</th>\n",
              "      <th>21D</th>\n",
              "      <th>30D</th>\n",
              "      <th>60D</th>\n",
              "      <th>factor</th>\n",
              "      <th>factor_quantile</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th>asset</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1017360000.00000000</th>\n",
              "      <th>AEP</th>\n",
              "      <td>-0.00911264</td>\n",
              "      <td>0.01583884</td>\n",
              "      <td>0.03905489</td>\n",
              "      <td>0.03862047</td>\n",
              "      <td>-0.00260313</td>\n",
              "      <td>-0.01084693</td>\n",
              "      <td>0.33885528</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1048118400.00000000</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.00043820</td>\n",
              "      <td>0.00394247</td>\n",
              "      <td>0.00306608</td>\n",
              "      <td>0.05124845</td>\n",
              "      <td>0.10249620</td>\n",
              "      <td>0.14388747</td>\n",
              "      <td>0.24424699</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1078963200.00000000</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.00298373</td>\n",
              "      <td>0.00954671</td>\n",
              "      <td>-0.01282786</td>\n",
              "      <td>-0.01014278</td>\n",
              "      <td>-0.04236293</td>\n",
              "      <td>-0.12464086</td>\n",
              "      <td>0.20137915</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1109635200.00000000</th>\n",
              "      <th>AEP</th>\n",
              "      <td>0.01601905</td>\n",
              "      <td>0.02729199</td>\n",
              "      <td>0.00563636</td>\n",
              "      <td>-0.02313818</td>\n",
              "      <td>0.01038291</td>\n",
              "      <td>0.04479391</td>\n",
              "      <td>0.52455370</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1141171200.00000000</th>\n",
              "      <th>AEP</th>\n",
              "      <td>-0.00218934</td>\n",
              "      <td>-0.01724132</td>\n",
              "      <td>-0.02162027</td>\n",
              "      <td>-0.04050390</td>\n",
              "      <td>-0.06896574</td>\n",
              "      <td>-0.08429156</td>\n",
              "      <td>0.62423398</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1392940800.00000000</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>-0.00447991</td>\n",
              "      <td>0.00655557</td>\n",
              "      <td>0.01966771</td>\n",
              "      <td>0.01409490</td>\n",
              "      <td>0.04818622</td>\n",
              "      <td>0.09462386</td>\n",
              "      <td>0.67321487</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1456272000.00000000</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.00781239</td>\n",
              "      <td>0.01981726</td>\n",
              "      <td>0.01791162</td>\n",
              "      <td>0.02334218</td>\n",
              "      <td>0.03191720</td>\n",
              "      <td>0.07964932</td>\n",
              "      <td>0.97402353</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1488153600.00000000</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.01004890</td>\n",
              "      <td>0.01070256</td>\n",
              "      <td>0.03496729</td>\n",
              "      <td>0.04632366</td>\n",
              "      <td>0.02058814</td>\n",
              "      <td>0.00874170</td>\n",
              "      <td>0.98509581</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1519171200.00000000</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>0.01624216</td>\n",
              "      <td>0.00617113</td>\n",
              "      <td>-0.00033636</td>\n",
              "      <td>0.02507359</td>\n",
              "      <td>-0.03085932</td>\n",
              "      <td>-0.01877412</td>\n",
              "      <td>0.97516889</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1581984000.00000000</th>\n",
              "      <th>JNJ</th>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>-0.00134089</td>\n",
              "      <td>0.99263108</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>206 rows × 8 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                   3D          7D  ...     factor  factor_quantile\n",
              "date                asset                          ...                            \n",
              "1017360000.00000000 AEP   -0.00911264  0.01583884  ... 0.33885528                3\n",
              "1048118400.00000000 AEP    0.00043820  0.00394247  ... 0.24424699                3\n",
              "1078963200.00000000 AEP    0.00298373  0.00954671  ... 0.20137915                1\n",
              "1109635200.00000000 AEP    0.01601905  0.02729199  ... 0.52455370                5\n",
              "1141171200.00000000 AEP   -0.00218934 -0.01724132  ... 0.62423398                3\n",
              "...                               ...         ...  ...        ...              ...\n",
              "1392940800.00000000 JNJ   -0.00447991  0.00655557  ... 0.67321487                3\n",
              "1456272000.00000000 JNJ    0.00781239  0.01981726  ... 0.97402353                5\n",
              "1488153600.00000000 JNJ    0.01004890  0.01070256  ... 0.98509581                3\n",
              "1519171200.00000000 JNJ    0.01624216  0.00617113  ... 0.97516889                3\n",
              "1581984000.00000000 JNJ   -0.00134089 -0.00134089  ... 0.99263108                3\n",
              "\n",
              "[206 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PeRxC1K05_Np"
      },
      "source": [
        "### Factor Returns\n",
        "Let's view the factor returns over time. We should be seeing it generally move up and to the right."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cetSMjpoXBUP"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_JkYgU9DjLU8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "4764a730-0af2-4854-f245-18d1a0c5d893"
      },
      "source": [
        "ls_factor_returns = pd.DataFrame()\n",
        "for factor_name, data in factor_data.items():\n",
        "    ls_factor_returns[factor_name] = al.performance.factor_returns(data).iloc[:, 0]\n",
        "ls_factor_returns.index = pd.to_datetime(ls_factor_returns.index)\n",
        "\n",
        "ls_factor_returns.head()"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>negative</th>\n",
              "      <th>positive</th>\n",
              "      <th>uncertainty</th>\n",
              "      <th>litigious</th>\n",
              "      <th>constraining</th>\n",
              "      <th>interesting</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>date</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2000-03-29</th>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-12-20</th>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2000-12-22</th>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2001-03-09</th>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2001-03-13</th>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "      <td>0.00000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             negative   positive  ...  constraining  interesting\n",
              "date                              ...                           \n",
              "2000-03-29 0.00000000 0.00000000  ...    0.00000000   0.00000000\n",
              "2000-12-20 0.00000000 0.00000000  ...    0.00000000   0.00000000\n",
              "2000-12-22 0.00000000 0.00000000  ...    0.00000000   0.00000000\n",
              "2001-03-09 0.00000000 0.00000000  ...    0.00000000   0.00000000\n",
              "2001-03-13 0.00000000 0.00000000  ...    0.00000000   0.00000000\n",
              "\n",
              "[5 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UWAUPGmBmV9a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 286
        },
        "outputId": "b751c3de-c480-4b49-d063-3ccf10e8914f"
      },
      "source": [
        "(1 + ls_factor_returns).cumprod().plot()"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f95be516910>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AB6H8Sbf5_Ns"
      },
      "source": [
        "### Basis Points Per Day per Quantile\n",
        "It is not enough to look just at the factor weighted return. A good alpha is also monotonic in quantiles. Let's looks the basis points for the factor returns."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S4gTq26B5_Ns"
      },
      "source": [
        "qr_factor_returns = pd.DataFrame()\n",
        "\n",
        "for factor_name, data in unixt_factor_data.items():\n",
        "    qr_factor_returns[factor_name] = al.performance.mean_return_by_quantile(data)[0].iloc[:, 0]"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "id": "jtYeprMGP8Cm",
        "outputId": "b1c762ad-48ef-4fab-e97f-245b229e4451"
      },
      "source": [
        "qr_factor_returns"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>negative</th>\n",
              "      <th>positive</th>\n",
              "      <th>uncertainty</th>\n",
              "      <th>litigious</th>\n",
              "      <th>constraining</th>\n",
              "      <th>interesting</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>factor_quantile</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.00015190</td>\n",
              "      <td>0.00015190</td>\n",
              "      <td>0.00015190</td>\n",
              "      <td>0.00015190</td>\n",
              "      <td>0.00015190</td>\n",
              "      <td>0.00015190</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.00168218</td>\n",
              "      <td>0.00168218</td>\n",
              "      <td>0.00168218</td>\n",
              "      <td>0.00168218</td>\n",
              "      <td>0.00168218</td>\n",
              "      <td>0.00168218</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.00005741</td>\n",
              "      <td>0.00005741</td>\n",
              "      <td>0.00005741</td>\n",
              "      <td>0.00005741</td>\n",
              "      <td>0.00005741</td>\n",
              "      <td>0.00005741</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.01344760</td>\n",
              "      <td>-0.01344760</td>\n",
              "      <td>-0.01344760</td>\n",
              "      <td>-0.01344760</td>\n",
              "      <td>-0.01344760</td>\n",
              "      <td>-0.01344760</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>-0.00049972</td>\n",
              "      <td>-0.00049972</td>\n",
              "      <td>-0.00049972</td>\n",
              "      <td>-0.00049972</td>\n",
              "      <td>-0.00049972</td>\n",
              "      <td>-0.00049972</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                   negative    positive  ...  constraining  interesting\n",
              "factor_quantile                          ...                           \n",
              "1                0.00015190  0.00015190  ...    0.00015190   0.00015190\n",
              "2                0.00168218  0.00168218  ...    0.00168218   0.00168218\n",
              "3                0.00005741  0.00005741  ...    0.00005741   0.00005741\n",
              "4               -0.01344760 -0.01344760  ...   -0.01344760  -0.01344760\n",
              "5               -0.00049972 -0.00049972  ...   -0.00049972  -0.00049972\n",
              "\n",
              "[5 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aOS3h4C1jUpn",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        },
        "outputId": "3ea91f37-451f-4fa2-8e7c-96537e00e391"
      },
      "source": [
        "(10000*qr_factor_returns).plot.bar(\n",
        "    subplots=True,\n",
        "    sharey=True,\n",
        "    layout=(5,3),\n",
        "    figsize=(14, 14),\n",
        "    legend=False,\n",
        "    ylabel='Mean Daily Returns (bps)');"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x1008 with 15 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9o5a-c2F5_Nt"
      },
      "source": [
        "### Turnover Analysis\n",
        "Without doing a full and formal backtest, we can analyze how stable the alphas are over time. Stability in this sense means that from period to period, the alpha ranks do not change much. Since trading is costly, we always prefer, all other things being equal, that the ranks do not change significantly per period. We can measure this with the **Factor Rank Autocorrelation (FRA)**."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7x2pjJnB5_Nt"
      },
      "source": [
        "ls_FRA = pd.DataFrame()\n",
        "\n",
        "for factor, data in unixt_factor_data.items():\n",
        "    ls_FRA[factor] = al.performance.factor_rank_autocorrelation(data)\n",
        "ls_FRA.index = pd.to_datetime((ls_FRA.index*1e9).astype('int64'))\n"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C56X0wNkjtRJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 285
        },
        "outputId": "7cae5645-a10b-4e2f-e1d7-a962c0b774dc"
      },
      "source": [
        "ls_FRA.plot(title=\"Factor Rank Autocorrelation\");"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QVCm4v4BkwWg",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 271
        },
        "outputId": "063b7a1f-7b32-424a-93d6-e4978c608ad7"
      },
      "source": [
        "ls_FRA.loc['2020':,:].resample('Q').plot(title=\"Factor Rank Autocorrelation\");"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZX0XzB6D8jgC"
      },
      "source": [
        ""
      ],
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i51q_WuIplV4"
      },
      "source": [
        "# Write-up\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ccurzzdMpsVt"
      },
      "source": [
        "The notebook didn't run all the way through out-of-the-box. I made some changes to make it work. Unfortunately, I spent too much time debugging a library (alphalens) that isn't updated anymore and was lacking core functionality to do the analysis without making the bias that the notebook author made.\n",
        "\n",
        "The alphalens documentation seems to show that all data points must have the same frequency ex pd.datetime frequency ='D', or 'M', or Y. However, we know the SEC filings are nearly never filed on a constant cadence, and filings across companies do not happen on the same day.. The original author of the notebook compromises, introducing bias, by: \n",
        "setting all companies SEC filing date to the same date\n",
        "setting all SEC filings to the first date of the year.\n",
        "\n",
        "There may be a better-equipped library that already does what I'm looking for.\n",
        "\n",
        "## Changes\n",
        "\n",
        "The changes I made include:\n",
        "\n",
        "* Update requirements.txt to use newest versions, when compatible\n",
        "\n",
        "* Install retrying module, and add @retry decorators to get_sec() so that the function retries checking the SEC RSS feed a few times until it gets all the required data, instead of gathering NoneType data, then continuing to throw an exception later.\n",
        "\n",
        "* Import jaccard_score from sklearn.metrics, instead of jaccard_similarity. The function name as changed in sklearn 0.24, and the old function name was removed. \n",
        "\n",
        "* Add working data source (yfinance, via Yahoo Finance) for price data. The author had left the notebook user to get their own data, possibly due to licensing issues.\n",
        "\n",
        "* Make cosine_similarities_dfs variable fit the alphalens documentation's expected parameters (Multi-index series).\n",
        "\n",
        "* Change min period for alpha calulations to 3, instead of 1, to avoid unpredictable look-back bias due to ambiguity on if sec filings were made before market hours, or after market hours, on any given trading day.\n",
        "\n",
        "* Delete the section titled, \"Sharpe Ratio of the Alphas\". The math makes no sense, as-implemented, and neither does the title.\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CndlrvlorIxo"
      },
      "source": [
        "## Future work:\n",
        "\n",
        "* Find a better price data source for EOD data. Yahoo Finance doesn't go back vary far. QuantConnect  is where I'll look next as they have a python library target algo-traders who are migrating from Quantopian. Also, maybe SimFin, or a python bridge to TWS client or TOS client via DDE on Windows.\n",
        "\n",
        "* Try alternative data sources. I'm interested in NLP on the textual data in the monthly PMI releases (full name: ISM Report on Business, Purchasing Manager's Index).\n",
        "\n",
        "* Using 'ffill' on the non-filing days introduces major bias, by treating each forward-filled day as if it were a fresh SEC filing with identical text to the most recent filing. A proper approach could consider a factors such as 'factor[0], factor[1], ... interesting[80]' which could be interpretted as \"factor-on-day-of-sec-filing, factor-1-day-after-sec-filing, ... factor-n-days-after-sec-filing.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YynNlIk4pnmJ"
      },
      "source": [
        ""
      ],
      "execution_count": 50,
      "outputs": []
    }
  ]
}