docs/example-usage.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyM5Jnt/7VDwBipbtQjQiujc",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/patcon/red-dwarf/blob/main/docs/example-usage.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kEyVHx6y7zpu",
        "outputId": "50a86923-1e1d-4216-cf4f-325186b5cdb9"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "  Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m201.1/201.1 kB\u001b[0m \u001b[31m6.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m31.6/31.6 MB\u001b[0m \u001b[31m177.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m61.4/61.4 kB\u001b[0m \u001b[31m149.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m66.4/66.4 kB\u001b[0m \u001b[31m134.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Building wheel for reddwarf (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n"
          ]
        }
      ],
      "source": [
        "!pip install --quiet --no-cache-dir git+https://github.com/patcon/red-dwarf.git@main"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We're going to try to use this library to reproduce the visualization in this Polis report using matplotlib: https://pol.is/report/r8xhmkwp6shm9yfermteh\n",
        "\n",
        "Note: This easy reproduction currently only works for conversations with the following features:\n",
        "\n",
        "- < 100 participants\n",
        "- no comments tags as metadata\n",
        "\n",
        "<img src=\"https://imgur.com/0CGarhi.png\" width=\"50%\"/>"
      ],
      "metadata": {
        "id": "UDHsPs8sM-qn"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from reddwarf.polis_pandas import PolisClient\n",
        "\n",
        "client = PolisClient()\n",
        "client.load_data(report_id=\"r8xhmkwp6shm9yfermteh\")\n",
        "# Generate the raw vote matrix\n",
        "client.get_matrix(is_filtered=True)\n",
        "client.run_pca()\n",
        "client.scale_projected_data()\n",
        "included_participant_count = client.projected_data.shape[0]\n",
        "print(f\"Included participant count: {included_participant_count}\")\n",
        "client.generate_figure(coord_dataframe=client.projected_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "id": "NkGdoHCy8RdA",
        "outputId": "a582895a-dc40-40a1-89d2-e9a229cd2291"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Included participant count: 36\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 560x400 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFVCAYAAABIAi6iAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAMTgAADE4Bf3eMIwAAKMlJREFUeJzt3WlsW+ed7/HfISmSkqjNtqyNlEVbclzYWZzeIHa2ThMYQYMADtLiziAzQJ1B4kwAY1qoBQrcwbxsgAEyApIJJp0Y0+ReBCkCNEEwmA6mWTptmtT1TVC1dRrb8SJZpCxqsWhRO5fD+0JDXikWZcs6h0ckvx9AQLT4nL8Z2z8+z3me/2Nks9msAACALVxOFwAAQDkjaAEAsBFBCwCAjQhaAABsRNACAGAjghYAABt5nC7A5/OpubnZ6TIAbBIzMzMKBAJOlwHcsPHxcS0uLhb8vuNB29zcrGg06nQZADaJ3t5e9fX1OV0GcMOCweCa32fqGAAAGxG0AADYiKAFAMBGjj+jBVBcAwMDSqfTMgxDLpdLbW1tqq6udrosoGwRtECF6ezslNvtliQlEgkNDw+ru7vb4aqA8sXUMVBhciErSZlMxsFKgMrAiBaoQNFoVDMzM5Kkrq4uZ4sByhxBC1Sg3L6/eDyuWCxG2AI2YuoYqGBNTU2anZ1VOp12uhSgbBG0QAXJZDJKpVL5zxOJhNxu94rntgCsxdQxUKYW0xkd/3BA/UNx7e9s0tMPhGWYGUUiEZmmKcMw5Ha7tWPHDhmG4XS5QNkiaIEydfzDAb1x8pK8HpdOjyQkScce7NauXbscrgyoLEwdA2Wqfygur8elOn+VvB6X+iNxp0sCKhJBC5Sp/Z1NSqZNTS+klEyb2h9qcrokoCIxdQyUqacfCEuS+iNx7Q815T8HUFwELVCmfB63jj1Ia0XAaUwdAwBgI4IWAAAbEbQAANiIoAUAwEYELQAANrIsaBcWFvTYY49p9+7duv3223Xo0CGdP3/eqssDAFCSLN3ec/ToUX3jG9+QYRh66aWX9NRTT+mXv/yllbcAAGwily9f1vT0tFKplHbt2qXq6mql02kNDg7mf8Y0TSWTSe3Zs0ceT+XtKrVsROv3+/XII4/km5MfOHBgxQsNACg/DQ0N2rlzp6qqqvJf83g86u7uzn9s2bJFdXV1FRmyko3PaF944QUdPnz4mq/39fUpGAzmP2ZmZuwqAQBgs9ra2hUhu5p4PK6mpsptAWrL24vnnntO58+f1wcffHDN93p7e9Xb25v/PBgM2lECAGATmJubUyaTUV1dndOlOMbyoH3++ef19ttv6/3331dNTY3VlwcAlJDJyUk1NjZW9JnHlk4d9/X16Sc/+Ynee+89NTY2WnlpAECJyWQySiQSFT1tLFk4oo1Go/re976nnTt36utf/7okyefz6eTJk1bdAkARmaapSCSixcVFuVwuud1utbe3y+fzKRqNan5+XoZhyDAMtbS0KBAIOF0yNpmpqSn5/X75fD6nS3GUZUEbDAaVzWatuhyATWDLli0KBAIyDENXrlzR8PCwdu7cqba2NrndbknS/Py8BgcHtWfPnoqeHix3i+mMjn84oP6huPZ3Lh276PO4NTw8rOnp6fyWHrfbrd27d0taWgS1ZcsWhyt3XmWutQZwXS6Xa8UClpqaGk1MTEhSPmSlpelBlL/jHw7ojZOX5PW4dHokIUk69mC3Ojo6Cv6aXbt2Fau8TY2gBXBDJiYmVF9fn/88FotpampKpmkqFAoxmi1z/UNxeT0u1fmrNL2QUn8k7nRJJYNexwCua2xsTMlkUi0tLfmvtba26pZbblEoFNLo6KhM03SwQthtf2eTkmlT0wspJdOm9ocqe4HTejCiBbCmiYkJJRIJhcNhuVzXvjcPBAK6fPmyFhcXVV1d7UCFKIanHwhLkvojce0PNeU/x/URtAAKmpiY0NWrVxUOh/PPZbPZrJLJZH4laa4hgdfrdbJUrGK1PsSSdPbsWRmGkX/j1NzcrIaGhjWv5fO4dezBbttrLkcELYBVV5S6sqZisZiqqqo0MDAgSTIMQ+FwWMPDw8pkMvl/rEOh0IoFUqWuUEAV+vrNXGtgYEDpdDr/Gra1tVk+I9DQ0KDm5mZdvHjxmu+FQiFmIIqEoAVQcEXpvn37Vv35nTt3FrO8oisUUGsF13qv1dnZmX9zkkgkNDw8rO5ua0eMtbW1ll4PN4egBcCK0i8pFFA3E1yFfo3TW6Si0agkqbq6Wq2trRV7sk4x8MoC0P7OJp0eSbCitMii0Wj+BLOurq6i3TccDsvr9SqbzWp0dFTRaLSo9680BC2ATbWi9O6779bZs2fX9Ry0VOVOL4vH44rFYkULu9zCNcMwtHXrVp07d64o961UBC2ATbWidHBwUI8//vi6noOWuqamJl2+fFnpdNr2KVzTNJXNZvNT17l+xLAPQQtgUxkdHc0fJG6api5durTqwQbZbFZjY2OampqSYRjyeDwKh0tjb2cmk5FpmvnfZyKRkNvttnzl9mp9iLu6ujQ0NJTvTe/1ejkX3GYELYBNrdDBBleuXNHi4qK6u7vlcrmUSqUsu2ehRvlrNdBfT9P9rq4uRSIRmaYpwzDkdru1Y8eODbWxXO3+hfoQW726GWsjaAFsWi6Xa8Wq3eUHG0xMTKzoVpUbHa7XegJqrQb66226b3XD/UL3h/MIWgAlI3ewQSaTUTqdViKRUCKxFCrbtm27bnej1VgVUE5vkXL6/iiMQwUAlITlBxvkni9ms1nt2rVLoVBIIyMjmp+fX/d1lweU1+O66YByuum+0/dHYYxoAViuUNvB6elpjY6OSloKycYtW/TmH+Irpm3vuecenTlzZsUzzS1btqw42CD30djYKGlpQU9NTY3m5+fXvR3Iqj3ETm+Rcvr+KIygBWC51doOZrNZRaNRhcNh+f1+JZNJnTn7hd76dExZw8hP2178zW/0rW99K//rVjvYIHeP6elpbd26Vel0WvPz89q2bdu6a7UqoJzeIuX0/VEYQQvAcmu1Ksy1GzRNU3OprFwuQzW+KiXmk3q7P6qxP8X00i/Or3mwwa5du9TS0qLh4WFNTk5KWnpGW1NTs+5avxxQa20pSqfTikajSiaTMgxD7e3t9BPGdRG0AIrCMAyFQiENDQ3J5XIpk8nowqxX8ylTmWxKk7NJyTB0dT6lN05ekrT2wQYej0c7duywpdZCW4pisZiqq6vV1dWlubk5DQ0N6ZZbbtnQthyUP4IWQFHkGkx0dnaqtrZWc3NzMoxL+vaBTv3foatyGYYyZlazbteGFiVtlMvlUl1dXf7z5VuKEomEenp68l+vqqrS7OysAoGAI7XCOoWOLbyZs3u/jKAFUBQLCwtKp9P5qdaamhp5vV795V0teubrPXrpF+f1xslLSmbMTbVqNrelKJ1OK5vNrtivW1VVZWmjDDhnrWMLN3p2L9t7ABRFVVWV0um0FhYWJEmLi4tKJpP5BvdPPxDWE3fvUGNNlZ64e8emWDW7fEsRypudxxYyogVw09bTdnD37t1qb29XJBKRYRjKZrNqa2vLB21uUdLFf2/dFKtnJyYmrtlSJEmpVCo/ql3+3yh9hY4t3OjZvQQtgJu23raDjY2N+b2vm9laW4omJyfV0tKiubk5pVIpVh2XkdWOLbTi7F6CFsBNK8e2f6lUquCWotbWVkUiEX3xxRdKJpPyeDy6cOGCpJtbJIPNafmxhVac3UvQArhpVnVVcsqqU99VVWtuKcodxXf27Fl1dnaW9cH0laLQsYW5bWgbPbuXoAUqWO4Zao5pmkomk9qzZ88NPYcq9bZ/nHhT2gq1+lxcXFQ0Gs2HZEdHRz4gV3tzZZiZVY8tTKfTlpzdS9ACFczj8aw4m3RiYkKzs7M3vNij1Nv+bXTqe6OLZLAxq7X6lJYCeMuWLWpqatLU1JSGh4fzxxIWenNV6NhCK87uZXsPgLx4PK6mptKa/t2IjZx4Ew6H1dPTo+7ubnk8nnzoonhqa2uvWfWd63udW3RXX1+vVCqlxcVFSdad1rQevP0CIEmam5tTJpNZ0RVpPUzTVCQSWbVHcM7MzIwGBwfV2tp6UwcAWG0jU99WLJKB9VKplDweT74tpmEY+cYiPp/PkXUFBC0ASdLk5KQaGxs31Le3UI9gaWnByejo6E0HuR1udurbNE1ls9kNL5JB8TmxroCgBaBMJqNEIlHwOdWNWKtHsLT03Ky5uVmJRGJDtRZToYYcVi2SgfVyHciy2Wy+McryxiJOrCsgaAHkR2TLp3k3KtcjOHd9wzBUX19fUkFbaOGM1+u1ZJEMrOfxeOT3+3X16lU1NTUpkUjI4/FY+md73TU5dmcAm0Y8HteWLVssu16uR3BHR4dSqZTGx8fz+09LSTk25CgnhVp9dnR0KBqNanx8XC6Xy/HZBoIWqBCFpkElbWjK+Mu+3CN4dnZWqVRK58+fl7Q0TT09Pa1MJrPpm/WXekOOclHoz26hVp8+n8/SP9MbRdACFaIYzRlW6xFcV1enr3zlK/mfiUaj8vv916w6zp0HevjwYV28eDF/HmihpgTFUOoNOcpFqTcWIWiBTWCtTjZWsXIadLURhitrFuwRfCNy54G+/PLL+v73v68LFy6os7NTDQ0Nmp+fVyqVUiQSkcvlUmNjY1G2B5V6Q45yUepT+AQtsAms1cnGKlZOgxYaYRTqEbxcoedluRFwIBDQzMxMfptRbW1tvi3eRg/gRmkq9Sl8ghYo4EYaMFgh18kmd/RWfX29RkZGtLi4aOm9rJwGtWuEEYlE9PjjjyuTyTi6ShSbS6lP4RO0wBrWasBglet1srGKldOgdo0wqqurderUKT366KMaGRlZ8b1MZqnxu9/vV2tra74zE8pfqU/h0+sYKCDXgCEXgDU1NUqlUg5XtTk8/UBYT9y9Q7u2B/TE3TssGWEsLCxoampKv//979XU1CTTNJXJZCQtTTdXVVUpGAyqtrZWly5d2vD9gGJhRAvcoOUNGKx0vU42m5HVI4xcZ6pUKqVvfetbOn36tCQpFovJNE1t3bpV0v/vKxyLxZROpzktByWBP6XADVjegMFqm7GTTbHl9ta63W6lUin5/X5lMhlt375diURCp0+fViaT0eDgoAzDkMfjIWRRMviTClzHlxsw2GGzdbKxg2kubf+Znp5WfD6tyFRKo5nAUvMBrze/yvrll19WX19f/ozR9vZ2Xbx4Mf8c2+12q7W11cnfCrAuBC2whtUaMGxEoQ43m62TjR1GR0clSe8Nu/TGyZi2Vrs1GI/p334f1f/62na1BTxyu116+OGHtbi4qJ07d+rixYsaGRmR2+2WYRiW7p/NNcgwDEMulyvfIGN6elqjo6PKZrNyuVxqb29nSxE2hKAFZE8DhtWUeoebm2WapuLxuG655Rb1v/s7eT0ujc2mNZcyFZua1xu/v6Lbwq069mC3/uu//mvF6u62tjZbno3nGmRIUiKR0PDwsMLhsKLRqMLhsPx+v2ZnZxWNRtXT02P5/VE5CFpAG2vAsB6l3uHmZiWTSbndbo2Pj+up22sUS7j1xh+vamI2q2qfV19MppXxXJW09Dy8GKu7l89Q5FY35+rMdeWqra1VKpXS/Pw8o1rcNLb3AFoZgF6Py7YA3N/ZpGTaLNkONzcrt5La5/Ppzlv3aMoI6Hv3blNbnVd+j3vFa7F3794VI9hYLKZz585paGhIyWTS0rqi0ajOnDmj0dFRBYNBeb1eZTIZzc3NSVoa6Zqmafl9UVkY0QIqXou3Uu9wc7NyW5UaGxtlGIaOPLBb586f17P3deoXFxP512JsbEz19fX5U31y4ZfNZjU5OalLly5ZOo2bW3QWj8cVi8XU1dWlUCiU31ZUU1Mjn8+X30sN3AyCFlDxArDUO9xIWvU0neu1q/R4PKqtrdXMzIzq6uqUTCaVTqX0Pw906y/vXwrh3Orud999Vw8++KAk5bs/2b1/tqmpSZcvX1Y6nVYgEFAgEJC09Gz5zJkzFbXVCtYjaAGVRwAWS0NDg5qbm/Pbb3Jy7SqTGVP/8ck5nYmd0YX52hVnhw4PDysWi8kwDLW3t+dHustXd+emabPZrDKZTD5Up6amLNs/m8lkZJpm/v6JREJutzu/jzf39fHxcQUCAYIWG0LQAliX2traa76Wa1cpLS0s+/jMqP7mfzTqjZNLrRKPPdgtr9er9lDnstXd0tMPBK5Z3X348GFduHBB4XBYg4OD+Y5ZbrdbnZ2d6653tRXlhrnUN9k0zfy1d+zYIcMwNDY2ptnZWUlLvZftaFKCykLQArBU/1Bch3bV6syV9DULy25kdfc///M/q6+vT5LU3b3xWYZC9yy0TasYwVroMPtcU4/cMYF+v1+hUMj2emAvghaApQ7vqVM2OaeXPpm6ZmGZE9ubNuOWqkLT77mmHj09PTIMg0MsygTbewBYZmJiQl/Z5tFIpk6d26492ceJ7U2bcUtVbW3tNYdG5Jp6tLS05Fc5b+aDJXDjLB3Rnjt3Tt/+9rc1MTGhhoYGvfbaa9q7d6+VtwBgo7VWD0ejUc3Nzcnlcsnlcsk0zRW/NregaWc4rJ5ut/5mles7sb2pVLZULW/qMTMzI5fLpe3bt+dXQKN0WRq0zzzzjI4ePaojR47opz/9qY4cOaJPPvnEylsAsFmhw+7r6+vV0dGhy5cva2pqSqZpanBwUG63W+Fw+IbaVTqxurtUVpQvb+rR2tqq+fl5DQ4Oqqenh5OKSpxl//fGxsb06aef6t1335UkffOb39SxY8d0/vx5SxY0ALDf8tXDkuTx+jQ5PacfvvbJf6/YrVVHR4daWlp05swZ7dmzJz/NaXW7ykqzvKmHtLTi2ev1amFhgVFtibPsGW0kElFbW1v+nZdhGOrs7NTQ0JBVtwBQZJ+cHdLHl+Z0YXxGb5y8pOMfLo1Wr1y5orq6OjomWWh5Uw9paSo5mUyyh7cMFH0+oq+vL790X1r6C9vb21vsMgBcx2233ab6ba168YX/raxpKpkx9aNfVck4fa/uuOMO/exnP9PCwoLl9/3Vr37l+L8JDz/8sKqrq/PTub/97W81OTmp+vp63X///fL7/Uomk/r1r3+tq1evrnmtYDCoO++8M38c36lTp7R9+3aFQiFVV1frs88+UyqV0k9/+lPV1dXpvvvuy4drf3+/jh8/XoTfMexkZLPZrBUXGhsbU3d3tyYnJ+XxeJTNZtXW1qaPPvpozanjYDCoaDRqRQkALJJb2PR+JKv/89sheT0uJdOmvvNAUPubDXV1deXbI1qtt7d3xZtxJ2QymRVH6OX+fRsYGFBjY6Oampo0NTWliYmJNY9NzGazOn36tE6Oe/TRwJQOdjXoYHNKe/bsseR8Y2wO18sxy6aOt2/frjvvvFOvv/66JOmtt95SMBjk+SxQYpa3Q3zy/p164u4d2rU9oO88ENQd22RryG4Wqx2hl06nNT8/n3+GWl9fr1QqpcXFxTWvtZg29cszMV0Yn9HPPxvRQjrLlHuFsXTq+F/+5V905MgRPffcc6qvr9err75q5eUBWOhGD7v/xg5Dxx68S5999pmyWc+KdRddXV1luyI2Go3mn5d2dXUplUrJ4/HkQ9IwDFVVVeVXCq/GMAz927lFPXtXo1Km5PdI/35hUV+9nRYGlcTSvyG33HKLTpw4YeUlAdhkvYfdV9qq4i8foZc7um89stmsHgr79fInV3QpkVFbrUvfv3erLScQYfPibRVQoYp12H2pa2pq0uzsrKqqqpROp5Vb1pJbKLVW96aFhQU111bprp527doe0L17OlRf67dlERk2L95SARWqWIfdl5q1jtDz+/26evWqmpqalEgk5PF41tx+U1VVpUwmrafu2Sm/v1uLi4u6ePFi2T/jxkoELVChSqU1oZ3We4ReR0eHotGoxsfH5XK58tPLhXg8HrW3tysSicgwjPxuDIK2shC0QIUqldaEdlrvEXo+n6/g91YLbZ/HrcbGxvxKZVQmghZAxbLyCL1CoQ2wGApAxbLyCD0Wl6EQRrQAKpaVz6lZXIZCCFoAFcvK59QsLkMhBC0AWIDFZSiEZ7QAANiIoAUAwEYELQAANiJoAQCwEUELAICNCFoAAGzE9h7Y4vLly5qenlYqldKuXbtUXV0tSZqentbo6Kiy2axcLpfa29vz3wOAcsSIFrZoaGjQzp07V5zVmclkFI1GFQwG1dPTo9bWVkWjUQerBAD7EbSwRW1t7TUHYieTyfyZnrmfSaVSmp+fd6JEACgKghZF4/V6lclkNDc3J2npQG3TNJVMJh2uDADswzNaFI3b7VYoFFIsFpNpmqqpqZHP55NhGE6XBgC2IWhRVIFAQIFAQJJkmqbOnDkjn8/ncFUAYB+mjlFUqVQq/9/j4+MKBAIELYCyxogWthgeHtb09LTS6bQGBwfldru1e/dujY2NaXZ2VpJUXV2tjo4OhysFAHsRtGXCNE1FIhEtLi7K5XLJ7Xarvb1dPp9Pc3NzGhkZUTabVTabVWNjo5qbmy2572I6o+MfDqh/KK79nUtncPo87oIBSrACqDQEbRnZsmWLAoGADMPQlStXNDw8rJ07d+ry5cvavn276uvrlU6nde7cOdXV1eW32WzE8Q8H9MbJS/J6XDo9kpAkzuQEgGV4RlsmXC6X6urq8it4a2pqVjwPzWQykpZGvoZhyO12W3Lf/qG4vB6X6vxV8npc6o/ELbkuAJQLRrRlamJiQvX19ZKWpmuHhoY0NjamdDqt9vb2a5pJ3Kz9nU06PZLQ9EJKybSp/aEmS64LAOWCoC1DY2NjSiaT+eehExMTamlpUWNjo5LJpC5evKjq6mpLpo6ffiAsSeqPxLU/1JT/HACwhKAtMxMTE0okEgqHw3K5XEqn00okEgqFQpKWujPV1NRobm7OkqD1edw8kwWANfCMtoxMTEzo6tWrCofD+WewbrdbLpdLMzMzkqR0Oq35+XlLQhYAcH2MaMtEKpVSLBZTVVWVBgYGJEmGYWjXrl35toeSlM1mtXXrVtXU1DhZLkrYwMCA0um0DMOQy+VSW1ubfD5fwe1lQKUjaEvQqntXq6q0b9++VX8+EAiou7t8p3cLnX2bE4/HNTw8rM7OzvwCMdy8zs7O/IxJIpHIbyMrtL0MqHQEbQli7+pKDQ0Nam5u1sWLF6/5XjKZ1OTkJIfLW2j51rDctrHc9rKcmpoaTUxMFL22QlYbhVdVVWlwcDD/M7mTpPbs2SOPh38aYR3+NJWg5XtXpxdSFb93tba2dtWvZ7NZDQ8Pq729XSMjI0WuqrxFo9H8c/+urq5rvr98e9lmsNoovLu7e8VMz8TEhGZnZwlZWI7FUCVof2eTkmmTvavXceXKFdXU1DCatUEwGNSePXvU0tKSf/6fk9te1tLS4lB111ptFP5l8XhcTU38XYL1eOtWgti7en0LCwuampriGaHNmpqadPnyZaXTaXk8nmu2l20ma43C5+bmlMlkVkx/A1YhaEsQe1evb3Z2VqlUSl988YWkpW1Nw8PDSqVS2rp1q8PVla5MJiPTNPOdxRKJhNxut9xu96rbyzaTYDAoaWnkGovFVoTt5OSkGhsb8y1MASsRtChLW7duXRGoFy9e1LZt2zbVc8NSlMlkFIlEVvTM3rFjh9LpdMHtZZvNl0fhmUxGiURiU9aK8kDQomQUOpKv0Nm32JhVX2+vt2AgFdpe5rS1RuGSNDU1Jb/fz55f2KZigpa9lKWv0LamGznjlme161eK28hWe3NgmKuPwnPTxPF4XFu2bHG4cpSzigha9lKWB7Y1FVcpvt6F3hysNS3MlDHstrmWBdpg+V5KFjqUNrY1FVcpvt6cj4zNqOxHtOylLB9sayquUny9OR8Zm1FZBy17KcsL25qKqxRf71J8c4DiWa0V5/JBmF1reco6aNlLCVSWUnxzgOIp1IpTsnctT1kHLXspAQA5hVpx2t0XvWyCttAeSwAAclZrxWn3Wp6yCdob2fPHs1oAqGxfbsXZ2tpq+1qesgnaUtzzBwBwRq4VZyKRsH0tT9kELcv6AQCFFGrF2dzcrO3bt+d/zo61PGUTtCzrBwAUUuhAjGI0MiqboGVZPwBAWv+BGMvZ8ay2bIIWAABp8x2IUfa9jgEAlWWz9bwmaAEAZWWzHYjB1DEAoKystji2UJ9j0zQVi8U0MzMjwzDk9/sVCoUsrYegBQCUldUWxxbqczw6OipJ6unpkWEYSqVSltdjydTxiy++qH379unWW2/Vbbfdptdff92KywIAYInV+hybpql4PK6Wlpb8Np/cPlsrWTKi3bt3rz7++GM1NDQoEolo//79Onjw4A0tpQYAoBi+3Oc4mUzK7XZrfHxcMzMzcrlc2r59uwKBgKX3tWRE+9BDD6mhoUGSFAqF1NraqkgkYsWlAQCwRDAY1J49e9TS0qJYLKZsNqtUKiWfz6fu7m61tbUpEokonU5bel/LVx2///77isfjuuuuu1b9fl9fn4LBYP4j9+4CAIBiaGpq0uzsbH6auLGxUZJUXV0tr9erhYUFS+93Q0F78OBBbdu2bdWP5SPXU6dO6cknn9Sbb76p2traVa/V29uraDSa/7B6iA4AwHKZTGbFIqdcn2O3263a2tr8gC+ZTCqZTMrn81l6/xt6RnvixInr/sznn3+uRx99VD/+8Y913333bbgwAACssFaf446ODg0PDysWi8kwDLW3t1u+IMqSxVCnT5/WI488oldeeUWHDh2y4pIAAKzbevsce71ehcP2HkJjyTPav/3bv9XU1JR+8IMf6I477tAdd9yhn//851ZcGgCAG5brc3xhfEZvnLyk4x8OOF2SNSPa9957z4rLAACwIcv7HE8vpBzvcyzR6xgAUEY2W59jiRaMAIAyslqfY6cRtACAsrFan2OnMXUMAICNCFoAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGBC0AADYiaAEAsBFBCwCAjQhaAABsRNACAGAjghYAABsRtAAA2IigBQDARgQtAAA2ImgBALARQQsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANiJoAQCwEUELAICNCFoAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGBC0AADYiaAEAsBFBCwCAjQhaAABsRNACAGAjghYAABsRtAAA2IigBQDARgQtAAA2ImgBALARQQsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANiJoAQCwEUELAICNCFoAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGlgbt2NiYWlpa9Nhjj1l5WQAASpalQfvMM8/o0UcftfKSAACUNMuC9l//9V8VDod1//33W3VJAABKniVBOzAwoB/96Ef64Q9/eN2f7evrUzAYzH/MzMxYUQIAAJuS50Z+6ODBgzp37tyq3+vv79df//Vf66WXXlJ1dfV1r9Xb26ve3t7858Fg8AZLBQCg9NxQ0J44caLg96ampvTHP/5Rf/7nfy5JmpmZ0dzcnB566CF98MEH1lQJAECJuqGgXUtDQ4OuXLmS//y1117TO++8o3feeWejlwYAoOSxjxYAABtZHrRHjhxhNAsAwH9jRAsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANiJoAQCwEUELAICNCFoAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGBC0AADYiaAEAsBFBCwCAjQhaAABsRNACAGAjghYAABsRtAAA2IigBQDARgQtAAA2ImgBALARQQsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANiJoAQCwEUELAICNCFoAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGBC0AADYiaAEAsBFBCwCAjQhaAABsRNACAGAjghYAABsRtAAA2IigBQDARgQtAAA2ImgBALARQQsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANiJoAQCwkWVB+9Zbb+nWW2/Vvn37tG/fPg0ODlp1aQBABZmentb58+d17tw5XbhwQfPz806XtCEeKy7S39+vv/u7v9MvfvELtbe3a3p6Wm6324pLAwAqSCaTUTQaVTgclt/v1+zsrKLRqHp6epwu7aZZErT/+I//qN7eXrW3t0uS6urqrLgsAKDCJJNJud1u+f1+SVJtba1SqZTm5+dVXV3tcHU3x5Kp488//1xDQ0P62te+pv379+vv//7vlclkVv3Zvr4+BYPB/MfMzIwVJQAAyoDX61Umk9Hc3JwkKZFIyDRNJZNJhyu7eTcUtAcPHtS2bdtW/YhEIkqn0+rv79d//ud/6qOPPtJvfvMbvfzyy6teq7e3V9FoNP8RCAQs/Q0BAEqX2+1WKBRSLBbT+fPnNTMzI5/PJ8MwnC7tpt3Q1PGJEyfW/H5nZ6cef/zx/LD+8ccf14kTJ3Ts2LGNVwgAqCiBQCA/CDNNU2fOnJHP53O4qptnydTxE088oXfffVemaSqdTuvdd9/V7bffbsWlAQAVJpVK5f97fHxcgUCgpIPWksVQf/EXf6Hf/e532rt3r9xut+6//3595zvfseLSAIAytZjO6PiHA+ofimt/Z5OefiAsn8etsbExzc7OSpKqq6vV0dHhcKUbY0nQulwuPf/883r++eetuBwAoAIc/3BAb5y8JK/HpdMjCUnSsQe7Sz5Yv4zOUAAAR/QPxeX1uFTnr5LX41J/JO50SbYgaAEAjtjf2aRk2tT0QkrJtKn9oSanS7KFJVPHAACs19MPhCVJ/ZG49oea8p+XG4IWAOAIn8etYw92O12G7Zg6BgDARgQtAAA2ImgBALARQQsAgI0IWgAAbETQAgBgI4IWAAAbEbQAANiIoAUAwEYELQAANjKy2WzWyQJ8Pp+am5udLKEkzMzMKBAIOF1GxeF1Lz5ec2fwut+88fFxLS4uFvy+40GLGxMMBhWNRp0uo+Lwuhcfr7kzeN3tw9QxAAA2ImgBALARQVsient7nS6hIvG6Fx+vuTN43e3DM1oAAGzEiBYAABsRtAAA2IigBQDARgRtiXj11VdlGIbeeecdp0spewsLC3rssce0e/du3X777Tp06JDOnz/vdFkV4dy5c7rnnnu0e/du3XXXXfrTn/7kdEkV5cUXX9S+fft066236rbbbtPrr7/udEllgaAtAYODgzp+/LgOHDjgdCkV4+jRozp79qz+8Ic/6PDhw3rqqaecLqkiPPPMMzp69Ki++OIL/eAHP9CRI0ecLqmi7N27Vx9//LFOnTqln/3sZ/rud7+rCxcuOF1WySNoNznTNPXUU0/pn/7pn+Tz+ZwupyL4/X498sgjMgxDknTgwAENDg46W1QFGBsb06effqq/+qu/kiR985vfVCQSYTahiB566CE1NDRIkkKhkFpbWxWJRByuqvQRtJtcX1+f7r33Xn31q191upSK9cILL+jw4cNOl1H2IpGI2tra5PF4JEmGYaizs1NDQ0MOV1aZ3n//fcXjcd11111Ol1LyPE4XgMI+++wzvfXWW/rwww+dLqViPffcczp//rw++OADp0sBNuzgwYM6d+7cqt/r7+9XKBSSJJ06dUpPPvmk3nzzTdXW1hazxLJE0G5iv/71rzU4OKienh5JUiwW09GjRzUyMqJnn33W4erK3/PPP6+3335b77//vmpqapwup+yFQiGNjIwonU7L4/Eom81qaGhInZ2dTpdWNk6cOHHdn/n888/16KOP6sc//rHuu+++IlRV/pg63sSeffZZjYyMaHBwUIODgzpw4IBeeeUVQrYI+vr69JOf/ETvvfeeGhsbnS6nImzfvl133nlnfqXrW2+9pWAwqO7ubocrqxynT5/WI488oldeeUWHDh1yupyyQQvGEvJnf/Zn+u53v6vHHnvM6VLKWjQaVSgU0s6dO1VXVydp6dzkkydPOlxZ+Tt79qyOHDmiK1euqL6+Xq+++qpuvfVWp8uqGIcOHdKnn36qHTt25L/2D//wD3r44YcdrKr0EbQAANiIqWMAAGxE0AIAYCOCFgAAGxG0AADYiKAFAMBGBC0AADYiaAEAsBFBCwCAjf4fmYatfUS4ausAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This looks good and has the right amount of participants that matches the report page.\n",
        "\n",
        "Let's see is clusters look right."
      ],
      "metadata": {
        "id": "r2dPXQuhNahy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# This uses a random seed, and so will find slightly different things each time.\n",
        "# Future iterations will need to seed cluster centers and do something that Polis platform calls \"k-smoothing\" to ensure matches.\n",
        "client.find_optimal_k()"
      ],
      "metadata": {
        "id": "akOA-OHON4xf",
        "outputId": "acff4b78-06dc-4401-882b-a1960e57116e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 514
        }
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "K=2, silhouette_K=0.5359866103230353\n",
            "K=3, silhouette_K=0.4820296002710926\n",
            "K=4, silhouette_K=0.43371262546361994\n",
            "K=5, silhouette_K=0.3462947329064977\n",
            "Optimal clusters for K=2\n",
            "Plotting PCA embeddings with K-means, K=2\n",
            "Calculating convex hulls around clusters...\n",
            "Hull 0, bounding 3 points\n",
            "Hull 1, bounding 33 points\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 560x400 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFVCAYAAABIAi6iAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAMTgAADE4Bf3eMIwAAO6JJREFUeJzt3XlsZNd9L/jvvbdWFoss7ju7m81utdTdWiOPlEiQE0HPY0GANLbfxAiMpO0IEgxkEqMzGE8QBMj8ERl5UBrPjgEnytjyZAQ4fog8wgwySeSW4efYblvSWLGlXsnu5lLclyJrr7vOH/3O7WI1yeZyi3ep7wcgpGaRtw6LxfrV75zf+R3JsiwLREREVBey2wMgIiIKMgZaIiKiOmKgJSIiqiMGWiIiojpioCUiIqojBloiIqI6Crk9gGg0iq6uLreHQUQekc/n0dzc7PYwiHZsaWkJlUply9tdD7RdXV1Ip9NuD4OIPOLs2bM4d+6c28Mg2rHBwcFtb+fUMRERUR0x0BIREdURAy0REVEdub5GS0QH6+bNm9B1HZIkQZZl9PX1IR6Puz0sosBioCVqMMPDw1AUBQCQzWYxMzOD0dFRl0dFFFycOiZqMCLIAoBhGC6OhKgxMKMlakDpdBr5fB4AcPjwYXcHQxRwDLREDUjs+8tkMpifn2ewJaojTh0TNbC2tjYUCgXouu72UIgCi4GWKMAsy4KqqrAsC8CtNVlN0+zbs9ksFEXZsG5LRM7i1DFRQGWzWfzwhz9EPp9He3s7Pv7xj0OWZUxPT8M0TUiSBEVRcOjQIUiS5PZwiQKLgZYooN59913kcjlEIhEsLy/jV7/6FT72sY/h6NGjbg+NqKEw0BIFVKFQgGmaKJfL0HUdU1NTiEQiiMfjiMViiMViiEQiCIfD9ocsczWJyGkMtEQBY1kWlpeXEY1Goes6ZFmGoijo7+9HKBRCqVRCPp+/Yw+tLMuIRCKIxWJ2MK4OwuFwmFPMRHvAQEsUIOVyGVNTU1hdXUV/fz9SqRRKpRKam5vR0tICAAiHw5t+r2EYMAwDxWIR2WwWhmFsCKyKoiAajdqBOBqNbgjCoVCIgZhoEwy0RAEgstjJyUkAQCqVgiRJiEajaGtr29E1tqs+tiwLpmlC13XkcjlkMhm7klkIhUL2lHR1IA6FQvZ/iRoRn/lEPledxTY3N2+Zse6HqFDeLhCLrUPlchnLy8sbvhe4lUlHo1HE43HE43GuD1PDYKAl8inLsrC0tISpqSlIkmRnsW6QJAmhUGjLrNU0TRiGAVVVUSwWYZrmhoy4en04m81icXHxjkDMaWnyKwZaIh8qlUqYmppCJpOpWxbrJFmWIcvyluMU09LFYhGlUgkTExMbbuf6MPkZAy2Rj3gpi3WSyGiBW1PMra2t9m07XR8W09KxWGzTQEzkFj77iHyiVCphcnISa2trvshinbLT9WHDMLC2tgbDMDYEYjGtXV2oVTstzRaUVE+OBdpyuYzPfvazuHTpEuLxOLq7u/GNb3yDB0oT7ZNpmlhaWsL09HSgslinVK8PR6PRO26vXR+u3T8sSdJd9w+zUIv2w9GM9qWXXsInP/lJSJKEr3/963jxxRfxwx/+0Mm7IGoojZrFOmmn68OlUgm5XM7uAw3cypZlWbbXh0Uw5vrwbbOzs8jlctA0DUePHkU8Hoeu6xvW2U3ThKqqOHHiRENO4zv2E8diMTz77LP2vx977DG8+uqrTl2eqKGYponFxUVmsQegen14M2JaOp/PY319HaZpbrhdZNJbFWopihLo311rayu6urpw48YN+3OhUGjDbOby8jIKhUJDBlmgjmu0X/3qV/H888/f8flz587h3Llz9r/z+Xy9hkDkSyKLXV9fRyKRYBbrMrE+vFkwrl4fXl9fx+rqasOtDycSibt+TSaTQU9PzwGMxpvqEmhfeeUVjI+P45133rnjtrNnz+Ls2bP2vwcHB+sxBCLfqc5iZVlGa2troDOhILjb+rAIxKqqolQqYWlp6Y7vD4fDgT7oQayLJ5NJt4fiGscD7auvvorvfe97OH/+PJqampy+PFEgFYtFO4tNJpMNO8UWNDtt5BHkgx5WV1cbfunD0b/mc+fO4Tvf+Q7Onz+PVCrl5KWJAsk0TSwsLCCdTkOW5YZ/QWo0dyvUqj3ooXZ92OuNPAzDQDabbfgzkB0LtOl0Gn/8x3+MkZER/OZv/iYAIBqN4uc//7lTd0EUKF7PYsW0p1hzFPtZJUmCrusbPi8CBjmrXgc97Gd92DTNHf+u19fX7QYijcyxv+zBwcE7fslEdCc/ZbGyLEOSJEiSZGdXoVBoQyWtZVnQdd3+OjoY+znoQXx/OBzeMC293frw+vo6fvSjHyGXy6Gvrw9PPPEEwuEwZmZmkMvl7C09iqLg+PHjAG4VQbW3t9fvQfAJb72FJgq4QqGAyclJZLNZT2ax1WoDpyRJ9tRl9ef5Btubdro+XKlUUCgUtj3oIRaL4d///d+RzWYRiUQwOzuLS5cu4YEHHsDAwMCWY2j0KWPBu3/lRAHipyx2K7VThoZh2IE36HtFg2g3Bz1ks1l7D3GlUoFpmshmswc8Yv9ioCWqM5HF5nI5NDc3ezqL3Yqohq0OtGLa0jRNu5sSg21w1Dby6O3txezsrP1c0HUdc3Nz6Orq8uVz+iDx0SGqE8Mw7Cw2FAr5thJfFERtlbXKsnzHthQKnqGhIcTjcZTLZbS0tCCRSGBqagoLCwsYGhpCe3s732htgYGWqA4KhQImJiaQy+U8vxa7HTE9XL1NpLraGMAdW07IO6qn96t/h5qmbfg6RVHuWkksSRK6uro2fC6VSqFSqWB8fBwtLS0YGhpCc3Ozgz9BMPjzr5/IowzDwPz8PGZmZhAKhdDW1ub2kO7KsixMTU1hZWUFsVgMIyMjiMVi9vYR4NY0IXC70nWrbT9BsVWA2urze7lW7RapejyGoiBK/P6qObXHNhqNIhKJoFgs4uLFi+ju7kZ/f3/Db+mpxkBL5JB8Po+JiQkUCgVfrcUuLi5ibm4Osiwjm81ibGwMp0+ftrd/bMYvP9tebRWgtgtcu71WdWAVhUdO97U+qL3NkiQhkUjANE0sLy9jZWUF/f396O7uDvxzZSf4CBDtU20W67e12FKptGENtlQquT0k120VoPYSuLb6HrdnAGpnKZwYjyzLaGlpga7rSKfTWFxc5PotGGiJ9qU6i00mk748iaW1tRXz8/P2uh0bDByc6unjg8z8xLSxWB4QjUicvH5raysqlQrGxsbQ2tqKwcHBhj1YgIGWaA8Mw8Dc3BxmZ2cRDod9l8VWa2trw7Fjx7C6uopYLIb+/n5Xx/Nbv/VbdtB3u1dvvYngVo9gtx3xmIr2mbuZCt+N6vXbS5cuoaurC/39/YjFYnW5P69ioCXapSBksbU6OjrQ0dHh9jAAANeuXcPHPvaxur34e5HYImVZVt3fWGxWNV7P+6xev11ZWcHq6mrDrd82xk9J5IAgZbFelk6nN2wl2qrCWUx7VreF9MsLdz22SOm6jpWVFUiShPb2doRCoQ0Vz+KNS21hlnhM6616/XZ6erqh1m/98awkclkul8PExASKxWJgsli/2OpgA9Gbt3q9cbcsy8LCwgLW19eRSCTQ399vZ5e1ASocDm/5eQDIZrO4fv06NE1DT08PhoeH7TFvFuxq30DsZ5rcMAxcvHjRLmSbn5/HqVOntjx0wOnq5t0QBYONtH7LQEu0DZHFzszMIBKJMIs9YNsdbFDbSGMvQWphYQETExMAbp00o+s6Dh8+vGWA2urzpmni2rVrMAwDsixjbm4OyWQS7e3tW36Pk9l3Pp9HuVy2rykOkm9paXHsPpzWSOu3DLREW2AW6z3iYAORCVafOLOXM3HX1tbsbNIwDKytre1pXIZhQNd1O/CLU3EOSigUgmVZG7J6P0yjV6/frq6u2vtve3p6fDH+neJJzUQ1dF3H1NQULl26BMMwkEqlGGTraKdTvpsdbADAPh+3eip2pxKJhL0ObJomEonErr6/egxi/VFVVciyjNbW1j1day8SiQQGBgbsn2NwcBBNTU0Hdv/7JcsykskkmpqakE6n8dFHH2F5eTkw7T2D85aByAHZbBaTk5PMYvdpq7aDYhsLcHt9dHZ2FolEAsePH0ckEsEzzzxjb+8Ra5oii61trCCCrphi3m3V7sDAADRNs9doDx8+vKefV5Ik3HPPPVhYWICmaejs7DzwQDc0NGRvzfLr87Z6/ba6f7Lf128ZaIlw6wV9dnYWc3NziEajXIvdp83aDorMUQTemzdvor+/H4uLi8jn85icnMTo6CjefvttPPLII3f0F64tFhLrtYqi3DFtulOyLGNkZGT/PzBuBTe39yD7NcDWCtr6LQMtNbxsNouJiQmUy2VmsQ7Zbq1UZJ26rkPXdTsbzefz+MUvfoGFhQVcuXIFx48fhyzLmx5sUD1dLLLfvazRbjW+7bYUVd+2k1NvaG+CtH7rr9ESOag2iz3INbVGVH3yj2EYOHToEMbGxqCqKoBbR7eJ5gnr6+tYWFhAf3//lltR6rlvdqstRYZh2PcrpsF54H19ifVbXdcxMzOzYf+tX97kMNBSQ2IWe/BEgwmRBZqmidHRUSwuLqKpqQnXr1/fsAZbe2bqQdluS5FYJwZg/wwH0c2JNvZPvn79OhYXFzE4OOjYFqabN2/aMyyyLKOvrw/xeBxXr161PwcAXV1du35TzkBLDUW8K56fn2cWe8DEOqp4wZJlGYqiYGBgALIsI5/PY25uzt7C45WWkLVbihhU3SXWb0ulEi5fvozOzk709/cjHo/v67rDw8P2m6hsNouZmRmMjo4CuFVotp/rM9BSw1hfX8fExAQqlQpaWlp8M+3kdbquo1wuIxaLbTuVW91WUax3VmeDw8PDaG5uRjKZxMmTJ/e81cZJW20pIndJkoSmpibEYjFkMhm7f/J+1m+rZ7XE790pDLQUeNVZbCwWYxbroHw+j8uXL9trmPfddx+ampq2bFWoKMqGSuTqqWJJktDR0YFEIuGZILvZliJOFXuH0+u36XQa+XweADZs9Uqn0wCAeDyO3t7eXQdzBloKNGax9TU1NQXTNBEOh6FpGqanp3HPPfds2XbQqcrgetvJliKuz3qHWL9VVRXj4+NIJpP2/tvd/H4GBwcB3GrHOT8/j8OHD+PIkSOIRCL2vu90Or3r/dbef8YT7YGmaZiYmMCVK1cA3Drc3A8v8H5T27nHb5181tbWMD4+jqmpKTvTFkVbwK1sXNM0+zaxZ1fTNHt6UXyN3372IIpEImhra4Oqqrh8+TKuX79uH7SwG21tbSgUCtB1HZFIBMDtGZdisbjr6zGjpcBZX1/HzZs3oaoqs9g6GxgYwLVr16DrOmRZxsDAgNtD2rFcLoerV6/aRU65XA733XcfJEna0ZYiTdMCfzC9X4n127W1tQ3rt+L3WjsTIWYwqk9iEtXxhmHYszPr6+t7aprBQEuBoWkaZmZmsLCwwLXYHbIsa9MD1ncaQNra2nD//fejVCrZL25+kcvlYFkWwuEwLMtCLpezp4XJH7Zq9SlmJcTzcW5uDktLS+ju7sbVq1exuLiItrY2PPnkk3ZNwfT0tL2PW1EUHDp0yO57Lt6MRSIRe3p5NxhoKRDW1tYwMTHBLHaXarM3UQC0mywtHo/ve2uFG8SLsHixjkaju37eVHerqi2aovrbrNUnAPu4QrHXuampCaZp4r333sPa2hri8TiWl5fx3nvv4amnnkIkEsHRo0c3vQ+xxWc/GGjJ1zRNQzqdxsLCAuLxOLPYfWqkjK6trQ2Dg4NYXFxEOBzG0aNHdxUoqw+cF12i/NYa0O82e2NUu21M/I7C4bBdwCae57lc7kDGyWcF+ZJlWVhbW8Pk5CRUVWWxkwPEFNxes7LtegRX34fINtwO6JIkYXBwcE9TgeL7xX9lWd50Cp7cU/37AW5tRWtpaUGxWLSniI8cOXIgY2GgJd+pzmKbmpqYxTpEdEDaz/TnVj2CgdvrZn6bXq1UKiiVSojH44hGowBwR5coP/5cjcA0TRQKBYTDYXR0dOD+++/H4uIilpaWkEql9nws4m4x0JJvWJaFTCaDyclJ6LrOLNZBYrptP1nmdj2CgdvrZn7aBpPNZnHlyhV7KvLEiRN2b93qDFZk7+QdhUIBmqaho6MDmqahv7/fLmba6yzGXvFVinxBVVXcvHkTY2NjUBSFBU8OE4HEyaxMZMji/wH/tTJMp9N24wrTNO0OQaKITHxwm493qKoKwzAQjUZx7733oqurC+Fw2J6NcAMzWvK02iyWAXZvLMvC/Pw81tbW0NTUhMHBwQ0ZWHVQdEJ1j+DqA9/9ZrMsnQ6eZVlYWVlBuVxGS0uLPatQ2+rTNE2Ypom+vj7k83nMz89DluUDz2Br+e+ZTw1DVVWk02n7GLWmpia3h+RbCwsLmJyctM961TRtw7YFJ4NgbY/g2p7HwO01Tq9Ptw4ODuLKlSvQdR2Korj+gt2o0uk0ZmZmANx6s3PPPfcglUrZz7F8Pg9JkjAwMICuri6EQiH09PS4POrbGGjJc7gW67xsNgsA9uHl4t9O26xHcG1/Y3HmZ22Q1XUdlmXhd3/3d+3AJoqqNmtKcBCSySQefPBBuxhqq45RVF+Li4t2pbqmaVheXkZraysKhQIMw0B3dzf6+vpcnR7eDl+9yFNUVcWNGzdw7dq1hlqLFR2aRF9dkfE5pbm5GcDtQCj+vVdra2v46KOPcPHiRXsv4nY9gndCURSEw2H8/d//vX2N2mpecV2njzHbTjgcRktLC4Osi6LRqL0HFriV1a6trSGRSODkyZM4fPiwZ4MswIyWPMKyLKyurmJychKGYTRcFlvbycbpNc3e3l5omma/OB06dGjP1yqXy7h27ZpdqXzlyhU8+OCDdmHQ3Wz1c4mA2tLSsuGNhnhMLMti0VGDGhkZwdWrV1GpVBCPx9HX14cjR46gtbXVF88HBlpynaqqmJ6extLSEpqamjxxFimwswYMTt3PZp1snDyCTZZlHDp0aF8BViiVSnbQA24f/O5ExqdpGr7whS/s+zoULNFoFEeOHIGiKBgeHkZHR4fn1/erMdCSa/yQxW7XgMFp1YFWBFsvvltvamqCJEnQNA3ArQzVqV7Hsizj3XffxZNPPrnp9HD1+q4XHxtylmg4YZom+vv70dvb68spfAZackWlUsH09DSWl5eRSCQ8k8VWu1sDhkYVjUZx4sQJzM7O2m0MnXjzIbL4n/3sZ3jqqac2zCYoimIXSIn1bD++4NLOWJaFUqkEVVXR0dGBgYEBXx5cITDQ0oES++EmJydhmiZSqZRvMhOn95rWEhms09PG9VC9l9EJopDKsiz8/u//vp0tizc2YppQ9BUWX+vlx4j2plKpoFgsoqWlBUePHkUymfT975mBlg5MpVLB1NQUVlZWkEgkEIlE3B7SjlU3YHBa7VRxPbo0eYlpmpifn0ehUEAqlUJnZyeA23trK5WK/TgoigLTNO3AW9v2kIJD13Xk83lEo1GMjo6ivb3dc0tJexWMn4I8zbIsLC8v46OPPsL6+jpSqZTvgmx1A4b9qFQqGBsbw6VLl7CysmJ/vjqgBPWoOlFcVi6XkUgk0NzcjKtXr2J8fBzArbVesb2nurq49nEXZ5BSMJimiWw2i2KxiKGhIZw6dQqdnZ2BCbIAM1qqM79ksZZlQVVVhMPhDX/gmzVg2M99XL58GeVyGZIkIZfLIRKJ2FNjQQ8eYhr4ypUr9mNgGAYWFhbQ29trH8T+mc98xq5qrt1T7OTxetXXri6wEm38xMwCC6/qw7IsFAoF6LqO7u5u9Pf3e3ov7H4E+y+bXFO9FmtZlqfXYjVNw+XLl1EsFhEKhXDPPfcgmUze0YAB2F82JbbBiKCt6zoKhQKSyaRjP4tXiccyFAqhqanJriQVgWxxcREjIyOQJAnXr1/HkSNH7MdZUZS6ZDfVAdQ0Tei6bnfOEr8j8XkWXjmrXC6jXC6jtbUVg4OD+26g4nUMtOQ4v2SxwuzsrB1kdV3HzZs3cf/999sntDglFAohGo3aa5AAGq5/s2maGBkZQU9PDyYnJ1EoFADcmjkQj8ns7KzjnbE2s9kbv9pzZmVZtpcOvPpG0U80TUOhUEA8Hsfx48c9/QbcSQy05BixFjs5OQkAvvkjqs5WZVneVdvA3RDnmU5OTkLTNPT29jpauesHkiQhEonYj8Xly5cRjUY3HMD98MMP3zF9LwKx09O41dPH1TMVosJczGgw0O6PYRjI5/NQFAWHDx9GZ2dnIOsQtsJAS44ol8uYmprC6uoqmpubfTXV1t3djZWVFbuyta+vr273FY/HceLEibpd3+tEsAqHw5AkCQ888MAdQbWtrc3+nAisYurZ6WlcEVyr216KwjTRFpP2zjRNFItFmKaJ3t5e9Pb2en6Gqx4YaGlf/JrFVksmkzh16hRyuRxisRhaW1vdHpKnbXaaTm27SmDj2upmW5hqs0Tx/W+++SbOnj1rf5/4rwiA9cguq6eIq08bEgHeb89ptwWt4cR+MdDSnvk5i63F8253ThSE1U6xi0Pex8bGEA6H0d7ebp/ABNzu7qRpmn0NEcCqq7srlQqAO9dLRXDfS9ATQVRksFtdW9xW/fkg72muB9FwIplMBqbhxH4x0NKuWZaFpaUlTE1NQZIkX2axtHebTaeKYDQ/P4+1tTUkk0mEw2Fcu3YNDz30kF3Be/XqVeTzeQDAwMAAhoaG7qjurj6PtvZQh71UfM/Pz9vV7319fRgeHgaATa8tqsFrt/3Q3YmGE5FIJHANJ/aLjwLtSqlUwtWrV3Hjxg3EYjE0NzczyJJN0zRYloWuri5ks1l73RsAMpkM8vm8vQ46MzNjZ7fiiL3ahhWiiUU4HN7TXuZKpYLJyUl7/+3c3Bxyudy2197vfe6EYRjQNM1+vAQxBS9uq1dhnpNqG06cPn06cA0n9osZLe0Is1jaifb2dpimiXA4jJs3b6Krq8t+nhzElp1aIjutXuv1QvDaavp9s7Vvr2qkhhP7xUBLd1UqlTA5OYm1tTXfr8VSfcViMfT29mJ1dRVHjx5FR0eHfVt7ezvm5ubs82z7+vrq/lxqampCMplELpcDcOvkIS9sqdos26tu6lH9xsCLyuUySqUSUqlUQzSc2C9HA+3Y2Bh+7/d+D8vLy2htbcW3v/1tnDx50sm7oANkmiaWlpYwPT0NwJ8VxbQ72x12X7t2WUsUNEUikU23SCmKgpMnT9r7KQ/iaESxX3dlZQWWZaG9vd3zrS5FZTWADRXQXlDbcKKtrY2vCTvg6DPu5ZdfxksvvYQzZ87gH//xH3HmzBm89957Tt4FHRBmsY2r+rB7XddRqVRQqVTQ3Nxsb7GpbU0ZCoV21K5SUZQD3z6lKAq6u7sP9D73Q7y5EefueqHqubrhxKFDh9DV1cUisV1w7K3S4uIi3n//fXzuc58DAHz605/G9PS0fTIH+YM4wuyjjz5CPp9HKpVikG0gojuWaPh/48YN6LqOy5cvY2JiAsCtwCUCqCgcqi1oEoVEtHvV08Zur9NaloVcLod8Po/e3l6cPn0avb29DLK75NhfwvT0NPr6+uw/LkmSMDw8jKmpKYyOjjp1N1RHIotdX19HIpFggG1wa2traGlpQTabRSgUwtLSEgYGBhCLxbi/tA520tTjIBWLRVQqFbvhBPeZ792Bv+U8d+4czp07Z/97ZWXF7gJD7rAsC8Vi0d72wABLAPDII4/gyJEjeO211+xtKOfPn8epU6fw+OOP4x/+4R9QLBYdv9/33nsPX/nKVxy/7m585jOfQSKRsI9P/MEPfoDFxUWkUil88pOfRDweh6qq+Od//ucN5wpv5tChQ3jiiSfsqfT33nsP/f39GBkZQSKRwPr6OlRVxTe/+U20trbiE5/4BOLxOCzLwoULFzA2NnZAP/Ut4lzkcDiMZDJp96amvZMsh+YlFhcXMTo6itXVVYRCIbuq8Mc//vG2Ge3g4CDS6bQTQ6A9KBaLdhabTCY53UcAbhc23bhxA5lMBsCtBhMDAwMbjpGrh6985Sv4kz/5k7pce6dqu0MZhoFwOAxd1+0CJbFWvd3fTKVSgWmauHbtGkqlEjo6OnD48OG6Pn57Vd1wYmhoiA0nduFuccyxV9Xu7m48/PDDeOONN3DmzBm8+eabGBwc5LSxR5mmiYWFBaTTaciyzIriBiTW3wzDQEtLi73uVt0O8fjx4ygWi5BlGdFotO5B1iu2OkKvdk/u3aZ3xRvYaDQKTdNQLBY9tzfWNE3k83lIkoTBwUH09PTwDbfDHH00//Zv/xZnzpzBK6+8gpaWFrz++utOXp4cwiyWAGBychLz8/OQJAmxWAwnT560q4qB29XD0WgUoVDI7vJU3WQhyEF3qyP0NitW2uoxCIVCuHHjBkZGRmCapn3QghceM7FkpGkaurq60N/fj1gs5vawAsnRV9h77rkHFy5ccPKS5CBmsSRomob5+Xl7j2ypVEImk0FXV9eWa/SNtnZfe4TeXiptRdX+9evXUSqVMDo6iubmZtfPt2XDiYPFVKZBFAoFTE5OIpfLobm5mVlsg9vsRZ5vujYnjtATdlsVHIvFcO+999qZbG1byIPEhhPu4KttwFVnsaJZAP+wKBQKYWhoyO76lUwm0dbW5vKo3CG6YVUfMg9sfoRe7fabu21xEreJIOvWlh02nHAXA22AMYul7QwMDKCjowO6riORSDTkG7BsNotr165B13WkUikcO3ZswyHwwMYj9MTRfYZh3HGEnqqqyGQyUBQFHR0ddhAW5/AKIqAfBMuykM/nYZoment70dvbi0gkciD3TbfxlTeADMOws9hQKIRUKuX2kMijGrn4xbIsjI+P25XUa2trWFpaQm9v75ZvSrc6E1dVVXz44Yd2wdjq6iqOHTtmd9pyY5tMqVRCpVJBe3s7G064jIE2YAqFAiYmJpDL5VhRTHQXmqZtmNatPj93NzKZjN3kwbIsrK6uQtd1VwrIVFVFoVBAc3Mzjhw5gpaWloacrfASvgoHhGEYmJ+fx8zMDEKhUMOutxHtlCRJ6O7uxsLCgr1G297evqdriSlkcdSdG1msrusoFAoIh8P2EYVsOOENDLQBkM/nMTExYb+LZRZLtDOHDx9GMpmEpmlobW3d8/RqR0cHMpkMVlZWIMsyRkZGDqzYyDRNFAoFWJaFgYEBdHd3N9xWLK/jK7KP1WaxXIsl2h1JktDZ2enIdUZHR3H48GHIsnwgQZYNJ/yDgdanqrPYZDLJUn0ilx3kgRzlchnlchktLS04duwYksnkgdwv7Q0Drc9UZ7HhcJhZLFEDEQ0nYrEYjh07xoYTPsFA6yPMYokak2EYKBQKkGWZDSd8iIHWBwzDwNzcHGZnZ5nFEjWQ6oYTPT096OvrY8MJH2Kg9bhcLoeJiQkUi0VmsUQNRDScaGtrw8DAABKJhNtDoj1ioPUokcXOzMwgEokwiyVqEKqqolgsIpFIsOFEQDDQehCzWKLGU91wYmRkhA0nAoSB1kN0Xcfs7Czm5uYQjUZ9ncUahmGfelJ9OLhpmjBN0z7B5CAbrBN5UXXDif7+fvT09LDhRMAw0HpE0LJY0Xy9+tQScRyZCLymabrWD5bIbWw40TgYaF0WpCy22mZTXrXnfFYfR8aslhpJpVJBsVhEa2srG040AAZaF2WzWUxMTKBcLgcii72b6uljWZbtqWUGWmoUouFENBrFsWPH0N7ezud+A2CgdUFtFtva2ur2kA6EWJM1TROGYbDQgxqGaDghSRKGh4fR3d0d+DfWdBsD7QFrtCy2VvXxYeJIMb6jp6CyLAuFQgGGYaC7uxt9fX2IRqNuD4sOGAPtAdF1HTMzM5ifn2+oLLZW9TSxCLIMtBREpVIJ5XIZ7e3tbDjR4BhoD8D6+jomJiZQqVTQ0tISqClTy7KwuroKXdeRSqXsd+vV23tE5XE4HLaLn4DbU8lEQaKqKgqFAhKJBE6cOIHW1la+mWxwDLR1VJ3FxmKxumaxYutMbRAT22jE3lUAjp6Xef36dSwvLwO4tV/29OnTiEajUBRl0/vgofQUVIZhIJ/PIxwO4+jRo2w4QTa+6tWJG1msLMv2VKxhGPaeVcMwoCgKZFmGZVnQdd3+2v3QdR3Ly8v2vlhN05DJZNDb2+vQT0TkfWw4QXfDQOswTdMwMzODhYWFumex1WrXOkUmW0+isEls1wHAqWBqGNUNJzo7O9Hf3494PO72sMiDGGgdtL6+jps3b0JVVdfXYquDn+jQZBgGADjW9lCWZRw9ehTXr1+Hrutoa2tDR0fHvq9L5HWi4URLSwsbTtBdMdA6wK0sdisioIpAu9nUsVPVvh0dHUilUjAMA+FwmEUfFGi6riOfz7PhBO0KA+0+ra2tYWJiwhNZLAC7IEpkrZZlwbIse1wiwDrZjWmrwieioDBNE/l83m440dXVxcI+2jE+U/ZI0zSk02ksLCwgHo+7nsUCt7fUVJ+WI4ipZBF4ieju2HCCnMBAu0uWZWFtbQ2Tk5NQVRWtra0HnsXmcjlkMhlEIhF0d3fbAbR236o4QUdRFLsKGdjYnYlot3Rd33Qv9Fbby/yKDSfIKQy0uyCy2MXFRdey2Hw+j0uXLtldlXK5HI4dOwZJkrbcUhD0wLrV2beC6K0s1qlpf6oDqDjqMBQKbbm9zG/YcIKc5r+/AhdYloVMJoPJyUnouu7qWmwmk4FpmohEIjBNE6urqw1/+s1mZ98K7KfsvM0eSze2l+3GVll47XPGsixomoaRkRF0dHSw9oAcwUB7F6qqIp1OY2lpCfF4HE1NTa6OJxqN2i9ihmEgEom4Oh4v2OpNj+iWJabOyTnVgWuzrLV6e5kXbJaFh8NhhMNhu+GELMtoamrCiRMn2HCCHMVAuwUvZbHVOjs7kcvlsLy8bG8xYLa2OR5aUD8iuIo3fNXBtnZ7mRds9hywLAulUgmqqqKjowOGYaCvr49BlhzHQLsJkcUuLi6iqanJ9Sy2mmgSMTIywgCyDVFdzam/+pJl2S6CEmuz1dvLvKQ6CzdNE2tra2hpacHRo0ehKAqmp6fZeILqgoG2Sm0W60ZF8U557UXMa8QhCtVrcNVBgPamep0TwIZ12O22l3mBWMcvl8sIh8MYHR1Fe3s7ZFlGOp1GKpXy5LjJ/xho/xtVVTE9PY2lpSXPZbF0m6ZpdjHYdi+KtU00xEEKXn3j5EXiCETDMJBKpex6gNptPNXTyMCd28u8oLrhRH9/PzKZDFKplJ2RZ7NZHD161O1hUkB546/AReLFZHJyEoZheDqLbXSzs7OYmpoCcKv14+joqD1dudnZt7R3lmVhfHwcKysrAIBIJIJTp04hEolsGTy98phXKhUYhmE3+C8UCjBNE11dXejr60OlUkE2m7XfiK2vryMWi7ERBdVNwwTaTCaDmZkZDA8Po6WlBcDtLHZ5eRnxeJwb0j1MVVVMTU3Za38rKyvo7OxEW1vbjlpAeiWz8gtd17GysmJPA6uqikwmg56eHreHtq35+XlMTk7CsiwkEgn09/fbzxFN0zA9PQ1FUXDo0CF7RiSTyaC9vd3lkVOQNcSrj6qqWF1dtd/h1maxXqkopq2JjFW8OIpMlupDNJ9w6whEscYOYEPbUPH/m/3bMAzcvHnTHns+n0dLSwtOnDix7TIDp4yp3gIfaC3LwszMDPr7+zE3NwdN03D9+nUsLy8jkUgwi/WJaDSKtrY2ZDIZAEAsFkMqlXJ3UAGwXQAbHh7esKQSj8dRKpW2DXY7VRv4qr9XVVV7PbX6Q6yxiw/ReEL8W6zDV7cdjcViLHAi1wU+0K6srKCpqQmxWAyapmFqagq6rrPC0GckScLx48ftzlipVMqX08FbBantbtvJ11Zn+tX3tRPV3ysCmiRJaG9vRyqVgmVZiEaj9rT9ZgFvs89t9XG327u7u/Hwww9v+NxOra+vY2xsDKZporm5GYODgzv+XqJ68d8r1S6Uy2Wsr69jYGAA4+PjKJfLCIVCrCj2KfHivxvVAcmJ4LaXMVd/X21Q2y5bk2XZzthqv2azf+/3o/ooRTeJn3svfu3Xfg1DQ0NQVRW9vb3snEYb3Lx50z6PW5Zl9PX12UuKwOa1PE4IdKAtFAqoVCoYGxsDgA2njHAvpfP2ko1td7v4/9oeuuLzO7FVtlYb2LbLzHaTre0mc2PXKudJkoTe3l63h0EeNTw8bL/2Z7NZzMzMYHR0FMCdtTxOClSgrVQqAG6t51UqFayuriKbzSKRSCASiQRuL+VepxoPKlsDsOOA1ijZGhG5pzrBqi6mrK3lcVpgAu0vf/lLXLx4EQBw+PBhe8rIqbXYg8zWqu9zO5tNQ+40eDFbI6JGlE6nkc/nAdyKFcDtWp56ZLNAQAJtJpPBxYsXEQqFoGkarl69isOHD6O5uRm5XG5P2VqtvWRr1f+/2VrbXgIXszUior0TBXKZTAbz8/Po7e3F+vo6RkZG6nafgQi0mqYBgF3SrygKOjo60NHRwWyNiIju0NbWhtnZWWSzWWiahmvXrgG41axlZmYGmqaho6PDkfsKRKDt6OhAe3s7VldXAQBdXV247777fLn9g4iInCdatYpWoaINZ1dXF7q7u+2vu3HjBjo7O1l1XEtRFDz99NOYnp4GAAwNDTHIEhE1IMuyUPq//x+o776H0H33IvE//kdI4TAMw8D09LR9TnVtK856Ckw0CofDdZ1jJyIi7yt+7/9C7j/9J0CSgO+/DXNlBS1/+D8hEonsqN1mPeJIMPa5EBERAaj86N8AWYGZbAGiMVT+6391e0gMtEREFBzh48dg6Tq0fB5WqYTQf2tI4SYGWiIiCozEFz6P0H94BohGEHniCbR8+ctuDyk4a7RERERyPI7Q//zHyPzHz2D4gQfsYqet+hybpon5+Xn7xKhYLIahoSFHx8RAS0REgVIul2EYxoaK4q36HC8sLAAAjh07BkmS7L4MTnJk6vhrX/saTp06hdOnT+P+++/HG2+84cRliYiIdq1UKt3R036zPsemaSKTyaCnp8cOymKfrZMcyWhPnjyJn/zkJ2htbcX09DQeeughPP744zsqpSYiInJSsVjc9IS22j7HqqpCURQsLS0hn89DlmV0d3ejubnZ0fE4ktE+/fTTaG1tBXCrWURvb6/dPIKIiOigmKaJcrm86Sltg4ODOHHiBHp6ejA/Pw/LsqBpGqLRKEZHR9HX14fp6Wnouu7omByvOj5//jwymQweffTRTW8/d+4cBgcH7Q/x7oKIiGi/VFW1j0TdSltbGwqFgj1NnEqlAADxeByRSATlctnRMe0o0D7++OPo7Ozc9KM6c/3www/x+c9/Ht/97neRSCQ2vdbZs2eRTqftD6dTdCIialwi0FYXQhmGsaHISfQ5VhQFiUTCTvhUVYWqqohGo46OaUdrtBcuXLjr11y6dAnPPfccvvWtb+GJJ57Y98CIiIh2Sy0WEbvwM0RjMRjt7VB6erbtczwwMICZmRnMz89DkiT09/c7XhDlSDHU5cuX8eyzz+K1117DM88848QliYiIdsWyLJT//H9D54WfQVEUrPzT/4uO//P/QKSra8vi3EgkgiNHjtR1XI6s0f7hH/4h1tfX8eUvfxkPPvggHnzwQfzrv/6rE5cmIiLaEXN5GdbPfw6zuRlIpWCuZVD56U/dHpYzGe33v/99Jy5DRES0Z1I8DksJQVJVyJEwYAGyg+fK7hV7HRMRUSCYsRjWfu93IVkWpFwesU/+94g+9ZTbw2ILRiIiCgZVVVH+7z6G8oMP4N7RUST7+tweEgBmtEREFBCqqsKyLCAaRayz0+3h2BhoiYgoECqVCgzDQCQSQSjknQlbBloiIgqEYrEIy7LQ1NS0oWGF2xhoiYgoEEqlEizL2rIzoVsYaImIyPcsy0KpVIKiKIjH424PZwMGWiIi8j1VVWEYBhRFQSQScXs4GzDQEhGR71UfGsBAS0RE5DBxak84HHb8UID9YqAlIiLfK5fLMAzDcxXHAAMtEREFQPXWHq9hoCUiIt8rFouQZZmBloiIyGmGYaBSqSAUCnmuEApgoCUiIp8TPY5lWWagJSIicprYQ8uMloiIqA4qlQp0XUc8HvdcxTHAQEtERD5XKpVgmqYnC6EABloiIvK5YrEISZIYaImIiJxmWRbK5TJCoRCi0ajbw9kUAy0REfmWruvQNM2ThwkIDLRERORbqqrCNE0GWiIionpQVRWapiEej0OWvRnSvDkqIiKiHSiXy9B13bOFUAADLRER+VipVAIAJBIJl0eyNQZaIiLyrWKxiHA47NmKY4CBloiIfMo0TZTLZU8XQgEMtERE5FOqqkLXdc/2OBYYaImIyJdEoI3H41AUxe3hbImBloiIfEnTNM9XHAMMtERE5FOlUgmWZXm64hhgoCUiIp8qFotQFMXTFccAAy0REflUsVj0fCEUwEBLREQ+JA4TCIfDDLREREROEz2Oo9EoQqGQ28PZFgMtERH5jtja4/WKY4CBloiIfKhSqcAwDAZaIiKieigWi5BlGfF43O2h3BUDLRER+U6pVPJFIRTAQEtERD5jWRbK5bIvtvYADLREROQzmqZB0zREIhGEw2G3h3NXDLREROQrfqo4BhhoiYjIZ0Sg9XqPY4GBloiIfKVcLgOALyqOAQZaIiLyGb/0OBYYaImIyFcYaImIiOrEMAxUKhXfVBwDDLREROQj1RXHkiS5PZwdYaAlIiLf8NvWHoCBloiIfERVVZimyUBLRERUD6VSCYqi+KYQCmCgJSIiHykUCr6qOAYYaImIyCfEYQKRSISBloiIyGm6rkPTNMTjcd9UHAMMtERE5BN+rDgGGGiJiMgnVFWFYRi+OUxAYKAlIiJfKJfLkCTJV+uzgMOBdnFxET09PXjhhRecvCwRERFKpZLvKo4BhwPtyy+/jOeee87JSxIREQG4dZhAOBxu3ED7zW9+E0eOHMGTTz7p1CWJiIgAAKZpolwuIxaLQZb9terpyGhv3ryJv/mbv8Ff/MVf3PVrz507h8HBQfsjn887MQQiIgowUXHst0IoAAjt5Isef/xxjI2NbXrbBx98gC984Qv4+te/vqPT7s+ePYuzZ8/a/x4cHNzhUImIqFGJimO/be0BdhhoL1y4sOVt6+vr+NWvfoXf/u3fBgDk83kUi0U8/fTTeOedd5wZJRERNTRN0wAA0WjU5ZHs3o4C7XZaW1uxsrJi//vb3/423nrrLbz11lv7vTQRERGAWxXHsiz7rhAK4D5aIiLyAb9WHAN1CLRnzpxhNktERI4qFouIxWJQFMXtoewaM1oiIvI0cZiAHwuhAAZaIiLyOD9v7QEYaImIyONUVYVlWb6sOAYYaImIyOMqlYovDxMQGGiJiMjTisUiQqEQM1oiIqJ6KJVKiEajvqw4BhhoiYjIwyzLQrlc9m3FMcBAS0REHqZpGjRN823FMcBAS0REHiYqjmOxmNtD2TMGWiIi8ixVVQHAtxXHAAMtERF5WLlchqIovq04BhhoiYjIw8RhAqHQvg+bcw0DLREReVaxWPR1IRTAQEtERB5lGAYqlYqvt/YADLRERORRouI4Ho+7PZR9YaAlIiJPUlXV1z2OBQZaIiLyJLG1x88VxwADLREReVSpVEIoFPJ1xTHAQEtERB5VLBbR1NQESZLcHsq+MNASEZHnWJaFUqnk+4pjgIGWiIg8SNd1GIbBQEtERFQPQehxLDDQEhGR54itPX6vOAYYaImIyIMqlQokSUI4HHZ7KPvGQEtERJ5TLBYRj8d9X3EMMNASEZEHBeEwAYGBloiIPMU0TaiqGoiKY4CBloiIPEbTNADBqDgGGGiJiMhjgtLjWGCgJSIiTxFbe4JQcQww0BIRkceUSiVEo1HIcjBCVDB+CiIiCgxxmEBQMNASEZGnBOUwAYGBloiIPEPXdZimGZhCKICBloiIPCRIPY4FBloiIvKMIJ3aIzDQEhGRZ1QqFYRCocBUHAMMtERE5CFBK4QCGGiJiMhDSqVSYA4TEBhoiYjIEyzLgqZpgSqEAhhoiYjII8RhAgy0REREdRDEimOAgZaIiDxCVVUoigJFUdweiqMYaImIyBPK5TJisZjbw3AcAy0REXlC0A4TEBhoiYjIE1RVZUZLRERUD4ZhBO4wAYGBloiIXBfUimOAgZaIiDxAVByHQiG3h+I4BloiInKdpmkIh8NuD6MuGGiJiMh1Qa04BhhoiYjIAyqVSiArjgEGWiIicllQDxMQGGiJiMhVuq4DCGbFMcBAS0RELlNVFZIkBbLiGGCgJSIil6mqimg0CkmS3B5KXTgWaN98802cPn0ap06dwqlTpzAxMeHUpYmIKMAqlcqGaeNcLofx8XGMjY3h+vXrKJVKLo5u/xzJ0z/44AP86Z/+KX7wgx+gv78fuVwucMccERFRfZTLZSSTSQC3WjGm02kcOXIEsVgMhUIB6XQax44dc3mUe+dIoP2rv/ornD17Fv39/QBgP2BERER3U11xLDpEia0+iUQCmqahVCohHo+7Ocw9c2Tq+NKlS5iamsJTTz2Fhx56CH/2Z38GwzA2/dpz585hcHDQ/sjn804MgYiIfMg0Tei6bk8dRyIRGIaBYrEIAMhmszBN0+6F7Ec7CrSPP/44Ojs7N/2Ynp6Gruv44IMP8C//8i/48Y9/jJ/+9Kf4xje+sem1zp49i3Q6bX80Nzc7+gMREZF/aJoGWZbt9ouKomBoaAjz8/MYHx9HPp/3faHUjqaOL1y4sO3tw8PD+NSnPmWn9Z/61Kdw4cIF/MEf/MH+R0hERIGlqirC4fCGQNrc3GwnYaZp4sqVK75uZuHI1PHv/M7v4O2337anAN5++2088MADTlyaiIgCyrIsVNbWEK05TEDTNPv/l5aW0Nzc7OtA60gx1Gc/+1n84he/wMmTJ6EoCp588kn80R/9kROXJiKiADILBaz9L/8rKj/7GaSOdmj/+T8jfN+9AIDFxUUUCgUAQDwex8DAgJtD3TdHAq0sy3j11Vfx6quvOnE5IiIKuOJ3/wsq774Ls7kZyto61v78z9H1X74LAL4PrLWC2e+KiIg8zVhZAWAhEovBkiSYyytuD6lu2IKRiIgOXPw/PAMpEoWZyQClEpr+hxfcHlLdMKMlIqIDF3ngAXT8769Bff//g9LXh+hv/abbQ6obBloiInJF+N57Eb73XreHUXecOiYiIqojBloiIqI6YqAlIiKqIwZaIiKiOmKgJSIiqiMGWiIiojpioCUiIqojBloiIqI6YqAlIiKqIwZaIiKiOpIsy7LcHEA0GkVXV5ebQ/CFfD6P5uZmt4fRcPi4Hzw+5u7g4753S0tLqFQqW97ueqClnRkcHEQ6nXZ7GA2Hj/vB42PuDj7u9cOpYyIiojpioCUiIqojBlqfOHv2rNtDaEh83A8eH3N38HGvH67REhER1REzWiIiojpioCUiIqojBloiIqI6YqD1iddffx2SJOGtt95yeyiBVy6X8cILL+D48eN44IEH8Mwzz2B8fNztYTWEsbEx/Pqv/zqOHz+ORx99FBcvXnR7SA3la1/7Gk6dOoXTp0/j/vvvxxtvvOH2kAKBgdYHJiYm8Hd/93d47LHH3B5Kw3jppZdw9epV/PKXv8Tzzz+PF1980e0hNYSXX34ZL730Eq5du4Yvf/nLOHPmjNtDaignT57ET37yE3z44Yf4p3/6J3zpS1/C9evX3R6W7zHQepxpmnjxxRfx13/914hGo24PpyHEYjE8++yzkCQJAPDYY49hYmLC3UE1gMXFRbz//vv43Oc+BwD49Kc/jenpac4mHKCnn34ara2tAIChoSH09vZienra5VH5HwOtx507dw6/8Ru/gUceecTtoTSsr371q3j++efdHkbgTU9Po6+vD6FQCAAgSRKGh4cxNTXl8sga0/nz55HJZPDoo4+6PRTfC7k9ANraRx99hDfffBM/+tGP3B5Kw3rllVcwPj6Od955x+2hEO3b448/jrGxsU1v++CDDzA0NAQA+PDDD/H5z38e3/3ud5FIJA5yiIHEQOth//Zv/4aJiQkcO3YMADA/P4+XXnoJc3Nz+OIXv+jy6ILv1Vdfxfe+9z2cP38eTU1Nbg8n8IaGhjA3Nwdd1xEKhWBZFqampjA8POz20ALjwoULd/2aS5cu4bnnnsO3vvUtPPHEEwcwquDj1LGHffGLX8Tc3BwmJiYwMTGBxx57DK+99hqD7AE4d+4cvvOd7+D73/8+UqmU28NpCN3d3Xj44YftStc333wTg4ODGB0ddXlkjePy5ct49tln8dprr+GZZ55xeziBwRaMPvLxj38cX/rSl/DCCy+4PZRAS6fTGBoawsjICJLJJIBb5yb//Oc/d3lkwXf16lWcOXMGKysraGlpweuvv47Tp0+7PayG8cwzz+D999/HoUOH7M/95V/+JT7xiU+4OCr/Y6AlIiKqI04dExER1REDLRERUR0x0BIREdURAy0REVEdMdASERHVEQMtERFRHTHQEhER1REDLRERUR39/90K4Vig4bE2AAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}