UCI_mushroom_dataset_project.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "eK3SyK8TTqn5"
      },
      "outputs": [],
      "source": [
        "from sklearn.preprocessing import OneHotEncoder\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.feature_selection import SelectKBest, chi2\n",
        "from sklearn.feature_selection import f_classif\n",
        "from sklearn.feature_selection import RFECV\n",
        "from sklearn.model_selection import KFold\n",
        "from sklearn.metrics import f1_score\n",
        "\n",
        "\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.linear_model import SGDClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.linear_model import Perceptron\n",
        "from sklearn.svm import NuSVC\n",
        "from sklearn.neural_network import MLPClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "\n",
        "\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from re import X\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ddvKv9L9TzMB",
        "outputId": "00cbf5ec-087f-4128-abf7-359df1af3658"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Initial exploration of the dataset\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "PEdK2IdqT7GS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_train = pd.read_csv(\"/content/drive/MyDrive/EE559/Project/Mushroom_datasets/mushroom_train.csv\")\n",
        "df_test = pd.read_csv(\"/content/drive/MyDrive/EE559/Project/Mushroom_datasets/mushroom_test.csv\")"
      ],
      "metadata": {
        "id": "lD9H-SyeT4T8"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_df = pd.read_csv(\"/content/mushroom_train.csv\")\n",
        "test_df = pd.read_csv(\"/content/mushroom_test.csv\")"
      ],
      "metadata": {
        "id": "qXabxmqvUEtx"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_train.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 268
        },
        "id": "HQKiOWNiUHC3",
        "outputId": "cf63d1d9-159b-4732-9901-be8e2963706f"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   cap-diameter cap-shape cap-surface cap-color does-bruise-or-bleed  \\\n",
              "0          4.98         c           i         y                    f   \n",
              "1          2.84         x           y         y                    f   \n",
              "2         11.44         x           y         y                    f   \n",
              "3          8.77         s           t         r                    t   \n",
              "4          7.55         x           d         n                    t   \n",
              "\n",
              "  gill-attachment gill-spacing gill-color  stem-height  stem-width stem-color  \\\n",
              "0               a            c          n         6.04        6.21          w   \n",
              "1               a            c          w         5.66        3.55          y   \n",
              "2               a            c          w         7.03       25.29          n   \n",
              "3               d            c          g         4.44       13.61          r   \n",
              "4               p            c          y         8.41       18.44          y   \n",
              "\n",
              "  has-ring ring-type habitat season class  \n",
              "0        f         f       d      a     p  \n",
              "1        t         r       h      u     p  \n",
              "2        t         e       d      w     e  \n",
              "3        f         f       d      a     p  \n",
              "4        f         f       d      a     e  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-807d90a5-d16e-4433-978b-62343b48be3b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>cap-diameter</th>\n",
              "      <th>cap-shape</th>\n",
              "      <th>cap-surface</th>\n",
              "      <th>cap-color</th>\n",
              "      <th>does-bruise-or-bleed</th>\n",
              "      <th>gill-attachment</th>\n",
              "      <th>gill-spacing</th>\n",
              "      <th>gill-color</th>\n",
              "      <th>stem-height</th>\n",
              "      <th>stem-width</th>\n",
              "      <th>stem-color</th>\n",
              "      <th>has-ring</th>\n",
              "      <th>ring-type</th>\n",
              "      <th>habitat</th>\n",
              "      <th>season</th>\n",
              "      <th>class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4.98</td>\n",
              "      <td>c</td>\n",
              "      <td>i</td>\n",
              "      <td>y</td>\n",
              "      <td>f</td>\n",
              "      <td>a</td>\n",
              "      <td>c</td>\n",
              "      <td>n</td>\n",
              "      <td>6.04</td>\n",
              "      <td>6.21</td>\n",
              "      <td>w</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>d</td>\n",
              "      <td>a</td>\n",
              "      <td>p</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2.84</td>\n",
              "      <td>x</td>\n",
              "      <td>y</td>\n",
              "      <td>y</td>\n",
              "      <td>f</td>\n",
              "      <td>a</td>\n",
              "      <td>c</td>\n",
              "      <td>w</td>\n",
              "      <td>5.66</td>\n",
              "      <td>3.55</td>\n",
              "      <td>y</td>\n",
              "      <td>t</td>\n",
              "      <td>r</td>\n",
              "      <td>h</td>\n",
              "      <td>u</td>\n",
              "      <td>p</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>11.44</td>\n",
              "      <td>x</td>\n",
              "      <td>y</td>\n",
              "      <td>y</td>\n",
              "      <td>f</td>\n",
              "      <td>a</td>\n",
              "      <td>c</td>\n",
              "      <td>w</td>\n",
              "      <td>7.03</td>\n",
              "      <td>25.29</td>\n",
              "      <td>n</td>\n",
              "      <td>t</td>\n",
              "      <td>e</td>\n",
              "      <td>d</td>\n",
              "      <td>w</td>\n",
              "      <td>e</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>8.77</td>\n",
              "      <td>s</td>\n",
              "      <td>t</td>\n",
              "      <td>r</td>\n",
              "      <td>t</td>\n",
              "      <td>d</td>\n",
              "      <td>c</td>\n",
              "      <td>g</td>\n",
              "      <td>4.44</td>\n",
              "      <td>13.61</td>\n",
              "      <td>r</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>d</td>\n",
              "      <td>a</td>\n",
              "      <td>p</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>7.55</td>\n",
              "      <td>x</td>\n",
              "      <td>d</td>\n",
              "      <td>n</td>\n",
              "      <td>t</td>\n",
              "      <td>p</td>\n",
              "      <td>c</td>\n",
              "      <td>y</td>\n",
              "      <td>8.41</td>\n",
              "      <td>18.44</td>\n",
              "      <td>y</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>d</td>\n",
              "      <td>a</td>\n",
              "      <td>e</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-807d90a5-d16e-4433-978b-62343b48be3b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-807d90a5-d16e-4433-978b-62343b48be3b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-807d90a5-d16e-4433-978b-62343b48be3b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_train.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "jFMGdzH5ULxn",
        "outputId": "47b20f89-b338-403d-e6e7-ca3766539599"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       cap-diameter   stem-height    stem-width\n",
              "count  42748.000000  42748.000000  42748.000000\n",
              "mean       6.714149      6.583224     12.117692\n",
              "std        5.220008      3.368333     10.004874\n",
              "min        0.380000      0.000000      0.000000\n",
              "25%        3.480000      4.640000      5.180000\n",
              "50%        5.865000      5.960000     10.200000\n",
              "75%        8.530000      7.750000     16.540000\n",
              "max       62.340000     33.920000    103.910000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1174bc56-3bf4-474e-8ca0-c33305b19df9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>cap-diameter</th>\n",
              "      <th>stem-height</th>\n",
              "      <th>stem-width</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>42748.000000</td>\n",
              "      <td>42748.000000</td>\n",
              "      <td>42748.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>6.714149</td>\n",
              "      <td>6.583224</td>\n",
              "      <td>12.117692</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>5.220008</td>\n",
              "      <td>3.368333</td>\n",
              "      <td>10.004874</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.380000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.480000</td>\n",
              "      <td>4.640000</td>\n",
              "      <td>5.180000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>5.865000</td>\n",
              "      <td>5.960000</td>\n",
              "      <td>10.200000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>8.530000</td>\n",
              "      <td>7.750000</td>\n",
              "      <td>16.540000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>62.340000</td>\n",
              "      <td>33.920000</td>\n",
              "      <td>103.910000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1174bc56-3bf4-474e-8ca0-c33305b19df9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1174bc56-3bf4-474e-8ca0-c33305b19df9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1174bc56-3bf4-474e-8ca0-c33305b19df9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_test.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "QlcHfl1SGWct",
        "outputId": "dd43d189-1f81-48c6-ca60-b6a13fa8183d"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       cap-diameter   stem-height    stem-width\n",
              "count  18321.000000  18321.000000  18321.000000\n",
              "mean       6.779834      6.577604     12.223417\n",
              "std        5.367869      3.374032     10.107991\n",
              "min        0.440000      0.000000      0.000000\n",
              "25%        3.490000      4.630000      5.260000\n",
              "50%        5.860000      5.940000     10.160000\n",
              "75%        8.580000      7.720000     16.660000\n",
              "max       58.300000     33.720000    102.480000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1156b6c1-e8dc-458f-ad58-e0a6d1766160\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>cap-diameter</th>\n",
              "      <th>stem-height</th>\n",
              "      <th>stem-width</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>18321.000000</td>\n",
              "      <td>18321.000000</td>\n",
              "      <td>18321.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>6.779834</td>\n",
              "      <td>6.577604</td>\n",
              "      <td>12.223417</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>5.367869</td>\n",
              "      <td>3.374032</td>\n",
              "      <td>10.107991</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.440000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.490000</td>\n",
              "      <td>4.630000</td>\n",
              "      <td>5.260000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>5.860000</td>\n",
              "      <td>5.940000</td>\n",
              "      <td>10.160000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>8.580000</td>\n",
              "      <td>7.720000</td>\n",
              "      <td>16.660000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>58.300000</td>\n",
              "      <td>33.720000</td>\n",
              "      <td>102.480000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1156b6c1-e8dc-458f-ad58-e0a6d1766160')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1156b6c1-e8dc-458f-ad58-e0a6d1766160 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1156b6c1-e8dc-458f-ad58-e0a6d1766160');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cols = list(df_train.columns)\n",
        "cols.remove('cap-diameter')\n",
        "cols.remove('stem-height')\n",
        "cols.remove('stem-width')\n",
        "for col in cols:\n",
        "  print(f\"Unique values and value counts for {col}:\")\n",
        "  print(df_train[col].value_counts())\n",
        "  print(\"-\"*100)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jkk6e706UQbC",
        "outputId": "8364e5ab-5adb-471d-b225-cc73a7eab2b8"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Unique values and value counts for cap-shape:\n",
            "x    18808\n",
            "f     9289\n",
            "s     5079\n",
            "b     4043\n",
            "o     2417\n",
            "p     1824\n",
            "c     1288\n",
            "Name: cap-shape, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for cap-surface:\n",
            "t    15569\n",
            "s     5341\n",
            "y     4460\n",
            "h     3482\n",
            "g     3280\n",
            "d     3125\n",
            "e     1789\n",
            "k     1602\n",
            "i     1599\n",
            "w     1494\n",
            "l     1007\n",
            "Name: cap-surface, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for cap-color:\n",
            "n    16960\n",
            "y     6026\n",
            "w     5402\n",
            "g     3072\n",
            "e     2781\n",
            "o     2581\n",
            "r     1227\n",
            "u     1181\n",
            "p     1161\n",
            "k      907\n",
            "b      870\n",
            "l      580\n",
            "Name: cap-color, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for does-bruise-or-bleed:\n",
            "f    35279\n",
            "t     7469\n",
            "Name: does-bruise-or-bleed, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for gill-attachment:\n",
            "a    15844\n",
            "d     7133\n",
            "x     5174\n",
            "p     4185\n",
            "e     3998\n",
            "s     3913\n",
            "f     2501\n",
            "Name: gill-attachment, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for gill-spacing:\n",
            "c    34791\n",
            "d     5456\n",
            "f     2501\n",
            "Name: gill-spacing, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for gill-color:\n",
            "w    12893\n",
            "y     6743\n",
            "n     6742\n",
            "p     4178\n",
            "g     2860\n",
            "f     2501\n",
            "o     2041\n",
            "k     1697\n",
            "r      962\n",
            "e      756\n",
            "u      713\n",
            "b      662\n",
            "Name: gill-color, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for stem-color:\n",
            "w    15996\n",
            "n    12678\n",
            "y     5547\n",
            "g     1841\n",
            "o     1512\n",
            "e     1431\n",
            "u     1031\n",
            "f      745\n",
            "p      710\n",
            "k      599\n",
            "r      383\n",
            "l      150\n",
            "b      125\n",
            "Name: stem-color, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for has-ring:\n",
            "f    32085\n",
            "t    10663\n",
            "Name: has-ring, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for ring-type:\n",
            "f    35581\n",
            "e     1703\n",
            "z     1418\n",
            "r     1013\n",
            "l      998\n",
            "p      907\n",
            "g      888\n",
            "m      240\n",
            "Name: ring-type, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for habitat:\n",
            "d    30945\n",
            "g     5486\n",
            "l     2244\n",
            "m     2041\n",
            "h     1444\n",
            "w      259\n",
            "p      245\n",
            "u       84\n",
            "Name: habitat, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for season:\n",
            "a    21053\n",
            "u    16065\n",
            "w     3669\n",
            "s     1961\n",
            "Name: season, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n",
            "Unique values and value counts for class:\n",
            "p    23595\n",
            "e    19153\n",
            "Name: class, dtype: int64\n",
            "----------------------------------------------------------------------------------------------------\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                            'does-bruise-or-bleed', 'gill-attachment', \n",
        "                            'gill-spacing', 'gill-color','stem-color', \n",
        "                            'has-ring', 'ring-type', 'habitat', 'season']\n",
        "\n",
        "cols_vals = {}\n",
        "for col in categorical_cols_names:\n",
        "  print(f\"Number of unique values in {col}: {len(df_train[col].unique())}\")\n",
        "  cols_vals[col] = len(df_train[col].unique())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WBArGRWRVG1o",
        "outputId": "6fb463ce-5edc-4ce9-b26f-0b53a7a69578"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of unique values in cap-shape: 7\n",
            "Number of unique values in cap-surface: 11\n",
            "Number of unique values in cap-color: 12\n",
            "Number of unique values in does-bruise-or-bleed: 2\n",
            "Number of unique values in gill-attachment: 7\n",
            "Number of unique values in gill-spacing: 3\n",
            "Number of unique values in gill-color: 12\n",
            "Number of unique values in stem-color: 13\n",
            "Number of unique values in has-ring: 2\n",
            "Number of unique values in ring-type: 8\n",
            "Number of unique values in habitat: 8\n",
            "Number of unique values in season: 4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Functions\n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "D4_uATWIUauv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def onehotencode(training_data):\n",
        "\n",
        "  cat_cols = ['cap-shape','cap-surface','cap-color','does-bruise-or-bleed','gill-attachment','gill-spacing','gill-color','stem-color','has-ring','ring-type','habitat','season']\n",
        "  num_cols = ['cap-diameter','stem-height','stem-width','class']\n",
        "\n",
        "  train_data = training_data[cat_cols]\n",
        "  one_hot_encoded_data = pd.get_dummies(train_data)\n",
        "  train_num_data = training_data[num_cols]\n",
        "  df_train = pd.concat([one_hot_encoded_data, train_num_data], axis=1)\n",
        "\n",
        "  return df_train"
      ],
      "metadata": {
        "id": "JsQ1q_HKUUoO"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def labels(data_train_labels):\n",
        "\n",
        "  train_labels = np.zeros(data_train_labels.shape)\n",
        "  for j in range(data_train_labels.shape[0]):\n",
        "    if data_train_labels[j] == 'e':\n",
        "      train_labels[j] = 1\n",
        "    else:\n",
        "      train_labels[j] = 0\n",
        "  \n",
        "  return train_labels"
      ],
      "metadata": {
        "id": "z9_tGPeQUfxX"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def dataset(training_data, testing_data):\n",
        "\n",
        "  df_train = onehotencode(training_data)\n",
        "  df_test = onehotencode(testing_data)\n",
        "\n",
        "  data_train_values = df_train.values\n",
        "  data_test_values = df_test.values\n",
        "\n",
        "  data_train_labels = data_train_values[:,-1]\n",
        "  data_test_labels = data_test_values[:,-1]\n",
        "\n",
        "  train_labels = labels(data_train_labels)\n",
        "  test_labels = labels(data_test_labels)\n",
        "\n",
        "  return data_train_values, data_test_values, train_labels, test_labels"
      ],
      "metadata": {
        "id": "21sSL3TJUikc"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def one_hot_encoder(df_col, encoder):\n",
        "  \n",
        "  col = df_col.to_numpy()\n",
        "  col = col.reshape(-1, 1)\n",
        "  col_enc = encoder.fit_transform(col)\n",
        "  col_arr = col_enc.toarray()\n",
        "\n",
        "  return col_arr"
      ],
      "metadata": {
        "id": "qVMcliMaUlM3"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def catTonum(df, cat_cols):\n",
        "  encoder = OneHotEncoder()\n",
        "  temp = []\n",
        "  for col in cat_cols:\n",
        "    temp.append(one_hot_encoder(df[col], encoder))\n",
        "  \n",
        "  data = np.hstack((temp[0], temp[1], temp[2], temp[3], temp[4], temp[5], temp[6], temp[7],\n",
        "                  temp[8], temp[9], temp[10], temp[11]))\n",
        "\n",
        "  return data\n"
      ],
      "metadata": {
        "id": "3DGy0ZYcUnpb"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def pre_processing(df_train, df_test):\n",
        "  train_labels = df_train['class'].apply(lambda x: 1 if x == 'e' else 0)\n",
        "  test_labels = df_test['class'].apply(lambda x: 1 if x == 'e' else 0)\n",
        "\n",
        "  numeric_cols_train = df_train[['cap-diameter', 'stem-height', 'stem-width']]\n",
        "  numeric_cols_test = df_test[['cap-diameter', 'stem-height', 'stem-width']]\n",
        "\n",
        "  categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                            'does-bruise-or-bleed', 'gill-attachment', \n",
        "                            'gill-spacing', 'gill-color','stem-color', \n",
        "                            'has-ring', 'ring-type', 'habitat', 'season']\n",
        "  \n",
        "  train_data = catTonum(df_train, categorical_cols_names)\n",
        "  test_data = catTonum(df_test, categorical_cols_names)\n",
        "\n",
        "  train_data = np.hstack((train_data, numeric_cols_train))\n",
        "  test_data = np.hstack((test_data, numeric_cols_test))\n",
        "\n",
        "  return train_data, train_labels, test_data, test_labels\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "WBGUwqdVUqDH"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def addColumns1(df, groupby_col, numeric_col, stat):\n",
        "\n",
        "  categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                            'does-bruise-or-bleed', 'gill-attachment', \n",
        "                            'gill-spacing', 'gill-color','stem-color', \n",
        "                            'has-ring', 'ring-type', 'habitat', 'season']\n",
        "\n",
        "  temp_df = df.copy()\n",
        "  mean_df = temp_df.groupby([groupby_col])[numeric_col].agg(stat).reset_index()\n",
        "\n",
        "  for index, row in mean_df.iterrows():\n",
        "    temp_df[f'{index}_{stat}_{groupby_col}_{numeric_col}'] = 0\n",
        "    temp_df.loc[(temp_df[groupby_col] == mean_df[groupby_col][index]), f'{index}_{stat}_{groupby_col}_{numeric_col}'] = mean_df[numeric_col][index]\n",
        "\n",
        "  temp_df = temp_df.drop(columns=categorical_cols_names)\n",
        "  train_labels = temp_df['class'].apply(lambda x: 1 if x == 'e' else 0)\n",
        "  temp_df['labels'] = train_labels\n",
        "  temp_df = temp_df.drop(columns=['class'])\n",
        "  \n",
        "  return temp_df\n",
        "\n"
      ],
      "metadata": {
        "id": "XOT1M7pPVgJW"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def addColumns2(df):\n",
        "\n",
        "  categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                            'does-bruise-or-bleed', 'gill-attachment', \n",
        "                            'gill-spacing', 'gill-color','stem-color', \n",
        "                            'has-ring', 'ring-type', 'habitat', 'season']\n",
        "\n",
        "  temp_df = df.copy()\n",
        "  mean_df = temp_df.groupby(['stem-color', 'gill-attachment'])['stem-height'].agg('median').reset_index()\n",
        "\n",
        "  for index, row in mean_df.iterrows():\n",
        "    res = temp_df.loc[(temp_df['stem-color'] == mean_df['stem-color'][index]) & (df['gill-attachment'] == mean_df['gill-attachment'][index])]\n",
        "    value = mean_df['stem-height'][index]\n",
        "    res[f'avg_cd_sccs{index}'] = value\n",
        "    temp_df[f'avg_cd_sccs{index}'] = 0\n",
        "    temp_df.loc[(temp_df['stem-color'] == mean_df['stem-color'][index]) & (temp_df['gill-attachment'] == mean_df['gill-attachment'][index])] = res\n",
        "\n",
        "  temp_df = temp_df.drop(columns=categorical_cols_names)\n",
        "  train_labels = temp_df['class'].apply(lambda x: 1 if x == 'e' else 0)\n",
        "  temp_df['labels'] = train_labels\n",
        "  temp_df = temp_df.drop(columns=['class'])\n",
        "  \n",
        "  return temp_df\n",
        "\n"
      ],
      "metadata": {
        "id": "ZC6lXG4PViws"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def genParams():\n",
        "  categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                              'does-bruise-or-bleed', 'gill-attachment', \n",
        "                              'gill-spacing', 'gill-color','stem-color', \n",
        "                              'has-ring', 'ring-type', 'habitat', 'season']\n",
        "\n",
        "  numeric_cols_names = ['cap-diameter', 'stem-height', 'stem-width']\n",
        "\n",
        "  statistics = ['mean', 'max', 'min', 'median']\n",
        "\n",
        "  params = []\n",
        "  for i in categorical_cols_names:\n",
        "    for j in numeric_cols_names:\n",
        "      for k in statistics:\n",
        "        params.append([i, j, k])\n",
        "  \n",
        "  return params\n"
      ],
      "metadata": {
        "id": "uECR-V-iV9u7"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def decodeIndexes(indexes, df):\n",
        "  categorical_cols_names = ['cap-shape', 'cap-surface', 'cap-color', \n",
        "                            'does-bruise-or-bleed', 'gill-attachment', \n",
        "                            'gill-spacing', 'gill-color','stem-color', \n",
        "                            'has-ring', 'ring-type', 'habitat', 'season']\n",
        "\n",
        "  unique_col_vals = []\n",
        "  for col in categorical_cols_names:\n",
        "    unique_col_vals.append(df_train[col].unique().tolist())\n",
        "\n",
        "  flattened_list = [item for sublist in unique_col_vals for item in (sublist if isinstance(sublist, list) else [sublist])]\n",
        "\n",
        "  cols_vals = {}\n",
        "  for col in categorical_cols_names:\n",
        "    cols_vals[col] = len(df_train[col].unique())\n",
        "  \n",
        "  feats = []\n",
        "  for index in indexes:\n",
        "    if index == 89:\n",
        "      feats.append(\"cap-diameter\")\n",
        "    elif index == 90:\n",
        "      feats.append(\"stem-height\")\n",
        "    elif index == 91:\n",
        "      feats.append(\"stem-width\")\n",
        "    else:\n",
        "      count = 0\n",
        "      for key, value in cols_vals.items():\n",
        "        count += value\n",
        "        if count >= index:\n",
        "          feats.append(key)\n",
        "          break\n",
        "\n",
        "  features_map = {}\n",
        "  for i in range(len(indexes)):\n",
        "    if indexes[i] in [89, 90, 91]:\n",
        "      features_map[i] = [feats[i], 0]\n",
        "    else:\n",
        "      features_map[i] = [feats[i], flattened_list[indexes[i]]]\n",
        "\n",
        "  return features_map"
      ],
      "metadata": {
        "id": "dEwCnxBn4RWW"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def top3feats_of_poison_mushrooms(test_data, pred_val, original_indexes, df_train):\n",
        "\n",
        "  cat_feat_map = {}\n",
        "  for i in range(len(pred_val)):\n",
        "    if pred_val[i] == 0:\n",
        "      key = tuple(test_data[i])\n",
        "      if key not in cat_feat_map:\n",
        "        cat_feat_map[key] = 0\n",
        "      else:\n",
        "        cat_feat_map[key]+=1\n",
        "  \n",
        "  sorted_dict = dict(sorted(cat_feat_map.items(), key=lambda item: item[1], reverse=True))\n",
        "  top_3 = list(sorted_dict.items())[:3]\n",
        "\n",
        "  cat_inds = [i for i in original_indexes if i not in [89, 90, 91]]\n",
        "\n",
        "  for i in range(3):\n",
        "    temp_val = list(top_3[i][0])\n",
        "    temp_ind = []\n",
        "    for j in range(len(temp_val)):\n",
        "      if temp_val[j] == 1:\n",
        "        temp_ind.append(cat_inds[j])\n",
        "    \n",
        "    feats = decodeIndexes(temp_ind, df_train)\n",
        "    print(feats)\n"
      ],
      "metadata": {
        "id": "ucLeO6404ZHF"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Trivial system\n",
        "class trivialSytem2Class():\n",
        "\n",
        "  def __init__(self):\n",
        "    self.ps0 = 0\n",
        "    self.ps1 = 0\n",
        "\n",
        "  def fit(self, X, Y):\n",
        "    class0_count = len(Y[Y==0])\n",
        "    class1_count = len(Y[Y==1])\n",
        "\n",
        "    self.ps0 = class0_count / len(Y)\n",
        "    self.ps1 = class1_count / len(Y)\n",
        "  \n",
        "  def predict(self, X):\n",
        "    predictions = np.random.choice([0, 1], size=len(X), p=[self.ps0, self.ps1])\n",
        "    return predictions\n",
        "\n",
        "  def get_params(self):\n",
        "    return self.ps0, self.ps1    "
      ],
      "metadata": {
        "id": "MIZGpJ01U882"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def feature_selection_ufs(classifier, traindata, trainlabels, testdata, testlabels):\n",
        "\n",
        "  feat_num = traindata.shape[1]\n",
        "  accuracy = np.zeros(feat_num)\n",
        "\n",
        "  for i in range(1, feat_num, 1):\n",
        "    fs = SelectKBest(score_func=chi2, k=i)\n",
        "    X_selected = fs.fit_transform(Xtrain, Ytrain)\n",
        "\n",
        "    imp_ind = np.argsort(-fs.scores_)\n",
        "\n",
        "    tempTrain = traindata[:,np.sort(imp_ind[:i])] \n",
        "    tempTest = testdata[:,np.sort(imp_ind[:i])]\n",
        "\n",
        "    classifier.fit(tempTrain, trainlabels)\n",
        "    preds = classifier.predict(tempTest)\n",
        "    acc = accuracy_score(testlabels, preds)\n",
        "    accuracy[i] = acc\n",
        "  \n",
        "\n",
        "  return accuracy\n"
      ],
      "metadata": {
        "id": "bE4am5zcbTn4"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def radialSVC(Xtrain, Ytrain, Xtest, Ytest):\n",
        "\n",
        "  num_folds = 5\n",
        "  kf = KFold(n_splits=num_folds)\n",
        "  gamma = [0.01, 0.1, 1, 10]\n",
        "\n",
        "  accuracy_fold = []\n",
        "  accuracy_run = []\n",
        "  accuracy_test = []\n",
        "  f1_list = []\n",
        "\n",
        "  for g in gamma:\n",
        "\n",
        "    clf = NuSVC(gamma = g)\n",
        "    accuracy_scores = []\n",
        "\n",
        "    for fold, (train_indices, val_indices) in enumerate(kf.split(Xtrain)):\n",
        "\n",
        "        Xtrain_fold = Xtrain[train_indices]\n",
        "        Ytrain_fold = Ytrain[train_indices]\n",
        "        Xval_fold = Xtrain[val_indices]\n",
        "        Yval_fold = Ytrain[val_indices]\n",
        "\n",
        "        \n",
        "        clf.fit(Xtrain_fold, Ytrain_fold)\n",
        "\n",
        "        \n",
        "        pred_val = clf.predict(Xval_fold)\n",
        "        accuracy = accuracy_score(Yval_fold, pred_val)\n",
        "        accuracy_scores.append(accuracy)\n",
        "    accuracy_fold.append(accuracy_scores)\n",
        "    accuracy_run.append(np.array(np.mean(accuracy_scores)))\n",
        "\n",
        "    preds = clf.predict(Xtest) \n",
        "    acc = accuracy_score(Ytest, preds)\n",
        "    accuracy_test.append(acc)\n",
        "    f1 = f1_score(Ytest, preds)\n",
        "    f1_list.append(f1)\n",
        "\n",
        "  print(f\"Best Accuracy of RBF Kernel: {max(accuracy_test)*100:.3f}% for gamma = {gamma[np.argmax(accuracy_test)]}\")\n",
        "  print(f\"F1 Score of RBF is:{f1_list[np.argmax(accuracy_test)]}\")\n",
        "\n",
        "  return gamma, accuracy_fold, accuracy_run, accuracy_test"
      ],
      "metadata": {
        "id": "LGp4Nnm_XXmf"
      },
      "execution_count": 91,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Testing K Neighbors Classifier\n",
        "\n",
        "def knearest(Xtrain, Ytrain, Xtest, Ytest):\n",
        "  \n",
        "  num_folds = 5\n",
        "  kf = KFold(n_splits=num_folds)\n",
        "  neighbors = [1,3,50,100]\n",
        "\n",
        "  accuracy_fold = []\n",
        "  accuracy_run = []\n",
        "  accuracy_test = []\n",
        "  f1_list = []\n",
        "\n",
        "  for n in neighbors:\n",
        "\n",
        "    neigh = KNeighborsClassifier(n_neighbors=n)\n",
        "    accuracy_scores = []\n",
        "\n",
        "    for fold, (train_indices, val_indices) in enumerate(kf.split(Xtrain)):\n",
        "\n",
        "        Xtrain_fold = Xtrain[train_indices]\n",
        "        Ytrain_fold = Ytrain[train_indices]\n",
        "        Xval_fold = Xtrain[val_indices]\n",
        "        Yval_fold = Ytrain[val_indices]\n",
        "\n",
        "        \n",
        "        neigh.fit(Xtrain_fold, Ytrain_fold)\n",
        "\n",
        "        \n",
        "        pred_val = neigh.predict(Xval_fold)\n",
        "        accuracy = accuracy_score(pred_val, Yval_fold)\n",
        "        accuracy_scores.append(accuracy)\n",
        "    \n",
        "    accuracy_fold.append(accuracy_scores)\n",
        "    accuracy_run.append(np.array(np.mean(accuracy_scores)))\n",
        "\n",
        "    preds = neigh.predict(Xtest) \n",
        "    acc = accuracy_score(Ytest, preds)\n",
        "    accuracy_test.append(acc)\n",
        "    f1 = f1_score(Ytest, preds)\n",
        "    f1_list.append(f1)\n",
        "\n",
        "  print(f\"Best Accuracy of KNN Classifier: {max(accuracy_test)*100:.3f}% for number of neighbors = {neighbors[np.argmax(accuracy_test)]}\")\n",
        "  print(f\"F1 Score of KNN is:{f1_list[np.argmax(accuracy_test)]}\")\n",
        "\n",
        "  return neighbors, accuracy_fold, accuracy_run, accuracy_test"
      ],
      "metadata": {
        "id": "zSKbLrCkXeSt"
      },
      "execution_count": 92,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def mlp_classifier(Xtrain,Ytrain,Xtest,Ytest,optim):\n",
        "  lr = [0.0001,0.001,0.01,0.1]\n",
        "  accuracy_run = []\n",
        "  accuracy_test = []\n",
        "  accuracy_fold = []\n",
        "  f1_list = []\n",
        "\n",
        "  for j in lr:\n",
        "\n",
        "    mlp = MLPClassifier(hidden_layer_sizes = (20,5), activation = 'relu', solver = optim ,max_iter = 1000, learning_rate = 'constant', learning_rate_init = j)\n",
        "    num_folds = 5\n",
        "    kf = KFold(n_splits=num_folds)\n",
        "    accuracy_scores = []\n",
        "\n",
        "\n",
        "    for fold, (train_indices, val_indices) in enumerate(kf.split(Xtrain)):\n",
        "\n",
        "        Xtrain_fold = Xtrain[train_indices]\n",
        "        Ytrain_fold = Ytrain[train_indices]\n",
        "        Xval_fold = Xtrain[val_indices]\n",
        "        Yval_fold = Ytrain[val_indices]\n",
        "\n",
        "        \n",
        "        mlp.fit(Xtrain_fold, Ytrain_fold)\n",
        "\n",
        "        \n",
        "        pred_val = mlp.predict(Xval_fold)\n",
        "        accuracy = accuracy_score(Yval_fold, pred_val)\n",
        "        accuracy_scores.append(accuracy)\n",
        "\n",
        "\n",
        "    mean_accuracy = np.mean(accuracy_scores)\n",
        "    accuracy_run.append(mean_accuracy)\n",
        "    accuracy_fold.append(accuracy_scores)\n",
        "\n",
        "    pred_test = mlp.predict(Xtest)\n",
        "\n",
        "\n",
        "    accuracy = accuracy_score(Ytest, pred_test)\n",
        "    accuracy_test.append(accuracy)\n",
        "\n",
        "    f1_list.append(f1_score(Ytest, pred_test))\n",
        "\n",
        "  return lr, accuracy_run, accuracy_test, accuracy_fold, f1_list\n"
      ],
      "metadata": {
        "id": "HdJY_77AXshv"
      },
      "execution_count": 93,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def random_forest(Xtrain, Ytrain, Xtest, Ytest):\n",
        "\n",
        "  n_estim = [10,50,100,1000]\n",
        "  accuracy_run = []\n",
        "  accuracy_test = []\n",
        "  accuracy_fold = []\n",
        "  f1_list = []\n",
        "\n",
        "  for n in n_estim:\n",
        "      \n",
        "    rfc = RandomForestClassifier(n_estimators= n, random_state=42)\n",
        "    num_folds = 5\n",
        "    kf = KFold(n_splits=num_folds)\n",
        "    accuracy_scores = []\n",
        "\n",
        "\n",
        "    for fold, (train_indices, val_indices) in enumerate(kf.split(Xtrain)):\n",
        "\n",
        "        Xtrain_fold = Xtrain[train_indices]\n",
        "        Ytrain_fold = Ytrain[train_indices]\n",
        "        Xval_fold = Xtrain[val_indices]\n",
        "        Yval_fold = Ytrain[val_indices]\n",
        "\n",
        "        rfc.fit(Xtrain_fold,Ytrain_fold)\n",
        "        preds = rfc.predict(Xval_fold)\n",
        "        accuracy = accuracy_score(preds, Yval_fold)\n",
        "        accuracy_scores.append(accuracy)\n",
        "    \n",
        "    accuracy_run.append(np.mean(np.array(accuracy_scores)))\n",
        "    accuracy_fold.append(accuracy_scores)\n",
        "\n",
        "    y_pred = rfc.predict(Xtest)\n",
        "    acc_test = accuracy_score(Ytest, y_pred)\n",
        "    accuracy_test.append(acc_test)\n",
        "\n",
        "    f1 = f1_score(Ytest, y_pred, average = 'binary')\n",
        "    f1_list.append(f1)\n",
        "\n",
        "  return n_estim, accuracy_run, accuracy_fold, accuracy_test, f1_list"
      ],
      "metadata": {
        "id": "LDbsPKGlZMGn"
      },
      "execution_count": 90,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_fin(l1,accuracy_run,accuracy_test,accuracy_fold):\n",
        "  plt.plot(l1,accuracy_run, label = 'Training Accuracy')\n",
        "  plt.plot(l1,accuracy_test,label = 'Testing Accuracy')\n",
        "  plt.xlabel('Hyperparameter')\n",
        "  plt.ylabel('Accuracies')\n",
        "  plt.legend()\n",
        "  plt.title('Hyperparameters vs Accuracy')\n",
        "  plt.show()\n",
        "  print()\n",
        "  plot_val(l1,accuracy_fold)"
      ],
      "metadata": {
        "id": "TXwkmPOyZQzY"
      },
      "execution_count": 79,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_val(l1,accuracy_fold):\n",
        "  for j in range(len(l1)):\n",
        "    plt.plot(accuracy_fold[j], label = l1[j])\n",
        "    plt.legend()\n",
        "  plt.xlabel('Number of Folds')\n",
        "  plt.ylabel('Accuracy of every fold')\n",
        "  plt.title('Number of Folds vs Accuracy')\n",
        "  plt.show()\n"
      ],
      "metadata": {
        "id": "ApVlzu68ZQwr"
      },
      "execution_count": 71,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Main\n",
        "---"
      ],
      "metadata": {
        "id": "LgXPZ1zMb_s1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#Setting up Trivial and Baseline models\n",
        "---"
      ],
      "metadata": {
        "id": "7CZtpnYUcIYs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "Xtrain, Ytrain, Xtest, Ytest = pre_processing(df_train, df_test)"
      ],
      "metadata": {
        "id": "l2PdIWoRUr9e"
      },
      "execution_count": 111,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Xtrain.shape, Ytrain.shape, Xtest.shape, Ytest.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kuBy9TcpUxZJ",
        "outputId": "7e49d95c-2243-4896-be2a-b01eb8cbeb43"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((42748, 92), (42748,), (18321, 92), (18321,))"
            ]
          },
          "metadata": {},
          "execution_count": 112
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Performance of the trivial system\n",
        "trivial_model = trivialSytem2Class()\n",
        "trivial_model.fit(Xtrain, Ytrain)\n",
        "preds = trivial_model.predict(Xtest)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "f1 = f1_score(Ytest, preds, average = 'binary')\n",
        "\n",
        "print(f\"Accuracy of trivial system: {acc*100:.3f}%\")\n",
        "print(f\"F1 Score of trivial system: {f1*100:.3f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JMIS5T8-U15y",
        "outputId": "386feeb9-df8d-494e-9a4c-a553eb9054ca"
      },
      "execution_count": 113,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of trivial system: 50.227%\n",
            "F1 Score of trivial system: 43.790\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# checking if the predictions are following the class probabilites\n",
        "c0 = len(preds[preds==0])\n",
        "c1 = len(preds[preds==1])\n",
        "\n",
        "c0/len(preds), c1/len(preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "45RqWrPgVAPo",
        "outputId": "528b4269-0061-4c45-df74-c168a1ce9190"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(0.5548823754161891, 0.44511762458381093)"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Nearest Mean Classifier\n",
        "class_means = []\n",
        "\n",
        "for j in np.unique(Ytrain):\n",
        "  class_means.append(np.mean(Xtrain[Ytrain == j], axis = 0))\n",
        "\n",
        "y_pred = []\n",
        "for j in Xtest:\n",
        "  dist = [np.linalg.norm(j - m) for m in class_means]\n",
        "  y_pred.append(np.argmin(dist))\n",
        "\n",
        "accuracy = accuracy_score(np.array(y_pred),Ytest)\n",
        "f1 = f1_score(Ytest, y_pred, average = 'binary')\n",
        "\n",
        "print(f'The accuracy of the nearest means classifier is {100*accuracy:.3f}%')\n",
        "print(f\"F1 Score of nearest means classifier : {f1*100:.3f}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "meIp98UuXDTE",
        "outputId": "0ca11d8d-556a-4db8-c362-f231d9bc1b25"
      },
      "execution_count": 115,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy of the nearest means classifier is 60.346%\n",
            "F1 Score of nearest means classifier : 53.646\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Exploring Model performance using Logistic Regression model\n",
        "---"
      ],
      "metadata": {
        "id": "mGo2JYKccVUA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# trying logistic regression using all one hot encoded variables (92 features)\n",
        "clf = LogisticRegression(max_iter=1000)\n",
        "clf.fit(Xtrain, Ytrain)\n",
        "preds = clf.predict(Xtest)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Logistric regression accruacy: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vpon87JTXDP4",
        "outputId": "9e8a5871-1577-4a31-fe2c-f04842ce3b88"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Logistric regression accruacy: 77.616%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Applying PCA\n",
        "pca = PCA(n_components=30)\n",
        "pca.fit(Xtrain)\n",
        "reduced_Xtrain = pca.transform(Xtrain)\n",
        "reduced_Xtest = pca.transform(Xtest)\n",
        "print(reduced_Xtrain.shape, reduced_Xtest.shape)\n",
        "\n",
        "# trying logistic regression and PCA reduced features\n",
        "clf = LogisticRegression(random_state=5983, max_iter=1000)\n",
        "clf.fit(reduced_Xtrain, Ytrain)\n",
        "preds = clf.predict(reduced_Xtest)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Logistric regression accruacy: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FUFAAuVeaRvB",
        "outputId": "f5a114a2-40cf-4749-bf05-973872384a2e"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(42748, 30) (18321, 30)\n",
            "Logistric regression accruacy: 67.573%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Running by grouping on one feature at a time and instead of one hot encoding \n",
        "# we are checking by useing avg, max, min and median. \n",
        "# Example: group by cap shape and use avg to encode its different categorical \n",
        "# values\n",
        "accuracy = []\n",
        "params = genParams()\n",
        "for i in range(len(params)):\n",
        "  train_df = addColumns1(df_train, params[i][0], params[i][1], params[i][2])\n",
        "  test_df = addColumns1(df_test, params[i][0], params[i][1], params[i][2])\n",
        "\n",
        "  train_arr = train_df.to_numpy()\n",
        "  test_arr = test_df.to_numpy()\n",
        "\n",
        "  Xtrain = train_arr[:,:train_arr.shape[1]-1]\n",
        "  Ytrain = train_arr[:,train_arr.shape[1]-1]\n",
        "\n",
        "  Xtest = test_arr[:,:train_arr.shape[1]-1]\n",
        "  Ytest = test_arr[:,train_arr.shape[1]-1]\n",
        "\n",
        "  # trying logistic regression \n",
        "  clf = LogisticRegression(max_iter=5000)\n",
        "  clf.fit(Xtrain, Ytrain)\n",
        "  preds = clf.predict(Xtest)\n",
        "  acc = accuracy_score(Ytest, preds)\n",
        "  accuracy.append(acc)\n"
      ],
      "metadata": {
        "id": "lVQilLejVCsm"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking for which combination of parameters we are getting the highest accuracy\n",
        "# We are conisdering top 5 accuracies here\n",
        "acc_sorted = sorted(accuracy, reverse=True)[0:5]\n",
        "\n",
        "print(\"Logistic Regression model Accuracy for top 5 feature combinations\")\n",
        "for i in acc_sorted:\n",
        "  print(f\"{params[accuracy.index(i)]} : {i*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7Iqj5Mj0WCNX",
        "outputId": "6beb5df2-cf3e-49df-8814-599ccdb47cf2"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Logistic Regression model Accuracy for top 5 feature combinations\n",
            "['gill-attachment', 'stem-width', 'max'] : 63.670%\n",
            "['gill-attachment', 'stem-height', 'median'] : 63.610%\n",
            "['gill-spacing', 'cap-diameter', 'max'] : 63.594%\n",
            "['gill-attachment', 'stem-width', 'median'] : 63.583%\n",
            "['gill-attachment', 'cap-diameter', 'mean'] : 63.545%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Summary 1\n",
        "\n",
        "From the above observations we can see that:\n",
        "- Trivial system has a classification accuracy of 50.67% which is as good as tossing a coin\n",
        "- The baseline model of nearest mean classifier has an accuracy of 60.346% which is only slightly better than the trivial system\n",
        "- Training a logistic regression model on all one hot encoded features (92 features) resulted in a classification accuracy of 77.616% which is promising\n",
        "- PCA is tested to reduce the feature dimension from 92 to 30 which resulted in logestic regression model accuray of 68.561%, hence supporting use of other techniques to reduce feature dimensions\n",
        "- In the next step feature reduction by combining multiple features together has been tried. It can be seen from the results that the top 5 feature combinations based on classification accuracy, all have an accuracy that is less compared to training the model using all 92 one hot encoded features. \n",
        "- These observation formed the basis of motivation to explore other feature engineering methods for dimensionality reduction. \n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "ut-jcv82Ygr5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Exploring UFS and RFECV methods for dimension reduction\n",
        "---"
      ],
      "metadata": {
        "id": "kOU8S89lclfZ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "## Univariate Feature Selection (UFS)"
      ],
      "metadata": {
        "id": "Z0UDXTIHdDJg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "Xtrain, Ytrain, Xtest, Ytest = pre_processing(df_train, df_test)\n",
        "Xtrain.shape, Ytrain.shape, Xtest.shape, Ytest.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yyS_fQcndBqn",
        "outputId": "1e8ec86f-575a-4150-f71f-45b528607c5d"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((42748, 92), (42748,), (18321, 92), (18321,))"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "clf = LogisticRegression(max_iter=5000)\n",
        "accuracy = feature_selection_ufs(clf, Xtrain, Ytrain, Xtest, Ytest)"
      ],
      "metadata": {
        "id": "7z4E1nJNX8Hr"
      },
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Max accuracy: {np.max(accuracy)*100:.3f}%\")\n",
        "print(f\"Number of features to acheive this max accuracy: {np.argmax(accuracy)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YpVG0DNFeBRX",
        "outputId": "312a0f20-4796-4af2-8a31-fb429df9fc87"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Max accuracy: 77.971%\n",
            "Number of features to acheive this max accuracy: 46\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure()\n",
        "x = np.linspace(1, 92, 92)\n",
        "plt.plot(x, accuracy)\n",
        "plt.title(\"Accuracy vs K best features\")\n",
        "plt.xlabel(\"Number of Features\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "GG_1UXPEbcXL",
        "outputId": "f14c3750-6284-4816-d3fa-d32b2b3dbf9a"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Recursive Feature Elemination (RFE)"
      ],
      "metadata": {
        "id": "xkGEMTUCdPk8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "estimator = LogisticRegression(max_iter=5000)\n",
        "selector = RFECV(estimator, step=1, cv=5)\n",
        "selector = selector.fit(Xtrain, Ytrain)\n",
        "ranks = selector.ranking_\n",
        "indexes = [i for i in range(len(ranks)) if ranks[i] == 1]"
      ],
      "metadata": {
        "id": "OOHzpMr5dCMx"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Number of features selected by RFE: {len(indexes)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ge-ixqNVdfTr",
        "outputId": "ad5e0615-b15c-4bc8-972b-b8c9236827f7"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of features selected by RFE: 51\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking performance with 51 features selected by RFE\n",
        "temp_train = Xtrain[:,indexes]\n",
        "temp_test = Xtest[:,indexes]\n",
        "\n",
        "\n",
        "clf = LogisticRegression(max_iter=5000)\n",
        "clf.fit(temp_train, Ytrain)\n",
        "preds = clf.predict(temp_test)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Logistic regrssion accuracy with 51 best features: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_rI7ZwgvfOon",
        "outputId": "d2b9280b-59cf-47a2-9e2d-bdbb373057e0"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Logistic regrssion accuracy with 51 best features: 77.638%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Running RFE on 46 features selected by UFS"
      ],
      "metadata": {
        "id": "DabZ4pOFfxaw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Selecting the best 46 features that were given by UFS\n",
        "fs = SelectKBest(score_func=chi2, k=46)\n",
        "X_selected = fs.fit_transform(Xtrain, Ytrain)\n",
        "\n",
        "imp_ind = np.argsort(-fs.scores_)\n",
        "\n",
        "tempTrain = Xtrain[:,np.sort(imp_ind[:46])] "
      ],
      "metadata": {
        "id": "uaaHQ5LJfgLh"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Running RFE on the 46 features selected by UFS to see if there is any futher\n",
        "# feature selection possible to reduce the number further down from 46\n",
        "estimator = LogisticRegression(max_iter=5000)\n",
        "selector = RFECV(estimator, step=1, cv=5)\n",
        "selector = selector.fit(tempTrain, Ytrain)\n",
        "ranks = selector.ranking_\n",
        "ufs_rfe_ind = [i for i in range(len(ranks)) if ranks[i] == 1]"
      ],
      "metadata": {
        "id": "-MFF2DWZf79R"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Number of UFS features selected by RFE: {len(ufs_rfe_ind)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OFghYlv6gcCO",
        "outputId": "2e3057f3-a297-4b5d-d22a-a43e1d5be49c"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of UFS features selected by RFE: 46\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Testing model performance on features common to both UFS and RFE"
      ],
      "metadata": {
        "id": "pjnmpbXYgCel"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Getting features that were commonly selected by both RFE and UFS\n",
        "temp1 = imp_ind[:46]\n",
        "temp2 = np.array(indexes)\n",
        "\n",
        "temp3 = []\n",
        "for i in temp2:\n",
        "  if i in temp1:\n",
        "    temp3.append(i)\n",
        "\n",
        "temp3 = np.array(temp3)\n",
        "print(f\"Number of features common to UFS and RFE: {len(temp3)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1N3EobUzf898",
        "outputId": "5b1dad34-6b45-408c-ee64-dc52ee875bc7"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of features common to UFS and RFE: 32\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking performance with 32 features common to both RFE and UFS\n",
        "temp3Train = Xtrain[:,temp3]\n",
        "temp3Test = Xtest[:,temp3]\n",
        "\n",
        "clf = LogisticRegression(max_iter=5000)\n",
        "clf.fit(temp3Train, Ytrain)\n",
        "preds = clf.predict(temp3Test)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Logistic Regression accuracy with features common to UFS and RFE: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "196OjAn-gXAA",
        "outputId": "f458cd33-fbea-4b96-ca67-6a05912ec129"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Logistic Regression accuracy with features common to UFS and RFE: 75.804%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Summary 2\n",
        "\n",
        "- There were 46 features selected using UFS with chi2 statistic test\n",
        "- There were 51 features selected using RFE\n",
        "- There were 32 features common to both UFS and RFE\n",
        "- Running RFE on 46 features selected by UFS did not further reduce the features, REF considered all 46 features\n",
        "- Accuracy with 46 UFS features was 77.971%\n",
        "- Accuracy with 51 RFE features was 77.638%\n",
        "- Accuracy with 32 common features was 75.804%\n",
        "\n",
        "Result of analysis: UFS method gave the lowest number of features with highest accuracy. The model used for accuracy measurement is logistic regression.\n",
        "\n",
        "---"
      ],
      "metadata": {
        "id": "tPyh3Or-j99z"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Model Selection"
      ],
      "metadata": {
        "id": "p_CeTbYekVo0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "Xtrain, Ytrain, Xtest, Ytest = pre_processing(df_train, df_test)"
      ],
      "metadata": {
        "id": "tHZNlgB0j5ML"
      },
      "execution_count": 116,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Selecting the best 46 features that were given by UFS\n",
        "# These will be the final train and test sets on which \n",
        "# we will run different models and see which one performs the best\n",
        "fs = SelectKBest(score_func=chi2, k=46)\n",
        "fs.fit(Xtrain, Ytrain)\n",
        "\n",
        "imp_ind = np.argsort(-fs.scores_)\n",
        "\n",
        "Xtrain = Xtrain[:,np.sort(imp_ind[:46])] \n",
        "Xtest = Xtest[:,np.sort(imp_ind[:46])]\n",
        "Xtrain.shape, Xtest.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Moxfu4PDkgtm",
        "outputId": "20e56881-adf0-4f50-98b9-2ee120caeb92"
      },
      "execution_count": 117,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((42748, 46), (18321, 46))"
            ]
          },
          "metadata": {},
          "execution_count": 117
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Logistic Regression"
      ],
      "metadata": {
        "id": "YvTKG9D2lItT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking basic logistic regression \n",
        "clf = LogisticRegression(max_iter=1000)\n",
        "clf.fit(Xtrain, Ytrain)\n",
        "preds = clf.predict(Xtest)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Accuracy of Logistic Regression: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "A0RqNgFHkmHm",
        "outputId": "7b80f3c8-74c9-4ff8-ce4a-30c0721a0a02"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of Logistic Regression: 77.971%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hyperparameter search via crossbalidation for Logistic Regression classifier"
      ],
      "metadata": {
        "id": "Mm4fuDYlYgVw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking basic logistic regression to make sure everything so far is correct\n",
        "clf = LogisticRegression(max_iter=1000)\n",
        "accuracy_scores = []\n",
        "num_folds = 5\n",
        "kf = KFold(n_splits=num_folds)\n",
        "\n",
        "for fold, (train_indices, val_indices) in enumerate(kf.split(Xtrain)):\n",
        "\n",
        "      Xtrain_fold = Xtrain[train_indices]\n",
        "      Ytrain_fold = Ytrain[train_indices]\n",
        "      Xval_fold = Xtrain[val_indices]\n",
        "      Yval_fold = Ytrain[val_indices]\n",
        "\n",
        "      \n",
        "      clf.fit(Xtrain_fold, Ytrain_fold)\n",
        "\n",
        "      \n",
        "      pred_val = clf.predict(Xval_fold)\n",
        "      accuracy = accuracy_score(Yval_fold, pred_val)\n",
        "      accuracy_scores.append(accuracy)\n",
        "\n",
        "preds = clf.predict(Xtest)\n",
        "acc = accuracy_score(Ytest, preds)\n",
        "f1 = f1_score(preds, Ytest, average = 'binary')\n",
        "print(f\"Accuracy of Logistic Regression: {acc*100:.3f}%\")\n",
        "print(f'The F1 score of Logistic Regression:{f1*100}')"
      ],
      "metadata": {
        "id": "vyvdq5P8SzcQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0844f232-33fe-4a60-966f-150b89e22428"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of Logistic Regression: 77.965%\n",
            "The F1 score of Logistic Regression:75.14927670052325\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(accuracy_scores)\n",
        "plt.xlabel('Number of Folds')\n",
        "plt.ylabel('Validation Accuracy')\n",
        "plt.title('Number of Folds vs Validation Accuracy for Logistic Regression')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "vIWyzR13S9H6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "outputId": "98162707-b4d7-4abc-c98b-ad71d7456260"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlUAAAHHCAYAAACWQK1nAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACb2ElEQVR4nOzdeVhTZ/YH8G8SSMK+7yCouCGbgFitigsVFRcUFW2tlrp1sa2l7Yx0ptp2xtL+qlZrbTvtoLVjWxBFi1VRi/tWFETFBXfZN5F9T+7vj5ArEdAEE28C5/M8eVpvbu49AZKcvO97z+ExDMOAEEIIIYQ8FT7XARBCCCGEdAWUVBFCCCGEqAElVYQQQgghakBJFSGEEEKIGlBSRQghhBCiBpRUEUIIIYSoASVVhBBCCCFqQEkVIYQQQogaUFJFCCGEEKIGlFTpiCNHjoDH42H79u1ch6KUoqIizJgxA1ZWVuDxeFi3bt0zOS+Px8PHH3/8xP0+/vhj8Hg8zQekhUaNGoVRo0ax/7579y54PB5++umnJz72lVdegZubm1rj+emnn8Dj8XD37l21Hpc8O19++SV69eoFgUAAX19frsPRCGXfW5ShymuOKEf+GXnkyBFO46CkqhX5m7tYLEZeXl6b+0eNGgVPT08OItM97777Lvbv34/o6Gj873//w/jx4zvcl8fjtXuzt7d/hhFrn8TERPB4PPz3v//tcJ+DBw+Cx+Ph66+/foaRdc5nn32GXbt2cR1GuyQSCRwdHcHj8bBv3z6uw9EpBw4cwN/+9jc8//zz2Lx5Mz777DONnu+VV16BsbGxRs+hLr/++qvGv1DKEzT5jc/nw9LSEhMmTMDp06c1em7Slh7XAWijhoYGfP7559iwYQPXoeisQ4cOYerUqXj//feV2v+FF17AvHnzFLYZGBhoIjSdERoaCjMzM/z6669YuHBhu/v8+uuvEAgEmD17dqfP4+rqirq6Oujr63f6GMr47LPPMGPGDISFhSlsf/nllzF79myIRCKNnv9xDh06hIKCAri5ueGXX37BhAkTOItF1xw6dAh8Ph+xsbEQCoVch6MxdXV10NNT7SPz119/RWZmJpYtW6awXROvuTlz5mDixImQSCS4fv06vv32W4wePRpnz56Fl5eX2s6jrUaOHIm6ujrO/wYpqWqHr68vfvzxR0RHR8PR0ZHrcJ6pmpoaGBkZPfVxiouLYW5urvT+ffv2xdy5c5/6vF2JSCTCjBkzsHnzZuTn57f5W6yvr8fOnTvxwgsvwNbWttPnkY/OckUgEEAgEHB2fgDYunUr/Pz8MH/+fHz44Ydqex2oW3NzM6RSKecfHK0VFxfDwMBAbTExDIP6+nqt+1KlzteIJl5zfn5+Cu+hI0aMwIQJE/Ddd9/h22+/Veu5noSL1w+fz+f0fYyNg+sAtNGHH34IiUSCzz///LH7PW5e/NH5d/kanuvXr2Pu3LkwMzODjY0NPvroIzAMg5ycHEydOhWmpqawt7fHmjVr2j2nRCLBhx9+CHt7exgZGWHKlCnIyclps99ff/2F8ePHw8zMDIaGhggKCsLJkycV9pHHdOXKFbz44ouwsLDA8OHDH/ucb9++jZkzZ8LS0hKGhoZ47rnnsGfPHvZ++RQqwzDYuHEjOyT9tIqLi7FgwQLY2dlBLBbDx8cHW7ZsUeqxJ06cwODBgyEWi9G7d2/85z//aXe/gwcPYvjw4TA3N4exsTH69euHDz/88LHH9vT0xOjRo9tsl0qlcHJywowZM9htcXFx8Pf3h4mJCUxNTeHl5YX169c/9vhz586FVCpFXFxcm/v27NmDiooKvPTSSwCAzZs3Y8yYMbC1tYVIJIKHhwe+++67xx4f6PjveNeuXfD09IRYLIanpyd27tzZ7uNXr16NYcOGwcrKCgYGBvD392+z9o/H46GmpgZbtmxh/yZeeeUVAB2vqfr2228xcOBAiEQiODo64s0330R5ebnCPvIp+StXrmD06NEwNDSEk5MT/u///u+Jz1uurq4OO3fuxOzZszFr1izU1dXh999/b3ffffv2ISgoiP0dDh48GL/++qvCPn/99RcmTpwICwsLGBkZwdvbW+H3/OiaNrlH16vJfy+rV6/GunXr0Lt3b4hEIly5cgWNjY1YsWIF/P39YWZmBiMjI4wYMQKHDx9uc1ypVIr169fDy8sLYrEYNjY2GD9+PM6dOwcACAoKgo+PT7vPt1+/fggJCenwZ8fj8bB582bU1NSwv1f531FzczP+9a9/sXG7ubnhww8/RENDg8Ix3NzcMGnSJOzfvx8BAQEwMDDo8DWqioSEBPj7+8PAwADW1taYO3duu8s6EhIS4OHhofB33t7awUff06uqqrBs2TK4ublBJBLB1tYWL7zwAtLT0wHIfs979uzBvXv32J+N/JgdveauXbuGWbNmwcbGBgYGBujXrx/+8Y9/dOr5jxgxAgBw69Ythe3l5eVYtmwZXFxcIBKJ4O7uji+++AJSqVRhv/v37+Pll1+GqakpzM3NMX/+fFy4cKFN3PLp2Fu3bmHixIkwMTFh35OkUinWrVuHgQMHQiwWw87ODkuWLMGDBw8UznXu3DmEhITA2toaBgYG6NmzJ1599VWFfZ70/tnRmipl/g7kzyEvLw9hYWEwNjaGjY0N3n//fUgkEuV/6KCRqnb17NkT8+bNw48//ojly5erdbQqIiICAwYMwOeff449e/bg3//+NywtLfGf//wHY8aMwRdffIFffvkF77//PgYPHoyRI0cqPH7VqlXg8Xj4+9//juLiYqxbtw7BwcHIyMhgv9kdOnQIEyZMgL+/P1auXAk+n89+4B4/fhyBgYEKx5w5cyb69OmDzz77DAzDdBh7UVERhg0bhtraWrz99tuwsrLCli1bMGXKFGzfvh3Tpk3DyJEj8b///Q8vv/xyu1N6Hamvr0dpaanCNhMTE4hEItTV1WHUqFG4efMmli5dip49eyIhIQGvvPIKysvL8c4773R43EuXLmHcuHGwsbHBxx9/jObmZqxcuRJ2dnYK+12+fBmTJk2Ct7c3Pv30U4hEIty8ebNNIvqoiIgIfPzxxygsLFRYA3bixAnk5+ez03IHDx7EnDlzMHbsWHzxxRcAgKtXr+LkyZOPjX/kyJFwdnbGr7/+iqioKIX7fv31VxgaGrLTad999x0GDhyIKVOmQE9PD7t378Ybb7wBqVSKN99887HP41EHDhxAeHg4PDw8EBMTg/v37yMyMhLOzs5t9l2/fj2mTJmCl156CY2NjYiLi8PMmTPxxx9/IDQ0FADwv//9DwsXLkRgYCAWL14MAOjdu3eH5//444/xySefIDg4GK+//jqysrLw3Xff4ezZszh58qTCtMmDBw8wfvx4TJ8+HbNmzcL27dvx97//HV5eXkpN4yUlJaG6uhqzZ8+Gvb09Ro0ahV9++QUvvviiwn4//fQTXn31VQwcOBDR0dEwNzfH+fPnkZyczO578OBBTJo0CQ4ODnjnnXdgb2+Pq1ev4o8//njs7/lxNm/ejPr6eixevBgikQiWlpaorKzEf//7X8yZMweLFi1CVVUVYmNjERISgtTUVIXF4gsWLMBPP/2ECRMmYOHChWhubsbx48dx5swZBAQE4OWXX8aiRYuQmZmpsGb07NmzuH79Ov75z392GNv//vc//PDDD0hNTWXX/g0bNgwAsHDhQmzZsgUzZszAe++9h7/++gsxMTG4evVqmwQ9KysLc+bMwZIlS7Bo0SL069evUz8ruZ9++gmRkZEYPHgwYmJiUFRUhPXr1+PkyZM4f/48O4q+Z88eREREwMvLCzExMXjw4AEWLFgAJyenJ57jtddew/bt27F06VJ4eHjg/v37OHHiBK5evQo/Pz/84x//QEVFBXJzc/HVV18BwGPXgl28eBEjRoyAvr4+Fi9eDDc3N9y6dQu7d+/GqlWrVP4ZyL+kWFhYsNtqa2sRFBSEvLw8LFmyBD169MCpU6cQHR2NgoICdv2XVCrF5MmTkZqaitdffx39+/fH77//jvnz57d7rubmZoSEhGD48OFYvXo1DA0NAQBLlixhfxdvv/027ty5g2+++Qbnz59nX8fFxcXse/Ty5cthbm6Ou3fvIjExkT1+Z98/lf07AGQDFiEhIRgyZAhWr16NP//8E2vWrEHv3r3x+uuvK/+DZwhr8+bNDADm7NmzzK1btxg9PT3m7bffZu8PCgpiBg4cyP77zp07DABm8+bNbY4FgFm5ciX775UrVzIAmMWLF7PbmpubGWdnZ4bH4zGff/45u/3BgweMgYEBM3/+fHbb4cOHGQCMk5MTU1lZyW7ftm0bA4BZv349wzAMI5VKmT59+jAhISGMVCpl96utrWV69uzJvPDCC21imjNnjlI/n2XLljEAmOPHj7PbqqqqmJ49ezJubm6MRCJReP5vvvmmUscF0O5N/nNdt24dA4DZunUr+5jGxkZm6NChjLGxscLP49Gfe1hYGCMWi5l79+6x265cucIIBAKm9Z//V199xQBgSkpKlIpZLisriwHAbNiwQWH7G2+8wRgbGzO1tbUMwzDMO++8w5iamjLNzc0qHZ9hGOaDDz5gADBZWVnstoqKCkYsFiv87uTnai0kJITp1auXwragoCAmKCiI/Xd7f8e+vr6Mg4MDU15ezm47cOAAA4BxdXVVON6j521sbGQ8PT2ZMWPGKGw3MjJS+JuWk7/u7ty5wzAMwxQXFzNCoZAZN26cwt/UN998wwBgNm3apPBcADA///wzu62hoYGxt7dnwsPD25yrPZMmTWKef/559t8//PADo6enxxQXF7PbysvLGRMTE2bIkCFMXV2dwuPlr7Pm5mamZ8+ejKurK/PgwYN295HH3PrnLzd//nyFn63892JqaqoQi/xcDQ0NCtsePHjA2NnZMa+++iq77dChQwwAhfexR2MqLy9nxGIx8/e//13h/rfffpsxMjJiqqur2zz20biNjIwUtmVkZDAAmIULFypsf//99xkAzKFDh9htrq6uDAAmOTn5sed53Plaa2xsZGxtbRlPT0+F39Uff/zBAGBWrFjBbvPy8mKcnZ2ZqqoqdtuRI0fa/Tt/9L3FzMzsie9xoaGhbY7DMO2/5kaOHMmYmJgovFcxjOLfTnvkx/rkk0+YkpISprCwkDl+/DgzePBgBgCTkJDA7vuvf/2LMTIyYq5fv65wjOXLlzMCgYDJzs5mGIZhduzYwQBg1q1bx+4jkUiYMWPGtIl7/vz5DABm+fLlCsc8fvw4A4D55ZdfFLYnJycrbN+5cyf7udsRZd4/5Z+Rhw8fZhhGtb8D+XP49NNPFY45aNAgxt/fv8Nztoem/zrQq1cvvPzyy/jhhx9QUFCgtuO2XnAsEAgQEBAAhmGwYMECdru5uTn69euH27dvt3n8vHnzYGJiwv57xowZcHBwwN69ewEAGRkZuHHjBl588UXcv38fpaWlKC0tRU1NDcaOHYtjx461GeZ97bXXlIp97969CAwMVJgiNDY2xuLFi3H37l1cuXJFuR9CO6ZOnYqDBw8q3OTTDnv37oW9vT3mzJnD7q+vr4+3334b1dXVOHr0aLvHlEgk2L9/P8LCwtCjRw92+4ABA9pMaci/sfz+++9tfj6P07dvX/j6+iI+Pl7hvNu3b8fkyZPZ0UNzc3PU1NTg4MGDSh9bTr5OovU0044dO1BfX88OswOKC/srKipQWlqKoKAg3L59GxUVFUqfr6CgABkZGZg/fz7MzMzY7S+88AI8PDza7N/6vA8ePEBFRQVGjBjBToOo6s8//0RjYyOWLVsGPv/hW9SiRYtgamqqMN0MyP4GW68lEQqFCAwMbPf186j79+9j//79Cn9b4eHh4PF42LZtG7vt4MGDqKqqwvLly9us25BPb58/fx537tzBsmXL2qwnfJop8PDwcNjY2ChsEwgE7BomqVSKsrIyNDc3IyAgQOHnvmPHDvB4PKxcubLNceUxmZmZYerUqfjtt9/YkWqJRIL4+HiEhYV1am2M/P3o0dHV9957DwDa/A579uz52GlGVZw7dw7FxcV44403FH5XoaGh6N+/P3vu/Px8XLp0CfPmzVMYQQoKClJqYbe5uTn++usv5OfnP3XMJSUlOHbsGF599VWF9ypA+b+dlStXwsbGBvb29hgxYgSuXr2KNWvWKCxBSEhIwIgRI2BhYcF+NpSWliI4OBgSiQTHjh0DACQnJ0NfXx+LFi1iH8vn8x874v3oaE5CQgLMzMzwwgsvKJzL398fxsbG7FS1/LXyxx9/oKmpqd1jd+b9U9m/g9Ye/SwcMWKEUu8jrVFS9Rj//Oc/0dzc/MS1Vap49AVjZmYGsVgMa2vrNtsfnXcGgD59+ij8m8fjwd3dnR3qvXHjBgBg/vz5sLGxUbj997//RUNDQ5sP2J49eyoV+71799odlh8wYAB7f2c5OzsjODhY4ebg4MAet0+fPgofsMqct6SkBHV1dW1+ZgDaPI+IiAg8//zzWLhwIezs7DB79mxs27ZNqQQrIiICJ0+eZOfpjxw5guLiYkRERLD7vPHGG+jbty8mTJgAZ2dnvPrqq0hOTn7isQHA29sbnp6e+O2339htv/76K6ytrRU+iE6ePIng4GAYGRnB3NwcNjY27JowVZIq+c9TmZ8bIHszfO655yAWi2FpaQkbGxt89913Kp2zvfM/ei6hUIhevXq1+X07Ozu3+eCxsLBo9/XzqPj4eDQ1NWHQoEG4efMmbt68ibKyMgwZMgS//PILu598XcrjSqoos09ndPT63LJlC7y9vSEWi2FlZQUbGxt2nV3rmBwdHWFpafnYc8ybNw/Z2dk4fvw4AFliW1RUhJdffrlTMd+7dw98Ph/u7u4K2+3t7WFubt7md6jse5Cy5wba/1vt378/e7/8v4/G2NG2R/3f//0fMjMz4eLigsDAQHz88ccqfwDLyR/3NH87ixcvxsGDB7F79268++67qKura7Me6MaNG0hOTm7z2RAcHAxAtnYVkP1sHBwc2Gk8uY5+Lnp6em2WBty4cQMVFRWwtbVtc77q6mr2XEFBQQgPD8cnn3wCa2trTJ06FZs3b1ZYe9eZ909l/w7k5OsNW1P2faQ1WlP1GL169cLcuXPxww8/YPny5W3u7+gbxOMWtrV3lVNHVz4xj1nf1BF5EvDll192WITv0Xl9bbvKhgsGBgY4duwYDh8+jD179iA5ORnx8fEYM2YMDhw48Nir0yIiIhAdHY2EhAQsW7YM27Ztg5mZmUJtLltbW2RkZGD//v3Yt28f9u3bh82bN2PevHlKLbifO3culi9fjnPnzsHZ2RmHDx/GkiVL2Eu8b926hbFjx6J///5Yu3YtXFxcIBQKsXfvXnz11Vcqjb6p4vjx45gyZQpGjhyJb7/9Fg4ODtDX18fmzZvbLODWlKd5/cgTp+eff77d+2/fvo1evXp1Prh2yC/keFRH7xvtvT63bt2KV155BWFhYfjggw9ga2sLgUCAmJiYNguTlRESEgI7Ozts3boVI0eOxNatW2Fvb89+2HaWsqMsuvgeNGvWLIwYMQI7d+7EgQMH8OWXX+KLL75AYmIiJyU5+vTpw/6+Jk2aBIFAgOXLl2P06NEICAgAIPt8eOGFF/C3v/2t3WP07du3U+cWiURtvvRKpVLY2toqfDlpTZ7AyItanzlzBrt378b+/fvx6quvYs2aNThz5gyMjY2f+v1TGeq6AplGqp5APlolXxzXmnwB4KNXJD3NiM2TyEei5BiGwc2bN9mrSuSLf01NTduM/Mhvna2N4urqiqysrDbbr127xt6vCa6urrhx40abxOBJ55VfQfPozwxAu8+Dz+dj7NixWLt2La5cuYJVq1bh0KFD7V5R1VrPnj0RGBiI+Ph4NDc3IzExEWFhYW3qLgmFQkyePBnffvstbt26hSVLluDnn3/GzZs3H3t8QFaDhsfj4ddff0V8fDwkEonC1N/u3bvR0NCApKQkLFmyBBMnTkRwcHCnPqzkP09lfm47duyAWCxm3wgnTJjQ4Qexsh+w8vM/eq7GxkbcuXNHbX9nd+7cwalTp7B06VIkJCQo3OLj4yEUCtnEUP66yszM7PB4yuwDyN43Hn3PAFR739i+fTt69eqFxMREvPzyywgJCUFwcDDq6+vbxJSfn4+ysrLHHk8gEODFF1/E9u3b8eDBA+zatQtz5szp9AeNq6srpFJpm7+hoqIilJeXa+y9Qn5uoP3XeFZWFnu//L/tvf6UeU0CgIODA9544w3s2rULd+7cgZWVlcKicmX/5uWJ+5P+dlTxj3/8AyYmJgoXGvTu3RvV1dUdfjbIZ1JcXV1RUFCA2tpahWMq+3ORn+v+/ft4/vnn2z3Xo1ecPvfcc1i1ahXOnTuHX375BZcvX1a46lnV909l/w7UjZKqJ+jduzfmzp2L//znPygsLFS4z9TUFNbW1uw8tJwma4L8/PPPqKqqYv+9fft2FBQUsN+M/P390bt3b6xevRrV1dVtHl9SUtLpc0+cOBGpqakKVXpramrwww8/wM3Nrd31NuowceJEFBYWKqxbam5uxoYNG2BsbIygoKB2HycQCBASEoJdu3YhOzub3X716lXs379fYd/2PnTkI32PXgLenoiICJw5cwabNm1CaWmpwtQfIFu70xqfz4e3t7fSx+/RowdGjBiB+Ph4bN26FT179mSvsgIefstqPQJSUVGBzZs3P/HYj3JwcICvry+2bNmiMJV08ODBNuvmBAIBeDyewijL3bt3262cbmRk1G4y8ajg4GAIhUJ8/fXXCs8nNjYWFRUV7BWFT0v+Dfpvf/sbZsyYoXCbNWsWgoKC2H3GjRsHExMTxMTEtElc5DH6+fmhZ8+eWLduXZvn2fp59O7dG9euXVN4LV64cOGJV5q21t7v+6+//mpTQTs8PBwMw+CTTz5pc4xHR8tefvllPHjwAEuWLEF1dfVT1Y2bOHEiALSpJr527VoAUNvvsD0BAQGwtbXF999/r/Da2rdvH65evcqe29HREZ6envj5558V3iuPHj2KS5cuPfYcEomkzfS2ra0tHB0dFc5pZGSk1DS4jY0NRo4ciU2bNim8VwGdm7EAZOuQlixZgv379yMjIwOAbHTt9OnTbd7/ANngQHNzMwDZyGVTUxN+/PFH9n6pVIqNGzcqff5Zs2ZBIpHgX//6V5v7mpub2dfIgwcP2jzHR997O/P+qezfgbrR9J8S/vGPf+B///sfsrKyMHDgQIX7Fi5ciM8//xwLFy5EQEAAjh07huvXr2ssFktLSwwfPhyRkZEoKirCunXr4O7uzi4o5PP5+O9//4sJEyZg4MCBiIyMhJOTE/Ly8nD48GGYmppi9+7dnTr38uXL8dtvv2HChAl4++23YWlpiS1btuDOnTvYsWNHm+FfdVm8eDH+85//4JVXXkFaWhrc3Nywfft2nDx5EuvWrVNYuP+oTz75BMnJyRgxYgTeeOMNNhkbOHAgLl68yO736aef4tixYwgNDYWrqyuKi4vx7bffwtnZ+Ym1uwDZG8j777+P999/H5aWlm1GaxYuXIiysjKMGTMGzs7OuHfvHjZs2ABfX192bdiTzJ07F4sXL0Z+fn6b2jXjxo1jv8nJPxR//PFH2NradupCi5iYGISGhmL48OF49dVXUVZWxv7cWn8AhYaGYu3atRg/fjxefPFFFBcXY+PGjXB3d1f4+QKyhP/PP//E2rVr4ejoiJ49e2LIkCFtzm1jY4Po6Gh88sknGD9+PKZMmYKsrCx8++23GDx4sNqKxP7yyy/w9fWFi4tLu/dPmTIFb731FtLT0+Hn54evvvoKCxcuxODBg9m6bhcuXEBtbS22bNkCPp+P7777DpMnT4avry8iIyPh4OCAa9eu4fLly+wH2auvvoq1a9ciJCQECxYsQHFxMb7//nsMHDgQlZWVSsU+adIkJCYmYtq0aQgNDcWdO3fw/fffw8PDQ+H3M3r0aLz88sv4+uuvcePGDYwfPx5SqRTHjx/H6NGjsXTpUnbfQYMGwdPTEwkJCRgwYAD8/Pw6/bP18fHB/Pnz8cMPP6C8vBxBQUFITU3Fli1bEBYW1m5tN1U0NTXh3//+d5vtlpaWeOONN/DFF18gMjISQUFBmDNnDnspvZubG9599112/88++wxTp07F888/j8jISDx48ADffPMNPD092/1SKldVVQVnZ2fMmDEDPj4+MDY2xp9//omzZ88q1Bj09/dHfHw8oqKiMHjwYBgbG2Py5MntHvPrr7/G8OHD4efnh8WLF6Nnz564e/cu9uzZwyZFqnrnnXewbt06fP7554iLi8MHH3yApKQkTJo0Ca+88gr8/f1RU1ODS5cuYfv27bh79y6sra0RFhaGwMBAvPfee7h58yb69++PpKQk9sunMiNwQUFBWLJkCWJiYpCRkYFx48ZBX18fN27cQEJCAtavX48ZM2Zgy5Yt+PbbbzFt2jT07t0bVVVV+PHHH2Fqasom5515/9TX11f670CtVLpWsItrXVLhUfJLLluXVGAY2eXkCxYsYMzMzBgTExNm1qxZTHFxcYclFR69ZL+jy4MfLd8gv1z0t99+Y6KjoxlbW1vGwMCACQ0NbXMJLsMwzPnz55np06czVlZWjEgkYlxdXZlZs2YxKSkpT4zpcW7dusXMmDGDMTc3Z8RiMRMYGMj88ccfbfaDiiUVnrRvUVERExkZyVhbWzNCoZDx8vJSqpQFwzDM0aNHGX9/f0YoFDK9evVivv/+e/a5y6WkpDBTp05lHB0dGaFQyDg6OjJz5sxpc+nx4zz//PPtXkbOMAyzfft2Zty4cYytrS0jFAqZHj16MEuWLGEKCgqUPn5ZWRkjEokYAMyVK1fa3J+UlMR4e3szYrGYcXNzY7744gtm06ZNCuUKGEa5kgoMI7usesCAAYxIJGI8PDyYxMTENpf9MwzDxMbGMn369GFEIhHTv39/ZvPmzW1+vgzDMNeuXWNGjhzJGBgYMADY8gqPllSQ++abb5j+/fsz+vr6jJ2dHfP666+3KVXw6OtErr04W0tLS2MAMB999FGH+9y9e5cBwLz77rvstqSkJGbYsGGMgYEBY2pqygQGBjK//fabwuNOnDjBvPDCC4yJiQljZGTEeHt7tym5sXXrVqZXr16MUChkfH19mf3793dYUuHLL79sE5tUKmU+++wzxtXVlRGJRMygQYOYP/74o93n3dzczHz55ZdM//79GaFQyNjY2DATJkxg0tLS2hz3//7v/xgAzGeffdbhz+VRHb2HNTU1MZ988gnTs2dPRl9fn3FxcWGio6OZ+vp6hf1cXV2Z0NBQlc6HDsqw9O7dm90vPj6eGTRoECMSiRhLS0vmpZdeYnJzc9scLy4ujunfvz8jEokYT09PJikpiQkPD2f69++vsF/r95aGhgbmgw8+YHx8fNjfs4+PD/Ptt98qPKa6upp58cUXGXNzc4UyDR295jIzM5lp06ax76/9+vV77N9o62O193fCMAzzyiuvMAKBgLl58ybDMLIyONHR0Yy7uzsjFAoZa2trZtiwYczq1auZxsZG9nElJSXMiy++yJiYmDBmZmbMK6+8wpw8eZIBwMTFxSn8Ph5X4uKHH35g/P39GQMDA8bExITx8vJi/va3vzH5+fkMwzBMeno6M2fOHKZHjx6MSCRibG1tmUmTJjHnzp1jj6HM++ejJRXklPk76Og5tPc+9iQ8hunk2CIhhJAuZf369Xj33Xdx9+7dNlcqdye+vr6wsbHpVAmUrmzXrl2YNm0aTpw40eHFHd0drakihBAChmEQGxuLoKCgbpNQNTU1seuI5I4cOYILFy6020qoO6mrq1P4t0QiwYYNG2BqavpUU8NdHa2pIoSQbqympgZJSUk4fPgwLl261GHfw64oLy8PwcHBmDt3LhwdHXHt2jV8//33sLe3V7ooclf11ltvoa6uDkOHDkVDQwMSExNx6tQpfPbZZzpZAuNZoek/Qgjpxu7evYuePXvC3Nwcb7zxRqf6zOmqiooKLF68GCdPnkRJSQmMjIwwduxYfP7554/tTdkd/Prrr1izZg1u3ryJ+vp6uLu74/XXX1e4uIG0RUkVIYQQQoga0JoqQgghhBA1oKSKEEIIIUQNaKF6J0mlUuTn58PExOSpOtATQggh5NlhGAZVVVVwdHRUe9FqSqo6KT8/v8NKzIQQQgjRbjk5OXB2dlbrMSmp6iR5a5ScnByYmppyHA0hhBBClFFZWQkXF5fHtjjrLEqqOkk+5WdqakpJFSGEEKJjNLF0hxaqE0IIIYSoASVVhBBCCCFqQEkVIYQQQogaUFJFCCGEEKIGlFQRQgghhKgBJVWEEEIIIWpASRUhhBBCiBpQUkUIIYQQogaUVBFCCCGEqAElVYQQQgghakBJFSGEEEKIGlBSRQghhBCiBpRUaRmGYZCRU47K+iauQyGEEEKICrQiqdq4cSPc3NwgFosxZMgQpKamdrjvqFGjwOPx2txCQ0PZfYqKivDKK6/A0dERhoaGGD9+PG7cuKFwnPr6erz55puwsrKCsbExwsPDUVRUpLHnqKzXt6YjbONJ7L6Qz3UohBBCCFEB50lVfHw8oqKisHLlSqSnp8PHxwchISEoLi5ud//ExEQUFBSwt8zMTAgEAsycOROAbKQnLCwMt2/fxu+//47z58/D1dUVwcHBqKmpYY/z7rvvYvfu3UhISMDRo0eRn5+P6dOnP5Pn/Dj+rhYAgB1puRxHQgghhBCVMBwLDAxk3nzzTfbfEomEcXR0ZGJiYpR6/FdffcWYmJgw1dXVDMMwTFZWFgOAyczMVDimjY0N8+OPPzIMwzDl5eWMvr4+k5CQwO5z9epVBgBz+vRppc5bUVHBAGAqKiqU2l9ZRZV1TK/oPYzr3/9gbhZXqfXYhBBCSHenqc9vhmEYTkeqGhsbkZaWhuDgYHYbn89HcHAwTp8+rdQxYmNjMXv2bBgZGQEAGhoaAABisVjhmCKRCCdOnAAApKWloampSeG8/fv3R48ePZQ+r6bYmogxso81ACAxnUarCCGEEF3BaVJVWloKiUQCOzs7he12dnYoLCx84uNTU1ORmZmJhQsXstvkyVF0dDQePHiAxsZGfPHFF8jNzUVBQQEAoLCwEEKhEObm5kqft6GhAZWVlQo3TQn3dwYA7EzPg1TKaOw8hBBCCFEfztdUPY3Y2Fh4eXkhMDCQ3aavr4/ExERcv34dlpaWMDQ0xOHDhzFhwgTw+Z1/ujExMTAzM2NvLi4u6ngK7QoeYAdTsR7yK+px+vZ9jZ2HEEIIIerDaVJlbW0NgUDQ5qq7oqIi2NvbP/axNTU1iIuLw4IFC9rc5+/vj4yMDJSXl6OgoADJycm4f/8+evXqBQCwt7dHY2MjysvLlT5vdHQ0Kioq2FtOTo4Kz1Q1Yn0BJvs4AqAF64QQQoiu4DSpEgqF8Pf3R0pKCrtNKpUiJSUFQ4cOfexjExIS0NDQgLlz53a4j5mZGWxsbHDjxg2cO3cOU6dOBSBLuvT19RXOm5WVhezs7A7PKxKJYGpqqnDTJPkU4L7MQlQ3NGv0XIQQQgh5enpcBxAVFYX58+cjICAAgYGBWLduHWpqahAZGQkAmDdvHpycnBATE6PwuNjYWISFhcHKyqrNMRMSEmBjY4MePXrg0qVLeOeddxAWFoZx48YBkCVbCxYsQFRUFCwtLWFqaoq33noLQ4cOxXPPPaf5J62EQS7m6GVthNulNdh7qQCzAjQ33UgIIYSQp8d5UhUREYGSkhKsWLEChYWF8PX1RXJyMrt4PTs7u81aqKysLJw4cQIHDhxo95gFBQWIiopCUVERHBwcMG/ePHz00UcK+3z11Vfg8/kIDw9HQ0MDQkJC8O2332rmSXYCj8dDuL8zvtyfhR1puZRUEfIMVdU3QV/Ah1hfwHUohBAdwmMYhi4v64TKykqYmZmhoqJCY1OB+eV1eP6LQ2AY4PjfRsPF0lAj5yGEPFRS1YCQdcdgbyrGH28NB5/P4zokQogaafLzW6ev/uvqHM0NMKy3bHozMT2P42gI6R52pOeirKYRVwoqcfJWKdfhEEJ0CCVVWm5Gy4L1Hem5oEFFQjSLYRhsO/vwyt64s5q7ypcQ0vVQUqXlQgbaw0goQHZZLc7efcB1OIR0aWn3HuB2aQ30BbIpvwOXC1FW08hxVIQQXUFJlZYzFOphopcDAKpZRYimxbeMTE31dYKXkxmaJAy1iyKEKI2SKh0gr1m151IB6holHEdDSNdU3dCMPZdkrawiBrsgYrDsitv4szk09U4IUQolVTog0M0SzhYGqG5oxoErT+6JSAhR3d6LBahtlKCXtRECXC0wxdcRYn0+bhRXIz27nOvwCCE6gJIqHcDn8xDuJxut2k5TgIRoRPw52dTfzAAX8Hg8mIr1EeolaxcVfzaby9AIITqCkiodIU+qTtwsRUFFHcfRENK13CyuRtq9BxDweQj3c2K3zw6UTQHuvlCAqvomrsIjhOgISqp0RA8rQwS6WYJhgJ3nqWYVIeqU0DJKNbqfDWxNxez2AFcL9LIxQl2TBH9cLOAqPEKIjqCkSoeE+8u+Qe9Io5pVhKhLk0SKHS3FdWc+0g6Kx+MhomUb1awihDwJJVU6ZKKXA8T6fNwqqcGF3AquwyGkSzh8rRil1Q2wNhZiTH/bNvdP93OGHp+HCznluFpQyUGEhBBdQUmVDjER62P8QHsAwPY0+tZMiDpsOye7+GO6nzP0BW3fEm1MRAgeIGvwHk+jVYSQx6CkSsfIa1btvlCAhmaqWUXI0yiuqsfhrGIAwKwA5w73i2hZsL7zfB7qm+h1RwhpHyVVOmZYb2vYm4pRUdeElKvFXIdDiE5LTM+DRMrAr4c53G1NOtxvZB8bOJjJXncHrhQ9wwgJIbqEkiodI+DzMM3v4YJ1QkjnMAyDbS1X/c16ZIH6owR8HruInWpWEUI6QkmVDpLXrDpyvQQlVQ0cR0OIbkq79wC3S2pgoC/AJB/HJ+4/098ZPB5w8uZ9ZN+vfQYREkJ0DSVVOsjd1hi+LuaQSBn8nkE1qwjpDPkoVai3A4xFek/c38XSEMPdrRUeSwghrVFSpaPkC9apbQ0hqqtuaGaLeT5p6q+12YN7AAAS0nLQLJFqJDZCiO6ipEpHTfZ2gFDAx7XCKlzOp5pVhKhC3jy5p7URBrtZKP24YA9bWBoJUVTZgKPXSzQYISFEF1FSpaPMDYUI9pAVKtyRRlOAhKhiG9s82Rk8Hk/px4n0BJg+SHahCFVYJ4Q8ipIqHSZfsP57Rh6aaCqCEKXcLK7GuZbmyTP8Oq5N1ZGIwbLpwkPXilFcWa/u8AghOoySKh02sq8NrI1FuF/TiCNZNBVBiDISWroRjOqr2DxZWX3sTODvagGJlMH2dFrTSAh5iJIqHaYv4CPMV3YpONWsIuTJmiRSdrp81mDlF6g/KoKtWZVDzc0JISxKqnSc/CrAlGtFeFDTyHE0hGi3I1klj22erKxQbwcYCQW4d78WZ26XqTFCQoguo6RKxw1wMIWHgymaJAx2X8znOhxCtJq8IfK0QU7tNk9WlpFID1NaRompwjohRI6Sqi5APlpFU4CEdEyxeXLnp/7kIlpqVu3NLERFbdNTH48QovsoqeoCpvo6Qo/Pw4XcCtwoquI6HEK00s6W5smDepijj13HzZOV5eNshv72JmhslmIXdTYghICSqi7B2liEUf1k60PoaiRC2mIYBvEttaki1DBKBQA8Ho8trxBHC9YJIaCkqsuY4S8rSLjrvOzbOCHkofTsh82TQ70d1HbcaYOcINTj42pBJTLzKtV2XEKIbqKkqosY3d8W5ob6KKpswImbpVyHQ4hWkS9QD/V2gIlYX23HNTcUYvxAewBAHC1YJ6Tbo6SqixDpCTDFh2pWEfKomk42T1bW7JYpwKSMfNQ2Nqv9+IQQ3UFJVRcyo+UqwP2XC1FZT1cjEQIAey51rnmysp7rZYUeloaoamjG3kuFaj8+IUR3cJ5Ubdy4EW5ubhCLxRgyZAhSU1M73HfUqFHg8XhtbqGhoew+1dXVWLp0KZydnWFgYAAPDw98//33CscpLCzEyy+/DHt7exgZGcHPzw87duzQ2HN8VryczNDH1hgNzVLsaflmTkh3t+1s55onK4vPf7hgnWpWEdK9cZpUxcfHIyoqCitXrkR6ejp8fHwQEhKC4uLidvdPTExEQUEBe8vMzIRAIMDMmTPZfaKiopCcnIytW7fi6tWrWLZsGZYuXYqkpCR2n3nz5iErKwtJSUm4dOkSpk+fjlmzZuH8+fMaf86axOPxqGYVIa3cKpE1T+bzHjYg14QZ/s7g84Czdx/gZnG1xs5DCNFunCZVa9euxaJFixAZGcmOKBkaGmLTpk3t7m9paQl7e3v2dvDgQRgaGiokVadOncL8+fMxatQouLm5YfHixfDx8VEYATt16hTeeustBAYGolevXvjnP/8Jc3NzpKWlafw5a9q0QU7g84Bz9x7gbmkN1+EQwqltLWUURvezhV0nmicry85UzLa9kZ+TENL9cJZUNTY2Ii0tDcHBwQ+D4fMRHByM06dPK3WM2NhYzJ49G0ZGRuy2YcOGISkpCXl5eWAYBocPH8b169cxbtw4hX3i4+NRVlYGqVSKuLg41NfXY9SoUR2eq6GhAZWVlQo3bWRnKsbwPjYAgESqWUW6sdbNk2dqYIH6o+QV1nek5aKxWarx8xFCtA9nSVVpaSkkEgns7OwUttvZ2aGw8MmLPVNTU5GZmYmFCxcqbN+wYQM8PDzg7OwMoVCI8ePHY+PGjRg5ciS7z7Zt29DU1AQrKyuIRCIsWbIEO3fuhLu7e4fni4mJgZmZGXtzcdH8m3RnyRes70jPg5RqVpFuqnXz5LEDOt88WVmj+9nAxkSE+zWNSLlapPHzEUK0D+cL1TsrNjYWXl5eCAwMVNi+YcMGnDlzBklJSUhLS8OaNWvw5ptv4s8//2T3+eijj1BeXo4///wT586dQ1RUFGbNmoVLly51eL7o6GhUVFSwt5wc7R3iH+dhBxOxHvLK63Dmzn2uwyGEE/JpuKdtnqwsPQGf/UITd1Z73x8IIZqjx9WJra2tIRAIUFSk+I2uqKgI9vb2j31sTU0N4uLi8Omnnypsr6urw4cffoidO3eyVwR6e3sjIyMDq1evRnBwMG7duoVvvvkGmZmZGDhwIADAx8cHx48fx8aNG9tcKSgnEokgEok6+3SfKbG+AJO8HfBbag52pOVhWG9rrkMi5JkqrqrHoWvqa56srFkBLvjuyC0cu1GCvPI6OJkbPLNzE0K4x9lIlVAohL+/P1JSUthtUqkUKSkpGDp06GMfm5CQgIaGBsydO1dhe1NTE5qamsDnKz4tgUAAqVS2xqG2thYAHrtPVyC/0mlfZgFqGqggIele1N08WVk9rY3wXC9LMAyQQAvWCel2OJ3+i4qKwo8//ogtW7bg6tWreP3111FTU4PIyEgAstIH0dHRbR4XGxuLsLAwWFlZKWw3NTVFUFAQPvjgAxw5cgR37tzBTz/9hJ9//hnTpk0DAPTv3x/u7u5YsmQJUlNTcevWLaxZswYHDx5EWFiYxp/zs+LvagE3K0PUNkqQnEkFCUn3wTAMO/X3LEep5Ga3LFhPOJdLfTgJ6WY4m/4DgIiICJSUlGDFihUoLCyEr68vkpOT2cXr2dnZbUaUsrKycOLECRw4cKDdY8bFxSE6OhovvfQSysrK4OrqilWrVuG1114DAOjr62Pv3r1Yvnw5Jk+ejOrqari7u2PLli2YOHGiZp/wM8Tj8RDu54w1B69je1ouW7+KkK4uPfsBbrU0T56kxubJyhrvaQ/T32VrGk/eLMXIvjbPPAZCCDd4DMPQV6lOqKyshJmZGSoqKmBqasp1OO3KfVCL4V8cBgCc+PtoOFsYchwRIZr39+0XEX8uR/alYpYPJzGs/D0TW07fQ6iXAza+5MdJDISQ9mny81tnr/4jT+ZsYYihvWRTpDvT8ziOhhDNkzVPzgcAtnUMF+Q1qw5cKcT96gbO4iCEPFuUVHVx8mm/xPOyYqiEdGV7LhWgplECNytDjTRPVpaHoym8nc3QJGGw8zx9oSGku6Ckqoub4GkPQ6EAd0prkJ79gOtwCNEo+RV3MwNcNNI8WRXykbK4szn0hYaQboKSqi7OSKSHCZ6yxbrbqcky6cJulVTj7F1Z8+QZWnBhxhQfRxjoC3CzuJq+0BDSTVBS1Q2E+zsBAP64UID6JgnH0RCiGQnnZF8aRmm4ebKyTMT6CG25+jAulWpWEdIdUFLVDTzX0wpO5gaoamjGgSvUk4x0Pc0SKXa0NBDnojZVR2a3TAH+cbEAVfVNHEdDCNE0Sqq6AT6fh+l+stGqHTQFSLqgI1klKKlqgJWREGP6a755srL8XS3Q28YIdU0S7L5QwHU4hBANo6Sqm5je0rbm+I0SFFXWcxwNIeoV36p5slBPe97WeDweu2A9/mw2x9EQQjRNe959iEb1tDZCgKsFpAzoEm/SpSg0T+awNlVHpvs5Q4/Pw4XcClzJr+Q6HEKIBlFS1Y3Ia1btSMulS7xJl7HrvKx5sq+LOfo+w+bJyrI2FuEFD1nrrW3UZJmQLo2Sqm4k1NsBIj0+bhRX41JeBdfhEPLUGIZB/FlZosJlBfUnkce283weXYFLSBdGSVU3YirWx7iB9gBowTrpGtKzyzltnqysEX1s4GgmRkVdE/ZfLuQ6HEKIhlBS1c2Et1wF+PuFfDQ00zdmotu2tYxSTfRygIlYn+NoOibg8zAzQL5gnaYACemqKKnqZkb0sYGdqQjltU043LK4lxBd1Lp58qwA7iuoP8nMAGfweMCpW/dx734N1+EQQjSAkqpuRsDnIWyQbLRqexpdBUh0V+vmyYE9LbkO54mcLQwxoo8NAFqwTkhXRUlVNzSjpWbVkaxi3K9u4DgaQjpHm5onK0teYT3hXC6aJVKOoyGEqBslVd1QHzsTeDuboVnK4PeMfK7DIURlt1s1Tw730/6pP7ngAXawNBKiuKoBR7JKuA6HEKJmlFR1U/IPInm/NEJ0ybZWzZPtzbhvnqwsoR6fvVgkjhasE9LlUFLVTU3xcYS+gIfL+ZW4WkBVnonuUGyerDujVHLymlWHs4qpZRQhXQwlVd2UhZEQY/vLqjxTzSqiSxSbJ9txHY7K3G1NEOBqAYmUwXZ67RHSpVBS1Y3J29bsysinRbNEZ2zT0ubJqpCPVm07lwOplFpGEdJV6OY7ElGLUf1sYGUkRGl1A47doEWzRPuVVDVodfNkZYV6O8BYpId792tx5s59rsMhhKgJJVXdmL6Ajym+jgCAHVSziuiAnedz0azFzZOVZSjUw2Qf2WuPKqwT0nVQUtXNzWiZAjx4pQgVtU0cR0NIxxiGYa/6mxWgu6NUcvKaVfsyC+m1R0gXQUlVNzfQ0Qz97U3QKJEi6SLVrCLaKz27HDeLqyHW52Oyj/Y2T1aWt3PLa69Zil0ZNFJMSFdASRVhR6voKkCizeQV1LW9ebKyeDweO1r1W2o2GIYWrBOi6yipIpjq6wQBn4eMnHLcKqnmOhxC2qhpaMbuC7KR1IguMPUnF9ZyBeO1wipcyqvgOhxCyFOipIrAxkSEoL6yRq80WkW00V4da56sLHNDISZ42gOgCuuEdAWUVBEAD6cAd57Pg4Tq5hAts00HmycrS16zKikjH7WNzRxHQwh5GpRUEQDA2AG2MDPQR0FFPU7dKuU6HEJYuto8WVnP9bSCq5UhqhuasediAdfhEEKeAiVVBAAg0hOwV1TRFCDRJgktf49BfW10qnmysvh8HlsigmpWEaLbOE+qNm7cCDc3N4jFYgwZMgSpqakd7jtq1CjweLw2t9DQUHaf6upqLF26FM7OzjAwMICHhwe+//77Nsc6ffo0xowZAyMjI5iammLkyJGoq6vTyHPUFfJRgOTLhaiqp7o5hHvNEimb5EfocAX1J5nh7wwBn4dz9x7gZnEV1+EQQjqJ06QqPj4eUVFRWLlyJdLT0+Hj44OQkBAUFxe3u39iYiIKCgrYW2ZmJgQCAWbOnMnuExUVheTkZGzduhVXr17FsmXLsHTpUiQlJbH7nD59GuPHj8e4ceOQmpqKs2fPYunSpeDzOc8xOeXrYo5eNkaob5Ji36VCrsMhBEevl6BYh5snK8vOVIzR/WwB0GgVIbqM0yxi7dq1WLRoESIjI9kRJUNDQ2zatKnd/S0tLWFvb8/eDh48CENDQ4Wk6tSpU5g/fz5GjRoFNzc3LF68GD4+PgojYO+++y7efvttLF++HAMHDkS/fv0wa9YsiEQijT9nbcbj8dgF69vTaQqQcE+eYOhy82RlyWtW7UjPQ2MzNTgnRBdx9i7V2NiItLQ0BAcHPwyGz0dwcDBOnz6t1DFiY2Mxe/ZsGBkZsduGDRuGpKQk5OXlgWEYHD58GNevX8e4ceMAAMXFxfjrr79ga2uLYcOGwc7ODkFBQThx4oR6n6COmjbICTwekHqnDNn3a7kOh3RjrZsnz+xCtak6MqqfDWxNRCiracSfV4u4DocQ0gmcJVWlpaWQSCSws1Mc0rezs0Nh4ZOnnlJTU5GZmYmFCxcqbN+wYQM8PDzg7OwMoVCI8ePHY+PGjRg5ciQA4Pbt2wCAjz/+GIsWLUJycjL8/PwwduxY3Lhxo8PzNTQ0oLKyUuHWFTmYGWC4uzUAYAeNVhEOyZsn+7iYo5+97jZPVpaegI+ZAbKRYqpZRYhu0tnx9NjYWHh5eSEwMFBh+4YNG3DmzBkkJSUhLS0Na9aswZtvvok///wTACCVyobVlyxZgsjISAwaNAhfffUV+vXr1+G0IwDExMTAzMyMvbm4dN1vzvIF64nncyGlmlWEA62bJ3elCupPIr8K8PiNEuQ+oJFiQnQNZ0mVtbU1BAIBiooUh7mLiopgb2//2MfW1NQgLi4OCxYsUNheV1eHDz/8EGvXrsXkyZPh7e2NpUuXIiIiAqtXrwYAODjIygZ4eHgoPHbAgAHIzs7u8JzR0dGoqKhgbzk5XfebZMhAexiL9JBTVoezd8u4Dod0Q+dzHjZPntQFmicry9XKCEN7WYFhgIRzNFJMiK7hLKkSCoXw9/dHSkoKu00qlSIlJQVDhw597GMTEhLQ0NCAuXPnKmxvampCU1NTm6v4BAIBO0Ll5uYGR0dHZGVlKexz/fp1uLq6dnhOkUgEU1NThVtXZSAUINSrpWYVTQESDmw7+7B5smkXaJ6sitmBstGq7Wm51N2AEB3D6fRfVFQUfvzxR2zZsgVXr17F66+/jpqaGkRGRgIA5s2bh+jo6DaPi42NRVhYGKysrBS2m5qaIigoCB988AGOHDmCO3fu4KeffsLPP/+MadOmAZBd4fbBBx/g66+/xvbt23Hz5k189NFHuHbtWpuRr+4svOUqwD0XC6h1BnmmahsfNk+e1Y2m/uRCBtrDzEAfeeV1OHGTuhsQokv0uDx5REQESkpKsGLFChQWFsLX1xfJycns4vXs7Ow2o05ZWVk4ceIEDhw40O4x4+LiEB0djZdeegllZWVwdXXFqlWr8Nprr7H7LFu2DPX19Xj33XdRVlYGHx8fHDx4EL1799bck9Uxg90s0MPSENlltdh/uRDTBnW99iBEO+25+LB58pAu1DxZWWJ9AaYNcsJPp+4i/mw22+ycEKL9eAzD0PhyJ1RWVsLMzAwVFRVddipw3Z/Xse7PGxjubo2tC4dwHQ7pJmZ9fxqpd8vwQUg/vDnanetwOHG1oBIT1h+HvoCH09FjYW3cvWvoEaJOmvz81tmr/4jmya8CPHmrFPnl3buFD3k2bpdUI/VuWZdtnqysAQ6m8HE2Q5OEwc70PK7DIYQoiZIq0iEXS9n0C8MAO8/TGzvRvK7ePFkVEYN7AADizmaDJhQI0Q2UVJHHki9Y35GWS2/sRKNaN0/ujgvUHzXZxwEG+gLcKqlB2r0HXIdDCFECJVXksSZ6yd7Yb5fW4HxOOdfhkC5M3jzZ0kiIsQO6bvNkZZmI9THJW1bahCqsE6IbKKkij2Us0sN4T1kxVvkoAiGasO1c92merCx5zao9FwtQWd/EcTSEkCehdy7yRPIFw7sv5KO+ScJxNKQrKqlqQMpVWfNkmvp7yK+HBdxtjVHXJGFrdxFCtJfKSZW8ITHpPob2toKjmRiV9c3sBx8h6rTrfF63ap6sLB6Ph9mDZUlmPE0BEqL1VE6q3N3dMXr0aGzduhX19fWaiIloGQGfh2l+TgCA7Wn0xk7US9Y8WfZ3NSug+5ZR6Mi0QU7QF/BwMbcCl/MruA6HEPIYKidV6enp8Pb2RlRUFOzt7bFkyRKkpqZqIjaiRaa3TAEeu1GK4ipKpon6nM8px42W5smTfRy5DkfrWBmL8IKHbOH+NhqtIkSrqZxU+fr6Yv369cjPz8emTZtQUFCA4cOHw9PTE2vXrkVJSYkm4iQc621jjEE9zCGRMvj9PK3tIOqT0DJKNdGz+zVPVpa8ZtXO83m0rpEQLdbphep6enqYPn06EhIS8MUXX+DmzZt4//334eLignnz5qGgoECdcRItIF+wviOdalYR9ZA1T5a9V8waTAvUOzLc3RpO5gaorG/G/suFXIdDCOlAp5Oqc+fO4Y033oCDgwPWrl2L999/H7du3cLBgweRn5+PqVOnqjNOogUmeztCqMfHtcIqXM6v5Doc0gXsvVSI6oZmuHbT5snKEvB5mNmy3iwulaYACdFWKidVa9euhZeXF4YNG4b8/Hz8/PPPuHfvHv7973+jZ8+eGDFiBH766Sekp6drIl7CITNDfXZtx3aqWUXUQL5GaKa/M3g8HsfRaLeZAS7g8YDTt+/jbmkN1+EQQtqhclL13Xff4cUXX8S9e/ewa9cuTJo0CXy+4mFsbW0RGxurtiCJ9pjRMgWYdCEfjc1SjqMhukyhebI/XfX3JE7mBhjZxwbAw0KphBDtonJSdePGDURHR8PBwaHDfYRCIebPn/9UgRHtNKKPNWxMRCiracSRLKpZRTpPPto5sq8NHMwMOI5GN8hrViWk5aJZQl9qCNE2KidVmzdvRkJCQpvtCQkJ2LJli1qCItpLT8BHmK/ssvcd6TQFSDqnWSJlk6oIqqCutLED7GBlJERJVQMOZ9GV1oRoG5WTqpiYGFhbW7fZbmtri88++0wtQRHtJp+qOXStGGU1jRxHQ3TRsRvUPLkzhHp89vUXfzab42gIIY9SOanKzs5Gz54922x3dXVFdja9yLuD/vam8HQyRZOEQVJGHtfhEB0kb7lCzZNVJ++NeOhaMQorqBAvIdpE5XczW1tbXLx4sc32CxcuwMrKSi1BEe33sGYVJVVENaXV1Dz5abjbGmOwmwWkDE3BE6JtVE6q5syZg7fffhuHDx+GRCKBRCLBoUOH8M4772D27NmaiJFooSk+jtDj83AprwLXi6q4DofokJ3pLc2Tnc2oeXInySusx5/NgVRKhXgJ0RYqJ1X/+te/MGTIEIwdOxYGBgYwMDDAuHHjMGbMGFpT1Y1YGYswur8tAGAH1awiSlJonkwV1Dttopc9TER6yC6rxZnb97kOhxDSQuWkSigUIj4+HteuXcMvv/yCxMRE3Lp1C5s2bYJQKNREjERLyacAd57Po8u7iVIyqHmyWhgK9TCl5SrcOGqyTIjW0OvsA/v27Yu+ffuqMxaiY8b0t4WFoT6Kqxpw/GYpRvez5TokouW2UfNktYkY7IJf/spG8uVClNc2wtyQvtQSwrVOJVW5ublISkpCdnY2GhsVL6lfu3atWgIj2k+ox8dUXyf8dOoudqTlUlJFHqt18+SZtED9qXk5mWGAgymuFlRi1/k8vPJ826uyCSHPlspJVUpKCqZMmYJevXrh2rVr8PT0xN27d8EwDPz8/DQRI9Fi4X7O+OnUXRy4UoSKuiaYGdDoA2lf6+bJz/Wi5slPi8fjYfZgF6xMuoy4szmYP8yN+icSwjGV11RFR0fj/fffx6VLlyAWi7Fjxw7k5OQgKCgIM2fO1ESMRIt5Opmir50xGpul2HOxgOtwiBaTT/1R82T1CfOV1fm6VliFi7kVXIdDSLenclJ19epVzJs3DwCgp6eHuro6GBsb49NPP8UXX3yh9gCJduPxeK1qVtFVgKR9d0prkHqHmierm5mhPiZ62gOgBeuEaAOVkyojIyN2HZWDgwNu3brF3ldaWqq+yIjOmDbICXwekHbvAW6XVHMdDtFCCS2jVNQ8Wf3kNauSMvJQ09DMcTSEdG8qJ1XPPfccTpw4AQCYOHEi3nvvPaxatQqvvvoqnnvuObUHSLSfrakYI/vaAAASqcI6eUSzRMqOYlIFdfV7rpcl3KwMUdMowZ5LNAVPCJdUTqrWrl2LIUOGAAA++eQTjB07FvHx8XBzc0NsbKzaAyS6oXXNKqrwTFo7dqMERZWy5snB1DxZ7Xg8HltINZ6mAAnhlEpX/0kkEuTm5sLb2xuAbCrw+++/10hgRLe84GEHE7Ee8srrcOb2fQxzt+Y6JKIltp2VjVLJF1UT9Zvh54w1B64j7d4D3CiqQh87av9DCBdUeocTCAQYN24cHjx4oKl4iI4S6wswyVtW4Xk7LVgnLUqrG/Dn1SIAsmKVRDNsTcUY09I2ikarCOGOyl8bPT09cfv2bbUGsXHjRri5uUEsFmPIkCFITU3tcN9Ro0aBx+O1uYWGhrL7VFdXY+nSpXB2doaBgQE8PDw6HFFjGAYTJkwAj8fDrl271Pq8upsZLVd17WupR0TIrvPUPPlZmd2StCaez0NDs4TjaAjpnlROqv7973/j/fffxx9//IGCggJUVlYq3FQVHx+PqKgorFy5Eunp6fDx8UFISAiKi4vb3T8xMREFBQXsLTMzEwKBQKFGVlRUFJKTk7F161ZcvXoVy5Ytw9KlS5GUlNTmeOvWraOaOWri18McPa2NUNckwT5aMNvtMQzDjppQBXXNC+prAztTEcpqGvHnlfbfPwkhmqVyUjVx4kRcuHABU6ZMgbOzMywsLGBhYQFzc3NYWFioHMDatWuxaNEiREZGsiNKhoaG2LRpU7v7W1pawt7enr0dPHgQhoaGCknVqVOnMH/+fIwaNQpubm5YvHgxfHx82oyAZWRkYM2aNR2ei6hGVrPKCQDVrCIPmyeL9Phs81+iOXoCPmb6y5LXuLPZHEdDSPekcpuaw4cPq+3kjY2NSEtLQ3R0NLuNz+cjODgYp0+fVuoYsbGxmD17NoyMjNhtw4YNQ1JSEl599VU4OjriyJEjuH79Or766it2n9raWrz44ovYuHEj7O3tn3iehoYGNDQ0sP/uzKhcdzDNzxlrDl7HmdtlyCmrhYulIdchEY5sOydLrCd6UfPkZ2VWgAu+OXwTJ26W0uuPEA6onFQFBQWp7eSlpaWQSCSws1O8zNrOzg7Xrl174uNTU1ORmZnZppTDhg0bsHjxYjg7O0NPTw98Ph8//vgjRo4cye7z7rvvYtiwYZg6dapSscbExOCTTz5Rat/uzMncAEN7WeHUrfvYeT4Pb4/tw3VIhAOy5sn5AKg21bPUw8oQz7tb4eTN+0hIy0XUC325DomQbkXlpOrYsWOPvb914qJpsbGx8PLyQmBgoML2DRs24MyZM0hKSoKrqyuOHTuGN998E46OjggODkZSUhIOHTqE8+fPK32u6OhoREVFsf+urKyEiwt9WLRnhr8zTt26jx3puXhrjDutWeuG5Bcr9LA0xJCe1Dz5WZoV4IKTN+9j+7kcvDO2DwR8ev0R8qyonFSNGjWqzbbWH5oSifJXnVhbW0MgEKCoqEhhe1FR0ROn5GpqahAXF4dPP/1UYXtdXR0+/PBD7Ny5k70i0NvbGxkZGVi9ejWCg4Nx6NAh3Lp1C+bm5gqPDQ8Px4gRI3DkyJE25xOJRBCJREo/t+5svKc9PtqViXv3a3Hu3gMMdqMP1e4mvqUtzawAZ/DpQ/2ZChloDzMDfeRX1OP4jRKM6mfLdUiEdBsqL1R/8OCBwq24uBjJyckYPHgwDhw4oNKxhEIh/P39kZKSwm6TSqVISUnB0KFDH/vYhIQENDQ0YO7cuQrbm5qa0NTUBD5f8akJBAJIpVIAwPLly3Hx4kVkZGSwNwD46quvsHnzZpWeA2nLUKiHCV4OAIAdabRgvbuRN0/mUfNkToj1BZg2SHbBCNWsIuTZUnmkyszMrM22F154AUKhEFFRUUhLS1PpeFFRUZg/fz4CAgIQGBiIdevWoaamBpGRkQCAefPmwcnJCTExMQqPi42NRVhYGKysrBS2m5qaIigoCB988AEMDAzg6uqKo0eP4ueff8batWsBgL1y8FE9evRAz549VYqftC/czxnb03Kx52IBPp4yEGJ9AdchkWeEbZ7ch5oncyVisAt+OnUXB68UobS6AdbGNMpOyLOgclLVETs7O2RlZan8uIiICJSUlGDFihUoLCyEr68vkpOT2cXr2dnZbUadsrKycOLEiQ5HxuLi4hAdHY2XXnoJZWVlcHV1xapVq/Daa6+p/sRIpwzpaQkncwPklddh/+VCTPV14jok8gy0bp5MFdS5M8DBFD4u5riQU47E9FwsHtmb65AI6RZ4DMOo1P324sWLCv9mGAYFBQX4/PPP0dzcjBMnTqg1QG1VWVkJMzMzVFRUwNTUlOtwtNLag9fxdcoNjOhjjf8tGMJ1OOQZOHytGJE/nYWFoT7OfDgWIj0aoeTKb6nZiE68hF42RkiJCqILRghpocnPb5VHqnx9fcHj8fBoLvbcc89REU2iINzPCV+n3MDJm6UorKiHvZmY65CIhsnX8Ewb5EwJFccm+zjiX39cwe2SGrpghJBnROWk6s6dOwr/5vP5sLGxgVhMH5hEkauVEQa7WeDs3QfYeT4Pr4+iKYiu7H6r5smzBtMCda4Zi/QwydsB287lIi41h5IqQp4Bla/+c3V1Vbi5uLhQQkU6FO4n+3DdkZ7bZnSTdC07WzVP7m9PU+LaIGJwDwDAnkv5qKxv4jgaQro+lZOqt99+G19//XWb7d988w2WLVumjphIFzLR2wEiPT5uFlfjYm4F1+EQDaHmydrJr4c5+tgao75JiqSMfK7DIaTLUzmp2rFjB55//vk224cNG4bt27erJSjSdZiK9THeU1a+YjvVrOqyLuRWsM2TJ/tQ82RtwePx2KswqWYVIZqnclJ1//79dmtVmZqaorS0VC1Bka5FPgWYdCEfDc3KV9wnukP+gT3RywFmBtQ8WZtM93OGvoCHS3kVyMyj0WJCNEnlpMrd3R3Jyclttu/btw+9evVSS1Cka3ne3Rr2pmJU1DXh0NVirsMhalbXKGGbJ88MoAXq2sbSSIhxA2WjxdvO0WgVIZqkclIVFRWFv/3tb1i5ciWOHj2Ko0ePYsWKFVi+fDneffddTcRIdJyAz0NYS9sMeWFI0nXsvVTANk9+rqfVkx9AnrnZLVOAu87nob6JRosJ0RSVSyq8+uqraGhowKpVq/Cvf/0LAODm5obvvvsO8+bNU3uApGuY4e+E74/ewuGsEpRUNcDGhNpmdBXy0Y+Z/tQ8WVs939ua7XCQnFnIfskhhKiXyiNVAPD6668jNzcXRUVFqKysxO3btymhIo/lbmsCHxdzSKQMfs/I4zocoiZ3S2vwFzVP1np8Pg+zWq7KjDubzXE0hHRdKidVd+7cwY0bNwAANjY2MDY2BgDcuHEDd+/eVWtwpGuZ4SefAqSkqqtISHvYPNnRnJona7OZAc7g8YAzt8twp7SG63AI6ZJUTqpeeeUVnDp1qs32v/76C6+88oo6YiJd1GQfRwgFfFwtqMSV/EquwyFPSSJl2DIZs6g2ldZzNDdAUF8bALRgnRBNUTmpOn/+fLt1qp577jlkZGSoIybSRZkbCjF2gC0AWrDeFRy7XoKiygZYGOoj2MOW63CIEuQL1ren5aJJIuU4GkK6HpWTKh6Ph6qqqjbbKyoqIJHQVSXk8eQ1q37PyKM3dR0nH+0IG+REzZN1xJj+drA2FqKkqgGHr1F5E0LUTeWkauTIkYiJiVFIoCQSCWJiYjB8+HC1Bke6nqB+NrA2FqK0uhHHrpdwHQ7ppNbNk+UVu4n2E+rx2S82VGGdEPVTuaTCF198gZEjR6Jfv34YMWIEAOD48eOorKzEoUOH1B4g6Vr0BXxM9XVC7Ik72J6Wi7ED7LgOiXTCzvN5aJIw8KbmyTpn1mAX/OfYbRzOKkZhRT3szcRch0RIl6HySJWHhwcuXryIWbNmobi4GFVVVZg3bx6uXbsGT09PTcRIuhj5N+WUq8Uor23kOBqiKoZhHtamogXqOqe3jTEC3SwhZYDtaTRaRYg6dapOlaOjIz777DPs2bMH27dvx4oVK8Dn8/HNN9+oOz7SBXk4mmKAgykaJVK2vQnRHRdyK3C9SNY8eQo1T9ZJbJPlczmQShmOoyGk6+hUUtVaSkoKXnzxRTg4OGDlypXqiIl0A+EtNau2U80qnSMfpZrgaU/Nk3XURC8HmIj0kFNWh9O373MdDiFdRqeSqpycHHz66afo2bMnxo0bBwDYuXMnCgsL1Roc6brCBjlBj8/DhZxy3CxuezUp0U51jRLszpCNLs6iBeo6y0AowNRBslHGOFqwTojaKJ1UNTU1ISEhASEhIejXrx8yMjLw5Zdfgs/n45///CfGjx8PfX361kqUY20swqh+skKE29NotEpX7MssQFVDM1wsDah5so6bPbgHAGB/ZiEe1NDaRkLUQemkysnJCRs2bEB4eDjy8vKQmJiIGTNmaDI20sXJF6zvPJ8LCa3r0Anyy/Bn+btQ82Qd5+lkhoGOsrWNu6gfJyFqoXRS1dzcDB6PBx6PB4GACv2RpzdmgC3MDPRRVNmAkzdLuQ6HPAE1T+565AvW41JzwDD0xYaQp6V0UpWfn4/Fixfjt99+g729PcLDw7Fz507wePRtlXSOSE/AXj1GbWu0n7x58ghqntxlTPVxgkiPj6yiKlzIreA6HEJ0ntJJlVgsxksvvYRDhw7h0qVLGDBgAN5++200Nzdj1apVOHjwILWpISqb0TLisf9yISrrmziOhnSkdfPkCKpN1WWYGepjopcDACD+bDbH0RCi+zp19V/v3r3x73//G/fu3cOePXvQ0NCASZMmwc6OqmMT1Xg7m8Hd1hj1TVLsvVjAdTikA8duUPPkrko+BZiUkY+ahmaOoyFEtz1VnSo+n48JEyZg+/btyM3NxYcffqiuuEg3wePx2AXrNAWovbadpebJXdWQnpZwszJETaMEe+iLDSFP5amLf8rZ2NggKipKXYcj3ci0QU7g84Czdx/g3v0arsMhj2jdPHkWTf11OTweDxEt5RXiaAqQkKeitqSKkM6yNxPjeXdrAMAOqrCudVo3Tx7gQM2Tu6JwfycI+DykZ5fjehEV4yWksyipIlpBvmA9MT2XepFpEWqe3D3Ymogxtr9srVw8VVgnpNMoqSJaIWSgPUxEesh9UIe/7pRxHQ5pQc2Tu4/ZgbKkOTE9Fw3NdCU3IZ2hFUnVxo0b4ebmBrFYjCFDhiA1NbXDfUeNGsUWIW19Cw0NZfeprq7G0qVL4ezsDAMDA3h4eOD7779n7y8rK8Nbb72Ffv36wcDAAD169MDbb7+Nigqq08IVsb4Aod6yS7tpwbr2oObJ3cfIPjawNxXjQW0TDl4p4jocQnSSnqoPkEgk+Omnn5CSkoLi4mJIpVKF+w8dOqTS8eLj4xEVFYXvv/8eQ4YMwbp16xASEoKsrCzY2ra9dDsxMRGNjQ/7VN2/fx8+Pj6YOXMmuy0qKgqHDh3C1q1b4ebmhgMHDuCNN96Ao6MjpkyZgvz8fOTn52P16tXw8PDAvXv38NprryE/Px/bt29X8SdC1CXc3xlxZ3Ow71IBPp06EIZClf88iRopNE+mqb8uT0/Ax8wAZ2w4dBPxZ3MwyZtGJglRlcojVe+88w7eeecdSCQSeHp6wsfHR+GmqrVr12LRokWIjIxkR5QMDQ2xadOmdve3tLSEvb09ezt48CAMDQ0VkqpTp05h/vz5GDVqFNzc3LB48WL4+PiwI2Cenp7YsWMHJk+ejN69e2PMmDFYtWoVdu/ejeZmqtPClQBXC7i2XNqdnFnIdTjdnkLz5F7UPLk7kCfPx2+UIqesluNoCNE9Kg8FxMXFYdu2bZg4ceJTn7yxsRFpaWmIjo5mt/H5fAQHB+P06dNKHSM2NhazZ8+GkZERu23YsGFISkrCq6++CkdHRxw5cgTXr1/HV1991eFxKioqYGpqCj09Gh3hirxm1dqD17EjPRfT/ai/HJfYBerUPLnbcLE0xHB3a5y4WYqEczmIGteP65AI0Skqj1QJhUK4u7ur5eSlpaWQSCRtKrHb2dmhsPDJIxWpqanIzMzEwoULFbZv2LABHh4ecHZ2hlAoxPjx47Fx40aMHDmywzj+9a9/YfHixR2eq6GhAZWVlQo3on7TBjkBAE7duo+88jqOo+m+7t2vwZnbsubJM6h5crcir7CekJYLCV2JS4hKVE6q3nvvPaxfv14rOprHxsbCy8sLgYGBCts3bNiAM2fOICkpCWlpaVizZg3efPNN/Pnnn22OUVlZidDQUHh4eODjjz/u8FwxMTEwMzNjby4utMZEE1wsDfFcL0swDLCTFqxzJuGc7GdPzZO7n3ED7WBuqI+Cinocu1HCdTiE6BSV57pOnDiBw4cPY9++fRg4cCD09RWvCEpMTFT6WNbW1hAIBCgqUrzSpKioCPb29o99bE1NDeLi4vDpp58qbK+rq8OHH36InTt3slcEent7IyMjA6tXr0ZwcDC7b1VVFcaPHw8TExPs3LmzzXNpLTo6WqFifGVlJSVWGhLu54wzt8uwIz0Pb452B49HU0/PUuvmybMCaJSquxHpCTBtkBM2n7yL+NQcjO5HvR4JUZbKI1Xm5uaYNm0agoKCYG1trTB6Y2ZmptKxhEIh/P39kZKSwm6TSqVISUnB0KFDH/vYhIQENDQ0YO7cuQrbm5qa0NTUBD5f8akJBAKFKxUrKysxbtw4CIVCJCUlQSwWP/Z8IpEIpqamCjeiGRO8HGCgL8Cd0hqkZ5dzHU63c+xGCQor62FuqI8XPKhJencknwL882oRSqoaOI6GEN2h8kjV5s2b1RpAVFQU5s+fj4CAAAQGBmLdunWoqalBZGQkAGDevHlwcnJCTEyMwuNiY2MRFhYGKyvFq5JMTU0RFBSEDz74AAYGBnB1dcXRo0fx888/Y+3atQAeJlS1tbXYunWrwhopGxsbCATUMJZLxiI9TPCyR2J6Hnak58Lf1YLrkLqVhJYF6mG+1Dy5u+pvbwpfF3Nk5JQjMT0XS4J6cx0SITqh05e6lZSUICsrCwDQr18/2NjYdOo4ERERKCkpwYoVK1BYWAhfX18kJyezi9ezs7PbjDplZWXhxIkTOHDgQLvHjIuLQ3R0NF566SWUlZXB1dUVq1atwmuvvQYASE9Px19//QUAbRbd37lzB25ubp16LkR9Zvg5IzE9D7sv5GPFJA+I9enD/Vm4X93AFn6k2lTd2+zBLsjIKUf82RwsHtmLpuEJUQKPUXHFeU1NDd566y38/PPP7HSaQCDAvHnzsGHDBhgaGmokUG1TWVkJMzMzthQDUS+plMGI/zuMvPI6bJgzCJOpRcozEXviDv71xxV4OZlh91vDuQ6HcKi6oRmBq/5EbaME25YMRWBPS65DIkQtNPn5rfKaqqioKBw9ehS7d+9GeXk5ysvL8fvvv+Po0aN477331Boc6b74fB5bXoHa1jwbDMNgW0sz3VmDaZSquzMW6WFyS1X1uLPZHEdDiG5QOanasWMHYmNjMWHCBHbB9sSJE/Hjjz9SixeiVtP9ZEnVseslKK6s5ziaru9ibgWyiqqoeTJhRbQ0Wd57qQAVdU0cR0OI9lM5qaqtrW1TrBMAbG1tUVtLbQ2I+vSyMYa/qwWkDLArI4/rcLq8+JYF6uOpeTJpMcjFHH3tjFHfJEXShXyuwyFE66mcVA0dOhQrV65Eff3DkYO6ujp88sknTyyDQIiqwlta1WxPy9WKgrNdVevmyRG0QJ204PF4iBjcAwAQT1OAhDyRyknV+vXrcfLkSTg7O2Ps2LEYO3YsXFxccOrUKaxfv14TMZJuLNTbAUI9Pq4XVSMzj1oDaUryZWqeTNo3bZAThAI+MvMqkZlXwXU4hGg1lZMqT09P3LhxAzExMfD19YWvry8+//xz3LhxAwMHDtREjKQbMzPQx7iWApS0YF1z4s9S82TSPksjIcYNlL0G5X8nhJD2dapOlaGhIRYtWqTuWAhpV7i/M/64WIDfM/Lw4cQBEOqp/F2APEbr5snh1DyZtGP24B7442IBdmXk4R+hA6huHCEdUCqpSkpKwoQJE6Cvr4+kpKTH7jtlyhS1BEaI3Ah3a9iaiFBc1YDDWcUIGfj4vpBENfLmycPdreFEzZNJO4b1toKzhQFyH9RhX2YBpg2i5JuQ9iiVVIWFhaGwsBC2trYICwvrcD8ejweJRKKu2AgBAOgJ+Jg2yAn/OXYb29NyKalSo9bNkyOoNhXpAJ/PQ0SAC9YcvI641BxKqgjpgFLzKFKpFLa2tuz/d3SjhIpoinxa6vC1Ytyvpgav6kLNk4myZgQ4g88D/rpThtsl1VyHQ4hWUnlxys8//4yGhrYfao2Njfj555/VEhQhj+prZwIvJzM0Sxmql6NG1DyZKMvBzABBfWU9Xredo4tGCGmPyklVZGQkKiraXlZbVVWFyMhItQRFSHvC/ahtjTqV1TRS82SiEnnNqu1puWiSSDmOhhDto3JSxTBMu93Kc3NzYWZmppagCGnPFF8n6At4yMyrxLVCqln1tHaez0OThIGXkxk8HKkpOHmysQNsYW0sRGl1Aw5dK+Y6HEK0jtIlFQYNGgQejwcej4exY8dCT+/hQyUSCe7cuYPx48drJEhCAFm9nDH9bbH/chF2pOXiH6EeXIeksxiGYaf+ZgXQomOiHH0BH+H+zvjP0duIP5tDF40Q8gilkyr5VX8ZGRkICQmBsbExe59QKISbmxvCw8PVHiAhrYX7OWP/5SLsPJ+Pv4/vDz0B1azqjIu5FbhWWAWhHh9TfJy4DofokIgAF/zn6G0cySpGQUUdHMyoDAchckonVStXrgQAuLm5ISIiAmKxWGNBEdKRUf1sYWkkm344fqMUo/vbch2STtrWMko1wdMeZobUPJkor5eNMQJ7WiL1Thm2n8vFW2P7cB0SIVpD5a/58+fPp4SKcEY2suIIANhOC9Y7pa5RgqSW5sm0QJ10xuyWmmbx53IglVKjc0LkVE6qJBIJVq9ejcDAQNjb28PS0lLhRoimzWipWXXwShEqaps4jkb3yJsnO1sYYCg1TyadMMHTASZiPeQ+qMOpW/e5DocQraFyUvXJJ59g7dq1iIiIQEVFBaKiojB9+nTw+Xx8/PHHGgiREEUDHU3R394Ejc1S7L5INatUte2sbISPmieTzjIQChDmK1uLF3c2m+NoCNEeKidVv/zyC3788Ue899570NPTw5w5c/Df//4XK1aswJkzZzQRIyEKeDwewv1ko1VUs0o19+7X4PTt++DxZBWyCekseVujA5eL8KCmkeNoCNEOKidVhYWF8PLyAgAYGxuzhUAnTZqEPXv2qDc6QjowdZAjBHwezmeX4xa1zFCavM8fNU8mT8vTyQyeTqZolEix83we1+EQohVUTqqcnZ1RUFAAAOjduzcOHDgAADh79ixEIpF6oyOkA7YmYozsYw0ASKTRKqW0bp5MC9SJOsgrrMefzQHD0IJ1QlROqqZNm4aUlBQAwFtvvYWPPvoIffr0wbx58/Dqq6+qPUBCOjLDX5YYJKbnQUJXID3R8RslKKiQNU8eN5CaJ5OnN8XHEWJ9PrKKqpCRU851OIRwTuk6VXKff/45+/8RERHo0aMHTp8+jT59+mDy5MlqDY6Qxxk7wBamYj0UVNTj9K37GN4yckXal9DSBJeaJxN1MTPQx0RPBySez0P82RwM6mHBdUiEcOqpy1EPHToUUVFRlFCRZ06sL8DklppVtGD98cpqGnHgSiEAmvoj6iVfsJ50IR/VDc0cR0MIt5QaqUpKSlL6gFOmTOl0MISoKtzfGb/8lY3kzEL8K6wZxiKVB1+7hV0tzZM9nUypeTJRq8CeluhpbYQ7pTXYczGfXWdFSHek1CeQvO+fHI/Ha7MokceT1buRSCTqiYwQJQxyMUcvayPcLq3B3ksFNArTDoZh2LY0EfTzIWrG4/EQMdgFn++7hrizOZRUkW5Nqek/qVTK3g4cOABfX1/s27cP5eXlKC8vx759++Dn54fk5GRNx0uIAh6Ph/CWCus70mgKsD2X8qh5MtGs6X5O0GspcZJVWMV1OIRwRuU1VcuWLcP69esREhICU1NTmJqaIiQkBGvXrsXbb7+tiRgJeazpfk7g8YC/7pQhp6yW63C0TvxZ2SjV+IHUPJlohq2JGGMHyJqby//eCOmOVE6qbt26BXNz8zbbzczMcPfuXTWERIhqHMwM8Hxv2ZV/tGBdUevmyfIFxYRowuyWab/E87loaKZlIKR7UjmpGjx4MKKiolBUVMRuKyoqwgcffIDAwEC1BkeIssL9ZdNaiel5VISwlf2XC6l5MnkmRva1gb2pGOW1TThwuejJDyCkC1I5qdq0aRMKCgrQo0cPuLu7w93dHT169EBeXh5iY2M1ESMhTxQy0B5GQgGyy2px9u4DrsPRGvKpGGqeTDRNwOdhVks/SZoCJN2VykmVu7s7Ll68iN27d+Ptt9/G22+/jT/++AOXLl2Cu7t7p4LYuHEj3NzcIBaLMWTIEKSmpna476hRo8Dj8drcQkND2X2qq6uxdOlSODs7w8DAAB4eHvj+++8VjlNfX48333wTVlZWMDY2Rnh4uMLoG9EthkI9hHo7AKAF63LZ92upeTJ5pmYGuIDHA07cLKX1jaRb6lTxTx6Ph3HjxrFJ1QsvvMCWVFBVfHw8oqKisHLlSqSnp8PHxwchISEoLi5ud//ExEQUFBSwt8zMTAgEAsycOZPdJyoqCsnJydi6dSuuXr2KZcuWYenSpQr1tt59913s3r0bCQkJOHr0KPLz8zF9+vROPQeiHcL9ZInDnksFqGukNR0JabLRAmqeTJ4VF0tDDHeXrW9MOEejVaT74TFKLED5+uuvsXjxYojFYnz99deP3VfVKwCHDBmCwYMH45tvvgEgK9/g4uKCt956C8uXL3/i49etW4cVK1agoKAARkZGAABPT09ERETgo48+Yvfz9/fHhAkT8O9//xsVFRWwsbHBr7/+ihkzZgAArl27hgEDBuD06dN47rnnnnjeyspKmJmZoaKiAqamVExRG0ilDIJWH0ZOWR3WRfgibFD3LR8gkTIY/sUhFFTUY8OcQWzleUI07Y+L+Vj663nYm4pxcvkYCGjamWgZTX5+K1X886uvvsJLL70EsViMr776qsP9eDyeSklVY2Mj0tLSEB0dzW7j8/kIDg7G6dOnlTpGbGwsZs+ezSZUADBs2DAkJSXh1VdfhaOjI44cOYLr16+zsaelpaGpqQnBwcHsY/r378/2MWwvqWpoaEBDQwP778rKSqWfJ3k2+Hwepg9yxvqUG9iRntutkyp582QzA3284EHNk8mz84KHHSwM9VFYWY9j10swur8t1yER8swolVTduXOn3f9/WqWlpZBIJLCzU3zTt7Ozw7Vr1574+NTUVGRmZrZZIL9hwwYsXrwYzs7O0NPTA5/Px48//oiRI0cCAAoLCyEUCtuUhrCzs0NhYWG754qJicEnn3yiwrMjXAj3kyVVJ26WoqCiDg5m3XPaS948edogJ4j1qXkyeXZEegJM93NG7Ik7iDubTUkV6VaeuqEyl2JjY+Hl5dWmlMOGDRtw5swZJCUlIS0tDWvWrMGbb76JP//8s9Pnio6ORkVFBXvLyaH1Atqoh5UhAntagmGAnefzuA6HE62bJ8+kBeqEA/KaaClXi1FcVc9xNIQ8O0qNVEVFRSl9wLVr1yq9r7W1NQQCQZur7oqKimBvb//Yx9bU1CAuLg6ffvqpwva6ujp8+OGH2LlzJ3tFoLe3NzIyMrB69WoEBwfD3t4ejY2NKC8vVxitetx5RSIRRCKR0s+NcGeGnzNS75Rhe1ouXg/q3emLKHRV6+bJAx3NuA6HdEN97UwwqIc5zmeXIzE9D68F9eY6JEKeCaWSqvPnzyt1MFU/vIRCIfz9/ZGSksI2bZZKpUhJScHSpUsf+9iEhAQ0NDRg7ty5CtubmprQ1NQEPl9xEE4gEEAqlQKQLVrX19dHSkoKwsPDAQBZWVnIzs7G0KFDVXoORPtM8LLHiqRM3C6pQUZOOQb1sOA6pGemdfNkai5NuDR7sAvOZ5cj/mwOlozs1e2+3JDuSamk6vDhwxoLICoqCvPnz0dAQAACAwOxbt061NTUIDIyEgAwb948ODk5ISYmRuFxsbGxCAsLg5WVYpVoU1NTBAUF4YMPPoCBgQFcXV1x9OhR/Pzzz+wompmZGRYsWICoqChYWlrC1NQUb731FoYOHarUlX9Eu5mI9TF+oD12ZeRjR3put0qqWjdPnkrNkwmHJnk74tPdV3CntAapd8owhCr6k25AqaRKkyIiIlBSUoIVK1agsLAQvr6+SE5OZhevZ2dntxl1ysrKwokTJ3DgwIF2jxkXF4fo6Gi89NJLKCsrg6urK1atWoXXXnuN3eerr74Cn89HeHg4GhoaEBISgm+//VZzT5Q8U+H+ztiVkY/dFwrw0SQPiPS6x2Jt+SgVNU8mXDMS6WGyjyPizuYg/mwOJVWkW1CqTtWjzp07h23btiE7OxuNjY0K9yUmJqotOG1Gdaq0m0TK4PnPD6Gwsh7fvuSHiV4OXIekcfVNEgxe9Seq6puxdcEQDO9jzXVIpJs7n/0A0749BZEeH6n/CIaZASX6hHua/PxW+eq/uLg4DBs2DFevXsXOnTvR1NSEy5cv49ChQzAzo0WxRDsI+DxM95NNf23vJm1rkjMLUVXfDCdzAwzrTaMChHu+LuboZ2eChmYpkjK659W4pHtROan67LPP8NVXX2H37t0QCoVYv349rl27hlmzZqFHjx6aiJGQTgn3l5UTOHq9BCVVDU/YW/fJp/5mBjhT82SiFXg8HlteIY6aLJNuQOWk6tatW2ypAqFQiJqaGvB4PLz77rv44Ycf1B4gIZ3V28YYvi7mkEgZ/N7FvyVn36/FqVstzZP9qTYV0R7TBjlBKODjcn4lMvMquA6HEI1SOamysLBAVVUVAMDJyQmZmZkAgPLyctTWUldyol3ko1VdfQpwe6vmyc4WhhxHQ8hDFkZChHjK6v/Fnc3mOBpCNEvlpGrkyJE4ePAgAGDmzJl45513sGjRIsyZMwdjx45Ve4CEPI3J3g4QCvi4VliFy/ld81uyRMogoSVppNpURBvNbpkC/D0jH3WNEo6jIURzlE6q5CNS33zzDWbPng0A+Mc//oGoqCgUFRUhPDy8TQ8+QrhmbihkGwp31dEqWZ9Dap5MtNfQXlZwsTRAVX0z9mUWcB0OIRqjdFLl7e2NIUOGYMeOHTAxMZE9mM/H8uXLkZSUhDVr1sDCovsUWSS6I9xfdhVgUkY+miRSjqNRv20tC4DDfB2peTLRSnw+DxEBtGCddH1KJ1VHjx7FwIED8d5778HBwQHz58/H8ePHNRkbIWoxso8NrI1FuF/TiCNZJVyHo1atmyfPGkxTf0R7zfB3AZ8HpN4pw+2Saq7DIUQjlE6qRowYgU2bNqGgoAAbNmzA3bt3ERQUhL59++KLL75AYWGhJuMkpNP0BHyE+ToCAHZ0sSnA3zNkzZMHOlLzZKLd7M3EGNXPFgAQf45Gq0jXpPJCdSMjI0RGRuLo0aO4fv06Zs6ciY0bN6JHjx6YMmWKJmIk5KnJrwJMuVaEBzWNT9hbNzAMg/iWqZQIGqUiOkD+d7ojLbdLTsUTonJS1Zq7uzs+/PBD/POf/4SJiQn27NmjrrgIUasBDqYY6GiKJgmD3RfzuQ5HLTLzKtnmyVN8HLkOh5AnGtPfFtbGIpRWNyLlajHX4RCidp1Oqo4dO4ZXXnkF9vb2+OCDDzB9+nScPHlSnbERolbhfl2rZlX8OVnNn5CB9jA3FHIcDSFPpi/gs8Vp46lmFemCVEqq8vPz8dlnn6Fv374YNWoUbt68ia+//hr5+fn48ccf8dxzz2kqTkKe2lRfR+jxebiYW4EbRVVch/NU6psk+D1DNuIWQbWpiA6RTwEevV6C/PI6jqMhRL2UTqomTJgAV1dXbNiwAdOmTcPVq1dx4sQJREZGwsjISJMxEqIWVsYidqHs9nTdHq2i5slEV/W0NsKQnpaQMl1n1JgQOaWTKn19fWzfvh25ubn44osv0K9fP03GRYhGzGipWbXrfB4kUobjaDqPmicTXTY7UDZaFX82B1Idfh0S8iilk6qkpCRMnToVAgEVFyS6a0x/O1gY6qOosgEnbpZyHU6n5JRR82Si2yZ4OsBErIe88jqcvKWbr0NC2vNUV/8RomtaXymnq1MPCeeoeTLRbWJ9AaYNko0aU4V10pVQUkW6HXnNqgOXC1FZ38RxNKqRSBk2GZxJC9SJDpMvWD9wuRBlXaR2HCGUVJFux8vJDH1sjdHQLMWei7rV3PXEzVLktzRPHkfNk4kOG+hoBi8nMzRJGOw8n8d1OISoBSVVpNvh8XjsaJWuta2RL1Cn5smkK5CPVsWfzQbD0IJ1ovsoqSLd0rRBTuDzgHP3HuBuaQ3X4SjlQU0jDl4uAkBTf6RrmOLrCLE+H9eLqnE+p5zrcAh5apRUkW7JzlSMEX1sAAA7dKRm1a6MPDRKpBjoaApPJ2qeTHSfqVgfoV6yC0fiU2nBOtF9lFSRbks+BZiYnqf1tXJaN0+eRaNUpAuRTwHuvpiP6oZmjqMh5OlQUkW6rXEedmytnDN37nMdzmO1bp481ZeaJ5OuY7CbBXpZG6G2UYI/LnSNZuek+6KkinRbYn0BJnk7AAB2pGn31UfyBerUPJl0NTwejx2toppVRNdRUkW6NXlF8n2ZBajR0qmH+iYJdmXIkr5ZAVRBnXQ90/2cocfnISOnHNcKK7kOh5BOo6SKdGt+PSzQs2XqYV9mIdfhtGv/5YfNk5/vbc11OISonY2JCMEDZHXX4mm0iugwSqpIt8bj8TC9pV2Gttaskk/9zfCn5smk64poabK883we6pskHEdDSOdQUkW6vWl+sqTq9O37yH1Qy3E0inLKanHypqx58kya+iNd2Mg+NnAwE6O8tgkHrhRxHQ4hnUJJFen2nC0MMbSXFQBgZ7p2LVhPaBk9e743NU8mXZuAz2OL2safzeY4GkI6h5IqQvBwwfqO9FytaZchkTLY3jL1N2sw1aYiXd9Mf2fweMDJm/eRfV+7Ro0JUYZWJFUbN26Em5sbxGIxhgwZgtTU1A73HTVqFHg8XptbaGgou0979/N4PHz55ZfsPtevX8fUqVNhbW0NU1NTDB8+HIcPH9bo8yTaa7ynPQyFAty9X4u0ew+4DgcAcLKlebKpWI+aJ5NuwcXSEMPdZRdjJKTRgnWiezhPquLj4xEVFYWVK1ciPT0dPj4+CAkJQXFxcbv7JyYmoqCggL1lZmZCIBBg5syZ7D6t7y8oKMCmTZtkTXTDw9l9Jk2ahObmZhw6dAhpaWnw8fHBpEmTUFionVeAEc0yEulhgmdLzSotaVsTL2+ePMiJmieTbmP24B4AgIRzuWiWSDmOhhDVcJ5UrV27FosWLUJkZCQ8PDzw/fffw9DQEJs2bWp3f0tLS9jb27O3gwcPwtDQUCGpan2/vb09fv/9d4wePRq9evUCAJSWluLGjRtYvnw5vL290adPH3z++eeora1FZmbmM3neRPuE+8sWrP9xoYDzq49aN0+mtjSkOwn2sIWlkRCFlfU4dqOE63AIUQmnSVVjYyPS0tIQHBzMbuPz+QgODsbp06eVOkZsbCxmz54NIyOjdu8vKirCnj17sGDBAnablZUV+vXrh59//hk1NTVobm7Gf/7zH9ja2sLf3//pnhTRWc/1tIKTuQGqGpo5v/ro95bmyR4O1DyZdC8iPQFb5iSOmiwTHcNpUlVaWgqJRAI7O8X1InZ2dkpNw6WmpiIzMxMLFy7scJ8tW7bAxMQE06dPZ7fxeDz8+eefOH/+PExMTCAWi7F27VokJyfDwsKi3eM0NDSgsrJS4Ua6Fj6fh3A/7mtWMQyD+HOy80fQAnXSDcn/7lOuFaO4qp7jaAhRHufTf08jNjYWXl5eCAwM7HCfTZs24aWXXoJYLGa3MQyDN998E7a2tjh+/DhSU1MRFhaGyZMno6CgoN3jxMTEwMzMjL25uNCHXVc03U92FeDxGyUoquTmzfxyfiWuFlRS82TSbfWxM4G/qwUkUkbr+3IS0hqnSZW1tTUEAgGKihSnWoqKimBvb//Yx9bU1CAuLk5hWu9Rx48fR1ZWVpuRrEOHDuGPP/5AXFwcnn/+efj5+eHbb7+FgYEBtmzZ0u6xoqOjUVFRwd5ycmhYuityszZCgKsFpIyssjMX5G06qHky6c4iWtWs0pYyJ4Q8CadJlVAohL+/P1JSUthtUqkUKSkpGDp06GMfm5CQgIaGBsydO7fDfWJjY+Hv7w8fHx+F7bW1svonfL7i0+fz+ZBK27/aRCQSwdTUVOFGuqZwec2qtGdfs6q+SYLfqXkyIQj1doBRS5mTv+6UcR0OIUrhfPovKioKP/74I7Zs2YKrV6/i9ddfR01NDSIjIwEA8+bNQ3R0dJvHxcbGIiwsDFZWVu0et7KyEgkJCe2utxo6dCgsLCwwf/58XLhwAdevX8cHH3yAO3fuKNS7It1TqLcDRHp83CiuxqW8imd67v2XC1FJzZMJgZFID1Napr+pyTLRFZwnVREREVi9ejVWrFgBX19fZGRkIDk5mV28np2d3WadU1ZWFk6cOPHYqb+4uDgwDIM5c+a0uc/a2hrJycmorq7GmDFjEBAQgBMnTuD3339vM6pFuh9TsT5CBsqmn5/1gnVqnkzIQxEtNav2XipARW0Tx9EQ8mQ8hiarO6WyshJmZmaoqKigqcAu6Oj1EszflApzQ3389eFYiPQ0X3wzp6wWI/7vMHg84NgHo+FiSb3+SPfGMAwmrD+Oa4VV+HTqQMwb6sZ1SKQL0OTnN+cjVYRoo+Hu1rAzFaG8tgmHr7Vf3V/dWjdPpoSKEFn5G3l5hd9Sc2jBOtF6lFQR0g4Bn4ewlgKE25/BJd2tmyfPpAXqhLCmDXKCUI+PqwWVyMyj+oBEu1FSRUgHZrTUrDqSVYz71Q0aPVfr5sny9VyEEMDcUIjxLa+JuLPZHEdDyONRUkVIB/rYmcDH2QzNUga/Z+Rr9FzbqHkyIR2a3TIFmJSRj7pGbvtyEvI4lFQR8hjymlXbNXgV4IOaRhyg5smEdOi5XlboYWmIqoZm7L3UftcLQrQBJVWEPMZkb0foC3i4UiBrHaMJ1DyZkMfj8x8uWKeaVUSbUVJFyGNYGAkxtr+sZpqmalZta2meTBXUCenYDH9n8HlA6t0y3Cqp5jocQtpFSRUhTyCfAtyVkY9mSfttjDorM68CVwoqIRTw2asNCSFt2ZmKMaa/LQBgG41WES1FSRUhTzCqnw2sjIQorW7AsRslaj22fIH6uIF21DyZkCeQV1jfkZ6Lxmb1fsEhRB0oqSLkCfQFfEz1ldesUt8UYH2TBLvOy5sn0wJ1Qp5kdD8b2JiIUFrdiEPXirgOh5A2KKkiRAnh/rKk6s8rxSivbVTLMRWaJ7tT82RCnkRPwMeMlun4OJoCJFqIkipClDDQ0Qz97U3QKJFi90X1XNItn/oL93eGgJonE6IU+aju0eslyC+v4zgaQhRRUkWIkuTfkNVxFWBOWS1O3rwPAJjpT1f9EaKsntZGeK6XJRgGSDinufpxhHQGJVWEKGmqrxMEfB4ycspxs/jpLumWr8163t2KmicToqLZLQvWt53LgURKTZaJ9qCkihAl2ZiIMKqvDQDZ1UedJZEybFJFC9QJUd14T3uYivWQV16HkzdLuQ6HEBYlVYSoQF6zamd6Xqe/IZ+6VYq88jpqnkxIJ4n1BZjWUteNKqwTbUJJFSEqGDvAFmYG+iisrMepW537hiz/EJjqS82TCeksec2qA1cKUVajnityCXlalFQRogKRngCTfRwAdG7Benntw+bJ8l5mhBDVeTiawtvZDE0SBolPMR1PiDpRUkWIimb4y5Kh5MuFqKpvUumxu87LmicPcDDFQEdTTYRHSLfRuskyw9CCdcI9SqoIUZGPsxl62xihvkmKvZdUq1klb54cEeAMHo9qUxHyNKb4OMJAX4AbxdVIzy7nOhxCKKkiRFU8Ho9dsL4jLU/px7Vunixve0MI6TwTsT5CvWXT8fFnszmOhhBKqgjplGmDnMDjAal3y5B9v1apx7RunmxhRM2TCVGH2S1TgLsvFKg8HU+IulFSRUgnOJgZYHhLvz5lalZR82RCNMPf1QK9bYxQ1yTBH2pqIUVIZ1FSRUgnydvWJJ7PhfQJNavkzZMdzcTUPJkQNeLxeOyCdWqyTLhGSRUhnTTOwx7GIj3klNUh9W7ZY/eV9yibEeBCzZMJUbPpfs7Q4/NwIaccVwsquQ6HdGOUVBHSSQZCAUK9nlyzKqesFidaWmlQ82RC1M/aWIQXPOwAUIV1wi1Kqgh5CvKrAPdeKkBtY3O7+1DzZEI0Tz4FuPN8HuqbJBxHQ7orSqoIeQqD3SzQw9IQNY0S7L9c2OZ+KTVPJuSZGNHHBo5mYlTUNbX7WiTkWaCkipCnwOPxEO7Xcc2qk9Q8mZBnQsDnYWbAwwrrhHCBkipCntJ0P1khz5O3SpFfXqdwn7yCOjVPJkTzZgY4g8cDTt26j3v3a7gOh3RDlFQR8pRcLA0xpKclGEa2nkOuvLaRnYagqT9CNM/ZwhAj+tgAeHjFLSHPEiVVhKjBw7Y1uWxj198z8tHYLGue7OlEzZMJeRbkFdYT0nLQLJFyHA3pbrQiqdq4cSPc3NwgFosxZMgQpKamdrjvqFGjwOPx2txCQ0PZfdq7n8fj4csvv1Q41p49ezBkyBAYGBjAwsICYWFhmnqKpIub6OUAA30BbpfW4HxOOYCHbWlmUfNkQp6Z4AF2sDQSoqiyAXtUbHhOyNPiPKmKj49HVFQUVq5cifT0dPj4+CAkJATFxcXt7p+YmIiCggL2lpmZCYFAgJkzZ7L7tL6/oKAAmzZtki0oDg9n99mxYwdefvllREZG4sKFCzh58iRefPFFjT9f0jUZi/QwwVO2EH1HWi4y8ypwOV/WPDmMmicT8swI9fgIb1nn+E5cBoLXHsXag9eRVVjFjiIToik8huO/siFDhmDw4MH45ptvAABSqRQuLi546623sHz58ic+ft26dVixYgUKCgpgZGTU7j5hYWGoqqpCSkoKAKC5uRlubm745JNPsGDBgk7FXVlZCTMzM1RUVMDUlKZ2CHDyZile+u9fMBXrYaKXA+LO5iDU2wEbX/TjOjRCupWK2iYsT7yIlKvFaGw1BdjbxggTvRww0csB/e1NaAS5m9Lk57eeWo+mosbGRqSlpSE6OprdxufzERwcjNOnTyt1jNjYWMyePbvDhKqoqAh79uzBli1b2G3p6enIy8sDn8/HoEGDUFhYCF9fX3z55Zfw9PRs9zgNDQ1oaGhg/11ZSa0QiKKhvazgaCZGfkU924MsghaoE/LMmRnq47u5/qisb8KfV4qw91Ihjl0vwa2SGmw4dBMbDt1EL2sjTPCyx0QvB3g4mFKCRdSC0+m/0tJSSCQS2NnZKWy3s7NDYeGTi7elpqYiMzMTCxcu7HCfLVu2wMTEBNOnT2e33b59GwDw8ccf45///Cf++OMPWFhYYNSoUSgra7+HW0xMDMzMzNibiwt9WBJFfD4P0/weTvVR82RCuGUq1sd0P2f8d34A0j4KxroIX4zzsINQj4/bpTXYePgWQr8+gdGrj+CL5GvIzKugKULyVDhfU/U0YmNj4eXlhcDAwA732bRpE1566SWIxWJ2m1QqGw7+xz/+gfDwcPj7+2Pz5s3g8XhISEho9zjR0dGoqKhgbzk5VFyOtDXd72Fvvxn+ztQ8mRAtYSLWR9ggJ/wwLwDpH72A9bN9MX6gPUR6fNy9X4vvjtzCpA0nEPTlEcTsu4qLueWUYBGVcTr9Z21tDYFAgKKiIoXtRUVFsLd/fPXpmpoaxMXF4dNPP+1wn+PHjyMrKwvx8fEK2x0cZE1wPTw82G0ikQi9evVCdnZ2u8cSiUQQiUSPjYmQ3jbGCBloh7N3H2B2YA+uwyGEtMNYpIepvk6Y6uuEmoZmHLpWjH2ZBTh0rRjZZbX4z9Hb+M/R23C2MGDXYPk4m9EUIXkiTpMqoVAIf39/pKSksOUMpFIpUlJSsHTp0sc+NiEhAQ0NDZg7d26H+8TGxsLf3x8+Pj4K2/39/SESiZCVlYXhw4cDAJqamnD37l24uro+3ZMi3d73c/0hZUCjVIToACORHib7OGKyjyNqG5tx+FoJ9mYW4NDVYuQ+qMMPx27jh2O34WRugAme9pjo7QBfZ3Pw6fVN2sFpUgUAUVFRmD9/PgICAhAYGIh169ahpqYGkZGRAIB58+bByckJMTExCo+LjY1FWFgYrKys2j1uZWUlEhISsGbNmjb3mZqa4rXXXsPKlSvh4uICV1dXtoZV69IMhHQGj8eDgN5vCdE5hkI9hHo7INTbAXWNEhzJKsbezEKkXC1CXnkd/nviDv574g4czMSY4OmAUG97DHKxoASLsDhPqiIiIlBSUoIVK1awV+ElJyezi9ezs7PB5ysu/crKysKJEydw4MCBDo8bFxcHhmEwZ86cdu//8ssvoaenh5dffhl1dXUYMmQIDh06BAsLC/U9OUIIITrJQCjABC8HTPByQH2TBEeySrAvswB/XilCQUU9Np28g00n78DeVIzxnvYI9XaAfw9KsLo7zutU6SqqU0UIId1PfZMEx66XYF9mIf68UoSqhmb2PlsTkWyK0MsBAW6WtARAS2ny85uSqk6ipIoQQrq3hmYJTtwoxZ5LBTh4pQhV9Q8TLBsTEcYPtMcEL3sM6WlFCZYWoaRKC1FSRQghRK6hWYJTN+9jz6UCHLhciMpWCZa1sRAhA2UjWEN6WkJPoNPVjHQeJVVaiJIqQggh7WlsluLUrVLsvVSAA1eKUF7bxN5nZSTEuIH2mOhlj6G9rCjB4gAlVVqIkipCCCFP0iSR4vSt+9h7qQD7LxfiQasEy8JQHyED7THBywHDeltBnxKsZ4KSKi1ESRUhhBBVNEmk+Ot2Gfa0JFhlNY3sfeaG+hjnYYcJXg54vrc1hHqUYGkKJVVaiJIqQgghndUskSL1zsMEq7T6YYJlKtZjpwiHu9tQgqVmlFRpIUqqCCGEqINEyiD1Thn2XirAvsxClFY3sPeZiPXwgocdJno6YERfa4j0BBxG2jVQUqWFKKkihBCibhIpg3N3HyZYxVWtEiyRHsYOsMVELweM7GsDsT4lWJ1BSZUWoqSKEEKIJkmlDNKyH2DPxQIkZxaisLKevc9IKMDYAXaY6OWAUf0owVIFJVVaiJIqQgghz4pUyuB8zgPsuViIfZkFKKh4mGAZCgUY098WoV4OGNXPFgZCSrAeh5IqLURJFSGEEC5IpQwycsux96JsijCvvI69z0BflmBN9HLA6P42MBRy3uJX61BSpYUoqSKEEMI1hmFwIbcCey8VYO+lAuQ+eJhgifX5GN1PlmCN6W8LIxElWAAlVVqJkipCCCHahGEYXMqrwN5Lhdh7qQDZZbXsfSI9Pkb1s8FELweMHWAH426cYFFSpYUoqSKEEKKtGIbB5fxKdgTr7v2HCZZQj4+gvjYI9XLA2AG2MBHrcxjps0dJlRaipIoQQoguYBgGVwoqsa9lBOt2aQ17n1DAx8i+1pjo5YBgDzuYdoMEi5IqLURJFSGEEF3DMAyuFVZh36UC7LlUgFslDxMsfQEPI/rIpghf8LCDmUHXTLAoqdJClFQRQgjRZQzD4HpRNTtFeKO4mr1PX8DD8+6yEaxxHnYwNxRyGKl6UVKlhSipIoQQ0pXcKKpiF7lnFVWx2/X4PAxzt0aolz3GedjDwki3EyxKqrQQJVWEEEK6qpvF1ewU4bXChwmWgM/DsN5W7AiWlbGIwyg7h5IqLURJFSGEkO7gdkk19mUWYs/FAlwpqGS3C/g8PNfLEhO9HBAy0B7WOpJgUVKlhSipIoQQ0t3cLa3B3kzZGqzMvIcJFp8HDOlphYneDggZaAdbEzGHUT4eJVVaiJIqQggh3Vn2/Vo2wbqYW8Fu5/GAQDdLhHo7YPxAe9iaaleCRUmVFqKkihBCCJHJKavFvswC7LlUiAs55ex2Hg8Y7GqJiV72GO/pAHsz7hMsSqq0ECVVhBBCSFu5D2qRnFmIPZcKcD67XOG+AFcLTPRywAQveziYGXASHyVVWoiSKkIIIeTx8svrsC9TVqYh7d4Dhfv8epi3JFgOcDJ/dgkWJVVaiJIqQgghRHmFFfXY17IG69y9B2idffi6mGOilz0meDrAxdJQo3FQUqWFKKkihBBCOqeosh7JLSNYqXfLFBIsH2czTPByQKiXZhIsSqq0ECVVhBBCyNMrrqrH/sxC7L1UiL/u3Ie0VVby/ri+WDqmj1rPp8nPbz21Ho0QQgghRAW2JmK8PNQNLw91Q0lVA/ZfLsS+zAKcvnUfg3pYcB2eSmikqpNopIoQQgjRnPvVDTAz0IeegK/W49JIFSGEEEK6FV3sK6je9K+TNm7cCDc3N4jFYgwZMgSpqakd7jtq1CjweLw2t9DQUHaf9u7n8Xj48ssv2xyvoaEBvr6+4PF4yMjI0MTTI4QQQkg3wHlSFR8fj6ioKKxcuRLp6enw8fFBSEgIiouL290/MTERBQUF7C0zMxMCgQAzZ85k92l9f0FBATZt2gQej4fw8PA2x/vb3/4GR0dHjT0/QgghhHQPnCdVa9euxaJFixAZGQkPDw98//33MDQ0xKZNm9rd39LSEvb29uzt4MGDMDQ0VEiqWt9vb2+P33//HaNHj0avXr0UjrVv3z4cOHAAq1ev1uhzJIQQQkjXx2lS1djYiLS0NAQHB7Pb+Hw+goODcfr0aaWOERsbi9mzZ8PIyKjd+4uKirBnzx4sWLCgzfZFixbhf//7HwwNNVtojBBCCCFdH6dJVWlpKSQSCezs7BS229nZobCw8ImPT01NRWZmJhYuXNjhPlu2bIGJiQmmT5/ObmMYBq+88gpee+01BAQEKBVrQ0MDKisrFW6EEEIIIXKcT/89jdjYWHh5eSEwMLDDfTZt2oSXXnoJYvHDztgbNmxAVVUVoqOjlT5XTEwMzMzM2JuLi8tTxU4IIYSQroXTpMra2hoCgQBFRUUK24uKimBvb//Yx9bU1CAuLq7NtF5rx48fR1ZWVpuRrEOHDuH06dMQiUTQ09ODu7s7ACAgIADz589v91jR0dGoqKhgbzk5Oco8RUIIIYR0E5wmVUKhEP7+/khJSWG3SaVSpKSkYOjQoY99bEJCAhoaGjB37twO94mNjYW/vz98fHwUtn/99de4cOECMjIykJGRgb179wKQXYm4atWqdo8lEolgamqqcCOEEEIIkeO8+GdUVBTmz5+PgIAABAYGYt26daipqUFkZCQAYN68eXByckJMTIzC42JjYxEWFgYrK6t2j1tZWYmEhASsWbOmzX09evRQ+LexsTEAoHfv3nB2dlbH0yKEEEJIN8N5UhUREYGSkhKsWLEChYWF8PX1RXJyMrt4PTs7G3y+4oBaVlYWTpw4gQMHDnR43Li4ODAMgzlz5mg0fkIIIYQQgHr/dRr1/iOEEEJ0jyY/v3X66j9CCCGEEG1BSRUhhBBCiBpwvqZKV8lnTakIKCGEEKI75J/bmlj9RElVJ1VVVQEAFQElhBBCdFBVVRXMzMzUekxaqN5JUqkU+fn5MDExAY/HU9txKysr4eLigpycnC67AL6rP8eu/vyArv8c6fnpvq7+HOn5dR7DMKiqqoKjo2Ob6gJPi0aqOonP52u0plV3KDDa1Z9jV39+QNd/jvT8dF9Xf470/DpH3SNUcrRQnRBCCCFEDSipIoQQQghRA0qqtIxIJMLKlSshEom4DkVjuvpz7OrPD+j6z5Gen+7r6s+Rnp92ooXqhBBCCCFqQCNVhBBCCCFqQEkVIYQQQogaUFJFCCGEEKIGlFQRQgghhKgBJVUc2LhxI9zc3CAWizFkyBCkpqY+dv+EhAT0798fYrEYXl5e2Lt37zOKtPNUeY4//fQTeDyewk0sFj/DaFVz7NgxTJ48GY6OjuDxeNi1a9cTH3PkyBH4+flBJBLB3d0dP/30k8bj7CxVn9+RI0fa/P54PB4KCwufTcAqiomJweDBg2FiYgJbW1uEhYUhKyvriY/TlddhZ56frr0Gv/vuO3h7e7OFIYcOHYp9+/Y99jG68vsDVH9+uvb7e9Tnn38OHo+HZcuWPXY/XfgdUlL1jMXHxyMqKgorV65Eeno6fHx8EBISguLi4nb3P3XqFObMmYMFCxbg/PnzCAsLQ1hYGDIzM59x5MpT9TkCsqq5BQUF7O3evXvPMGLV1NTUwMfHBxs3blRq/zt37iA0NBSjR49GRkYGli1bhoULF2L//v0ajrRzVH1+cllZWQq/Q1tbWw1F+HSOHj2KN998E2fOnMHBgwfR1NSEcePGoaampsPH6NLrsDPPD9Ct16CzszM+//xzpKWl/X979x4UVeHFAfy7LCwQpIgYj0DQ0AURAVESqCAlMY1smrIah4GohICEkWioaQbCErZUfITJlKGTTdTIgNoDA0EYGRQEVhcChtAaTB5Z+KAckd3z+6Ph/lpgkV1X1vv7nc/MnWHvPffuOXvmMod7l12cOXMGy5cvx9q1a9Ha2jpuvJj6B+hfHyCu/v1bQ0MDCgoKsGjRognjRNNDYlMqKCiIkpKShMdqtZpcXFwoJydn3Ph169bRmjVrtNY9/PDDFB8ff1fzvBP61lhYWEjTp0+fouyMCwCVlJRMGPPWW2+Rj4+P1roXXniBIiMj72JmxjGZ+qqqqggADQwMTElOxtbf308AqLq6WmeMGM/DEZOpT8zn4IgZM2bQZ599Nu42MfdvxET1ibV/169fp3nz5lF5eTmFhYVRSkqKzlix9JCvVE2hoaEhNDY2IiIiQlhnZmaGiIgI1NXVjbtPXV2dVjwAREZG6ow3NUNqBIDBwUG4u7vDzc3ttn+RiY3Yemgof39/ODs744knnkBtba2p05m0q1evAgDs7e11xoi5h5OpDxDvOahWq1FUVIS//voLwcHB48aIuX+TqQ8QZ/+SkpKwZs2aMb0Zj1h6yEPVFLp8+TLUajUcHR211js6Oup8/0lvb69e8aZmSI1yuRyff/45Dh8+jIMHD0Kj0SAkJAQXL16cipTvOl09vHbtGm7cuGGirIzH2dkZe/fuRXFxMYqLi+Hm5obw8HA0NTWZOrXb0mg0SE1NRWhoKBYuXKgzTmzn4YjJ1ifGc1ClUsHW1haWlpZISEhASUkJFixYMG6sGPunT31i7F9RURGampqQk5MzqXix9NDc1AkwFhwcrPUXWEhICLy9vVFQUIDNmzebMDM2GXK5HHK5XHgcEhKCrq4u5OXl4YsvvjBhZreXlJSElpYWnDx50tSp3BWTrU+M56BcLodSqcTVq1dx6NAhxMTEoLq6WufgITb61Ce2/nV3dyMlJQXl5eWiekP9ZPBQNYUcHBwglUrR19entb6vrw9OTk7j7uPk5KRXvKkZUuNoFhYWCAgIwM8//3w3Upxyuno4bdo0WFtbmyiruysoKOieH1SSk5Px7bffoqamBq6urhPGiu08BPSrbzQxnIMymQyenp4AgMDAQDQ0NGDnzp0oKCgYEyvG/ulT32j3ev8aGxvR39+PxYsXC+vUajVqamrw8ccf4+bNm5BKpVr7iKWHfPtvCslkMgQGBuL48ePCOo1Gg+PHj+u8Vx4cHKwVDwDl5eUT3ls3JUNqHE2tVkOlUsHZ2flupTmlxNZDY1Aqlfds/4gIycnJKCkpQWVlJebMmXPbfcTUQ0PqG02M56BGo8HNmzfH3Sam/ukyUX2j3ev9W7FiBVQqFZRKpbAsWbIE69evh1KpHDNQASLqoanfKf//pqioiCwtLWn//v30008/0YYNG8jOzo56e3uJiCg6OpoyMjKE+NraWjI3N6etW7dSW1sbZWZmkoWFBalUKlOVcFv61vjee+/RsWPHqKurixobG+nFF18kKysram1tNVUJE7p+/To1NzdTc3MzAaDt27dTc3Mz/frrr0RElJGRQdHR0UL8+fPn6b777qP09HRqa2uj/Px8kkqlVFZWZqoSJqRvfXl5eVRaWkqdnZ2kUqkoJSWFzMzMqKKiwlQlTOj111+n6dOn04kTJ6inp0dY/v77byFGzOehIfWJ7RzMyMig6upqunDhAp07d44yMjJIIpHQjz/+SETi7h+R/vWJrX/jGf3ff2LtIQ9VJrB7926aPXs2yWQyCgoKolOnTgnbwsLCKCYmRiv+m2++ofnz55NMJiMfHx/67rvvpjhj/elTY2pqqhDr6OhIq1evpqamJhNkPTkjHyEwehmpKSYmhsLCwsbs4+/vTzKZjObOnUuFhYVTnvdk6VufQqGghx56iKysrMje3p7Cw8OpsrLSNMlPwni1AdDqiZjPQ0PqE9s5GBcXR+7u7iSTyWjWrFm0YsUKYeAgEnf/iPSvT2z9G8/ooUqsPZQQEU3ddTHGGGOMsf9N/J4qxhhjjDEj4KGKMcYYY8wIeKhijDHGGDMCHqoYY4wxxoyAhyrGGGOMMSPgoYoxxhhjzAh4qGKMMcYYMwIeqhhjovTLL79AIpFAqVSaOhVBe3s7li1bBisrK/j7+9+V5wgPD0dqauqEMR4eHtixY8ddeX7GmG48VDHGDBIbGwuJRILc3Fyt9aWlpZBIJCbKyrQyMzNhY2ODjo6OMd9TNmLkdRu93KtffssYmzweqhhjBrOysoJCocDAwICpUzGaoaEhg/ft6urCI488And3d8ycOVNn3KpVq9DT06O1GPLFx4yxewsPVYwxg0VERMDJyQk5OTk6Y7KyssbcCtuxYwc8PDyEx7GxsXjmmWewZcsWODo6ws7ODtnZ2RgeHkZ6ejrs7e3h6uqKwsLCMcdvb29HSEgIrKyssHDhQlRXV2ttb2lpwZNPPglbW1s4OjoiOjoaly9fFraHh4cjOTkZqampcHBwQGRk5Lh1aDQaZGdnw9XVFZaWlvD390dZWZmwXSKRoLGxEdnZ2ZBIJMjKytL5mlhaWsLJyUlrkUqlAIDq6moEBQXB0tISzs7OyMjIwPDwsM5j9ff3IyoqCtbW1pgzZw6+/PJLre1EhKysLMyePRuWlpZwcXHBxo0bdR6PMWY4HqoYYwaTSqXYsmULdu/ejYsXL97RsSorK3Hp0iXU1NRg+/btyMzMxFNPPYUZM2bg9OnTSEhIQHx8/JjnSU9PR1paGpqbmxEcHIyoqCj88ccfAIArV65g+fLlCAgIwJkzZ1BWVoa+vj6sW7dO6xgHDhyATCZDbW0t9u7dO25+O3fuxLZt27B161acO3cOkZGRePrpp9HZ2QkA6OnpgY+PD9LS0tDT04M333xT79fgt99+w+rVq7F06VKcPXsWn3zyCfbt24f3339f5z6xsbHo7u5GVVUVDh06hD179qC/v1/YXlxcjLy8PBQUFKCzsxOlpaXw9fXVOzfG2CSY+AudGWMiFRMTQ2vXriUiomXLllFcXBwREZWUlNC/f7VkZmaSn5+f1r55eXnk7u6udSx3d3dSq9XCOrlcTo8++qjweHh4mGxsbOirr74iIqILFy4QAMrNzRVibt26Ra6urqRQKIiIaPPmzbRy5Uqt5+7u7iYA1NHRQUREYWFhFBAQcNt6XVxc6IMPPtBat3TpUkpMTBQe+/n5UWZm5oTHiYmJIalUSjY2NsLy3HPPERHRO++8Q3K5nDQajRCfn59Ptra2wmsTFhZGKSkpRETU0dFBAKi+vl6Ib2trIwCUl5dHRETbtm2j+fPn09DQ0G1rZIzdGb5SxRi7YwqFAgcOHEBbW5vBx/Dx8YGZ2X9/JTk6OmpdUZFKpZg5c6bWVRgACA4OFn42NzfHkiVLhDzOnj2Lqqoq2NraCouXlxeAf97/NCIwMHDC3K5du4ZLly4hNDRUa31oaKhBNT/++ONQKpXCsmvXLgBAW1sbgoODtd7oHxoaisHBwXGvBLa1tcHc3Fwrfy8vL9jZ2QmPn3/+edy4cQNz587Fa6+9hpKSkglvJzLGDGdu6gQYY+L32GOPITIyEm+//TZiY2O1tpmZmYGItNbdunVrzDEsLCy0HkskknHXaTSaSec1ODiIqKgoKBSKMducnZ2Fn21sbCZ9TGOwsbGBp6fnlDyXm5sbOjo6UFFRgfLyciQmJuKjjz5CdXX1mNeXMXZn+EoVY8wocnNzcfToUdTV1WmtnzVrFnp7e7UGK2N+ttSpU6eEn4eHh9HY2Ahvb28AwOLFi9Ha2goPDw94enpqLfoMUtOmTYOLiwtqa2u11tfW1mLBggXGKQSAt7c36urqtF6r2tpa3H///XB1dR0T7+XlJdQ8oqOjA1euXNGKs7a2RlRUFHbt2oUTJ06grq4OKpXKaHkzxv7BQxVjzCh8fX2xfv164VbWiPDwcPz+++/48MMP0dXVhfz8fPzwww9Ge978/HyUlJSgvb0dSUlJGBgYQFxcHAAgKSkJf/75J1566SU0NDSgq6sLx44dw8svvwy1Wq3X86Snp0OhUODrr79GR0cHMjIyoFQqkZKSYrRaEhMT0d3djTfeeAPt7e04fPgwMjMzsWnTJq1boyPkcjlWrVqF+Ph4nD59Go2NjXj11VdhbW0txOzfvx/79u1DS0sLzp8/j4MHD8La2hru7u5Gy5sx9g8eqhhjRpOdnT3m9py3tzf27NmD/Px8+Pn5ob6+3qD/jNMlNzcXubm58PPzw8mTJ3HkyBE4ODgAgHB1Sa1WY+XKlfD19UVqairs7OzGHVImsnHjRmzatAlpaWnw9fVFWVkZjhw5gnnz5hmtlgcffBDff/896uvr4efnh4SEBLzyyit49913de5TWFgIFxcXhIWF4dlnn8WGDRvwwAMPCNvt7Ozw6aefIjQ0FIsWLUJFRQWOHj064edoMcYMI6HRb3ZgjDHGGGN64ytVjDHGGGNGwEMVY4wxxpgR8FDFGGOMMWYEPFQxxhhjjBkBD1WMMcYYY0bAQxVjjDHGmBHwUMUYY4wxZgQ8VDHGGGOMGQEPVYwxxhhjRsBDFWOMMcaYEfBQxRhjjDFmBDxUMcYYY4wZwX8AMG/TqiYSVv8AAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## SVM Classifier (Linear Kernel)"
      ],
      "metadata": {
        "id": "98RVRIe4VtZP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "svm = SVC(kernel='linear', C=1.0)\n",
        "svm.fit(Xtrain, Ytrain)\n",
        "y_pred = svm.predict(Xtest)\n",
        "accuracy = accuracy_score(Ytest, y_pred)\n",
        "f1 = f1_score(Ytest, y_pred, average = 'binary')\n",
        "print(\"Accuracy of SVM classifier with linear kernel:\", accuracy)"
      ],
      "metadata": {
        "id": "cAFUNDymV2Lk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d9746f71-0d16-4a40-a65a-567f20586ff2"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of SVM classifier with linear kernel: 0.7494678238087441\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'The F1 Score of the SVM Classifier (linear kernel) is {100*f1:.3f}%')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YzsL7YxlxFFh",
        "outputId": "0b01c474-02bf-4a45-eed3-c9ff8a6e353e"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The F1 Score of the SVM Classifier (linear kernel) is 71.277%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## SVM classifier (RBF kernel)"
      ],
      "metadata": {
        "id": "YPpfoTCHlQIT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Testing non linear SVM classifier with RBF kernel\n",
        "clf = NuSVC()\n",
        "clf.fit(Xtrain, Ytrain)\n",
        "preds = clf.predict(Xtest) \n",
        "acc = accuracy_score(Ytest, preds)\n",
        "f1 = f1_score(Ytest, preds, average = 'binary')\n",
        "print(f\"Accuracy of nonlinear SVM classifier using RBF kernel: {acc*100:.3f}%\")\n",
        "print(f'The F1 Score of the SVM Classifier (RBF kernel) is {100*f1:.3f}%')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bTvTdeYSk-B1",
        "outputId": "6aa55dfe-5d19-4c97-fec1-f8c1e3b6c8b0"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of nonlinear SVM classifier using RBF kernel: 86.868%\n",
            "The F1 Score of the SVM Classifier (RBF kernel) is 84.830%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hyperparameter search via crossvalidation for SVM classifer with RBF kernel"
      ],
      "metadata": {
        "id": "mm0zLq8MYByt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "gamma, accuracy_fold, accuracy_run, accuracy_test = radialSVC(Xtrain, Ytrain, Xtest, Ytest)"
      ],
      "metadata": {
        "id": "ygq-CnryTRn2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "28a469f7-bd19-4495-8366-99a760a53c3a"
      },
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best Accuracy of RBF Kernel: 98.766% for gamma = 1\n",
            "F1 Score of RBF is:0.9860061919504645\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_fin(l1=gamma, accuracy_run=accuracy_run, accuracy_test=accuracy_test, accuracy_fold=accuracy_fold)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 944
        },
        "id": "AzSIS0HpV7_j",
        "outputId": "5cec9d32-eda2-47e1-ee97-26db5c163d0a"
      },
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## K Nearest Neighbours Classifier"
      ],
      "metadata": {
        "id": "Lws_jpPllxnj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Testing K Neighbors Classifier\n",
        "neigh = KNeighborsClassifier(n_neighbors=3)\n",
        "neigh.fit(Xtrain, Ytrain)\n",
        "preds = neigh.predict(Xtest) \n",
        "acc = accuracy_score(Ytest, preds)\n",
        "print(f\"Accuracy of KNeighborsClassifier: {acc*100:.3f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "otslb9inl3Yz",
        "outputId": "c6f8e7a9-90e3-43f7-b388-86893fcadaf5"
      },
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy of KNeighborsClassifier: 99.203%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hyperparameter search via cross validation for K Neighbors Classifier"
      ],
      "metadata": {
        "id": "0ooknikdX8Ui"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "neighbors, accuracy_fold, accuracy_run, accuracy_test = knearest(Xtrain, Ytrain, Xtest, Ytest)"
      ],
      "metadata": {
        "id": "AmgZIAC-Tl_q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "321f3717-ef3a-4d5b-a6e2-d6404eab9d52"
      },
      "execution_count": 98,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Best Accuracy of KNN Classifier: 99.236% for number of neighbors = 1\n",
            "F1 Score of KNN is:0.99128160418483\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_fin(l1=neighbors, accuracy_run=accuracy_run, accuracy_test=accuracy_test, accuracy_fold=accuracy_fold)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 944
        },
        "id": "Ln3qawbf1oIH",
        "outputId": "2cb1875b-1310-47e4-c032-7655de4ac239"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Multilayer Perceptron (ANN)"
      ],
      "metadata": {
        "id": "iGB-KqJcln68"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hyperparameter search via crossvalidation for Multilayer Perceptron (MLP)"
      ],
      "metadata": {
        "id": "lodcvYtlXyCN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "lr, accuracy_run, accuracy_test, accuracy_fold, f1_list = mlp_classifier(Xtrain,Ytrain,Xtest,Ytest,'sgd')"
      ],
      "metadata": {
        "id": "plXqnmHSmFp6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9164b97c-f703-4816-944c-a4a154a5ca3c"
      },
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:686: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  warnings.warn(\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_fin(l1=lr, accuracy_run=accuracy_run, accuracy_test=accuracy_test, accuracy_fold=accuracy_fold)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 944
        },
        "id": "zJu4lx2s4rVa",
        "outputId": "b4db30d0-498d-4e55-8680-8f70445116b5"
      },
      "execution_count": 101,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'The accuracy obtained is {max(accuracy_test)*100} with learning rate = {lr[np.argmax(np.array(accuracy_test))]} and F1 Score = {f1_list[np.argmax(np.array(accuracy_test))]} for SGD')"
      ],
      "metadata": {
        "id": "6O3RIP-imT2p",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "23a4f56f-edff-4d9b-9732-894344965e00"
      },
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy obtained is 97.45101249931773 with learning rate = 0.001 and F1 Score = 0.9708616709303051 for SGD\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "lr, accuracy_run, accuracy_test, accuracy_fold, f1_list = mlp_classifier(Xtrain,Ytrain,Xtest,Ytest,'adam')"
      ],
      "metadata": {
        "id": "J36ZFXQdmhD_"
      },
      "execution_count": 103,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_fin(l1=lr, accuracy_run=accuracy_run, accuracy_test=accuracy_test, accuracy_fold=accuracy_fold)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 944
        },
        "id": "ywY1GWyp7tar",
        "outputId": "678b8fe2-ec6e-4a9e-d93c-ae2b8d24948e"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjcAAAHHCAYAAABDUnkqAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACMN0lEQVR4nO3dd1wT9/8H8NdlsxJEZLlAxW1dFYqjLhQXwy7t0lqtHVq1dvy0S2v7rdrham3tsra1w2pb3CjiVtQ6qHvhVoaIJCGMrM/vj5BIZEgw4TLez8fjBC6f3L2PSPK+z30+7+MYYwyEEEIIIW5CwHcAhBBCCCH2RMkNIYQQQtwKJTeEEEIIcSuU3BBCCCHErVByQwghhBC3QskNIYQQQtwKJTeEEEIIcSuU3BBCCCHErVByQwghhBC3QskNIU5m+/bt4DgOq1at4juUGsnJycFjjz2G+vXrg+M4LFiwoE72y3EcZs6cec92M2fOBMdxjg+IEOI0KLkhHmnZsmXgOA4ymQzXr1+v8HifPn3Qvn17HiJzPa+99ho2bdqE6dOn45dffsGgQYOqbMtxXKVLSEhIHUbs/AoKCiCTycBxHE6dOsV3OIS4HBHfARDCp9LSUsyZMwdffPEF36G4rK1btyIxMRFvvPFGjdoPGDAAo0aNslrn5eXliNBc1sqVKy1J36+//oqPPvqI75AIcSmU3BCP1qlTJ3z33XeYPn06wsLC+A6nTmk0Gvj4+Nz3dnJzc+Hv71/j9i1btsQzzzxz3/t1Z8uXL8eQIUPQtGlT/Pbbb06b3JSUlEAikUAgoIsAxLnQ/0ji0d5++20YDAbMmTOn2naXLl0Cx3FYtmxZhcfuHvthHuNx9uxZPPPMM1AoFGjQoAHee+89MMZw9epVJCYmQi6XIyQkBJ9//nml+zQYDHj77bcREhICHx8fJCQk4OrVqxXa7d+/H4MGDYJCoYC3tzd69+6NPXv2WLUxx3Ty5Ek89dRTqFevHnr27FntMV+4cAGPP/44AgIC4O3tjYceegjr16+3PG6+tMcYw+LFiy2XmO5Xbm4uxo4di+DgYMhkMnTs2BE//fRTjZ67e/dudOvWDTKZDM2bN8c333xTabvU1FT07NkT/v7+8PX1RatWrfD2229Xu+327dujb9++FdYbjUY0bNgQjz32mGXdH3/8ga5du8LPzw9yuRwdOnTAwoULa3QMV65cwa5duzBy5EiMHDkSFy9exN69eyttu3z5ckRFRcHb2xv16tXDww8/jM2bN1u12bhxI3r37m2JpVu3bvjtt98sj4eHh+O5556rsO0+ffqgT58+lp/NY8H++OMPvPvuu2jYsCG8vb2hUqmQn5+PN954Ax06dICvry/kcjkGDx6M//77r8J2S0pKMHPmTLRs2RIymQyhoaF45JFHkJmZCcYYwsPDkZiYWOnzFAoFXnzxxRr9Holno54b4tEiIiIwatQofPfdd5g2bZpde29GjBiBNm3aYM6cOVi/fj0++ugjBAQE4JtvvkG/fv0wd+5c/Prrr3jjjTfQrVs3PPzww1bP/9///geO4/B///d/yM3NxYIFCxAbG4uMjAzLZZytW7di8ODB6Nq1K2bMmAGBQIAff/wR/fr1w65duxAVFWW1zccffxyRkZH4+OOPwRirMvacnBx0794dRUVFmDRpEurXr4+ffvoJCQkJWLVqFYYPH46HH34Yv/zyC5599tlKLzVVpaSkBHl5eVbr/Pz8IJVKUVxcjD59+uD8+fOYOHEiIiIisHLlSjz33HMoKCjA5MmTq9zusWPHMHDgQDRo0AAzZ86EXq/HjBkzEBwcbNXuxIkTGDZsGB544AHMmjULUqkU58+fr5AQ3m3EiBGYOXMmsrOzrcYI7d69Gzdu3MDIkSMBmBKnJ598Ev3798fcuXMBAKdOncKePXuqjd/s999/h4+PD4YNGwYvLy80b94cv/76K7p3727V7oMPPsDMmTPRvXt3zJo1CxKJBPv378fWrVsxcOBAAKYE9Pnnn0e7du0wffp0+Pv748iRI0hJScFTTz11z1gq8+GHH0IikeCNN95AaWkpJBIJTp48ieTkZDz++OOIiIhATk4OvvnmG/Tu3RsnT560/F0ZDAYMGzYMaWlpGDlyJCZPngy1Wo3U1FQcP34czZs3xzPPPINPPvkE+fn5CAgIsOx37dq1UKlU1OtHaoYR4oF+/PFHBoD9+++/LDMzk4lEIjZp0iTL471792bt2rWz/Hzx4kUGgP34448VtgWAzZgxw/LzjBkzGAA2fvx4yzq9Xs8aNWrEOI5jc+bMsay/ffs28/LyYqNHj7as27ZtGwPAGjZsyFQqlWX9n3/+yQCwhQsXMsYYMxqNLDIyksXFxTGj0WhpV1RUxCIiItiAAQMqxPTkk0/W6PczZcoUBoDt2rXLsk6tVrOIiAgWHh7ODAaD1fFPmDChRtsFUOli/r0uWLCAAWDLly+3PEer1bKYmBjm6+tr9fu4+/eelJTEZDIZu3z5smXdyZMnmVAoZOXf6ubPn88AsJs3b9YoZrMzZ84wAOyLL76wWv/KK68wX19fVlRUxBhjbPLkyUwulzO9Xm/T9s06dOjAnn76acvPb7/9NgsMDGQ6nc6y7ty5c0wgELDhw4dbvRaMMcv/hYKCAubn58eio6NZcXFxpW0YY6xp06ZW///MevfuzXr37m352fz/slmzZpZjNSspKakQx8WLF5lUKmWzZs2yrFu6dCkDwObNm1dhf+aYzL/nr7/+2urxhIQEFh4ebhU7IVWhy1LE4zVr1gzPPvssvv32W2RlZdltu+PGjbN8LxQK8eCDD4IxhrFjx1rW+/v7o1WrVrhw4UKF548aNQp+fn6Wnx977DGEhoZiw4YNAICMjAycO3cOTz31FG7duoW8vDzk5eVBo9Ggf//+2LlzJ4xGo9U2X3rppRrFvmHDBkRFRVlduvL19cX48eNx6dIlnDx5sma/hEokJiYiNTXVaomLi7PsNyQkBE8++aSlvVgsxqRJk1BYWIgdO3ZUuk2DwYBNmzYhKSkJTZo0saxv06aNZdtm5vFBq1evrvD7qU7Lli3RqVMnrFixwmq/q1atQnx8vKU3zd/fHxqNBqmpqTXettnRo0dx7Ngxq+N/8sknkZeXh02bNlnWJScnw2g04v33368w3sV8aTA1NRVqtRrTpk2DTCartE1tjB49usIAcKlUaonDYDDg1q1blst9hw8ftrT766+/EBgYiFdffbXCds0xtWzZEtHR0fj1118tj+Xn52Pjxo14+umnaVo/qRFKbggB8O6770Kv199z7I0tyn/IAoBCoYBMJkNgYGCF9bdv367w/MjISKufOY5DixYtcOnSJQDAuXPnAJg+bBo0aGC1fP/99ygtLYVSqbTaRkRERI1iv3z5Mlq1alVhfZs2bSyP11ajRo0QGxtrtYSGhlq2GxkZWeED+177vXnzJoqLiyv8zgBUOI4RI0agR48eGDduHIKDgzFy5Ej8+eefNUp0RowYgT179ljKB2zfvh25ubkYMWKEpc0rr7yCli1bYvDgwWjUqBGef/55pKSk3HPbgGkMjY+PD5o1a4bz58/j/PnzkMlkCA8Pt/qwz8zMhEAgQNu2bavcVmZmJgDYvaRBZf+HjEYj5s+fj8jISEilUgQGBqJBgwY4evSo1f/BzMxMtGrVCiJR9SMiRo0ahT179lhe75UrV0Kn0+HZZ5+167EQ90XJDSEw9d4888wzVfbeVHW2aDAYqtymUCis0ToA1Y5/qYr5w/jTTz+t0BNiXnx9fa2eQ1OuTb+DnTt3YsuWLXj22Wdx9OhRjBgxAgMGDKj29QRMyQ1jDCtXrgQA/Pnnn1AoFFa1fYKCgpCRkYE1a9YgISEB27Ztw+DBgzF69Ohqt80Yw++//w6NRoO2bdsiMjLSsly6dAmrV69GYWHh/f8C7mLr/+3K/g99/PHHmDp1Kh5++GEsX74cmzZtQmpqKtq1a2dT75jZyJEjIRaLLQnd8uXL8eCDD1aacBNSGRpQTEiZd999F8uXL7cMAi2vXr16AEzF1cq7nx6MezH3zJgxxnD+/Hk88MADAIDmzZsDAORyOWJjY+2676ZNm+LMmTMV1p8+fdryuCM0bdoUR48ehdFotOq9udd+GzRoAC8vrwq/MwCVHodAIED//v3Rv39/zJs3Dx9//DHeeecdbNu2rdrfZUREBKKiorBixQpMnDgRf//9N5KSkiCVSq3aSSQSxMfHIz4+HkajEa+88gq++eYbvPfee2jRokWl296xYweuXbuGWbNmWXqqzG7fvo3x48cjOTkZzzzzDJo3bw6j0YiTJ0+iU6dOlW7P/P/j+PHjVe4TMP3fvvv/NWD6v92sWbMqn1feqlWr0LdvX/zwww9W6wsKCqx6Kps3b479+/dDp9NBLBZXub2AgAAMHToUv/76K55++mns2bOnzipfE/dAPTeElDHP1Pjmm2+QnZ1t9ZhcLkdgYCB27txptf6rr75yWDw///wz1Gq15edVq1YhKysLgwcPBgB07doVzZs3x2effVbpGf3Nmzdrve8hQ4bgwIEDSE9Pt6zTaDT49ttvER4eXu3lkPsxZMgQZGdnW41r0ev1+OKLL+Dr64vevXtX+jyhUIi4uDgkJyfjypUrlvWnTp2yGqsCmMZv3M2cIJSWlt4zxhEjRmDfvn1YunQp8vLyrC5JAcCtW7esfhYIBJaEtLrtmy9Jvfnmm3jssceslhdeeAGRkZGWnoykpCQIBALMmjWrQs+IuRdw4MCB8PPzw+zZs1FSUlJpG8D0/37fvn3QarWWdevWrau07EBVhEJhhd7HlStXVqj+/eijjyIvLw9ffvllhW3c/fxnn30WJ0+exJtvvgmhUGiZjUZITVDPDSHlvPPOO/jll19w5swZtGvXzuqxcePGYc6cORg3bhwefPBB7Ny5E2fPnnVYLAEBAejZsyfGjBmDnJwcLFiwAC1atMALL7wAwPSh+f3332Pw4MFo164dxowZg4YNG+L69evYtm0b5HI51q5dW6t9T5s2Db///jsGDx6MSZMmISAgAD/99BMuXryIv/76y2FF28aPH49vvvkGzz33HA4dOoTw8HCsWrXKcuZefoD13T744AOkpKSgV69eeOWVVyxJUbt27XD06FFLu1mzZmHnzp0YOnQomjZtitzcXHz11Vdo1KjRPWv/AMATTzyBN954A2+88QYCAgIq9PSMGzcO+fn56NevHxo1aoTLly/jiy++QKdOnSr0yJiVlpbir7/+woABAyoM/jVLSEjAwoULkZubixYtWuCdd97Bhx9+iF69euGRRx6BVCrFv//+i7CwMMyePRtyuRzz58/HuHHj0K1bN0t9o//++w9FRUWW2kHjxo3DqlWrMGjQIDzxxBPIzMzE8uXLLT0/NTFs2DDMmjULY8aMQffu3XHs2DH8+uuvFXp+Ro0ahZ9//hlTp07FgQMH0KtXL2g0GmzZsgWvvPKKVX2boUOHon79+li5ciUGDx6MoKCgGsdDCE0FJx6p/FTwu40ePZoBsJoKzphpivXYsWOZQqFgfn5+7IknnmC5ublVTgW/e6rx6NGjmY+PT4X93T3t3Dzl9vfff2fTp09nQUFBzMvLiw0dOtRqmrPZkSNH2COPPMLq16/PpFIpa9q0KXviiSdYWlraPWOqTmZmJnvssceYv78/k8lkLCoqiq1bt65CO9g4FfxebXNyctiYMWNYYGAgk0gkrEOHDjWags8YYzt27GBdu3ZlEomENWvWjC1ZssRy7GZpaWksMTGRhYWFMYlEwsLCwtiTTz7Jzp49W6NjYIyxHj16MABs3LhxFR5btWoVGzhwIAsKCmISiYQ1adKEvfjiiywrK6vK7f31118MAPvhhx+qbLN9+3arUgCMmaZWd+7cmUmlUlavXj3Wu3dvlpqaavW8NWvWsO7duzMvLy8ml8tZVFQU+/33363afP7556xhw4ZMKpWyHj16sIMHD1Y5FXzlypUVYispKWGvv/46Cw0NZV5eXqxHjx4sPT29wjYYM/0dvfPOOywiIoKJxWIWEhLCHnvsMZaZmVlhu6+88goDwH777bcqfy+EVIZjrBYjGQkhhBAHe+211/DDDz8gOzsb3t7efIdDXAiNuSGEEOJ0SkpKsHz5cjz66KOU2BCb0ZgbQgghTiM3NxdbtmzBqlWrcOvWrRrdsoKQu1FyQwghxGmcPHkSTz/9NIKCgrBo0aIqp7oTUh0ac0MIIYQQt0JjbgghhBDiVii5IYQQQohb8bgxN0ajETdu3ICfnx/dXZYQQghxEYwxqNVqhIWF3bOQqMclNzdu3EDjxo35DoMQQgghtXD16lU0atSo2jYel9yYy7dfvXoVcrmc52gIIYQQUhMqlQqNGzeu9jYsZh6X3JgvRcnlckpuCCGEEBdTkyElNKCYEEIIIW6FkhtCCCGEuBVKbgghhBDiVii5IYQQQohb4TW52blzJ+Lj4xEWFgaO45CcnHzP52zfvh1dunSBVCpFixYtsGzZMofHSQghhBDXwWtyo9Fo0LFjRyxevLhG7S9evIihQ4eib9++yMjIwJQpUzBu3Dhs2rTJwZESQgghxFXwOhV88ODBGDx4cI3bL1myBBEREfj8888BAG3atMHu3bsxf/58xMXFOSpMQgghhLgQlxpzk56ejtjYWKt1cXFxSE9Pr/I5paWlUKlUVgshhBBC3JdLJTfZ2dkIDg62WhccHAyVSoXi4uJKnzN79mwoFArLQrdeIIQQQtybSyU3tTF9+nQolUrLcvXqVb5DIoQQQogDudTtF0JCQpCTk2O1LicnB3K5HF5eXpU+RyqVQiqV1kV4hBBCCHECLtVzExMTg7S0NKt1qampiImJ4SkiQgghhDgbXntuCgsLcf78ecvPFy9eREZGBgICAtCkSRNMnz4d169fx88//wwAeOmll/Dll1/irbfewvPPP4+tW7fizz//xPr16/k6BEIIIffByIzQGXXQGrTQGrTQGXUAALFADIlQYloEkhrdLJEQM16Tm4MHD6Jv376Wn6dOnQoAGD16NJYtW4asrCxcuXLF8nhERATWr1+P1157DQsXLkSjRo3w/fffO8c08MKbwMlkQCQFhFJAJAGEknLfl/8qNT1291eBkO+jIIS4KYPRYEoijKYkQm/UWxIKrVFrSTB0Bp1VO3PCYXncnIgYy7Ut93NlX632Zyy3TYMOeqavUfwigQgSwZ1kRywU3/m+LBESC8VVtpEIy7UT3LXe1udRwuX0OMYY4zuIuqRSqaBQKKBUKiGXy+223ZwTuxC8ctj9bYQT3pXwVJYY1SJpqrCuqueKK66jhIvwiDGGIq0B+RotCop0yC/S4rZGi3yNFreL7vqqMT1eWGL6sOQ4gAMsH0Bc2T9W68r9zJX9jLLWdx6zrKmwTcs2uDuPMxjBQQ8IDOCgBxPowaAHx+kBzghwOoDTA5wBgAHgdGDmx6ADgx7g9GCcoayd6fkMZeugK/tqXnfnq3mdsWwdoIexbJ1p+86Pg+k9h8HAcyT3JuJEkAjFkAhMi9jyVVS2TmRZJylbZ3pMBAlnflwEMScsW1f22N3rOBHEAmHZ40LTYxDe+V4ggoQTQAwBBBwHMAaAAcwIMJR9X/az+fsq1xnLvkcl62q6HQA+DYB2SXb9fdvy+e1SA4qdWa5OhkOGKEiggwR6SDnTV/PPYughKVsnhQ4SzvTVCjMAuiLT4iyqTLjuM2mq7rlCcfXbo4TLZZXoDBWSkcqTFZ1pfZEWWn1NP5QZICyCQKS2JA+cOYng9Hd9X+6rQG+17p7PE+gBGMCVPa9ie9c4X+SMAgjAQWAUQABB2VcOQiaAkHFWi4hxEDJAxDiIGMoWDmLGIC77WcwAMRjEjEHCAAljkDDTz1JmtPwshRFSdmeRMQYpM1jWSzgDBDCCATBwzJTucRz0nBF6joOOY9BzKFsYdBwHHQfoytbpOECHsq+Wxzirx7UcBx3HQVtuMa/XgrM8bllXtujv6qnRMz30ej2KUHkpEj6Iyn7Pd37/KHtNGCS4s866jfl702t497rK2knKXuuq2nmFdYG3nZMbm34PvO3ZzQQ374DTiUtxvezssqDozpt3QZEWt4tMb9Z6Y/k3PgYxDGUJUFkyZEmK9FbrvDg96kmBelIjFBIGhdgIuZjBT2yEr8gAb6ERPgIDvIR6yDjTIoYeAkMpYNAC+lLAoAMMpWXfa6v4Wmp9YM6ecFXWU1XbpMm8Pa6ScfaVdj9Xsq6m3dQV2tV0W87TTmdk0JTqUViqh7pEj8JSA9SlehSWmNYVluqgKjGYvi8xLaUG04dWddsXAwgGEGReJwBEAg5+MhG8ZEIIvIphkGqglWigEapRyKlwmymRZ7iNYlZa6db5JGGmN1rTB0fZ13IfOOYPE4mRQcKMZeuN5b6v/IPK9DxYPkysPqDKPsjurLf+4BGj0leeP9xdX+9W1kHANyPKkqNyCZCW41AKDqWcAKWWRMj6ey0HaGFap+c4088VkivTevM2TQmadXKmtUrYzAneXQlX2T74fseOYCVYw+P+KbmxE6FIgxPaHxDmH4a2DUMR6hOKUN9QBHkHQSwQAzB1sReW6k3d62VnqObvzQlQ+cQot8jUpkRXduZaVLbUEMcBCi8x6nlLUM/b9NVfLkGAjxj+3hLU87b+vp6PGP4yMSScwZTslE949Nq7vt6VGFWWJN39HIO2ku1Ut71yj5V/Z3PGhMsDiQH4ly01IixbqsEAqAQCXBOJcFUswjWRCNfKvp4WiZAlEsJofjNnACoZruFvMEDKKp6hinDXh3+5hEBcdiZ6J1G462wVd50B351sVNFjIYIDkwiByLRwQkAgNvVoCoXl1gvufC8Qli3m9ub1gru2U66d1TrRXetrs21RNdu/ax3HmfbBcabfoNX3Vayr8HhZm+qeU+65RgYYGKA3ctAzBr0B0DOUfc9BzwCDkZnWGTnojYDeyEyLgUFvNJZ9NX1vMDLoDAyGsvXMyMAZjBAYGURlOxMYjRAZGaRGBp2h4nPM27rzPYPeYLR8NRgZtAYDDEY9tEwLg1EHvVEPPdPCwHTQMx2MzPSzkelhgK6SHsqqejAr7+2E4O42FdtzAgOyjEGO+p9fI5Tc2MkV9RX8de6vCusFnAANvBogzDcMIT4hCPMJsyQ+DYNC0c03DD5in2q3Xaw1deVXmgyV+76gyNSNX6DRQV2qB2NAQZEOBUU6XLThWHylItTzKUuGvCUI8DYnQDIE+PhZkiF/HzECfEzfe0kceKmIMcCorzp5qjRpqqJXyqC7RxKmRYVTxEqHpVWy7r7a2W9bjJW9Id71Rmh5A7asN8JgYNAZGQxlPYpcJfuofN0dIgFnWoSmr0IBB5FAAJEA5dYJytoBBgGQAyOucgZc44y4xhnKFiOucUao73FZR8qARkxgWRob73wfxgTwAlfDD/F7fWCXfa1sO7X+EK9u2/dIEu5eR+xKULaI+Q7EgRgz/a1XTJQqJmd6g7ltVcmVdTtduYRObzAg0I/f3yQNKLaTq+qrWJe5Djc0N5ClyUJWYRayNFmWaY3V8ZP4WSU95b+G+YShvld9CCq7VFINrd6IgmJTMnS7rJfodpHO9FVTLhkqG6h5u0iLgmJdjT53KyMVCRDgIylLfMSo53Ont8jSK+QtQUC5xMhPKqLZBvfAGIOqRG8Zf2I9PqX8a2teb3pdjbV8Hf297/T0mRPXAB8J6vmUvXY+pt4+83q5TAyBwPo1ZIxBpVXhmvoarhZexTX1tTtL4TVkabJgZNWPo2ng1QCN/BqhkW8jNPJrhMZ+jS0/B3oF0v8bQjyQLZ/flNw4kJEZkV+Sj6zCLNzQ3EC2Jhs3CsuSn7JFWaq853bEAjFCfEJMSU9Z4hPmU9YTVNYjJBXefxVmg5FBVayzfFje1pT7vuxD897jiGpOJODuJENlCZClt6iKS2cKLzGEAtf8YGOMQaM1WBKUu5MV8++zfPJSUFT736+fVGRKMn1MvW/WCcqdxMWcrCi8xBAJa5ZE64w6ZBdm30leCq9ZJTFqnbra50uFUkvicncSE+YbBi9R5RXHCSGei5KbatRlclMTGp2mQtJzo7AsEdLcQG5R7j3PcgGgvqy+dY/PXZfBFFKFQ852bR1HdPvucUQ2qnQc0d1jhyw9R6bv/b0lkIjs341frDVUTFA0WuQX6Srtabmt0UFrqN1xe0uEd/WiVJ2s1PMRw9/r/o9ZWarEtcJruKq+atXzck19DdmabBhY9VN17+59Kf99oFegzb2RhBDPRslNNZwtubkXvVGP3KLcCklP+Utfxfp7T0P0EnlVuNxl7vkJ9TENfBYJ6m4I1t3jiMpfLqtw6azcOKLa8pWK4F92qcWqt6hcj1E9bwn8ZCJoSvXlEhNdJfVUTDHVNkGTiASoX+GSj7hiklIuNpnY/mOadEYdsjXZlqTl7iRGrb1370tD34Z3Lhv53klgGvo1pN4XQohdUXJTDVdLbu6FMQZlqdKU/Nx96ass+blVcuue2xFwAgR5B1VIesxLmG8YvMXedXBEVavrcUQ1IRZyVsmIueekfI+KZaxR2QBsL7GwzsaMmHtfyict5iSmJr0vgV6BVj0v5ZMY6n0hhNQlSm6q4W7JTU2U6EuQU5Rj3fNTmGU19kdvvHeviFwir3TWlzn5CZAFON2HnfU4ojtJUUEV9YhUxXr4yUSVDqCtOLhWDF+eB0XrjXpkabIqjHu5qr5ao94XiUBS6biXRr6NnCKhJYQQM0puquGJyc29GJkRt4pvVbjcZf56Q3Pjnh+SgOmD0jLw2Te00gHQEqGkDo7Ivai0qkrHvVxTm2Ye2dr7Uj6Jod4XQoiroOSmGpTc1E6htvBOT0/Z7K/yCdDN4ps1Gvgc6BVodbmrfM9PqE8o5BK5x03z1Rv1prEvheV6XcolMSqtqtrnV9b7Yv7a0Lch9b4QQtwCJTfVoOTGMXRGHXKLcu9c+rpryntWYRZKDCX33I63yLvaWV8NvBvU6cBnezHXfSmftJiTmJr0vtSX1a9Q78X8tYF3A+p9IYS4PbpxJqlzYoEYDX0boqFvw0ofZ4yhoLTANOi5sOKMryxNFvJL8lGkL0KmMhOZysxKtyPkhAjyDqq03o+5N4iPnoq7e1/uTmJq0vvS0K9hhXEv1PtCCCG2o+SG1AmO41BPVg/1ZPXQrn67StuU6EusenrK9/zcKLyBHE0O9ExvWYfcyvflL/VHqE9opbO+Qn1DUV9Wv1aXvtRadYVLRpaqu4VZ0LPqB2Wbe18qG7xLvS+EEGI/dFmKuAyD0YC84rwKSY95Blh2YfY9K+MCpl6SUN/QSmd9BXoFIq84z2rGkTmBuVc16bt7X+4exEu9L4QQUns05qYalNy4N7VWXWGmV/nLYDeLboJVdgPKGgqQBVQ67qWRXyMEeQdR7wshhDgIjbkhHstP4gc/iR9a1mtZ6eM6gw45RTkVe37KBkDnFeeZpk7fXXWXel8IIcRlUHJDPIpYKLYkK4QQQtwT9aETQgghxK1QckMIIYQQt0LJDSGEEELcCiU3hBBCCHErlNwQQgghxK1QckMIIYQQt0LJDSGEEELcCiU3hBBCCHErlNwQQgghxK1QckMIIYQQt0LJDSFlSk6fxqWRT0Kzbz/foRBCCLkPlNwQUubml1+iOCMDufPn8R0KIYSQ+0DJDSEA9Pn5KNy+AwBQ8t9RlF64yHNEhHim4v/+gyo1FUaNhu9QiAuju4ITAkC1bj2g11t+Vq5ejaDXpvAXECEeSH/zJi4/OwpMqwUnk8G3Vy/4DYqDX58+EPj48B0ecSHUc0MIgILkfwAA3lFRAADlmjVgRiOfIRHicZRr14FptYBQCFZSAnVqKm68/gbOdu+Ba6++CuXadTAUUo8OuTdKbojHKzlzBqUnTwFiMcI+/QQCuRz6rCwU7aeBxYTUFcYYlMnJAICQ995FxN9/of748RA3bQJWWgp16hbcePNNnOveHVcnTIRy7VoYCgv5DZo4LbosRTye8p9kAIBfnz4QBwdDPngwClasgDJ5NXxiYvgNjhAPUXrqFErPngUnkUA+eDCECgVkbduiwWtTUHrmDFQpKVCnbIL20iUUpqWhMC0NnEQCn549IY8bCN9+/SD08+P7MIiToJ4b4tGYTgflunUAAMXwJNPXpEQAoEGNhNShgrKTDN/+/SBUKCzrOY6DrHVrBE2ZgmYbNyBidTLqv/wSJBERYFotCrduxY3/m4Zz3Xvg6ksvoyA5GQaViqejIM6Cem6IRyvcvRuGvDwIAwLg26sXAMCrUydImjaF9vJlqDanwr8s6SGEOAbTaqEqO8nwHz68ynYcx0HWqhVkrVqhwaRJKD13DuqUTVBt2gRtZiYKt29H4fbtyBKL4dM9BvK4QfC7K1kinoF6bohHUyavBgAo4oeBE4sBmN5Azb035jEAhBDHKdy5E4bbtyFsEAif7t1r9ByO4yBr2RINJr2K5uvXodnaNQicOBHSyBaATgfNjp3IevttnO3ZC1fGj0fBX3/DUFDg2AMhToOSG+KxDAUFKNy6FQCguOtsUZGQAAAo2r8fuuvX6zw2QjxJQdlJhCIhAZyodhcUpJGRaDBxApqtXYtm69ch8NWJkEZGmhKdnbuQ9c47pkRn3AsoWLUK+tu37XgExNlQckM8lnLDBjCdDtLWrSFr3drqMXHDhnemha9dy0d4hHiE8gU0/ZOS7LJNafPmaDBhApqtXYNmG9ajweRJkLZqBej10Ozejax338O5nr1wZew43F65khIdN0TJDfFY5ktSVY2pUZS90Sr/SQZjrI6iIsSzmAtoytq1M/W02Jm0WTMEvvwymq1ORrONG9BgymRIW7cGDAZo9uxB9nvvmxKd55/H7RV/Qp+fb/cYSN3jmIe9a6tUKigUCiiVSsjlcr7DITwpzczEhaHDAJEIkTu2Q1S/foU2hkINzvXqBVZcjPA/fodXp051Hyghbu7CI4+g9OQpBL/7LgKeebrO9qu9dAmqTZuh2pRiqnNlJhTCO6ob5HFx8BswoNL3BsIPWz6/qeeGeCTzQGHfXr2qfPMS+vpAPnAAgDtjAggh9lO+gKZ86JA63bckPByBL45Hs7//RvNNKWgwdSpkbdsCBgOK0vche+YHONfrYVwe/Rxu//479Hl5dRofuT+U3BCPwwwGKFevAXCntk1VFIllNW82bISxtNTRoRHiUcoX0BTVq8dbHJKmTRE4/gVE/P0XmqduRtAbr0PWvj1gNKJo/35kfzAL5x7ujcujRiP/11+hv3mTt1hJzVByQzyOZm869Lm5ECoU8O3Tp9q23tHREIWEwKhSoXDb9jqJjxBPwPT6CgU0nYGkcWPUHzcOEatWovmWVAS9+QZkHTqYEp0DB5Dz4UemROeZZ5G//FfocnP5DplUgpIb4nGU/5hukikfNgwCiaTatpxQaJkWTjVvCLGfygpoOhtJo0aoP3YsIlb+iRZpWxD01luQdXwAYAxFBw8i56OPcL53H1x6+hnk//wLdDk5fIdMylByQzyKQaWCOi0NwJ3ZUPdiLuhXuGsXXXcnxE7Ml6TKF9B0ZuKGDVH/+TGIWLECLbamIWja/5kmGTCG4kOHkPPxx6ZE56mnkf/zz9BlZ/Mdskej5IZ4FNXGFLDSUkgjW0DWvl2NniNt1gyyBx4ADAao1q93cISEuL/qCmi6AnFYGOo/9xzC//gdLbZtRfD0afDq3BkAUHz4MHI+no3zffri0sgncWvZMuiysniO2PNQckM8ivnSkiIpCRzH1fh55t6bgrLaOISQ2quugKarEYeGImD0aIT//hta7NiO4LffhlfXrgDHoTgjA7lz5uJ83364NGIkbv24DLobN/gO2SNQnRviMbSXLiFz0GBAIECL7dsgDgqq8XP1t2/j3MO9AZ0OEauTIWvVyoGREuLeLj4xAiVHjyJ4+jQEjB7NdzgOocvJhXqzqY5O8aHDQLmPWtkDD5jq6MTFQdKoIY9Ruhaqc0NIJcy1anx69rApsQEAUb168CubWWUeK0AIsV1pZiZKjh4FRCLIhw3jOxyHEQcHIeDZZxC+fLmpR+e9d+HdrRvAcSg5ehS5n36KzNhYXHz8Cdz64Qdor13jO2S3QskN8QjMaLTUtvGv5TV+83RV5bp1YHq9vUIjxKPUpICmuxEHBSHg6afR9JefEblzB4Lff8907zqBACXHjiH308+QGTsAFx99DHnffQft1at8h+zyanf7VUJcTNH+/dBnZUEgl8O3X79abcO3Z08I69WDIS8Pmj174Nu7t52jJMS92VJA012JGjRAwFNPIeCpp6DPy4N6yxaoUjah6MABlJw4gZITJ3Dz83mQtW0Lv0GDII8bCEnTpnyH7XKo54Z4BPPZonzwYAik0lptg5NILN3odDsGQmxnSwFNTyAKDES9kSPRdNmPiNy1EyEzZ8KnewwgFKLk5EncnDcPmXGDcGH4I8hb8g20ly7xHbLLoJ4b4vYMhRqoNqcCqPoO4DWlSErE7V9+QWHaVhhUKghpUDohNWZLAU1PI6pfH/VGjkC9kSOgz8+HessWqFM2QbN/P0pPncLNU6dwc8ECSFu1gnxQHPziBkHaLILvsJ0W9dwQt6fetAmsuBiS8HDIOna8r23J2raFNDISTKuFamOKnSIkxP0Z1GqbC2h6KlFAAOo98QSaLP0Bkbt3IfSjD+HTsycgEqH0zBncXLgIF4YMwYWERNz86iuUXrjAd8hOh5Ib4vbMZ4uK4cNtqm1TGY7jLDVv6HYMhNScauNGmwtoEtNMTf/HHkOT779D5K6dCP3fR/Dp1cuU6Jw9i7xFX+DCkKG4EB+Pm18uRun583yH7BSozg1xa9qrV5E5YCDAcWixbSvEISH3vU1dbi7O9+kLGI1ovimFBvsRUgOXnnwKxUeOIOjNN1B/7Fi+w3F5BqUS6rStUG1KgWZvOqDTWR6TtGgOedwgyAfFQRoZyWOU9kV1bggpoyyrKOwTE2OXxAYwTev06dHDtP3VVLGYkHvRXrqE4iNHAIEA8vh4vsNxC0KFAv6PDEeTb75By927EDp7tmkGp1gM7flM5C1ejAvxCcgcOgw3Fy1CyZmz8KS+DEpuiNsy1bYxJR/2nnaqSDRfmloNZjTadduEuJv7KaBJ7k2oUMB/eBIaf7MELffsRtjcOfDt2xecWAxtZibyvvoaFxMTcWHIUOQuXIiS06fdPtGh2VLEbRUfOgTdtWsQ+PjALzbWrtv2i+0Pga8vdDduoOjgQfhERdl1+4S4C3sU0CQ1J5TLoUhMhCIxEQa1GoXbtkG1aTM0u3ZBe/Eibn29BLe+XgJJ06amOjqD4iBt3fq+xyM6G+q5IW6roOw2CX6DB0Hg5WXXbQtkMsgHDwJw59IXIaQiexTQJLUj9PODIiEBjRd/ici9exD26afwje0PTiKB9vJl3PrmG1wc/ggyBw1C7rz5KD5xwm16dCi5IW7JWFQEdYppqrajzhbNl6bUKSkwFhU5ZB+EuDp7FNAk90/o6wtF/DA0/vJLRO7di7DPP4PfgAHgpFLoLl/BrW+/xaVHH0Nm3CDkfv45io+7dqJDl6WIW1KnpsJYVARxkybw6tLFIfvw6toV4saNobt6FeotW6BISHDIfghxVfYsoEnsR+jrA8XQoVAMHQqjRoPCHTugStmEwp07obtyBbe++x63vvse4kaN4Bc3EPJBgyBr396lLl1Rzw1xS+YBjIrEBIf9QXIcZzWwmBBizZ4FNIljCHx8IB8yBI0WLUTLvXvQcMF8+A0aBM7LC7pr15D/w1JcevwJZMYOQM4nn6L46FGX6NHhPblZvHgxwsPDIZPJEB0djQMHDlTZVqfTYdasWWjevDlkMhk6duyIlBSqEkus6W7cQNG+/QAARWKSQ/elSDT11mjS06HLznbovghxNfYsoEkcT+DtDfmgQWi0YD5a7tmNhgsWwG9wWaJz/Tryly7FpSdG4Hz//siZMxfFGRlOm+jwmtysWLECU6dOxYwZM3D48GF07NgRcXFxyM3NrbT9u+++i2+++QZffPEFTp48iZdeegnDhw/HkSNH6jhy4syUa9YAjME7KgqSRg0dui9J48bwerArwBiUa9Y6dF+EuBLttWsoOngQ4DjLSQBxHaZEJw6N5s839egsWgj5kCHgvL2hv5GF/GXLcGnkkzjfrz9yZs9B0ZEjTlUWg9cKxdHR0ejWrRu+/PJLAIDRaETjxo3x6quvYtq0aRXah4WF4Z133sGECRMs6x599FF4eXlh+fLlNdonVSh2b4wxXBg0GNrLlxE6e3adXOcvWLUKWe++B0nz5mi2bi2doRIC4OaXi5H35Zfw6d4dTZb+wHc4xE6MJSUo3LUL6k2bUbh1q9VkClFICPwGDoB80CB4deoETmDf/hOXqFCs1Wpx6NAhxJarPyIQCBAbG4v09PRKn1NaWgqZTGa1zsvLC7t3765yP6WlpVCpVFYLcV/FRzKgvXwZnLc35AMH1Mk+/QYNAieVQpuZiZLjx+tkn4Q4M2Y0WmZJ2buAJuGXQCaDfMAANPzsU0Sm70WjxV9CHh8PgY8P9NnZuP3zL7j81NO4EJ/A6yUr3pKbvLw8GAwGBAcHW60PDg5GdhVjF+Li4jBv3jycO3cORqMRqamp+Pvvv5GVlVXlfmbPng2FQmFZGjdubNfjIM7FMu104EAIfHzqZJ9CX19LkUBlWW0dQjyZIwtoEuchkErh178/Gn76CSL37kGjrxZDkZgAga8vZG3a8NqLzfuAYlssXLgQkZGRaN26NSQSCSZOnIgxY8ZAUE3X1/Tp06FUKi3L1atX6zBiUpeMJSVQbdgAAFAkJdXpvs37U61fD6bV1um+CXE2jiygSZyTQCqFX79+CJs7F5F79yB42v/xGw9fOw4MDIRQKEROTo7V+pycHIRUcYPDBg0aIDk5GRqNBpcvX8bp06fh6+uLZs2aVbkfqVQKuVxutRD3pE5Lg7GwEOKwMHhHdavTfft0j4GoQQPTnXp37KjTfRPiTOqigCZxbgKJBKLAQH5j4GvHEokEXbt2RVpammWd0WhEWloaYmJiqn2uTCZDw4YNodfr8ddffyGxrNYI8WzmS0KKpES7D2S7F04ohDzBdLdjqnlDPFldFNAk5F54vSw1depUfPfdd/jpp59w6tQpvPzyy9BoNBgzZgwAYNSoUZg+fbql/f79+/H333/jwoUL2LVrFwYNGgSj0Yi33nqLr0MgTkKXkwPN3r0A6v6SlJl/2X4Ld+yAPj+flxgI4VtdFNAk5F54vf3CiBEjcPPmTbz//vvIzs5Gp06dkJKSYhlkfOXKFavxNCUlJXj33Xdx4cIF+Pr6YsiQIfjll1/g7+/P0xEQZ6FcswYwGuHVtSskTZrwEoM0MhKydu1QcuIEVOs3IODZZ3iJgxC+1GUBTUKqw/u9pSZOnIiJEydW+tj27dutfu7duzdOnjxZB1ERV8IYs1wKUiTxe4lSkZiIkhMnoExOpuSGeJy6LKBJSHVcarYUIZUpOXYM2sxMcDIZ5IMG8RqLfNhQQCRCyYkTKD13jtdYCKlLjLE7495oIDHhGSU3xOWZa9v4xcZC6OfHayyigAD49u4NAFCupoHFxHPwUUCTkKpQckNcmlGrhXJ9WW0bJ6mEar40plyzFsxg4DkaQuoGHwU0CakKJTfEpRVu3QajUglRcDB8HnqI73AAAL69e0OoUECfmwvN3spvJUKIO+GzgCYhlaHkhrg0y/1rEhPBCYX8BlNGIJFAPnQoALo0RTwDnwU0CakMJTfEZenz8lC4axcA5ztbNF8iU2/ZAkNhIb/BEOJgfBbQJKQy9L+QuCzl2nWAwQCvjh0hbRbBdzhWZO3bQ9KsGVhJiaUUPSHuyKqAJlWLJ06CkhvikkzTTv8B4DwDicvjOM7Sm0S3YyDuzKqAZtOmfIdDCABKboiLKj11CqVnz4KTSCAfPJjvcCqlSIgHOA5FBw9Ce+0a3+EQYnfOVECTkPIouSEuqaDsGr9v/34QKhT8BlMFcUgIfGJMM7io94a4I2cqoElIeZTcEJfDtFqo1q0DcOdmlc7Kcmlq9WowxvgNhhA7c6YCmoSUR8kNcTmFO3fCcPs2hA0C4dOjB9/hVMsvNhYCb2/orl5F8eHDfIdDiN04YwFNQswouSEup8Bc2yY+AZyI93u/Vkvg7Q2/su5681kuIe7AGQtoEmJGyQ1xKfr8fBRu3wHAdQYwmqfHqjamwFhSwnM0hNiHMxbQJMSsRqe9a9asqfEGExISah0MIfeiWrce0Osha9cOspYt+Q6nRry7PQhxWBh0N25AnZYGRVn1YkJclTMX0CQEqGFyk3TXf16O46wGR3IcZ/neQDcKJA5kOVscPpzfQGzACQRQJCUi76uvoUxeTckNcXnOXECTEKCGl6WMRqNl2bx5Mzp16oSNGzeioKAABQUF2LBhA7p06YIUqsRKHKjkzFmUnDwJiMWQDx3Cdzg2UZT1aGr27IEuJ5fnaAipPWcvoEkIUMOem/KmTJmCJUuWoGfPnpZ1cXFx8Pb2xvjx43Hq1Cm7BkiImWXaaZ8+ENWrx28wNpKEh8Orc2cUHzkC1bp1qD/2eb5DIqRWXKGAJiE2DyjOzMyEv79/hfUKhQKXLl2yQ0iEVMT0eijXrgXgumeLd27HkEw1b4jLcoUCmoTYnNx069YNU6dORU5OjmVdTk4O3nzzTURFRdk1OELMCnfvhiEvD8KAAPj26sV3OLUiHzwInESC0nPnTJfXCHExrlRAk3g2m5ObpUuXIisrC02aNEGLFi3QokULNGnSBNevX8cPP/zgiBgJgbLsbFERPwycWMxvMLUklMvh278fALodA3FNrlRAk3g2m8fctGjRAkePHkVqaipOnz4NAGjTpg1iY2OtZk0RYi+GggIUbt0KwPWnnfonJUG9MQWqdesQ/NabLpuoEc/kSgU0iWer1f9OjuMwcOBADBw40N7xEFKBcsMGMJ0O0tatIWvThu9w7otPjx4QBgbCUFYnxK9fP75DIqRGXLGAJvFcNUpuFi1aVOMNTpo0qdbBEFIZ8yUcd3hD5UQiKIYNQ/6yZVD+k0zJDXEZrlhAk3iuGiU38+fPr9HGOI6j5IbYVWlmJkqOHgVEIiji4/kOxy4Uw5OQv2wZ1Nu3w1BQAGElsw8JcTauWECTeK4aJTcXL150dByEVMr8hurbqxdE9evzG4ydyFq1grRNG5SeOgXlhg0IeOopvkMipFquXECTeKb7unEmY4zqdRCHYQYDlKtN9zVz1do2VVEkmioW06wp4gpcuYAm8Uy1Sm5+/vlndOjQAV5eXvDy8sIDDzyAX375xd6xEQ+n2ZsOfW4uhAoFfPv04Tscu1IMGwYIhSg5ehSlFy7wHQ4hVXKHAprE89ic3MybNw8vv/wyhgwZgj///BN//vknBg0ahJdeeqnGY3MIqQnz2aJ82DAIJBJ+g7EzUWCgpRgh9d4QZ+YOBTSJ57F5KvgXX3yBr7/+GqNGjbKsS0hIQLt27TBz5ky89tprdg2QeCaDWg31li0AXL+2TVUUSYko3L4dyjVr0GDyJHBCId8hEVKBOxTQJJ7H5p6brKwsdO/evcL67t27Iysryy5BEaLauBGstBTSyBaQtW/HdzgO4du3LwRyOfTZ2Sg6cIDvcAipwJ0KaBLPYnNy06JFC/z5558V1q9YsQKRkZF2CYoQy9liUpLbVr4WSKWQDzHdVdl8CY4QZ+JOBTSJZ7H5stQHH3yAESNGYOfOnehRdm+RPXv2IC0trdKkhxBbaS9dQvGRI4BAALmb1LapiiIxEQV/rIBqcyqC33sfQl8fvkMixMKdCmgSz2Jzz82jjz6K/fv3IzAwEMnJyUhOTkZgYCAOHDiA4VTcidiB+f41Pj17QBwUxG8wDubVqRMkTZuCFRdDvXkz3+EQYuGOBTSJ56hRcjN16lRoNBoAwM6dO9GxY0csX74chw4dwqFDh7B8+XJ07tzZoYESz8CMRijXmGrb+HvANX6O4yzTa5WradYUcR7uWECTeI4aJTdffPEFCgsLAQB9+/ZFfn6+Q4MinqvowAHob2RB4OcH3/79+Q6nTigSTAX9ivbvh+76dZ6jIcS9C2gSz1CjMTfh4eFYtGgRBg4cCMYY0tPTUa+KKpUPP/ywXQMknkX5zz8AAPmQIRBIpTxHUzfEYWHwjo5G0f79UK5Zg8CXX+Y7JOLh3LmAJvEMNUpuPv30U7z00kuYPXs2OI6rcmwNx3EwGAx2DZB4DkOhBqrNqQAAfw87W1QkJZmSm+TVqP/SS247Q4y4BncuoEk8Q40uSyUlJSE7OxsqlQqMMZw5cwa3b9+usNDlKnI/1Js2gRUXQxIeDlnHjnyHU6fkAweA8/KC9vJlFGdk8B0O8WCeUECTuD+bZkv5+vpi27ZtiIiIgEKhqHQhpLbMZ4uK4cM9rudC4OMD+cABAOh2DIRf5gKakhbN3baAJnF/Nk8F7927N0Qim8vjEFIt7bVrKPr3X4DjLHfM9jTms2TVxo0wlpbyGwzxWOYCmv4eeJJB3Eet7gpOiL2Zeyt8YmIgDgnhORp+eEdHQxQaCqNKhcJt2/gOh3ggTyqgSdwbJTeEd8xoLHdJKonXWPjECQSWYmnms2dC6pInFdAk7o2SG8K74kOHoLt2DQIfH/jFxvIdDq/Ml6YKd++GPi+P32CIR/G0AprEvdmc3Pz4448oKipyRCzEQxWU9VL4DR4EgZcXv8HwTNosArKODwAGA5Tr1vEdDvEgnlhAk7gvm5ObadOmISQkBGPHjsXevXsdERPxIMaiIqhTUgDQ2aKZItF0k0KaNUXqkicW0CTuy+bk5vr16/jpp5+Ql5eHPn36oHXr1pg7dy6ys7MdER9xc+otW2AsKoK4cWN4de3KdzhOQTFkCDixGKWnT6Pk9Gm+wyEewJMLaBL3ZHNyIxKJMHz4cKxevRpXr17FCy+8gF9//RVNmjRBQkICVq9eDaPR6IhYiRsqKDtbVCQl0rTTMkJ/f/j27QuAem9I3fDkAprEPd3XgOLg4GD07NkTMTExEAgEOHbsGEaPHo3mzZtj+/btdgqRuCvdjRso2rcfAKBITOI3GCdjHlisXLcOTK/nNxji9jy5gCZxT7VKbnJycvDZZ5+hXbt26NOnD1QqFdatW4eLFy/i+vXreOKJJzB69Gh7x0rcjHLNGoAxeEdFQdKoId/hOBXfXj0hDAiAIS8Phbt38x0OcWNUQJO4I5uTm/j4eDRu3BjLli3DCy+8gOvXr+P3339HbNkUXh8fH7z++uu4evWq3YMl7oMxZqnloqjiRqyejBOLIR82FACgXE2Xpojj3Cmg+ZDHFtAk7sfm+ygEBQVhx44diImJqbJNgwYNcPHixfsKjLi34iMZ0F6+DM7b23JPJWLNPykJt3/+BYVpW2FQKiGke7cRO7MuoEknGcR92NRzo9PpcOnSJQQGBlbbjuM4NG3a9L4CI+7N/IYqHzgQAh8ffoNxUtI2bSCNjATTaqHamMJ3OMQNUQFN4q5sSm7EYjGOHj3qqFiIhzCWlEC1YQOAOwNnSUUcx90ZWEyXpogDUAFN4q5sHnPzzDPP4IcffnBELMRDqNPSYCwshDgsDN5R3fgOx6nJ44cBAgGKjxyB9tIlvsMhboQKaBJ3ZvOYG71ej6VLl2LLli3o2rUrfO66pDBv3jy7BUfck3kAoyIpEZyAbm9WHXFQEHx69IBm1y4UrF6NoMmT+Q6JuAkqoEncmc3JzfHjx9GlSxcAwNmzZ60eo/oI5F50ObnQ7NkD4M5tBkj1FEmJ0OzaBeXq1Wjw6quUEBK7oAKaxJ3ZnNxs27bNEXEQD6FauwYwGuHVtSskNOi8Rvz694fA1xf6G1ko+vcgfKKj+A6JuDgqoEncXa1PAc+fP49NmzahuLgYgKluCSHVYYxZBjAqkqjXpqYEMhnkgwcBuDPLjJD7QQU0ibuzObm5desW+vfvj5YtW2LIkCHIysoCAIwdOxavv/663QMk7qPk+HFoMzPByWSQDxrEdzguxTxrSr1pE4xFRfwGQ1waFdAknsDm5Oa1116DWCzGlStX4O3tbVk/YsQIpKRQLQ5SNWXZNX6/2FgI/fx4jsa1eHXpAnGTJqYZLlu28B0OcWFUQJN4ApuTm82bN2Pu3Llo1KiR1frIyEhcvnzZboER92LUaqFcX1bbZngSv8G4IK7cfX/o0hS5H5YCmgMGUAFN4rZsTm40Go1Vj41Zfn4+pFKpXYIi7qdw6zYYlUqIgoPh89BDfIfjksyzyzTp+6AruxxMiC2sCmjSJSnixmxObnr16oWff/7Z8jPHcTAajfjkk0/Qt29fuwZH3Ifl/jWJieCEQn6DcVGSRo3g/eCDAGNQrl3HdzjEBVEBTeIpbJ4K/sknn6B///44ePAgtFot3nrrLZw4cQL5+fnYU1a/hJDy9Hl5KNy1CwDdbuF+KYYnoejgQSiTk1H/hXFUn4TYRFlutiLVSyLuzOb/3e3bt8fZs2fRs2dPJCYmQqPR4JFHHsGRI0fQvHlzmwNYvHgxwsPDIZPJEB0djQMHDlTbfsGCBWjVqhW8vLzQuHFjvPbaaygpKbF5v6TuKNeuAwwGeHXsCGmzCL7DcWl+cXHgZDJoL1xAybFjfIdDXIguJxeavXsBUAFN4v5s7rkBAIVCgXfeeee+d75ixQpMnToVS5YsQXR0NBYsWIC4uDicOXMGQUFBFdr/9ttvmDZtGpYuXYru3bvj7NmzeO6558BxHN32wUmZpp2WVUKlgcT3TejrC7/YWKjWrYMyeTW8HniA75CIi6ACmsST2Nxz06JFC8ycORPnzp27753PmzcPL7zwAsaMGYO2bdtiyZIl8Pb2xtKlSyttv3fvXvTo0QNPPfUUwsPDMXDgQDz55JP37O0h/Ck9dQqlZ8+Ck0ggHzyY73DcgvnSnmr9ehi1Wn6DIS6BCmgST2NzcjNhwgSsX78erVq1Qrdu3bBw4UJkZ2fbvGOtVotDhw4hNjb2TjACAWJjY5Genl7pc7p3745Dhw5ZkpkLFy5gw4YNGDJkSJX7KS0thUqlslpI3SkoG0js278fhAoFv8G4CZ+YhyAKCoJBqUTh9u18h0NcABXQJJ6mVkX8/v33X5w+fRpDhgzB4sWL0bhxYwwcONBqFtW95OXlwWAwIDg42Gp9cHBwlcnSU089hVmzZqFnz54Qi8Vo3rw5+vTpg7fffrvK/cyePRsKhcKyNG7cuMYxkvvDtFqoymb1+NNAYrvhhEIoEuIB3LnDOiHVoQKaxNPUerh8y5Yt8cEHH+Ds2bPYtWsXbt68iTFjxtgztgq2b9+Ojz/+GF999RUOHz6Mv//+G+vXr8eHH35Y5XOmT58OpVJpWa5everQGMkdhbt2wXD7NoQNAuHTowff4bgV86Wpwp07oc/P5zcY4tSogCbxRLUaUGx24MAB/Pbbb1ixYgVUKhUef/zxGj83MDAQQqEQOTk5VutzcnIQEhJS6XPee+89PPvssxg3bhwAoEOHDtBoNBg/fjzeeecdCCqZ2iiVSqm4IE8KzAOJ4xPAie7rvxq5i7RFC8jat0fJ8eNQrVuPgFHP8h0ScVJUQJN4Ipt7bs6ePYsZM2agZcuW6NGjB06dOoW5c+ciJycHf/zxR423I5FI0LVrV6SlpVnWGY1GpKWlISYmptLnFBUVVUhghGUF4eiu5M5Fn5+Pwu07ANAARkcxT+el2zGQ6lgKaCYkUAFN4jFsPp1u3bo1unXrhgkTJmDkyJEVxszYYurUqRg9ejQefPBBREVFYcGCBdBoNJbLW6NGjULDhg0xe/ZsAEB8fDzmzZuHzp07Izo6GufPn8d7772H+Ph4S5JDnINq3XpAr4esXTvIWrbkOxy3JB82FDmffIKSkydRcvYs/Z5JBVYFNOmSFPEgNic3Z86cQWRkpF12PmLECNy8eRPvv/8+srOz0alTJ6SkpFgSpitXrlj11Lz77rvgOA7vvvsurl+/jgYNGiA+Ph7/+9//7BIPsR/L2SLdv8ZhRPXqwbf3wyjckgbl6tWQvfkm3yERJ2MuoCnr+ACkzZrxHQ4hdYZjtbieU1BQgFWrViEzMxNvvvkmAgICcPjwYQQHB6Nhw4aOiNNuVCoVFAoFlEol5HI53+G4pZIzZ3ExMREQixG5cwdE9erxHZLbUqWm4vqrkyBq0AAttm2lsU3EgjGGi4lJKD17FiEzZ6DeyJF8h0TIfbHl89vmd8KjR4+if//+8Pf3x6VLl/DCCy8gICAAf//9N65cuWLTdHDinsy9Nn59+lBi42B+vXtD6O8P/c2b0KTvg2+vnnyHRJwEFdAknqxWdW7GjBmDc+fOQSaTWdYPGTIEO3futGtwxPUwvR7KtWsB0DX+usBJJJAPHQqABhYTa1RAk3gym5ObgwcP4sUXX6ywvmHDhrWqVEzcS+Hu3TDk5UEYEADfXr34DscjmGejqbdsgUGt5jka4gyogCbxdDYnN1KptNJbGJw9exYNGjSwS1DEdZkr5irih4ETi3mOxjPI2reHpHlzsNJSqFJS+A6HOAEqoEk8nc3JTUJCAmbNmgWdTgcA4DgOV65cwf/93//h0UcftXuAxHUYCgpQWFa3SEFni3WG4zhL741yNd2OgVABTUJsTm4+//xzFBYWIigoCMXFxejduzdatGgBPz8/mpLt4VQbN4LpdJC2bg1ZmzZ8h+NRFPHxAMeh+OAhaOkWIx6NCmgSUovZUgqFAqmpqdizZw/+++8/FBYWokuXLlZ39yaeqeCfZAD0hsoHcUgIfGJioNm7F8rk1Wjw6kS+QyI8oQKahNzHvaV69OiBHnQtl5QpzcxEydGjgEhk6kUgdU4xPMmU3KxejcCJE8BxHN8hER5YCmjSpWHiwWp9V3BCyjO/ofr26gVR/fr8BuOh/GJjIfD2hu7aNRQfOsR3OIQHJWfOouTkSUAshnzYUL7DIYQ3lNyQ+8YMBihXrwFAtW34JPDygt+gQQDu1DghnuVOAc3eVECTeDRKbsh906Tvgz43F0KFAr59+vAdjkez1LzZmAJjSQnP0ZC6ZF1Ak+7pRjxbjZKbqVOnQqPRAAB27twJvV7v0KCIa1GWTTuVDxsGgUTCczSezfvBByFu2BBGjQbqLWl8h0PqEBXQJOSOGiU3X3zxBQoLCwEAffv2RX5+vkODIq7DoFZDvWULABrA6Aw4gQCKxAQAdDsGT0MFNAm5o0azpcLDw7Fo0SIMHDgQjDGkp6ejXhXXcx9++GG7Bkicm2rjRrDSUkhaNIesfTu+wyEAFImJyPvqa2j27oUuJxfi4CC+QyIORgU0CbFWo+Tm008/xUsvvYTZs2eD4zgMr+J6LsdxMBgMdg2QODfz2aL/8OE09dhJSJo2hVeXLig+fBiqdWtRf+xYvkMiDkYFNAmxVqPLUklJScjOzoZKpQJjDGfOnMHt27crLHS5yrNoL11C8eHDgEAAOdW2cSqKxLLbMSQngzHGczTE0aiAJiHWbJot5evri23btiEiIgIKhaLShXiOgrL7GPn07AFxEF36cCbywYPASSQoPXceJSdO8h0OcSBLAU2hEIphw/gOhxCnYHOF4t69e8NgMOCvv/7CqVOnAABt27ZFYmIihEKh3QMkzokZjZabNPrTNX6nI5TL4RfbH6oNG6FcvRpeNB7KbVkKaD78MESBgfwGQ4iTsLnOzfnz59G2bVuMGjUKf//9N/7++288++yzaNeuHTIzMx0RI3FCRQcOQH8jCwI/P/j27893OKQS5oGlqnXrwLRafoMhDmFVQJNOMgixsDm5mTRpEpo1a4arV6/i8OHDOHz4MK5cuYKIiAhMmjTJETESJ2SpbTNkCARSKc/RkMr4dO8OYYNAGG7fRuGuXXyHQxxAszf9TgHNvn34DocQp2FzcrNjxw588sknCAgIsKyrX78+5syZgx07dtg1OOKcDIUaqDanAgD86XYLTosTiaAYZhrobZ7VRtyL+ZKUfOhQKqBJSDk2JzdSqRRqtbrC+sLCQkjoj8sjqDdvBisuhiQ8HLKOHfkOh1TDfKlCvX079Ldv8xsMsSurApp0uwVCrNic3AwbNgzjx4/H/v37wRgDYwz79u3DSy+9hISEBEfESJyM+ZKUgmrbOD1Zq5aQtmkD6HRQbdjAdzjEjqiAJiFVszm5WbRoEZo3b46YmBjIZDLIZDL06NEDLVq0wMKFCx0RI3Ei2mvXUPTvvwDHWcr8E+fmX1b7xDzwlLgHKqBJSNVsngru7++P1atX4/z585ap4G3atEGLFi3sHhxxPuY3VJ+YhyAOCeE5GlIT8mHDkPPJpyg5ehSlFy5A2qwZ3yGR+0QFNAmpns3JjVmLFi0oofEwzGi0DGCka/yuQ1S/Pnx79ULh9u1Q/pOMoNen8h0SuU9UQJOQ6tl8WYp4ruJDh6C7dg0CHx/4xcbyHQ6xgXlgsXLNGjC6/5tLowKahNwbJTekxgrKem38Bg+CwMuL32CITXz79YVALoc+JwdF+/fzHQ65D1RAk5B7o+SG1IixqAjqjSkA6GzRFQkkEsiHDAZwJ0klrokKaBJyb5TckBpRb9kCY1ERxI0bw6trV77DIbVgTkrVqVtgKNTwGwyplfIFNOkO4IRUzebkJjw8HLNmzcKVK1ccEQ9xUgXm2jZJiTTt1EXJOnaEJDwcrLgY6s2b+Q6H1IJ60yZLAU2vTp34DocQp2VzcjNlyhT8/fffaNasGQYMGIA//vgDpaWljoiNOAldVhaK9pnGaSgSk/gNhtQax3F3BhbTpSmXZJmtmJREJxmEVKNWyU1GRgYOHDiANm3a4NVXX0VoaCgmTpyIw4cPOyJGwjPl6jUAY/COioKkUUO+wyH3QZEQD3Acig4cgPbadb7DITagApqE1Fytx9x06dIFixYtwo0bNzBjxgx8//336NatGzp16oSlS5eCMWbPOAlPGGNWt1sgrk0cFgbv6GgAgGotVSx2JVYFNENDeY6GEOdW6+RGp9Phzz//REJCAl5//XU8+OCD+P777/Hoo4/i7bffxtNPP23POAlPijMyoL18GZy3N+QDB/AdDrED80BUZfJqOglxEVRAkxDb2Fyh+PDhw/jxxx/x+++/QyAQYNSoUZg/fz5at25taTN8+HB069bNroESfij/SQYAyAcMgMDHh99giF3IBwxA9qwPob18GcVHMuDdpTPfIZF7oAKahNjG5p6bbt264dy5c/j6669x/fp1fPbZZ1aJDQBERERg5MiRdguS8MNYUgLVxo0A6GzRnQh8fCAfYOqFM1e6Jc6NCmgSYhubk5sLFy4gJSUFjz/+OMRicaVtfHx88OOPP953cIRf6rQ0GNVq0ziNKOqJcyeK4UkAANXGjTDSbEenRgU0CbGdzclNbm4u9ldSvn3//v04ePCgXYIizsE8gFGRlAhOQPUe3Yl3VBREoaEwqlQo3LqV73BINaiAJiG2s/kTa8KECbh69WqF9devX8eECRPsEhThny4nF5o9ewAAikSqhOpuOIEAigTTdGK6HYNzowKahNjO5uTm5MmT6NKlS4X1nTt3xsmTJ+0SFOGfau0awGiEV9eukDRtync4xAHMSatm9x7ob97kORpSGd2NG1RAk5BasDm5kUqlyMnJqbA+KysLIpHNk6+IE2KMoaBslhTdv8Z9SZtFQNbxAcBggHLder7DIZVQrqECmoTUhs3JzcCBAzF9+nQolUrLuoKCArz99tsYMIDqoLiDkuPHoc3MBCeTQT5oEN/hEAfyp9sxOC1TAc1kALDcNoMQUjM2JzefffYZrl69iqZNm6Jv377o27cvIiIikJ2djc8//9wRMZI6Zn5D9YuNhdDPj99giEPJBw8GJxaj9MwZlJw+zXc4pJziI+UKaMYN5DscQlyKzclNw4YNcfToUXzyySdo27YtunbtioULF+LYsWNo3LixI2Ikdcio1UK53nSJwjxdmLgvob8/fPv1A3AnqSXOwdybRgU0CbFdrQbJ+Pj4YPz48faOhTiBwm3bYVQqIQoOhs9DD/EdDqkDisREqDdtgnLdOgS98Tq4KupXkbpDBTQJuT+1HgF88uRJXLlyBVqt1mp9QgLdrdaVWW6SmZAATijkORpSF3x79YQwIACGW7dQuGcP/Pr04Tskj0cFNAm5PzYnNxcuXMDw4cNx7NgxcBxnufGeuf6CwWCwb4Skzujz8lC4axcAuiTlSTixGIr4Ycj/6Wcok1dTcuMEqIAmIffH5r+ayZMnIyIiArm5ufD29saJEyewc+dOPPjgg9i+fbsDQiR1Rbl2HWAwQNbxAUibNeM7HFKHzDVvCtPSYCg3E5LUPSqgScj9szm5SU9Px6xZsxAYGAiBQACBQICePXti9uzZmDRpkiNiJHXANO3UdEnKn67xexxpmzaQtmwJptNBVXYfI8IPSwHNLl2ogCYhtWRzcmMwGOBXNj04MDAQN27cAAA0bdoUZ86csW90pM6UnjqF0rNnwUkkkA8ezHc4pI5xHGeppUI1b/hjVUCTLg0TUms2Jzft27fHf//9BwCIjo7GJ598gj179mDWrFloRpcyXJb5/kK+/ftBqFDwGwzhhXzYUEAgQHFGBkovXuQ7HI9kKaAplVIBTULug83Jzbvvvguj0QgAmDVrFi5evIhevXphw4YNWLRokd0DJI7HtFqo1q4DcKdiLfE84qAg+PTsAQBQrl7NczSeyXxp2G/AACqgSch9sHm2VFxcnOX7Fi1a4PTp08jPz0e9evXojrUuqnDXLhhu34awQSB8evTgOxzCI/+kJGh27oJyzRo0mDSJZurUIVMBzQ0A6HYLhNwvm965dDodRCIRjh8/brU+ICCAEhsXZh5joYhPAEc3P/Vovv36QeDnB/2NLBQd+JfvcDxK4dZtdwpoxlABTULuh03JjVgsRpMmTaiWjRvR374N9fYdAOgO4AQQlLtZKg0srluWkwwqoEnIfbO5z/mdd97B22+/jfz8fEfEQ+qYat16QKeDrF07yFq25Dsc4gTMs3RUmzfDWFTEbzAeggpoEmJfNl+D+PLLL3H+/HmEhYWhadOm8Lnrhm6HDx+2W3DE8Sy3W6Br/KSMV+fOEDdpAt2VK1CnplIhuTpABTQJsS+bk5sk+hB0GyVnzqLk5ElALDZNAyYEZTVvEhOQ98WXKEhOpuTGwaiAJiH2Z3NyM2PGDEfEQXhgvsbv16c3RPXq8RsMcSqKxCTkffElivbthy4rC+LQUL5DcltUQJMQ+6N5nh6K6fVQrl0LAFDQ2SK5i6RRQ3h36wYwBuWatXyH49YsBTT7UQFNQuzF5uRGIBBAKBRWuRDXULh7Nwx5eRAGBMC3Vy++wyFOyDx7TpmcDMYYz9G4J6sCmjSQmBC7sfmy1D9l14bNdDodjhw5gp9++gkffPCB3QIjjqVMNlWgVcQPAycW8xwNcUZ+cXHI/vAjaC9eRMnRo/Dq2JHvkNwOFdAkxDFsTm4SKxlc+Nhjj6Fdu3ZYsWIFxo4da5fAiOMYlEoUpqUBoFlSpGpCX1/4DRgA1dq1UK5eTcmNAxSYZytSAU1C7MpuY24eeughpJV9YBLnptqwAUyng7R1a8jatOE7HOLEzDOllOs3wKjV8hyNe9Hfvo3CHTsBUAFNQuzNLslNcXExFi1ahIYNG9pjc8TBzAMY6Q2V3ItPzEMQBQfDqFSicNt2vsNxK1RAkxDHsTm5qVevHgICAixLvXr14Ofnh6VLl+LTTz+tVRCLFy9GeHg4ZDIZoqOjceDAgSrb9unTBxzHVViGDqU6LTVReuECSv47CohEUMTH8x0OcXKcUAhFgun/Cd0p3L6ogCYhjmPzRd758+db3SRTIBCgQYMGiI6ORr1a1EpZsWIFpk6diiVLliA6OhoLFixAXFwczpw5g6CgoArt//77b2jLdY/funULHTt2xOOPP27zvj2R8p9kAIBvr14Q1a/PbzDEJSgSE3Hru+9RuHMn9Pn5EAUE8B2Sy6MCmoQ4ls3JzXPPPWfXAObNm4cXXngBY8aMAQAsWbIE69evx9KlSzFt2rQK7QPuemP9448/4O3tTclNDTCDwXL2TWeLpKakLVpA1r49So4fh2rdOgSMGsV3SC6PCmgS4lg2X5b68ccfsXLlygrrV65ciZ9++smmbWm1Whw6dAixsbF3AhIIEBsbi/T09Bpt44cffsDIkSMr3OPKrLS0FCqVymrxVJr0fdDn5kKoUMC3bx++wyEuxJwMm0sIkNqjApqEOJ7Nyc3s2bMRGBhYYX1QUBA+/vhjm7aVl5cHg8GA4OBgq/XBwcHIzs6+5/MPHDiA48ePY9y4cdXGq1AoLEvjxo1titGdmK/xy4cOhUAi4Tka4krkQ4cAYjFKTp5EydmzfIfj0qiAJiGOZ3Nyc+XKFURERFRY37RpU1y5csUuQdXUDz/8gA4dOiAqKqrKNtOnT4dSqbQsV69ercMInYdBrYZ6yxYAdLZIbCeqVw++vR8GQL0398v8+5MPG0oFNAlxEJuTm6CgIBw9erTC+v/++w/1bRygGhgYCKFQiJycHKv1OTk5CAkJqfa5Go0Gf/zxxz2LBkqlUsjlcqvFE6k2bgQrLYWkRXPI2rfjOxzigvzNl6bWrgHT6/kNxkUZCgosBTTpDuCEOI7Nyc2TTz6JSZMmYdu2bTAYDDAYDNi6dSsmT56MkSNH2rQtiUSCrl27WhX/MxqNSEtLQ0xMTLXPXblyJUpLS/HMM8/YeggeyXy26D98uNVsN0JqyvfhhyH094fhZh40NRwTR6ypNm40FdBs1YoKaBLiQDYnNx9++CGio6PRv39/eHl5wcvLCwMHDkS/fv1sHnMDAFOnTsV3332Hn376CadOncLLL78MjUZjmT01atQoTJ8+vcLzfvjhByQlJdncW+SJtJcvo/jwYUAggJxq25Ba4iQSyMvqSZlLChDbFJT93hR0k0xCHMrmqeASiQQrVqzARx99hIyMDHh5eaFDhw5o2rRprQIYMWIEbt68iffffx/Z2dno1KkTUlJSLIOMr1y5AoHAOgc7c+YMdu/ejc2bN9dqn57GXJHYp2cPiCupHURITSmSknD711+hTkuDQa2G0M+P75BcRumFCyg5ehQQCqEYNozvcAhxa7W+U1tkZCQiIyPtEsTEiRMxceLESh/bvn17hXWtWrUCY8wu+3Z3zGi01Lbxp9o25D7J2reDpEVzaM9nQpWSgnpUX6rGLAU0H34YokpmnBJC7Mfmy1KPPvoo5s6dW2H9J598QoX0nFDRgQPQ38iCwM8Pvv378x0OcXEcx90ZWEyzpmqMCmgSUrdsTm527tyJIUOGVFg/ePBg7Ny50y5BEfsxny3KhwyBQCrlNxjiFuTx8YBAgOJDh6Ct4/IProoKaBJSt2xObgoLCyGppACcWCz26Oq/zshQqIGqbFwS3QGc2Is4OBg+ZbMZlavX8ByNa6ACmoTULZuTmw4dOmDFihUV1v/xxx9o27atXYIi9qHevBmsuBiS8HB4derEdzjEjVhux7B6NZjRyG8wTs66gGYSv8EQ4iFsHlD83nvv4ZFHHkFmZib69esHAEhLS8Pvv/9e6T2nCH/MZ4uKpCSqbUPsyi+2PwQ+PtBdu4biQ4fg3a0b3yE5LesCmu35DocQj2Bzz018fDySk5Nx/vx5vPLKK3j99ddx7do1bNmyBUk0UM5paK9dQ9G//wIcB0ViAt/hEDcj8PKC36A4AEDBahpYXB1LAU06ySCkztic3ADA0KFDsWfPHmg0GuTl5WHr1q3o3bs3jh8/bu/4SC2ZZ2b4xDwEcWgoz9EQd2SeNaXemAJjcTG/wTgp7aVL5Qpo0kkGIXWlVslNeWq1Gt9++y2ioqLQsWNHe8RE7hNjzHK2SDfJJI7i1bUrxA0bwqjRQL0l7d5P8EDmXi2fHj0gDqYCmoTUlVonNzt37sSoUaMQGhqKzz77DP369cO+ffvsGRuppeJDh6C7ehUCHx/4xcbyHQ5xU5xAAEWiaRaesqwKNrnDqoAmDSQmpE7ZlNxkZ2djzpw5iIyMxOOPPw6FQoHS0lIkJydjzpw56EaDCp1CQdlAYr/BgyDw8uI5GuLOzCUGNOnp0OXk8ByNc6ECmoTwp8bJTXx8PFq1aoWjR49iwYIFuHHjBr744gtHxkZqwVhUBHXKJgB0uwXieJImTeDVpQtgNEK1di3f4TgVKqBJCH9qnNxs3LgRY8eOxQcffIChQ4dCKBQ6Mi5SS+otW2DUaCBu3BheXbvyHQ7xAObem4LkZLrnWxkqoEkIv2qc3OzevRtqtRpdu3ZFdHQ0vvzyS+Tl5TkyNlIL5rEPiqREmnZK6oR88GBwUim05zNRcuIk3+E4BSqgSQi/apzcPPTQQ/juu++QlZWFF198EX/88QfCwsJgNBqRmpoKtVrtyDhJDeiysqBJNw3qViQm8RsM8RhCPz/4lY0poYHFJlRAkxB+2TxbysfHB88//zx2796NY8eO4fXXX8ecOXMQFBSEhASq48An5eo1AGPwjoqCpFFDvsMhHsR86UW1bh2YVstzNPyiApqE8O++6ty0atUKn3zyCa5du4bff//dXjGRWmCMWZ0tElKXfLp3h7BBIAwFBSjctYvvcHhlrjFFBTQJ4c99F/EDAKFQiKSkJKxZQ3cI5ktxRga0ly+D8/aGPG4g3+EQD8OJRFCUVeD15EtTzGgsN+4tiddYCPFkdkluCP/MZ4vyAQMg8PHhORriicwF/dTbd0B/+zbP0fCj+NAh6K5dowKahPCMkhs3YCwpgWrDBgB0uwXCH1mrlpC2bQPodJb/j56moKzXxm9QHATe3vwGQ4gHo+TGDRRu3QqjWg1xWBi8o6hKNOGPuXCkuSfRkxiLiqDemAIA8KeTDEJ4RcmNGygoq4SqSEoEJ6CXlPBHPnQoIBKh5NgxlGZm8h1OnVJv2QJjUREV0CTECdAnoYvT5eRCs2cPgDtjHgjhi6h+ffj26gXA8wYWUwFNQpwHJTcuTrV2DWA0wqtrV0iaNuU7HEIss4SUa9aCGQz8BlNHqIAmIc6FkhsXxhizDGCk+9cQZ+Hbtw8ECgX0OTnQ7NvHdzh1ggpoEuJcKLlxYSXHj0N7PhOcVAr5oEF8h0MIAEAgkUA+ZDAAzxhYTAU0CXE+lNy4MGXZQGK/AQMg9PPjNxhCyjHPmlKnpsJQqOE3GAezFND08oLfQCqgSYgzEPEdAKkdo1YL5fr1AOhskTgf2QMPQBIRAe3Fi1Bv2gT/Rx/hOySHMZ9kyAcOhNCXCmi6A4PBAJ1Ox3cYHkkikUBgh1m/lNy4qMJt22FUKiEKDoZPzEN8h0OIFY7joEhMxM0FC6BMTnbb5MZYUgLVxo0AAMXwJH6DIfeNMYbs7GwUFBTwHYrHEggEiIiIgEQiua/tUHLjoizTThMSwAmF/AZDSCUUiQm4uXAhiv79F9pr191yoK06LQ1GtRqisFB4R0XxHQ65T+bEJigoCN7e3jSlv44ZjUbcuHEDWVlZaNKkyX39/im5cUH6vDwU7twJgM4WifMSh4bC+6FoFKXvg3LNajR45RW+Q7I784BpRSIV0HR1BoPBktjUr1+f73A8VoMGDXDjxg3o9XqIxeJab4f+Gl2Qct06wGCArOMDkDZrxnc4hFTJXFhSuXo1GGM8R2Nf5Qto+tO4N5dnHmPjTfcE45X5cpThPmtkUXLjgswDGOn+NcTZyQcMAOftDd3lKyg+coTvcOzKUkCzSxcqoOlG6FIUv+z1+6fkxsWUnDqF0jNnwEkkkA8ezHc4hFRL4OMDedn0aHeqeWNVQJMuDRPidCi5cTEFZcXCfPv3g1Ch4DkaQu7NXD1btXEjjCUlPEdjH1RAkxDnRsmNC2FaLVRr1wGga/zEdXhHRUEUFgqjWo3CrVv5DscuqIAmcTaLFy9GeHg4ZDIZoqOjceDAgWrbr1y5Eq1bt4ZMJkOHDh2wYcMGq8cZY3j//fcRGhoKLy8vxMbG4ty5c1Zt8vPz8fTTT0Mul8Pf3x9jx45FYWGh5fGSkhI899xz6NChA0QiEZLq8HOLkhsXUrhrFwy3b0PYIBA+PXrwHQ4hNcIJBFAkJAAACla7/qUpKqBJnM2KFSswdepUzJgxA4cPH0bHjh0RFxeH3NzcStvv3bsXTz75JMaOHYsjR44gKSkJSUlJOH78uKXNJ598gkWLFmHJkiXYv38/fHx8EBcXh5Jyva9PP/00Tpw4gdTUVKxbtw47d+7E+PHjLY8bDAZ4eXlh0qRJiI2NddwvoDLMwyiVSgaAKZVKvkOx2dWJE9nJVq1Z9txP+A6FEJuUXLjATrZqzU62bcd0ubl8h3NflCmb2MlWrdnZh3szo17PdzjEToqLi9nJkydZcXEx36HYLCoqik2YMMHys8FgYGFhYWz27NmVtn/iiSfY0KFDrdZFR0ezF198kTHGmNFoZCEhIezTTz+1PF5QUMCkUin7/fffGWOMnTx5kgFg//77r6XNxo0bGcdx7Pr16xX2OXr0aJaYmHjPY6nudbDl85t6blyE/vZtqLfvAEB3ACeuRxoRAa+OHQGDAcqyS6uuynKTTCqg6fYYYyjS6ut8YTaUTdBqtTh06JBVz4hAIEBsbCzS09MrfU56enqFnpS4uDhL+4sXLyI7O9uqjUKhQHR0tKVNeno6/P398eCDD1raxMbGQiAQYP/+/TWO31GoiJ+LUK1bD+h0kLVrB1nLlnyHQ4jNFMOTUPzff1CuXo36z4/hO5xa0efloXDXLgA0S8oTFOsMaPv+pjrf78lZcfCW1OzjOS8vDwaDAcHBwVbrg4ODcfr06Uqfk52dXWn77Oxsy+PmddW1CQoKsnpcJBIhICDA0oZP1HPjIiy3W6Br/MRFyQcPBicWo/TMGZScOsV3OLWiXEsFNAlxBdRz4wJKzp5FyYkTgFgM+bChfIdDSK0IFQr49usH9aZNUCYnQ9amDd8h2YQxZrkkRbMVPYOXWIiTs+J42W9NBQYGQigUIicnx2p9Tk4OQkJCKn1OSEhIte3NX3NychAaGmrVplOnTpY2dw9Y1uv1yM/Pr3K/dYl6blyAufiZX5/eENWrx3M0hNSeebyYct16sLJy966i9NQplJ49C04shnzIEL7DIXWA4zh4S0R1vthSpVcikaBr165IS0uzrDMajUhLS0NMTEylz4mJibFqDwCpqamW9hEREQgJCbFqo1KpsH//fkubmJgYFBQU4NChQ5Y2W7duhdFoRHR0dI3jdxTquXFyTK+Hcs0aAICCbrdAXJxvz54Q1q8Pw61bKNy9G359+/IdUo2ZKxL79u9PBTSJU5k6dSpGjx6NBx98EFFRUViwYAE0Gg3GjDGNbRs1ahQaNmyI2bNnAwAmT56M3r174/PPP8fQoUPxxx9/4ODBg/j2228BmJK6KVOm4KOPPkJkZCQiIiLw3nvvISwszFKrpk2bNhg0aBBeeOEFLFmyBDqdDhMnTsTIkSMRFhZmie3kyZPQarXIz8+HWq1GRkYGAFh6gByFkhsnp9mzB4a8PAgDAuDbqxff4RByXzixGIphQ5H/089QJq92meTGqoAmDSQmTmbEiBG4efMm3n//fWRnZ6NTp05ISUmxDAi+cuUKBOXuWt+9e3f89ttvePfdd/H2228jMjISycnJaN++vaXNW2+9BY1Gg/Hjx6OgoAA9e/ZESkoKZDKZpc2vv/6KiRMnon///hAIBHj00UexaNEiq9iGDBmCy5cvW37u3LkzADj8Rrocc/QenIxKpYJCoYBSqYRcLuc7nHu6NuU1qFNSEDB6FIKnT+c7HELuW8mpU7g4/BFwYjEid+2E0N+f75DuSZ2WhmsTJkLYIBCR27aBE9F5obspKSnBxYsXERERYfUBTupWda+DLZ/fNObGiRmUShSWXfOkWVLEXcjatIG0VSswnQ6qlBS+w6kRy2zF+ARKbAhxAZTcODHVhg1gOh2krVq53MwSQqqjSCwbWFx2jyZnRgU0CXE9lNw4MfMARioWRtyNIn4YIBSi+L//UHrhIt/hVMtSQLNtWyqgSYiLoOTGSZVeuICS/44CQiEUw4bxHQ4hdiVq0AA+PU03f1Wuce6baVput0CzFQlxGZTcOClzd73vww9DFBjIbzCEOIC5EJ5yzRowo5HfYKpQcuYsSk6epAKahLgYSm6cEDMY7tS2oYHExE359usHgZ8f9DeyUHTgAN/hVMo8kJgKaBLiWii5cUKa9H3Q5+SYytX37cN3OIQ4hEAqhXzwYAB3qnA7E6bXQ7l2LQA6ySDE1VBy44TMZ4vyoUMhkEj4DYYQBzInDarNm2HUaPgN5i6Fu3ffKaD58MN8h0MIsQElN07GoFZDnZoKgAYwEvfn1bkTxE2bgBUVQVX2/95ZmHuT5MOGghOLeY6GEGILSm6cjColBay0FJIWzSFr347vcAhxKI7j7tS8We08l6bKF9D0p5MMQlwOJTdOxjxLyn/4cJvuDEuIq1IkmJKbon37ocvK4jkaEyqgSVzN4sWLER4eDplMhujoaBy4xyD9lStXonXr1pDJZOjQoQM2bNhg9ThjDO+//z5CQ0Ph5eWF2NhYnDt3zqrN//73P3Tv3h3e3t7wd7LbqFBy40S0ly+j+PBhQCCAPD6e73AIqROSRg3h3a0bwBiUq9fwHQ4AKqBJXMuKFSswdepUzJgxA4cPH0bHjh0RFxeH3NzcStvv3bsXTz75JMaOHYsjR44gKSkJSUlJOH78uKXNJ598gkWLFmHJkiXYv38/fHx8EBcXh5KSEksbrVaLxx9/HC+//LLDj9FWdONMJ5K7cCFufb0EPr16ocl33/IdDiF1puCvv5H1zjuQhIej2cYNvPZall64gAtDhgJCISJ3bKc6Ux6i0hs2Mgboiuo+GLE3YMPfQHR0NLp164Yvv/wSAGA0GtG4cWO8+uqrmDZtWoX2I0aMgEajwbp16yzrHnroIXTq1AlLliwBYwxhYWF4/fXX8cYbbwAAlEolgoODsWzZMowcOdJqe8uWLcOUKVNQUFBQi4O1Zq8bZ9Id4JwEMxotYw786WyReBi/uDhkf/QRtJcuoeToUXh17MhbLJYCmr16UWLj6XRFwMdhdb/ft28AEp8aNdVqtTh06BCmT59uWScQCBAbG4v09PRKn5Oeno6pU6darYuLi0NyWY/lxYsXkZ2djdjYWMvjCoUC0dHRSE9Pr5DcOCO6LOUkig78C/2NLAj8/ODbvz/f4RBSp4S+PvAbYHojNV8S4gMzGCwnGTRbkbiCvLw8GAwGBAcHW60PDg5GdnZ2pc/Jzs6utr35qy3bdDbUc+MkzPevkQ8ZAoFUynM0hNQ9RWIiVGvWQrVhI4KnT+elxpMmfR/0ubkQUAFNApguD719g5/9kvtCyY0TMGo0lhofiqREnqMhhB8+Dz0EUXAw9Dk5KNy2HfK4gXUeg+UmmVRAkwCmcS81vDzEl8DAQAiFQuTk5Fitz8nJQUhISKXPCQkJqba9+WtOTg5CQ0Ot2nTq1MmO0TsOXZZyAqpNm8GKiiAJD4eXi/zHIcTeOKEQioQEAHeqdNclg1oN9ZYtAGiWFHEdEokEXbt2RVpZXSbANKA4LS0NMTExlT4nJibGqj0ApKamWtpHREQgJCTEqo1KpcL+/fur3KazoeTGCZjfyBVJSVTbhng0c89l4a5d0N+6Vaf7Vm3cWK6AZvs63Tch92Pq1Kn47rvv8NNPP+HUqVN4+eWXodFoMGbMGADAqFGjrAYcT548GSkpKfj8889x+vRpzJw5EwcPHsTEiRMBmIprTpkyBR999BHWrFmDY8eOYdSoUQgLC0NSufusXblyBRkZGbhy5QoMBgMyMjKQkZGBwsLCOj3+ytBlKZ5pr10z3RGZ46BITOA7HEJ4JW3eHLIOHVBy7BhU69cjYNSoOtu3+XYL/nSSQVzMiBEjcPPmTbz//vvIzs5Gp06dkJKSYhkQfOXKFQgEd/oyunfvjt9++w3vvvsu3n77bURGRiI5ORntyyX1b731FjQaDcaPH4+CggL07NkTKSkpVtOz33//ffz000+Wnzt37gwA2LZtG/r06ePgo64e73VuFi9ejE8//RTZ2dno2LEjvvjiC0RFRVXZvqCgAO+88w7+/vtv5Ofno2nTpliwYAGGDBlSo/05W52bm4sXI++LL+HTPQZNli7lOxxCeJf/66/I+fAjSNu2QbO//66TfWovX0Zm3CBAIECLbdsgDg6qk/0S51FdfRVSd+xV54bXy1K2VlXUarUYMGAALl26hFWrVuHMmTP47rvv0LBhwzqO3D4YY5azRZp2SoiJfMgQQCxG6clTKDlztk72aZ5+7tOjByU2hLgBXpObefPm4YUXXsCYMWPQtm1bLFmyBN7e3lhaRQ/G0qVLkZ+fj+TkZPTo0QPh4eHo3bs3OvJY8Ot+FB86BN3VqxD4+MCvXLEkQjyZqF49+PXpDaBuBhZTAU1C3A9vyY25qmL5Coj3qqq4Zs0axMTEYMKECQgODkb79u3x8ccfw2Aw1FXYdmU+W/QbFAeBlxe/wRDiRBRlgxaV69aC6fUO3VfRgQNUQJMQN8NbclObqooXLlzAqlWrYDAYsGHDBrz33nv4/PPP8dFHH1W5n9LSUqhUKqvFGRiLi6HemALAdAdwQsgdvr16QejvD8PNPGj27nXovsy3W5APHkwFNAlxEy41FdxoNCIoKAjffvstunbtihEjRuCdd97BkiVLqnzO7NmzoVAoLEvjxo3rMOKqqbdsgVGjgbhxY3h17cp3OIQ4FU4igXzYMACOvTRlKNRAtXkzAKptQ4g74S25qU1VxdDQULRs2RJCodCyrk2bNsjOzoZWq630OdOnT4dSqbQsV69etd9B3AdLJdSkRJp2SkglzJem1FvSYHBQj6t682aw4mJImjalApqEuBHekpvaVFXs0aMHzp8/D6PRaFl39uxZhIaGQlJFqXSpVAq5XG618E2XlQVN+j4AgCIxid9gCHFSsnZtIY1sAabVQpWS4pB9WE4yhg+nkwxC3Aivl6Vsrar48ssvIz8/H5MnT8bZs2exfv16fPzxx5gwYQJfh1ArytVrAMbgHRUFSSPXnMZOiKNxHAdFoqlisblkgj1pr11D0b//UgFNQtwQrxWKba2q2LhxY2zatAmvvfYaHnjgATRs2BCTJ0/G//3f//F1CDYz1bZJBnCn250QUjl5fAJy581H8eHD0F65AkmTJnbbtjlh8ol5COJyNwckhLg+3gcUT5w4EZcvX0ZpaSn279+P6Ohoy2Pbt2/HsmXLrNrHxMRg3759KCkpQWZmJt5++22rMTjOrjgjA9pLl8B5e/Ny12NCXIk4OAg+3bsDsG/vDWPMUtuGTjKIO1i8eDHCw8Mhk8kQHR2NAwcOVNt+5cqVaN26NWQyGTp06IANGzZYPf73339j4MCBqF+/PjiOQ0ZGhgOjtz/ekxtPY36Dlg8YAIGPD8/REOL8LJemVq8GKzfe7n5QAU3iTmyt9r937148+eSTGDt2LI4cOYKkpCQkJSXh+PHjljYajQY9e/bE3Llz6+ow7IqSmzpkLCmBqiw7ptstEFIzfrH9IfDxge76dRQfOmSXbRaUDST2GxQHgbe3XbZJCF9srfa/cOFCDBo0CG+++SbatGmDDz/8EF26dMGXX35pafPss8/i/ffftyq060roruB1qHDrVhjVaojDwuAd1Y3vcAhxCQIvL/gNHgTlqr9QkJwM727397djLCqCOmUTANMdwAmpCmMMxfriOt+vl8irxrP3zNX+y0++uVe1//T0dEydOtVqXVxcHJLr4HYndYWSmzpkvt2CPDEBnIA6zQipKf/ERChX/QV1yiYY3333vm5XQgU0SU0V64sR/Vv0vRva2f6n9sNbXLMexeqq/Z8+fbrS52RnZ9t0dwBXRJ+wdUSXkwvN7j0A6GyREFt5de0KcaNGMGo0UG/Zcl/bssxWTEykkwxC3BT13NQR1bq1gNEIry5dIGnalO9wCHEpnEAARWIi8hYvhjJ5NRTx8bXajlUBzaREe4ZI3JCXyAv7n9rPy35rqjbV/kNCQmxq74rotKUOMMYsAxjp/jWE1I45GdGkp0N31xtzTVkKaHbrBkmjRvYMj7ghjuPgLfau88WWatm1qfYfExNj1R4AUlNTq2zviii5qQMlx09Aez4TnFQK+aBBfIdDiEuSmMfIGI1Qrllj8/MZY1a3WyDEXdha7X/y5MlISUnB559/jtOnT2PmzJk4ePAgJk6caGmTn5+PjIwMnDx5EgBw5swZZGRkuMy4HEpu6oD5DdVvwAAI/fx4joYQ12XuvVGuXg3GmE3PLc7IgPbyZXBeXvAbSAU0ifsYMWIEPvvsM7z//vvo1KkTMjIyKlT7z8rKsrTv3r07fvvtN3z77bfo2LEjVq1aheTkZLRv397SZs2aNejcuTOGDh0KABg5ciQ6d+6MJUuW1O3B1RLHbH2HcHEqlQoKhQJKpbJObqJp1GpxvtfDMCiVaPz99/Dt2cPh+yTEXRnUapzr2QustBThK1fCq0P7ez+pTNb7M1Dw559QJCYibO4cB0ZJXFFJSQkuXryIiIgIyGQyvsPxWNW9DrZ8flPPjYMVbtsOg1IJUXAwfGIe4jscQlya0M8Pfv37A7gz66kmjCUlUG3cCIDGvRHiCSi5cTDLtNOEBHAudA8sQpyVOTlRrV8PptXW6DnmApqisFB4R0U5MDpCiDOg5MaB9Hl5KNy5EwCdLRJiLz4xMRA1aABDQYHl7+teCv5JBkC1bQjxFPRX7kDKdesAgwGyjg9A2qwZ3+EQ4hY4kQjysjo3BTW4NKXLyYVmT1kBzUSqbUOIJ6DkxoHMdwD3p2mnhNiVedZU4Y6d0N++XW1b1do1dwpohofXQXSEEL5RcuMgJadOofT0aXASCeSDB/MdDiFuRdayJWRt2wI6HVTrN1TZjjFm6d2hisSEeA5KbhzEPJDYt18/CBUKfoMhxA1Zat5Uc2mq5PjxOwU06SSDEI9ByY0DMJ0OyrXrAAD+NJCYEIeQDxsGiEQoOX4cpefPV9pGWTaQ2C82lgpoEuJBKLlxgMJdu2DIz4ewQSB8elDRPkIcQRQQAN+HHwZgqlh8N6NWC+X69QDodguEeBpKbhzAcv+a+ARwIrrxOiGOcud2DGvADAarxwq3bYeRCmgS4pEoubEz/e3bUG/fAYAGMBLiaL59+kCgUECfmwtN+j6rx6iAJvEkixcvRnh4OGQyGaKjo3HgwIEq2544cQKPPvoowsPDwXEcFixYUHeB1hFKbuxMtW49oNNB1q4dZC1b8h0OIW5NIJFAMXQIAOtLU1RAk3iSFStWYOrUqZgxYwYOHz6Mjh07Ii4uDrm5uZW2LyoqQrNmzTBnzhyEhITUcbR1g5IbO7OcLSYl8RoHIZ7C/LemTk2FobAQABXQJJ5l3rx5eOGFFzBmzBi0bdsWS5Ysgbe3N5YuXVpp+27duuHTTz/FyJEjIZVK6zjaukEDQuyo5OxZlJw4AYjFkA8bync4hHgEWYcOkEREQHvxItSbNsH/0Ucts6T86SSD3AfGGFhxcZ3vl/PyAsdxNWqr1Wpx6NAhTJ8+3bJOIBAgNjYW6enpjgrR6VFyY0fmisR+fXpDVK8ez9EQ4hk4joMiKQk358+HMnk1ZG3bovTMGXBiMdW2IfeFFRfjTJeudb7fVocPgfP2rlHbvLw8GAwGBAcHW60PDg7G6dOnHRGeS6DLUnbC9Hoo164BQNNOCalrioR4gONQ9O+/uPnlYgCAb//+EPr78xsYIYQX1HNjJ5o9e2C4mQdhQAB8e/XiOxxCPIo4NBTeD0WjKH0fCtPSANBsRXL/OC8vtDp8iJf91lRgYCCEQiFycnKs1ufk5LjtYOGaoOTGTqStWiFw4kRwUgk4sZjvcAjxOP5JSSgqmw4uDAyEb8+ePEdEXB3HcTW+PMQXiUSCrl27Ii0tDUllY8yMRiPS0tIwceJEfoPjESU3diIOCUGDiRP4DoMQj+U3YAC4D2aBFRVBER9PBTSJx5g6dSpGjx6NBx98EFFRUViwYAE0Gg3GjBkDABg1ahQaNmyI2bNnAzANQj558qTl++vXryMjIwO+vr5o0aIFb8dhT/TXTwhxCwJvbwS+/BJUa9ag3tNP8x0OIXVmxIgRuHnzJt5//31kZ2ejU6dOSElJsQwyvnLlCgSCO0Nsb9y4gc6dO1t+/uyzz/DZZ5+hd+/e2L59e12H7xAcY4zxHURdUqlUUCgUUCqVkMvlfIdDCCHECZSUlODixYuIiIiATCbjOxyPVd3rYMvnN82WIoQQQohboeSGEEIIIW6FkhtCCCGEuBVKbgghhBDiVii5IYQQQohboeSGEEIIKeNhE4idjr1+/5TcEEII8XjissryRUVFPEfi2bRaLQBAKBTe13aoiB8hhBCPJxQK4e/vj9zcXACAt7c3OI7jOSrPYjQacfPmTXh7e0N0nxXGKbkhhBBCAMuNJs0JDql7AoEATZo0ue/EkpIbQgghBKYbZYaGhiIoKAg6nY7vcDySRCKxulVEbVFyQwghhJQjFArve8wH4RcNKCaEEEKIW6HkhhBCCCFuhZIbQgghhLgVjxtzYy4QpFKpeI6EEEIIITVl/tyuSaE/j0tu1Go1AKBx48Y8R0IIIYQQW6nVaigUimrbcMzDak0bjUbcuHEDfn5+di/QpFKp0LhxY1y9ehVyudyu23YG7n58gPsfIx2f63P3Y6Tjc32OOkbGGNRqNcLCwu45Xdzjem4EAgEaNWrk0H3I5XK3/U8LuP/xAe5/jHR8rs/dj5GOz/U54hjv1WNjRgOKCSGEEOJWKLkhhBBCiFuh5MaOpFIpZsyYAalUyncoDuHuxwe4/zHS8bk+dz9GOj7X5wzH6HEDigkhhBDi3qjnhhBCCCFuhZIbQgghhLgVSm4IIYQQ4lYouSGEEEKIW6HkxkaLFy9GeHg4ZDIZoqOjceDAgWrbr1y5Eq1bt4ZMJkOHDh2wYcOGOoq0dmw5vmXLloHjOKtFJpPVYbS22blzJ+Lj4xEWFgaO45CcnHzP52zfvh1dunSBVCpFixYtsGzZMofHWVu2Ht/27dsrvH4cxyE7O7tuArbR7Nmz0a1bN/j5+SEoKAhJSUk4c+bMPZ/nSn+DtTlGV/o7/Prrr/HAAw9YirvFxMRg48aN1T7HlV4/W4/PlV67ysyZMwccx2HKlCnVtuPjNaTkxgYrVqzA1KlTMWPGDBw+fBgdO3ZEXFwccnNzK22/d+9ePPnkkxg7diyOHDmCpKQkJCUl4fjx43Ucec3YenyAqQJlVlaWZbl8+XIdRmwbjUaDjh07YvHixTVqf/HiRQwdOhR9+/ZFRkYGpkyZgnHjxmHTpk0OjrR2bD0+szNnzli9hkFBQQ6K8P7s2LEDEyZMwL59+5CamgqdToeBAwdCo9FU+RxX+xuszTECrvN32KhRI8yZMweHDh3CwYMH0a9fPyQmJuLEiROVtne118/W4wNc57W727///otvvvkGDzzwQLXteHsNGamxqKgoNmHCBMvPBoOBhYWFsdmzZ1fa/oknnmBDhw61WhcdHc1efPFFh8ZZW7Ye348//sgUCkUdRWdfANg///xTbZu33nqLtWvXzmrdiBEjWFxcnAMjs4+aHN+2bdsYAHb79u06icnecnNzGQC2Y8eOKtu42t/g3WpyjK78d8gYY/Xq1WPff/99pY+5+uvHWPXH56qvnVqtZpGRkSw1NZX17t2bTZ48ucq2fL2G1HNTQ1qtFocOHUJsbKxlnUAgQGxsLNLT0yt9Tnp6ulV7AIiLi6uyPZ9qc3wAUFhYiKZNm6Jx48b3PENxNa70+t2PTp06ITQ0FAMGDMCePXv4DqfGlEolACAgIKDKNq7+GtbkGAHX/Ds0GAz4448/oNFoEBMTU2kbV379anJ8gGu+dhMmTMDQoUMrvDaV4es1pOSmhvLy8mAwGBAcHGy1Pjg4uMoxCtnZ2Ta151Ntjq9Vq1ZYunQpVq9ejeXLl8NoNKJ79+64du1aXYTscFW9fiqVCsXFxTxFZT+hoaFYsmQJ/vrrL/z1119o3Lgx+vTpg8OHD/Md2j0ZjUZMmTIFPXr0QPv27ats50p/g3er6TG62t/hsWPH4OvrC6lUipdeegn//PMP2rZtW2lbV3z9bDk+V3vtAOCPP/7A4cOHMXv27Bq15+s19Li7ghP7iYmJsToj6d69O9q0aYNvvvkGH374IY+RkZpo1aoVWrVqZfm5e/fuyMzMxPz58/HLL7/wGNm9TZgwAcePH8fu3bv5DsVhanqMrvZ32KpVK2RkZECpVGLVqlUYPXo0duzYUWUC4GpsOT5Xe+2uXr2KyZMnIzU11ekHPlNyU0OBgYEQCoXIycmxWp+Tk4OQkJBKnxMSEmJTez7V5vjuJhaL0blzZ5w/f94RIda5ql4/uVwOLy8vnqJyrKioKKdPGCZOnIh169Zh586daNSoUbVtXelvsDxbjvFuzv53KJFI0KJFCwBA165d8e+//2LhwoX45ptvKrR1xdfPluO7m7O/docOHUJubi66dOliWWcwGLBz5058+eWXKC0thVAotHoOX68hXZaqIYlEgq5duyItLc2yzmg0Ii0trcrrqTExMVbtASA1NbXa6698qc3x3c1gMODYsWMIDQ11VJh1ypVeP3vJyMhw2tePMYaJEyfin3/+wdatWxEREXHP57jaa1ibY7ybq/0dGo1GlJaWVvqYq71+lanu+O7m7K9d//79cezYMWRkZFiWBx98EE8//TQyMjIqJDYAj6+hQ4cru5k//viDSaVStmzZMnby5Ek2fvx45u/vz7KzsxljjD377LNs2rRplvZ79uxhIpGIffbZZ+zUqVNsxowZTCwWs2PHjvF1CNWy9fg++OADtmnTJpaZmckOHTrERo4cyWQyGTtx4gRfh1AttVrNjhw5wo4cOcIAsHnz5rEjR46wy5cvM8YYmzZtGnv22Wct7S9cuMC8vb3Zm2++yU6dOsUWL17MhEIhS0lJ4esQqmXr8c2fP58lJyezc+fOsWPHjrHJkyczgUDAtmzZwtchVOvll19mCoWCbd++nWVlZVmWoqIiSxtX/xuszTG60t/htGnT2I4dO9jFixfZ0aNH2bRp0xjHcWzz5s2MMdd//Ww9Pld67apy92wpZ3kNKbmx0RdffMGaNGnCJBIJi4qKYvv27bM81rt3bzZ69Gir9n/++Sdr2bIlk0gkrF27dmz9+vV1HLFtbDm+KVOmWNoGBwezIUOGsMOHD/MQdc2Ypz7fvZiPafTo0ax3794VntOpUycmkUhYs2bN2I8//ljncdeUrcc3d+5c1rx5cyaTyVhAQADr06cP27p1Kz/B10BlxwbA6jVx9b/B2hyjK/0dPv/886xp06ZMIpGwBg0asP79+1s++Blz/dfP1uNzpdeuKncnN87yGnKMMebYviFCCCGEkLpDY24IIYQQ4lYouSGEEEKIW6HkhhBCCCFuhZIbQgghhLgVSm4IIYQQ4lYouSGEEEKIW6HkhhBCCCFuhZIbQkidunTpEjiOQ0ZGBt+hWJw+fRoPPfQQZDIZOnXq5JB99OnTB1OmTKm2TXh4OBYsWOCQ/RPiSSi5IcTDPPfcc+A4DnPmzLFan5ycDI7jeIqKXzNmzICPjw/OnDlT4T44Zubf292Ls97kkBBPRskNIR5IJpNh7ty5uH37Nt+h2I1Wq631czMzM9GzZ080bdoU9evXr7LdoEGDkJWVZbXU5uaWhBDHouSGEA8UGxuLkJAQzJ49u8o2M2fOrHCJZsGCBQgPD7f8/NxzzyEpKQkff/wxgoOD4e/vj1mzZkGv1+PNN99EQEAAGjVqhB9//LHC9k+fPo3u3btDJpOhffv22LFjh9Xjx48fx+DBg+Hr64vg4GA8++yzyMvLszzep08fTJw4EVOmTEFgYCDi4uIqPQ6j0YhZs2ahUaNGkEql6NSpE1JSUiyPcxyHQ4cOYdasWeA4DjNnzqzydyKVShESEmK1mO+EvGPHDkRFRUEqlSI0NBTTpk2DXq+vclu5ubmIj4+Hl5cXIiIi8Ouvv1o9zhjDzJkz0aRJE0ilUoSFhWHSpElVbo8QcgclN4R4IKFQiI8//hhffPEFrl27dl/b2rp1K27cuIGdO3di3rx5mDFjBoYNG4Z69eph//79eOmll/Diiy9W2M+bb76J119/HUeOHEFMTAzi4+Nx69YtAEBBQQH69euHzp074+DBg0hJSUFOTg6eeOIJq2389NNPkEgk2LNnD5YsWVJpfAsXLsTnn3+Ozz77DEePHkVcXBwSEhJw7tw5AEBWVhbatWuH119/HVlZWXjjjTds/h1cv34dQ4YMQbdu3fDff//h66+/xg8//ICPPvqoyuc899xzuHr1KrZt24ZVq1bhq6++Qm5uruXxv/76C/Pnz8c333yDc+fOITk5GR06dLA5NkI8ksNvzUkIcSqjR49miYmJjDHGHnroIfb8888zxhj7559/WPm3hBkzZrCOHTtaPXf+/PmsadOmVttq2rQpMxgMlnWtWrVivXr1svys1+uZj48P+/333xljjF28eJEBYHPmzLG00el0rFGjRmzu3LmMMcY+/PBDNnDgQKt9X716lQFgZ86cYYyZ7j7cuXPnex5vWFgY+9///me1rlu3buyVV16x/NyxY0c2Y8aMarczevRoJhQKmY+Pj2V57LHHGGOMvf3226xVq1bMaDRa2i9evJj5+vpafjfl75585swZBoAdOHDA0v7UqVMMAJs/fz5jjLHPP/+ctWzZkmm12nseIyHEGvXcEOLB5s6di59++gmnTp2q9TbatWsHgeDOW0lwcLBVD4NQKET9+vWteiUAICYmxvK9SCTCgw8+aInjv//+w7Zt2+Dr62tZWrduDcA0Psasa9eu1camUqlw48YN9OjRw2p9jx49anXMffv2RUZGhmVZtGgRAODUqVOIiYmxGpDdo0cPFBYWVtozdurUKYhEIqv4W7duDX9/f8vPjz/+OIqLi9GsWTO88MIL+Oeff6q9zEUIuUPEdwCEEP48/PDDiIuLw/Tp0/Hcc89ZPSYQCMAYs1qn0+kqbEMsFlv9zHFcpeuMRmON4yosLER8fDzmzp1b4bHQ0FDL9z4+PjXepj34+PigRYsWdbKvxo0b48yZM9iyZQtSU1Pxyiuv4NNPP8WOHTsq/H4JIdao54YQDzdnzhysXbsW6enpVusbNGiA7OxsqwTHnrVp9u3bZ/ler9fj0KFDaNOmDQCgS5cuOHHiBMLDw9GiRQurxZaERi6XIywsDHv27LFav2fPHrRt29Y+BwKgTZs2SE9Pt/pd7dmzB35+fmjUqFGF9q1bt7Ycs9mZM2dQUFBg1c7Lywvx8fFYtGgRtm/fjvT0dBw7dsxucRPirii5IcTDdejQAU8//bTlEotZnz59cPPmTXzyySfIzMzE4sWLsXHjRrvtd/Hixfjnn39w+vRpTJgwAbdv38bzzz8PAJgwYQLy8/Px5JNP4t9//0VmZiY2bdqEMWPGwGAw2LSfN998E3PnzsWKFStw5swZTJs2DRkZGZg8ebLdjuWVV17B1atX8eqrr+L06dNYvXo1ZsyYgalTp1pdsjNr1aoVBg0ahBdffBH79+/HoUOHMG7cOHh5eVnaLFu2DD/88AOOHz+OCxcuYPny5fDy8kLTpk3tFjch7oqSG0IIZs2aVeGyUZs2bfDVV19h8eLF6NixIw4cOFCrmURVmTNnDubMmYOOHTti9+7dWLNmDQIDAwHA0ttiMBgwcOBAdOjQAVOmTIG/v3+lyUJ1Jk2ahKlTp+L1119Hhw4dkJKSgjVr1iAyMtJux9KwYUNs2LABBw4cQMeOHfHSSy9h7NixePfdd6t8zo8//oiwsDD07t0bjzzyCMaPH4+goCDL4/7+/vjuu+/Qo0cPPPDAA9iyZQvWrl1bbR0eQogJx+6+qE4IIYQQ4sKo54YQQgghboWSG0IIIYS4FUpuCCGEEOJWKLkhhBBCiFuh5IYQQgghboWSG0IIIYS4FUpuCCGEEOJWKLkhhBBCiFuh5IYQQgghboWSG0IIIYS4FUpuCCGEEOJWKLkhhBBCiFv5fyJhv6hLOSw2AAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'The accuracy obtained is {max(accuracy_test)*100} with learning rate = {lr[np.argmax(np.array(accuracy_test))]} for adam')"
      ],
      "metadata": {
        "id": "t_KceObKmjKH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0882eba3-fc20-4961-876f-de839ef4b289"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy obtained is 99.1812673980678 with learning rate = 0.001 for adam\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Random Forest Classifier"
      ],
      "metadata": {
        "id": "FgkhRR9lY2zF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rfc = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rfc.fit(Xtrain, Ytrain)\n",
        "y_pred = rfc.predict(Xtest)\n",
        "accuracy = accuracy_score(y_pred, Ytest)\n",
        "print(f'The accuracy of the Random Forest Classifier is {100*accuracy:.3f}%')\n",
        "f1 = f1_score(Ytest, y_pred, average = 'binary')\n",
        "print(f'The F1 Score of the Random Forest Classifier is {100*f1:.3f}%')"
      ],
      "metadata": {
        "id": "TbxgOaD4Y-pV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5ef1162d-48da-47cd-a126-e5b068d9f16b"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy of the Random Forest Classifier is 99.602%\n",
            "The F1 Score of the Random Forest Classifier is 99.545%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Hyperparameter search via crossvalidation for Random Forest Classifier"
      ],
      "metadata": {
        "id": "T1a0CpEGZsx6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "n_estim, accuracy_run, accuracy_fold, accuracy_test, f1_list = random_forest(Xtrain, Ytrain, Xtest, Ytest)"
      ],
      "metadata": {
        "id": "RiupDpyBCOHv"
      },
      "execution_count": 107,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'The accuracy obtained is {max(accuracy_test)*100} with n_estim = {n_estim[np.argmax(np.array(accuracy_run))]}')"
      ],
      "metadata": {
        "id": "mARma_VWCOFN",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "307c180b-b4eb-44cd-94c6-707803a9fa1c"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy obtained is 99.53059330822553 with n_estim = 100\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'The f1 Score obtained is {f1_list[np.argmax(np.array(accuracy_run))]*100}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9nWbraXJZiou",
        "outputId": "086501ae-0fe5-4ae2-be94-0c3715f3ae02"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The f1 Score obtained is 99.46397407130392\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_fin(l1=n_estim, accuracy_run=accuracy_run, accuracy_test=accuracy_test, accuracy_fold=accuracy_fold)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 944
        },
        "id": "dpuOhj79CJ8M",
        "outputId": "ec9b804c-42a9-46f1-fa06-82e15132481e"
      },
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "9MHabsJTZqbk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "W0LjppUUZqY3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Common Features of Poisonous Mushrooms"
      ],
      "metadata": {
        "id": "wdx8SeD3D2S5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Best model outside scope of EE559 \n",
        "rfc = RandomForestClassifier(n_estimators=100, random_state=42)\n",
        "rfc.fit(Xtrain, Ytrain)\n",
        "preds = rfc.predict(Xtest)\n",
        "accuracy = accuracy_score(preds, Ytest)\n",
        "print(f'The accuracy of the Random Forest Classifier is {100*accuracy:.3f}%')\n",
        "f1 = f1_score(Ytest, preds, average = 'binary')\n",
        "print(f'The F1 Score of the Random Forest Classifier is {100*f1:.3f}')"
      ],
      "metadata": {
        "id": "AuydX7csCVwc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2f2a1e40-bec5-49c6-ce40-a19c9e8a44e3"
      },
      "execution_count": 118,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The accuracy of the Random Forest Classifier is 99.602%\n",
            "The F1 Score of the Random Forest Classifier is 99.545\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "top3feats_of_poison_mushrooms(Xtest[:,:43], preds, imp_ind[:46], df_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4vgjnRtPaNLb",
        "outputId": "a327aa81-799e-4036-b943-6039935707be"
      },
      "execution_count": 119,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{0: ['cap-shape', 'o'], 1: ['stem-color', 'r']}\n",
            "{0: ['gill-attachment', 'p'], 1: ['habitat', 'w']}\n",
            "{0: ['cap-surface', 'h'], 1: ['gill-attachment', 'p'], 2: ['does-bruise-or-bleed', 'a'], 3: ['stem-color', 'b'], 4: ['ring-type', 'l'], 5: ['cap-color', 'l']}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "We can say that if a mushroom satisfies more than 1 of the following properties:\n",
        "\n",
        "- cap-shape\t: o\n",
        "- stem-color : r/b\n",
        "- gill-attachment :\tp\n",
        "- habitat\t: w\n",
        "- cap-surface\t: h\n",
        "- does-bruise-or-bleed :\ta\n",
        "- ring-type\t: l\n",
        "- cap-color\t: l\n",
        "\n",
        "then it can be assumed to be posionous\n",
        "\n",
        "---\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "Sab7-dkybUB0"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "OQdZ0dTYbW2W"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}