NLP_Emotion_Classification.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "NLP-Emotion-Classification.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "65oQ8V2rmaqh"
      },
      "source": [
        "# NLP for Emotion Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8lxRd3htkmos"
      },
      "source": [
        "import pandas as pd\r\n",
        "import numpy as np\r\n",
        "import seaborn as sns\r\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": 627,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hAJf5YOcnLz5"
      },
      "source": [
        "## Reading text-based datasets into pandas"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sjCxUuIImIFI"
      },
      "source": [
        "df_train = pd.read_csv('/content/drive/MyDrive/Kaggle-Datasets/Emotions-NLP/train.txt', sep=\";\", header=None)\r\n",
        "df_train.columns = [\"Message\", \"Emotion\"]\r\n",
        "df_test = pd.read_csv('/content/drive/MyDrive/Kaggle-Datasets/Emotions-NLP/test.txt', sep=\";\", header=None)\r\n",
        "df_test.columns = [\"Message\", \"Emotion\"]\r\n",
        "df_val = pd.read_csv('/content/drive/MyDrive/Kaggle-Datasets/Emotions-NLP/val.txt', sep=\";\", header=None)\r\n",
        "df_val.columns = [\"Message\", \"Emotion\"]"
      ],
      "execution_count": 628,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bs_GNN4anuT0"
      },
      "source": [
        "## Exploratory Data Analysis (EDA)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 164
        },
        "id": "FPh2aYsumoms",
        "outputId": "a27b6ead-16eb-40dc-9ee9-ce1c84af5241"
      },
      "source": [
        "df_train.describe()"
      ],
      "execution_count": 629,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>16000</td>\n",
              "      <td>16000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>15969</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>i could feel myself hit this strange foggy wall</td>\n",
              "      <td>joy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>2</td>\n",
              "      <td>5362</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                Message Emotion\n",
              "count                                             16000   16000\n",
              "unique                                            15969       6\n",
              "top     i could feel myself hit this strange foggy wall     joy\n",
              "freq                                                  2    5362"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 629
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 106
        },
        "id": "Aq_A72o8sDpX",
        "outputId": "70b0f4bd-43d7-4b3f-e2a9-1f3ed6fab13d"
      },
      "source": [
        "df_train.loc[df_train['Message'] == 'i feel so weird about it',['Message','Emotion']]\r\n",
        "\r\n",
        "# Here we notice that 2 records with the same message have 2 different emotions in training dataset"
      ],
      "execution_count": 630,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>7669</th>\n",
              "      <td>i feel so weird about it</td>\n",
              "      <td>fear</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12562</th>\n",
              "      <td>i feel so weird about it</td>\n",
              "      <td>surprise</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                        Message   Emotion\n",
              "7669   i feel so weird about it      fear\n",
              "12562  i feel so weird about it  surprise"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 630
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 164
        },
        "id": "p0yAM1A5mq0G",
        "outputId": "a8d55e64-41f7-4bd3-9243-709941c6a7df"
      },
      "source": [
        "df_test.describe()"
      ],
      "execution_count": 631,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>2000</td>\n",
              "      <td>2000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>2000</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>i feel like im a pathetic little desperation</td>\n",
              "      <td>joy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>1</td>\n",
              "      <td>695</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                             Message Emotion\n",
              "count                                           2000    2000\n",
              "unique                                          2000       6\n",
              "top     i feel like im a pathetic little desperation     joy\n",
              "freq                                               1     695"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 631
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 164
        },
        "id": "-CZTM3rqnSsJ",
        "outputId": "3d9fc7a3-9f7d-4a75-ed19-8db44620bf3e"
      },
      "source": [
        "df_val.describe()"
      ],
      "execution_count": 632,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>2000</td>\n",
              "      <td>2000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>1998</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>i have had several new members tell me how com...</td>\n",
              "      <td>joy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>2</td>\n",
              "      <td>704</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                                  Message Emotion\n",
              "count                                                2000    2000\n",
              "unique                                               1998       6\n",
              "top     i have had several new members tell me how com...     joy\n",
              "freq                                                    2     704"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 632
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 106
        },
        "id": "HoFg1SOrq-dp",
        "outputId": "bf082185-3d30-4fb6-8b5d-0865c1ac3b5a"
      },
      "source": [
        "df_val.loc[df_val['Message'] == 'i feel so tortured by it',['Message','Emotion']]\r\n",
        "\r\n",
        "# We also found 2 records with the same message and 2 different emotions in validation dataset"
      ],
      "execution_count": 633,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>774</th>\n",
              "      <td>i feel so tortured by it</td>\n",
              "      <td>fear</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1993</th>\n",
              "      <td>i feel so tortured by it</td>\n",
              "      <td>anger</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                       Message Emotion\n",
              "774   i feel so tortured by it    fear\n",
              "1993  i feel so tortured by it   anger"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 633
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "zzwGAQl1ndF9",
        "outputId": "82277014-66a2-440a-edf1-a62cb8a8d4d1"
      },
      "source": [
        "df_train.groupby('Emotion').describe()"
      ],
      "execution_count": 634,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>2159</td>\n",
              "      <td>2159</td>\n",
              "      <td>i can never fall in love with anyone because m...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>1937</td>\n",
              "      <td>1937</td>\n",
              "      <td>i only have to think about a high school exper...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>5362</td>\n",
              "      <td>5361</td>\n",
              "      <td>i feel more adventurous willing to take risks ...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>1304</td>\n",
              "      <td>1304</td>\n",
              "      <td>i feel that language is nothing at all cheer u...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>4666</td>\n",
              "      <td>4666</td>\n",
              "      <td>i hold it for a day my arm will feel numb and ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>572</td>\n",
              "      <td>572</td>\n",
              "      <td>i admit to feeling bitterly surprised at how r...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Message                                                               \n",
              "           count unique                                                top freq\n",
              "Emotion                                                                        \n",
              "anger       2159   2159  i can never fall in love with anyone because m...    1\n",
              "fear        1937   1937  i only have to think about a high school exper...    1\n",
              "joy         5362   5361  i feel more adventurous willing to take risks ...    2\n",
              "love        1304   1304  i feel that language is nothing at all cheer u...    1\n",
              "sadness     4666   4666  i hold it for a day my arm will feel numb and ...    1\n",
              "surprise     572    572  i admit to feeling bitterly surprised at how r...    1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 634
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "cAYQ6rCcnodh",
        "outputId": "66b3fc46-32a7-4aef-d6c5-652641c4bb43"
      },
      "source": [
        "df_test.groupby('Emotion').describe()"
      ],
      "execution_count": 635,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>275</td>\n",
              "      <td>275</td>\n",
              "      <td>i feel so disgusted when i see blood and feel ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>224</td>\n",
              "      <td>224</td>\n",
              "      <td>i feel uncomfortable since i have a smaller ri...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>695</td>\n",
              "      <td>695</td>\n",
              "      <td>i wasn t feeling well but no specific issue</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>159</td>\n",
              "      <td>159</td>\n",
              "      <td>i try my best to love on them shed some light ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>581</td>\n",
              "      <td>581</td>\n",
              "      <td>i feel like im a pathetic little desperation</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>66</td>\n",
              "      <td>66</td>\n",
              "      <td>i would look up at the sky scrapers and feel a...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Message                                                               \n",
              "           count unique                                                top freq\n",
              "Emotion                                                                        \n",
              "anger        275    275  i feel so disgusted when i see blood and feel ...    1\n",
              "fear         224    224  i feel uncomfortable since i have a smaller ri...    1\n",
              "joy          695    695        i wasn t feeling well but no specific issue    1\n",
              "love         159    159  i try my best to love on them shed some light ...    1\n",
              "sadness      581    581       i feel like im a pathetic little desperation    1\n",
              "surprise      66     66  i would look up at the sky scrapers and feel a...    1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 635
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "IpdVD_zyq5hC",
        "outputId": "0052df78-c5b6-4cc6-c6d8-5ef6dd63c161"
      },
      "source": [
        "df_val.groupby('Emotion').describe()"
      ],
      "execution_count": 636,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>275</td>\n",
              "      <td>275</td>\n",
              "      <td>i am feeling too grouchy to be properly penite...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>212</td>\n",
              "      <td>212</td>\n",
              "      <td>i feel doubtful and afraid</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>704</td>\n",
              "      <td>704</td>\n",
              "      <td>i feel ok with nothing sexually or whatsoever ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>178</td>\n",
              "      <td>178</td>\n",
              "      <td>i was just telling you how i feel about you an...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>550</td>\n",
              "      <td>550</td>\n",
              "      <td>i posted i think it was about feeling sorta sh...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>81</td>\n",
              "      <td>81</td>\n",
              "      <td>i am feeling amazing</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Message                                                               \n",
              "           count unique                                                top freq\n",
              "Emotion                                                                        \n",
              "anger        275    275  i am feeling too grouchy to be properly penite...    1\n",
              "fear         212    212                         i feel doubtful and afraid    1\n",
              "joy          704    704  i feel ok with nothing sexually or whatsoever ...    1\n",
              "love         178    178  i was just telling you how i feel about you an...    1\n",
              "sadness      550    550  i posted i think it was about feeling sorta sh...    1\n",
              "surprise      81     81                               i am feeling amazing    1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 636
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "id": "7qdVIGQrpFED",
        "outputId": "2bafe1e2-af29-43e1-fe1a-0b04b83a3fb8"
      },
      "source": [
        "df_train.head()"
      ],
      "execution_count": 637,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>i didnt feel humiliated</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i can go from feeling so hopeless to so damned...</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>im grabbing a minute to post i feel greedy wrong</td>\n",
              "      <td>anger</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i am ever feeling nostalgic about the fireplac...</td>\n",
              "      <td>love</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i am feeling grouchy</td>\n",
              "      <td>anger</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                             Message  Emotion\n",
              "0                            i didnt feel humiliated  sadness\n",
              "1  i can go from feeling so hopeless to so damned...  sadness\n",
              "2   im grabbing a minute to post i feel greedy wrong    anger\n",
              "3  i am ever feeling nostalgic about the fireplac...     love\n",
              "4                               i am feeling grouchy    anger"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 637
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "id": "jZ8NllORp4dr",
        "outputId": "31590853-b251-490e-9400-825eecc98f26"
      },
      "source": [
        "df_test.head()"
      ],
      "execution_count": 638,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>im feeling rather rotten so im not very ambiti...</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>im updating my blog because i feel shitty</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>i never make her separate from me because i do...</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i left with my bouquet of red and yellow tulip...</td>\n",
              "      <td>joy</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i was feeling a little vain when i did this one</td>\n",
              "      <td>sadness</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                             Message  Emotion\n",
              "0  im feeling rather rotten so im not very ambiti...  sadness\n",
              "1          im updating my blog because i feel shitty  sadness\n",
              "2  i never make her separate from me because i do...  sadness\n",
              "3  i left with my bouquet of red and yellow tulip...      joy\n",
              "4    i was feeling a little vain when i did this one  sadness"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 638
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oaDspf78oHEs"
      },
      "source": [
        "# Convert Emotions (target) into numerical variables using Label Encoding\r\n",
        "\r\n",
        "from sklearn.preprocessing import LabelEncoder\r\n",
        "\r\n",
        "encoder = LabelEncoder()\r\n",
        "df_train['Emotion'] = encoder.fit_transform(df_train['Emotion'])\r\n",
        "df_test['Emotion'] = encoder.fit_transform(df_test['Emotion'])\r\n",
        "df_val['Emotion'] = encoder.fit_transform(df_val['Emotion'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wkep8smnqCMs"
      },
      "source": [
        "df_train['Message length'] = df_train['Message'].apply(lambda x : len(x))\r\n",
        "df_test['Message length'] = df_test['Message'].apply(lambda x : len(x))\r\n",
        "df_val['Message length'] = df_val['Message'].apply(lambda x : len(x))"
      ],
      "execution_count": 639,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "id": "mUxLqha6q8Aa",
        "outputId": "372bd6bc-4d8b-4aaa-9e68-0d79e2a81cc3"
      },
      "source": [
        "df_train.head()"
      ],
      "execution_count": 640,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "      <th>Message length</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>i didnt feel humiliated</td>\n",
              "      <td>sadness</td>\n",
              "      <td>23</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i can go from feeling so hopeless to so damned...</td>\n",
              "      <td>sadness</td>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>im grabbing a minute to post i feel greedy wrong</td>\n",
              "      <td>anger</td>\n",
              "      <td>48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i am ever feeling nostalgic about the fireplac...</td>\n",
              "      <td>love</td>\n",
              "      <td>92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i am feeling grouchy</td>\n",
              "      <td>anger</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                             Message  Emotion  Message length\n",
              "0                            i didnt feel humiliated  sadness              23\n",
              "1  i can go from feeling so hopeless to so damned...  sadness             108\n",
              "2   im grabbing a minute to post i feel greedy wrong    anger              48\n",
              "3  i am ever feeling nostalgic about the fireplac...     love              92\n",
              "4                               i am feeling grouchy    anger              20"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 640
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Altd-kE_ruf0"
      },
      "source": [
        "palette = {0:\"C0\",1:\"C1\",2:\"C2\",3:\"C3\",4:\"C4\",5:\"C5\"} #0. Anger, 1. Fear, 2. Joy, 3. Love, 4. Sadness, 5. Surprise"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 624
        },
        "id": "Lk_FM45Jq90V",
        "outputId": "e780f3f7-2176-49a7-e263-0b5361e367af"
      },
      "source": [
        "plt.figure(figsize=(15,10))\r\n",
        "sns.histplot(data=df_train, hue='Emotion', x='Message length', stat='count', multiple='layer', palette=palette,kde=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f5cee706710>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3sAAAJNCAYAAACImWznAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXzcZb33/9c1SybJZE+apM3elZYuKXRjldUqcoqIR3DngMef3iBy9PahgBzpUZEfP0Xxth701ls9bni8kV0QEBBaKC1dKV3omq3Z920ms1y/P1Iq2CZt08l8k5n38/GYB0nmu7xDkzzmM9d1fS5jrUVEREREREQSi8vpACIiIiIiIhJ7KvZEREREREQSkIo9ERERERGRBKRiT0REREREJAGp2BMREREREUlAKvZEREREREQSkMfpAKejoKDAVlZWOh1DRERERETEEZs2bWqz1k453nOTutirrKzk9ddfdzqGiIiIiIiII4wxNSM9p2mcIiIiIiIiCUjFnoiIiIiISAJSsSciIiIiIpKAJvWaPRERERERkZMRCoWor68nEAg4HWVMUlNTKS0txev1nvQ5KvZERERERCTh1dfXk5mZSWVlJcYYp+OcEmst7e3t1NfXU1VVddLnaRqniIiIiIgkvEAgQH5+/qQr9ACMMeTn55/yqKSKPRERERERSQqTsdB721iyq9gTEREREZGk5Ha7qa6uPvq45557Tvuahw4d4ne/+93Rz19//XVuueWW077uWGjNnoiIiIiIJKW0tDS2bt0a02u+Xex97GMfA2DJkiUsWbIkpvc4WRrZExEREREReYfKykpuu+02qqurWbJkCZs3b2blypXMmDGDBx54ABhumvKVr3yF+fPns2DBAv7whz8A8LWvfY2XX36Z6upqvv/97/Piiy9y5ZVXAtDR0cEHP/hBFi5cyIoVK9i+fTsAd911FzfccAMXXXQR06dP54c//GFMvg+N7ImIiIiISFIaHBykurr66Oe33XYb1157LQDl5eVs3bqVf/u3f+P6669n3bp1BAIB5s+fz+c+9zn+9Kc/sXXrVrZt20ZbWxtLly7lwgsv5J577uG73/0uTzzxBAAvvvji0et/4xvfYPHixTzyyCM8//zzfOpTnzo6srh7925eeOEFent7mTNnDp///OdPaZuF41GxJyIiIiIiSWm0aZyrVq0CYMGCBfT19ZGZmUlmZiY+n4+uri7Wrl3LRz/6UdxuN0VFRbznPe9h48aNZGVljXi/tWvX8tBDDwFwySWX0N7eTk9PDwAf+MAH8Pl8+Hw+CgsLaW5uprS09LS+P03jFBERERER+Qc+nw8Al8t19OO3Pw+Hw+N2PxhuHBOLe6jYExEREREROUUXXHABf/jDH4hEIrS2tvLSSy+xbNkyMjMz6e3tHfGc3/72t8Dw9M6CgoJRRwJPl6ZxioiIiIhIUvrHNXvve9/7Tnr7hauvvppXX32VRYsWYYzh3nvvpbi4mPz8fNxuN4sWLeL6669n8eLFR895uxHLwoULSU9P51e/+lXMv6d3Mtbacb3BeFqyZIl9/fXXnY4hIiIiIiIT3K5du5g7d67TMU7L8b4HY8wma+1x93bQNE4REREREZEEpGJPREREREQkAanYExERERERSUAq9kRERERERBKQij0REREREZEEpGJPREREREQkAanYExERERERccDTTz/NnDlzmDlz5knv73cqVOyJiIiIiEjSKyuvwBgTs0dZecWo94tEItx000089dRT7Ny5k9///vfs3Lkzpt+TJ6ZXE5mkyirKqK+tH9O5peWl1NXUxTiRiIiIiMRTfV0t9z2zJ2bX+9J754z6/IYNG5g5cybTp08H4LrrruPRRx9l3rx5McugYk8EqK+tZ82WNWM696bFN8U4jYiIiIgkuoaGBsrKyo5+XlpaymuvvRbTe2gap4iIiIiISAJSsSciIiIiIhJnJSUl1NX9fSlQfX09JSUlMb2Hij0REREREZE4W7p0KXv37uXgwYMMDQ3x4IMPsmrVqpjeQ2v2RERERERE4szj8fCjH/2IlStXEolEuOGGGzjzzDNje4+YXk1EJpXT6UIK6kQqIiIiiaO0rPyEHTRP9XoncsUVV3DFFVfE7J7/SMWeSBI7nS6koE6kIiIikjjqamucjhBzWrMnIiIiIiKSgFTsiYiIiIiIJCAVeyIiIiIiIglIxZ6IiIiIiEgCUrEnIiIiIiKSgFTsiYiIiIiIOOCGG26gsLCQ+fPnj8v1VeyJiIiIiEjSqywvxRgTs0dleekJ73n99dfz9NNPj9v3pH32REREREQk6dXUNWCfvztm1zOX3H7CYy688EIOHToUs3v+I43siYiIiIiIJCAVeyIiIiIiIglIxZ6IiIiIiEgCUrEnIiIiIiKSgFTsiYiIiIiIOOCjH/0o55xzDnv27KG0tJSf//znMb2+unGKiIiIiEjSqygrOakOmqdyvRP5/e9/H7P7HY+KPRERERERSXqHauudjhBzmsYpIiIiIiKSgFTsiYiIiIiIJCAVeyIiIiIiIglIxZ6IiIiIiEgCUrEnIiIiIiKSgFTsiYiIiIiIxFldXR0XX3wx8+bN48wzz+T++++P+T1U7ImIiIiISNIrqyjDGBOzR1lF2aj383g8fO9732Pnzp2sX7+eNWvWsHPnzph+T9pnT0REREREkl59bT1rtqyJ2fVuWnzTqM9PnTqVqVOnApCZmcncuXNpaGhg3rx5McugkT0REREREREHHTp0iC1btrB8+fKYXlfFnoiIiIiIiEP6+vq45ppr+MEPfkBWVlZMr61iT8RhFeUVpzUfvKK8wulvQURERETGIBQKcc011/Dxj3+cD33oQzG//rit2TPGlAH/BRQBFviptfZ+Y8xdwL8CrUcOvd1a++cj59wG3AhEgFustX8Zr3wiE0VtXS2vPbZ/zOcvXzUjhmlEREREJB6stdx4443MnTuXL33pS+Nyj/Fs0BIGvmyt3WyMyQQ2GWOePfLc9621333nwcaYecB1wJnANOA5Y8xsa21kHDOKTGhhG2bQDhCwg0QIEbYRIkQAixsPXuPFO8WLJ8NDZCAy/LaKiIiIiEx469at49e//jULFiyguroagLvvvpsrrrgiZvcYt2LPWtsINB75uNcYswsoGeWUq4AHrbVB4KAxZh+wDHh1vDKKOCUSjdA80Ex9bz05F+Tw8MBv6Yy202O76Ip20BPtYtD2M8TQCa815/+bA4DBkOpJJdWTSqY3k2xfNjm+HHJ8ORSkFZDqST3m3BN1iRIRERFJFqXlpTF9bVRaXjrq8+effz7Wju879XHZesEYUwksBl4DzgNuNsZ8Cnid4dG/ToYLwfXvOK2e0YtDkaPKKsqor613OsZxDYQG2NG2g53tO9nbtZe9nXs50H2AYCQIQOmNpTwZ+L9kmxyyXbnkuvKpcM8g3ZVBuvGTbtLxmTQ8eHAbD27cGAwRwoRtiG/88Mt85M6PEAgHGAwPMhgepHeol8P9hwlHw0dzZKdkU+Qvoji9mLLMMrJ8sV0AfKpO99+stLyUupq6GCYSERGRZJaIryvGvdgzxmQADwG3Wmt7jDH/CXyT4Qln3wS+B9xwCtf7LPBZgPLy8tgHlknpdPdFieW7OMFIkNcaX+PVw6+ypWULuzt2EzkyG3lK2hRm5c7i2uJrqcyupCyzjAsXXciLP9uBx4zt17Hr5S4WTVl0zNettfSH++kKdNE62EpzfzN1vXW81fkWALm+XIr+uYim/iaK0oswxoz9mx6DifRvJiIiIpKIxrXYM8Z4GS70fmut/ROAtbb5Hc//b+CJI582AO/cZr70yNfexVr7U+CnAEuWLNEKJZkQeod6ea7mOV6se5FXG19lMDxIqjuV+QXzuWH+DVQXVrOgYAG5qbnHnBtqDY250BuNMYYMbwYZ3gxKM4enEVhr6R7qpqanhpqeGgpWFvDwvofJTslmTt4c5uTOISMlI+ZZRERERCT+xrMbpwF+Duyy1t73jq9PPbKeD+BqYMeRjx8DfmeMuY/hBi2zgA3jlU/kdFlraRlooeSGEi7946UMhgcp9hezasYqLi67mKXFS0lxpzgd812MMcPr+KbksGjKIr5w7he4+Ymb2d2xmw1NG9jQtIHyzHKqC6uZ5p8W99E+EREREYmd8RzZOw/4JPCGMWbrka/dDnzUGFPN8DTOQ8D/A2CtfdMY89/AToY7ed6kTpwyEVlrOdB9gE3Nm2gPtJO1LIsrqq7gmlnXML9g/qQqkKKD0eERvbw59AR72N25m53tO3ls/2MUphVSXVhNVXYVLqMtOUVEREQmm/HsxrkWON6r3j+Pcs63gW+PVyaR02Gtpaa3ho2NG2kLtJHjy+E9pe/hP6/8T+4avMvpeKcty5fFsuJlnFV4Fm91vsXWlq08U/MMub5clk9dTmVW5aQqZEVERESSXVy6cYpMdh2BDv5W9zeaBprISsnikrJLmJU7C5dxEQ1EnY4XUx6Xh3n58zgj7wwOdh9kQ9MGnj70NMXpxZwz7RyK/cVORxQRERGZ9AKBABdeeCHBYJBwOMyHP/xhVq9eHdN7qNiTCaOivILaulqnY7xL1EbZ2rKVjc0b8bq8XFh6IWfknYHbuJ2ONu5cxsWMnBlUZVexu2M3G5s28vC+h5mZM5Nzp52L3+t3OqKIiIhIzFSWlVFTH7utvCpKSzlUN/J2Dj6fj+eff56MjAxCoRDnn38+73//+1mxYkXMMqjYkwmjtq6W1x7bP6Zzl6+aEeM0w6N5z9c+T+tgK9Ozp3NByQWke9Njfp+JzmVczMufx6ycWWxt3cqWli3U9tSybOqy40/UFhEREZmEaurrafnh/4rZ9Qpv+cKozxtjyMgY7oIeCoUIhUIxXzKjrgsix3Gg6wAP7X2I3lAv7614LysrVyZlofdOXreXpcVLuXbOtRSmF7K2YS0z/n0Gezr2OB1NREREZFKKRCJUV1dTWFjI5ZdfzvLly2N6fRV7Iu9grWVT8yb+UvMX8lLzuHb2tczIif2o4WSW7cvmyulXcnn55XjyPFz35HX87I2fEYmqea6IiIjIqXC73WzdupX6+no2bNjAjh07TnzSKdA0TpEjwtEwL9S9wL6ufczKmcVFZRfhccXhV8Q1PtNQx5Mxhpm5M9l3xz5uefwW7t98Py/Vv8S3z/s2ZVllTscTERERmVRycnK4+OKLefrpp5k/f37MrqtiTwQwKYYnDzzJ4f7DLC9ezuLCxfHbZiAKtz/4rTGffvd1X49hmFMT6Yvwvfd8jycOPMF3XvsO1zx+Df9+zr9z5fQrHcskIiIiMhm0trbi9XrJyclhcHCQZ599lq9+9asxvYeKPUl64WiYilsqaOxv5NLyS5mdO9vpSKcs1m16T4Uxhn+a8U8sLV7KV1/6Kre9fBubmjfxtWVfw+f2jd+NXYy5IC8tL6WuZuTuWCIiIiLjrbGxkU9/+tNEIhGi0Sgf+chHuPLK2L5hrmJPklokGuEvh/6Cf56fi8oumpSFHsBnrrtlTOfd/UjsRgWL/cX8fOXP+dGWH/HzHT9nR9sOvvee71GeVR6ze7xLFNZsWTOmU29afFOMw4iIiMhkV1FaesIOmqd6vdEsXLiQLVu2xOx+x6MGLZK0ItEIz9Q8Q21vLYd/dZgz8s5wOtKk53F5uPXsW1lz6Roa+xu59olrean+JadjiYiIiJzQobo6rLUxe4y2x168qNiTpGSt5aWGlzjUc4gLSi6g82+dTkdKKBeWXsgfr/wjZZll3PzXm/nljl9irXU6loiIiEhS0TROSRinsm4trSqNzEWZ9O/p56FHHhrHVMlrasZUfvm+X3Lnujv53qbv8VbnW3zj3G+M7zo+ERERETlKxZ4kjJNdt9YebeGVoRcodE1l2cLzMYtcMV27Jn+X7k3nu+/5Lj/Z/hPWbF1DTW8NP7rkR+Sm5jodTURERCThaRqnJJVB28/GoXWkmwzO8q7AmBj8ChzpCjnWR6IzxvC5RZ/jvovuY0/HHj751Cep63F+DruIiIhIotPIniSNiI2wcWgdUSIs856P16TE5sKn0RUSkqcz5OUVlzMlbQo3P38zn3jqE6RVpTkdSURERCShqdiTpLErvJ0u28FS7/lkurKPed7JveqSRXVhNb9+/6/5/HOfp+prVdT01FCRVeF0LBERERHHRCIRlixZQklJCU888URMr61iT5JCe7SFA5E9VLhnMtV9/D1PJsJedcmgKruK31zxG1bcu4KnU5/msvLLmJEzw+lYIiIikuQqyiuorauN2fXKy8qpqa054XH3338/c+fOpaenJ2b3fpuKPUl4YRtmS2gD6cbPmZ5FTscRoCCtgIP/70Eu+a9LeLbmWULRkPY5FBEREUfV1tXy2mP7Y3a95atO/GZ2fX09Tz75JHfccQf33XdfzO79NjVokYS3M7yNAdtHtXc5HuN1Oo4cER2McmXVlZRklPBC3Qu80faG05FERERE4urWW2/l3nvvxeUan7JMxZ4ktNZIE4cie5nunk2Bq9DpOPIPvG4vV1RdQVVWFWsb1rK1ZavTkURERETi4oknnqCwsJCzzz573O6hYk8SVtiG2BragN9kcoZnodNxZARul5vLKy9nZs5MXm18lW2t25yOJCIiIjLu1q1bx2OPPUZlZSXXXXcdzz//PJ/4xCdieg8Ve5Kw9oV3McgA1d5leIyWp05kbuPm0vJLmZ49nVcOv8L21u1ORxIREREZV9/5zneor6/n0KFDPPjgg1xyySX85je/iek9VOxJQhqI9rEvspsSVwX5rilOx5GT4DIuLqu4jKrsKtYdXseOth1ORxIRERGZ1DTcIQlpZ3gbBsM877HTN11DHtyDqXgG0vAEfJiIm+su+Ddyds7ERA3WZbGu6PDDEyGaEiLiDQ3/1zdEJDUILuvAd5X43MbN5eWX80zNM7zc8DJu42Zu/lynY4mIiEgSKC8rP6kOmqdyvZN10UUXcdFFF8Xs3m9TsScJpy3awuFoHXM880kzflxBL76OHFLbc0jpysYdendHzqg7wsLKc0npScMai4m6/v4IuzGYdx1vscNFX1qAcPogF8xbhW1Mg7wgxheN57eakNwuN++teC9PHXqKv9X/DZ/bx/Sc6U7HEhERkQR3MnviTTYq9iShWBtlR2gzaaQzv+08suvLSenNACCSMkQwv5OQf4Bw+iDh9ACR1AC44O7rvs7tD37rOBcEV8iDK+TFNeTFHfDhCfhwD6biDvhIayng2gu+CI8eOdwfgrzg8CM/CEWDkBXCmGMvLSNzu9ysrFjJ4wce59naZ/mA+wOUZpY6HUtERERkUlGxJwmlNlRDj+3ikoMfY0rTPEL+AXqm1xDI6yKcMQCnWnQZiKaEiaaEwT947PMWfvyZH/Ct330X2n3QceTRkA7RI0ti08LYokEoHoTSfsgPqvg7CW9vy/Do/kd56tBTrJq+iiJ/kdOxRERERCYNFXuSMLzNWbyV/ibFg9MpD8yhfeEugnldp17gnQoDXf1tmPJ+KO8/+mUbATp90JIGTWnQnAaHMoefTA1jSwagvA8q+8Yx3OSX6knlyulX8si+R3jy4JNcPfNqclNznY4lIiIiMimo2JNJLxQI86/vXc3hzmYGs/tYxnvoWLzL0UzGDRQEhx/zugCw/R6oT4cGP9T7YX8WuCz/44rvkH44n8Ep7VhvxNHcE5Hf6+fK6Vfyp31/4smDT/KhmR8i3ZvudCwRERGRCU9bL8ik1tHYzxsvHWZWRTWby5+lwBSRnZnpdKzjMv4wZk4P5pJG+OQ+uPoQLOxgSlYJOXtmUPzKEnJ2zSClKxPU7PNdsn3ZXFF1BYOhQZ46+BShSMjpSCIiIiITnoo9mbSaDnazf3MrqX4PP279EkHXAHO8852OdVKMAVMUwKxoZfWDn6J1yXYGiltIbc2jYMt8pmxcRFpTAUz05p4uMMaM6XGqitKLuKziMloGW3iu9rnxnZ4rIiIiEgeVlZUsWLCA6upqlixZEvPraxqnTDrWWur3dNK0v4fc4nRKFmWQOs1Lgato0m6gHsrsp3vOQXpm1JDWUoC/oZjcXbPIPFRKb0UDg0VtE3Nvvyis2bJmTKfetPimUz6nKruK80vOZ23DWqZ+bOqY7isiIiJyPOXlZdTV1cfsemVlpdTW1p3wuBdeeIGCgoKY3fedVOzJpGKjlkNvtNNW38eU8gwq5ufzTOBRPNke5ngmx6jeaKwnysC0FgamtpDalkvGoVJyd88ks6aU7pkHCRZ0OR3RcQsKFtA71Mu2y7exo20H8wsm/7+7iIiIOK+urp51//2bmF3vvI98ImbXGitN45RJw1rLwe1ttNX3MW1WNhXz8xliiKcDf6JvR9+kHdU7LgOBKZ20LXmD9gW7sSZK/htzyd0xG1cgxel0jlsxdQU9W3tY27CW+t7YvQMnIiIiEk/GGN773vdy9tln89Of/jTm11exJ5NG4/5u2hv6mTY7h5LZuRhjeDHwND22m5ZHW5yONz4MBAs6aV26nZ7pNfjacyjcUI2/bmpSN3FxGRf1D9ST48vhmZpn6A52Ox1JRERE5JStXbuWzZs389RTT7FmzRpeeumlmF5fxZ5MCp1N/TTs6SJvmp9pM7MBCNswzwQe4QzPQgb2DjiccJy5LH0Vh2ldto2hnG6y91WSt/0MXEPJOxM7Gojy/qr3A/DUwacIRoIOJxIRERE5NSUlJQAUFhZy9dVXs2HDhpheX8WevEtFecWYuytWlFeMS6aB7iAHtrbhz0mhamH+0U6Om4Zeoct2sDL1qnG570QUSQvSsWAPXbMP4OvKZsrGRcyYusDpWI7J9mWzsnIl3cFu/lrzV6xN4uFOERERmVT6+/vp7e09+vEzzzzD/Pmx7UWQvMMCcly1dbW89tj+MZ27fNWMGKeBUDDC3tdbcHtdzDy7EJf77+9PPBt4jCLXNOZ7z4r5fSc0AwMlzQxl9ZL35my+eOX36Kupp6/8cFJuR1CSUcJ5JefxcsPLvN78OkuLlzodSUREROSEmpubufrqqwEIh8N87GMf433ve19M76FiTyYsa4c7b4aGosw9t5iU1L//uO4P7+FgZC8fT/8sLpOcA9ThzAFal2zn8C8HONt1Me6gj+5ZB5Oy4Dsz/0xaBlp4vfl1CtMLqcgan1FmERERSVxlZaUx7aBZVlY66vPTp09n27ZtMbvf8STnq2SZFDqbBuhqHqBkdg7+bN+7nns28Bhpxs+5vkscSjcxWE+UXz73bfrKGob35tsxGyLJ92ttjOHC0gspSCvgudrn1LBFRERETlltbR3W2pg9TmaPvfGWfK8KZVIID0Wo2dFOenYKxVVZ73quI9LKpqFXuNB3OakmzaGEE4fF0jOzlu6ZB0ltyyN/21xMKPkG7T0uDysrVmIwPH3oaUKRkNORRERERBylYk8mpNqdHURCUaoWFmBc756X+Hzwz1jgEt8HnAk3QfWXNdF55l5SejLI3zoXE3I7HSnusnxZXFZ+GR2BDl5qeEkNW0RERCSpqdiTCaerZYD2hn6mzsgmPevdG4gHbZC/BZ/hLO9yCtyFDiWcuAKF7XQs2IO3P5387Wdgwsn3K16eVc7SoqW81fkWuzt2Ox1HREREJpDJ/EbwWLIn3ytBmdAi4Sg1b7STmuFl6sycY55/NfgCA7aPy1NXOZBucgjmd9E5by/enkzy3jgDIsnXseWsorMozSjl5YaXaR9sdzqOiIiITACpqam0t7dPyoLPWkt7ezupqamndF7yLeyRCa35YA9DgQhnnDMFl/vYIuWl4DOUuauY6ZnrQLrJI1DYQVd0H7m7ZpH35mw65r8Frsn3h22sXMbFpeWX8se3/shfav7Ch2d9mBR3yolPFBERkYRVWlpKfX09ra2tTkcZk9TUVEpLR+/w+Y9U7MmE4fdl0XSgm5yidDLzjn3XojZ8gJrIfj6W/tmjG6vLyAaL2zARNzlvTSf7rel0z9mfVNsypHvTuaziMh7f/zgv1b/EpeWX6udGREQkiXm9XqqqqpyOEVeaxikTxmXV1xIJW0rmHDt9E+Dl4HN48LIi5T1xTjZ5DZQ001tRj7+xEH99sdNx4q4ko4SlxUvZ27WXXR27nI4jIiIiElcq9mRC6O8O8p4zP0jeND/pmcdOtwvZIdYP/Y2zU87B78pwIOHk1VtVx2BBO1n7KvG1H7+QTmRnFZ5FSUYJ6w6vozPQ6XQcERERkbhRsScTwqY/H8Lt8lAy+/jFyOah9QzYPs73XRrnZAnAQNfcfYQzBsh9cxae/lNb2DvZGWO4tPxSPMbDczXPEYlGnI4kIiIiEhcq9sRxPW2DvLn2MK/ueZpUv/e4x7wcfJYCVyFneBbGOV1isJ4oHQt2Y11R8t44I+n24PN7/VxcdjFtgTbWN653Oo6IiIhIXKjYE8dtfPIgxhie3vzr4z7fGmlmV3g75/kuxWX0IztWkdQhOhbswR3wkbNnBiRPc04AKrMrmZ8/n+1t28lYoKnAIiIikvjUjVNixmVcp9ztMC+jiLs++mte3PEwXYNtLF8145hjCq8uZMo/TeHez9/FtzvuiFXcpBTK7qNnei3Z+ysJHi5yOk7cnTPtHA73Hyb8mTBtg20UpBU4HUlERERk3KjYk5iJ2iivPbb/lM6p29VB08Eebr31Vv507X9y+4Pfetfz1kZ5LvgEGa4srvrxR0e8zt3XfX1MmZNRf1kjvs5ssvdVUpI33ek4ceVxebis/DIe7H2Q1a+s5oeX/FDbMYiIiEjCUrEnsePiuCNzI/F6fHzr4w+yp2ELNz/wH8c9pjXazCADnOmujlVKOdKwZcrGRfzLZXdiQ4MYb/LM6cxPy6f5j828+LEXeXjfw3xo1oecjiQiIiIyLlTsSexEOWZkbjTphwvx78mi/MP53P6Zbx13dK4+UoMXL0WuklgmTXrRlDCd8/ZSGJwLa3vg4ianI8VV+7PtXPWlq7hnwz0sLVpKWVaZ05FEREREYk7dLsQZFvz1Uwll9DOU3XvcQyI2QlO0nqnuUtwmubpHxsNQbg9/2fxb2JODPZhkDUssfOu8b+ExHm5fe7u2YxAREZGEpGJPHJHSlYW3P52+0kYYYe3kzg8AACAASURBVMlUS7SRMGGmuSriGy6JPLX511AQgJeKsYHk+nMwNWMqty2/ja2tW/nFm79wOo6IiIhIzGkapzjCX19MxBtisLB9xGMaIjWk4KPAVRjHZGOzevVqpyOMSTQagYsb4aFKWFsElzU6HSmurpx+JS/WvciarWu4oOQC5uTNcTqSiIiISMyo2JO4cwdSSG3Lo6/8MLijxz0mbEM0Rw9T5q6aFHvrfea6W8Z87t2PONtJ1OQHsWe3wcYp2Bm9mKo+R/PEkzGGO1fcyabmTXx93df53Qd+h9fldTqWiIiISExM/FfRknDSG4oBGCgZuSlIU7SBCBFK3OXxipXcqtuTdjpnTmoOd55zJ7s7dvOz7T9zOo6IiIhIzCTXqzpxXtSQ3lhIoKCDSOrQiIc1RGpJJY08MyWO4ZKXcTM8nTPoHp7OmWQuLb+UK6qu4Kfbf8rujt1OxxERERGJCRV7Ele+jmzcIS8DU1tHPGbIDtESbaLEXa4Nr+PI5AfhrDbYl42tS3c6Ttzdtuw2clJzuGPtHYQioXG7T1lFGcaYMT/KKrRNhIiIiJwcrdmTuEprnkLUEyKY1zXiMU2ReixRStzqwhl3iztgbxa8XIz9yEGMJ3k2W89JzeHfV/w7t7xwCz/Z/hNuXnzzuNynvraeNVvWjPn8mxbfFMM0IiIiksg0sidxY8IuUttyhztwukYuIhqiNaSbDLJNbhzTCYBxW7igGXpSYEu+03Hi7uLyi7ly+pX8/I2fs6djj9NxRERERE6Lij2Jm9S2PFxRN4NFbSMeE7QBWqMtlLg0hdMppnQAZnXDljxsV4rTceLuq0u/SpYvi2+88g3C0bDTcURERETGTMWexE1a0xTCqQGGsntHPKYp0gBYprm1LslR57SAx8LLRdjkmckJDE/nvG35bbzZ/ia/2fkbp+OIiIiIjJmKPYkLV9CLrzN7eFRvlAG7pmgD6cZPlsmJXzg5hkmPwPJWaPAPr+FLMisrVnJx2cX8aOuPqO2pdTqOiIiIyJio2JO4SGvJx2BGncLp8rlojTZR7CrRFM6JYG4XFA7C+kLsUHL9qTDGcMfyO/C6vNz16l3YZBveFBERkYSQXK/gxDFpzVMYyugj7B8c8ZiM+RlEiVLsKoljMhmJcQHnNcOAJymbtRT5i/jyki+zsWkjD+19yOk4IiIiIqdMxZ6MO09/Kim9GQwWjzyqB5C5OBMvKeS5tJH6RGGKAjC7G7blYru9TseJu2tmXcPS4qXct+k+2gZH//kVERERmWi0z56Mu7TmKVgsg4Ujv1iO2iiZ1ZkUuabhMnoPIp5Wr1496vMZrhxuyr+Pg7/o4L+774tTqonBGMOdK+7kmseu4d6N93Lvhfc6HUlERETkpKnYk3GX2prHUE4PUV9oxGM6oq14MjwUuzWFM94+c90tJzwmWNPCGQeWctPyOwnmdR/9+t2PfH08o00IVdlV/OuCf+XH237MB2d8kHNLznU6koiIiMhJ0RCKjCv3QCregXQCUzpGPa4p2kA0FKXQVRynZHIq+soaCacFyNpbCdHka55z44Ibqcyq5Jvrv0kgHKCsogxjzJgeIiIiIvGikT0ZV6ntuQAE8jtHPMZaS1O0gf6d/XiWJ9+6sEnBZemZcYi8HWfgP1xEf2mT04niKsWdwp0r7uTGZ27kp9t/Sn1tPWu2rBnTtW5afFOM04mIiIgcn0b2ZFyltuUS8vcTSQuOeEyv7WbA9tOzuSeOyeRUBQo6CeZ2kXGoFBN2Ox0n7pZNXcaqGav4xY5f4JvmczqOiIiIyAmp2JNxY0IeUrqzCBSMPKoHw1M4AXq39sYjloyVgZ4ZtbhDXjJqpjmdxhH/c8n/xJ/iZ9qnpmnvPREREZnwVOzJuEltz8FYc+JiL9JArskn3B2OUzIZq1BmPwNFrWTUT8MVSHE6TtzlpuZy61m34j/Dz96uvU7HERERERmVij0ZN6ntuURShghl9o14TMAO0mU71IVzEumdXgsWsg6WOR3FER+a9SEG9g/w6uFXCUZGnp4sIiIi4jQVezI+ogZfe85wY5ZRGhC2RBoBKHRNjVMwOV2R1CH6S5tIa5rCtLwqp+PEncu4OPzrwwyEB9jYtNHpOCIiIiIjUrEn48LXlYUr4iFQMPqWC63RJnykkmVy4pRMYqG3oh7riXDV8s86HcURgUMB5uXPY0fbDtoG25yOIyIiInJcKvZkXPjacom6IgRzR+6waW2UlmgTha5i7T82yVhvhN6Kes4sX4atT3c6jiOWFy/H5/bxcv3LatYiIiIiE5KKPYk9C6lteQTzusEdHfGwLttJiCEK3ZrCORn1lzbR1tMI6wtJxlon1ZPKiqkraBpo4q3Ot5yOIyIiInIMbaouMefpT8cT9NFbVTfqcS3R4fV6U1zF8YglseayPLHx/3B91h08fO9feSOw1ulEcXdG3hns7NjJ+sb1VGVXkeJOvg6lIiIiMnGp2JOYS23LxWIJ5neNelxLpJEck0eK0QbVk9WmfS/wsau+yD/5PsPy5WeD++SH+O5+5OvjmCw+jDGcN+08Ht73MJtbNrNi6gqnI4mIiIgcpWmcEnO+jhxCGf1EU0IjHjNkh+i0HerCOclZLD0za/AEffgbknOEtthfzOzc2Wxr3UZ3sNvpOCIiIiJHqdiTmDIRFyk9GcPr9UbRFm0CrNbrJYCh3B4CeZ1k1pRiQsk5WWDF1BW4jItXDr/idBQRERGRo1TsSUyldGVhrIuh3NGLvZZoI1685Ji8OCWT8dQzowYTdpNZU+J0FEf4vX7OLjybQz2HqOsdfa2qiIiISLyMW7FnjCkzxrxgjNlpjHnTGPPFI1/PM8Y8a4zZe+S/uUe+bowxPzTG7DPGbDfGnDVe2WT8+DqzsSbKUHbviMdYa2mJNFHgKsZl9H5DIghnDDIwtRV/fTHuweRcg7lwykKyUrJY17COiI04HUdERERkXEf2wsCXrbXzgBXATcaYecDXgL9aa2cBfz3yOcD7gVlHHp8F/nMcs8k48XVmM5Tdix1ly4Ve202AQQrVhTOh9FbVgYHMg2VOR3GEx+Xh3Gnn0hnsZFf7LqfjiIiIiIxfsWetbbTWbj7ycS+wCygBrgJ+deSwXwEfPPLxVcB/2WHrgRxjjBZ0TSIZqdl4+/wETziFswlA6/USTNQ3RF9pI2nNBXh6k3Oj9cqsSqb5p7GxeSNDkSGn44iIiEiSi8scOmNMJbAYeA0ostY2HnmqCSg68nEJ8M7FLvVHviaTxOySxQAnUew1kmmySTPJWRAksr7yBqwnQtaBcqejOMIYwznTziEQDrClZYvTcURERCTJjXuxZ4zJAB4CbrXW9rzzOWutBU5+Y67h633WGPO6Meb11tbWGCaV0zWnZDFRT5hQZt+Ix0RshI5oqzZST1DWG6G3op7UjlxSOrOcjuOIwvRCZuXMYlvrNvqGRv5dEBERERlv41rsGWO8DBd6v7XW/unIl5vfnp555L8tR77eALxzsU/pka+9i7X2p9baJdbaJVOmTBm/8HLK5pScRTCnZ9Sfqs5oG1GiFLgK4xdM4qq/pImwL0jW/opTfCsncSybugyLZUPTBqejiIiISBIbz26cBvg5sMtae987nnoM+PSRjz8NPPqOr3/qSFfOFUD3O6Z7ygQXGAhRkDXthFM426ItgCHfpUI9YbktvVV1pPRmkNqanFtrZKVksbBgIXs699A22OZ0HBEREUlS4zmydx7wSeASY8zWI48rgHuAy40xe4HLjnwO8GfgALAP+N/A/xjHbBJjvW0BAIZyu0Y9ri3aTI7JxWtS4hFLHDJY3ErIPzC8di9qnI7jiLOKzsLn9vHK4VcYnrEuIiIiEl+e8bqwtXYtMNKrvEuPc7wFbhqvPDK+utsG6epvI5weGPGYsA3TaTuY4Z4Tx2TiCAM902vIf2Mu6Y2FDJQ0O50o7nxuH0uKlrDu8Dpqe2upyKpwOpKIiIgkGe1oLafNWktvW4A99ZtHLu+BjmgrVuv1kkYwv4tgdg+Zh0ox4eT8U3Nm/plkpWTxauOrRO3Ie0+KiIiIjIfkfAUmMTXYM0Q4FGVPw+ZRj2uLtmAw5LkK4pRMHGWgZ0YN7qEU/PXJuaei2+VmxdQVdAY62d2x2+k4IiIikmRU7Mlp6+kYnrr51uHR9xVrizaTa/LxGG88YskEEMruY7CgnYzaabiGxm3W+IQ2PXs6xenFbGzaSCgScjqOiIiIJBEVe3La+jqCpKS56eofuetgyA7RZTs1hTMJ9U6vxUTcZNSUOh3FEW9vtD4QHmBr61an44iIiEgSUbEnp8VaS19nkIzc1FGPa4+2ApZ8V1F8gsmEEfYHGJjagr+hCPegz+k4jij2FzMjewZbW7fiyU7OEU4RERGJPxV7clqCA2FCwQiZeaO/iG+LtuDCRZ4rP07JZCLprarHGkvmwTKnozhm+dTlRG2Uwg9pdFtERETiQ8WenJa+ziDASYzstZDrKsBtNKqRjKK+IfpLG0lvnoKnN93pOI7I9mVzZv6Z5F6QS2eg0+k4IiIikgRU7Mlp6esI4PYY0jJHbroyZIN0a71e0usrP0zUEyLrwATYb841vJZurI+xOqvwLKJDUTY2bYzhNyMiIiJyfBpmkdPSe2S93mgvgIfX60GB1uslNeuN0FvRQPb+SlI6spwNE4U1W9aM+fSbFt80pvPSvem0/6Wd/Vftp3WglSnpU8acQURERORENLInYxYeihDoC5FxwvV6zbhxk2vy4pRMJqr+kibCviBZByowjH2EbDJre7oNn9vHa02vOR1FREREEpyKPRmzt9frZZ5EJ85cVwEu445HLJnI3Jbe6bWk9GZQPf1Cp9M4IjoY5azCs6jrreNw32Gn44iIiEgCU7EnY9bbEcAY8OekjHhMyA7RY7vId2m6mgwbLGoj5O9n1bIbsRGn0zhjfsF8/F4/6xvXY611Oo6IiIgkKBV7MmZ9nUHSs3243CP/GHVG2wHIMwXxiiUTnYGeGbVMyS6BXTlOp3GEx+VhSdESmgeaqempcTqOiIiIJCgVezIm0UiU/u7gCffXa4+2YjDkan89eYdgXhd7D2+F1wuwweT8MzQnbw7ZKdm81vSaRvdERERkXCTnqyw5bf3dQ9joiffX64i2km1y8ZiRt2aQJGTgoVcfgIAbtiTnGwFu42Zp8VI6Ah3s7drrdBwRERFJQCr2ZEz6OgIAZOSOPLIXtRE6bQd5Lk3hlGPVt+2F2T2wPRfbk5xvBszMmUlBagEbmzYSiSbpAkYREREZNyr2ZEx6O4Ok+r14fSN32OyynUSJkKfmLDKSZa3Df4VeS86fEWMMy6Yuo2eoh10du5yOIyIiIglGxZ6cMmstfZ3BE+6v1xFtA1CxJyMyGWFY1AH7s7BNaU7HcUR5ZjnF/mI2NW8iFA05HUdEREQSiIo9OWWBvhCRUHTUKZwwvF7PbzJJNaOv65MkV90O6SF4pZBk7FNijGFF8QoGwgPsaNvhdBwRERFJICr25JT1dw1vpp6RM3KxZ62lI9qq9XpyQsZrYXkrtKTB3iyn4zhiasZUKjIr2NKyhWAk6HQcERERSRAq9uSU9XcP4fIYUjNGbqrRZ3sZYoh8oymcchJm98CUQVg/BRsyTqdxxLKpywhGgmxt2ep0FBEREUkQKvbklPV1BfFn+zBm5BflHdFWAI3syUkxBjivGQa8sDk5t2IoSCtgZs5MtrdtZyA04HQcERERSQAepwPIBOOC5atmjPi0x+Xluzc8zvPb/y+P3f2zEY9rj7aSgg+/yRyPlJKATHEAO7sbtuVhz+jGZCdfs5Jlxcs40HWATc2buKD0AqfjiIiIyCSnYk/eLQq3P/itEZ/29mTg2eRlySfOYv6X3n3c3dd9/ejHHbaVPNeUUUf/RI6xvAUOZsArRfD+eqfTxF22L5sz8s9gZ8dOFk1ZRJYvOdcwioiISGxoGqecEm9PBgBDWX0jHhOwgwzYfvI1hVNOkfFH4Ox2qMnA1vidjuOIs4vOxmDY2LzR6SgiIiIyyanYk1OS0usnkjJE1Dc04jF/X6+n5iwyBgs6ICcIrxRhI06Hib8MbwYLChbwVudbtA+2Ox1HREREJjEVe3JKvD0ZhDL7YJTZmR3RNty4yTa58QsmCcO4gfNaoDsFtiVns5bFhYtJcaWwoWmD01FERERkElOxJyfNhN14BtJGncIJ0BltJ9vk4jL68ZKxMWX9ML0HNuVju0fe4iNRpXpSqS6s5lDPIZr6m5yOIyIiIpOUXo3LSfP2+jGY4ZG9EURshG7bSa7W68npOq8FXBbWFmGt02Hib2HBQtI8aaxvXI9Nxv8BIiIictpU7MlJSznanKV/xGN6bBdRouS6knP6ncSO8YdhWRvUZcD+5NvCw+v2sqRoCY39jdT21jodR0RERCYhFXty0ry9GYTTBrHe8IjHdEaHG0qo2JOYOLMTpgwON2sJJt+fq7l5c8lKyWJ943qiNup0HBEREZlkku/Vk4xZSk8GQ5kjj+rBcLGXShppJj1OqSSRGRdwYRMMumFD8nV3dbvcLC9eTkegg72de52OIyIiIpOMNlWXk+IKenEHfYSyDo96XKdt06ienLTVq1ef1HErMz/Fsh0r+T8vf5/6UHIVPTNyZrC1dSsbmzYyM2em03FERERkElGxJyfF23vizdTdmW4GbD+Vbr0glZPzmetuOanjTNhFZOMQn/TfRuuS7eCOcvcjXx/ndBODMYYVU1fw+IHH2dG+w+k4IiIiMoloGqeclJSeDKyxhDMGRjwmffrw1E2N7EmsWU+Urjn78Q6kkXWw1Ok4cVeaWUppRimbmzfjStOfbRERETk5etUgJ8Xbk0HYP4B1j9wkIm1GGgZDtsmLYzJJFkN5PfRPa8JfNw1vd4bTceJuxdQVBCIBCt6vbU1ERETk5KjYkxOzkNKbwdAo++vB8MhelsnBYzQ7WMZHz4xaIr4hcnbPwONOrs3Wp6RPYWbOTApWFtA60Op0HBEREZkEVOzJCbkDPlxhz6ibqVtrSZuepimcMq6sJ0L3GfvxDqTzgSXXOx0n7pYVL8O4DQ9se8DpKCIiIjIJqNiTE/L2+QEIZY68Xq/P9uBOd5OjYk/GWTCvm/6pzVy68J+xjWlOx4mrbF82HS928NDehzjUfcjpOCIiIjLBqdiTE/L2+rHGEvKPXOy9vZl6nlGxJ+OvZ+Yh2nub4K/Tkm6z9ZbHWkhxp/DDLT90OoqIiIhMcMn1KknGxNvnJ5w+CKM0Z+m0bYT7wvhNZhyTSbKynii//Ou3YcADLxVjrdOJ4ifSE+H6M6/n2ZpneaP1DafjiIiIyASmYk9OyNvrJ5TRP+oxndEOBg8OYoyJUypJdjWte2BJG+zPgj3ZTseJq0+f+WnyUvP4/ubvY5Op0hUREZFTomJPRuUa8uAeShm12AvbED22m8H9g3FMJgJUt8PUAVhbhO1Onu6cfq+fzy36HBubNvJi3YtOxxEREZEJSsWejMrb+3ZzlpGLva5oB2AZ2D/ymj6R8WBcwKWHwW3huWnYcPKMLH949oeZnj2d777+XUKRkNNxREREZAJSsSejOtqJc5SRvS7bAcDgAY3sSfyZjDBc1AitabCu0Ok4ceN1efnK0q9Q21vL73b/zuk4IiIiMgGp2JNRefv8hFMDWG9kxGO6oh2kGz+R/pGPERlPpqpveErnrlzs7uRZv3d+yfmcX3I+P9n2EzoCHU7HERERkQlGxZ6M6mSas3TZDnJMXpwSiYxgWStM64eXi7BtPqfTxM1XlnyFgfAAP976Y6ejiIiIyASjYk9GZMIuPINpo67XG7JBBmw/2S4Ve+Is4wIuPwypEfhLSdLsvzc9ZzrXzrmWP771R/Z27nU6joiIiEwgyfFqSMbkpNbrRTsBNLInE4JJi8DlDdDvHW7YMvLWkAnl84s+T4Y3g3s33qutGEREROQoFXsyoqPFXubIXTa7jzRnyXHlxiWTyImY4gCc3wR1GbCuyOk4cZGTmsPNi29mfeN6nql5xuk4IiIiMkGo2JMReXv9RLwhoilDIx7TFe3AbzLxmpQ4JhMZnZnXDYva4c1c7BvJ8UbER2Z/hLl5c7l3470MhLQNioiIiKjYk1F4+440Zxll67KuaAc5JjleTMsks7wVKnvhlUJsjd/pNOPO7XJz+/LbaRlo4YFtDzgdR0RERCYAFXtyfFGDp3/05ixBG2CQAXLUnEUmoKMbrucHh9fvJUGHzurCaq6eeTW/3vlr9nftdzqOiIiIOEzFnhyXpz8NY10naM7y9no9FXsyMRmvhffVQ0oU/lyG7fY6HWnc3Xr2raR70/n2a99WsxYREZEkp2JPjuvvzVlGKfaONGfJ1jROmcBMRhg+UAcRA0+WYfvdTkcaV3mpeXzxrC+ysWkjfz74Z6fjiIiIiINU7MlxeXv9RN0RImmBEY/pinaQYbLwmMQfLZHJzeQNwRV1MOCBJ8tIS8lwOtK4umbWNczPn8+9G++lK9DldBwRERFxiIo9Oa6Tb86iKZwyOZiiwPCUzq4UPv/+u7GhUX64Jzm3y81d595FT7CHezfe63QcERERcYiKPTmWBW9fOuGMkdu3B+wgQQJaryeTiikdgMsaqSw8Y3gNXwIXfHPy5nDjght5/MDjrG1YG5NrllWUYYwZ06OsoiwmGUREROTkeZwOIBOPO+DDFfEQ8qs5iyQeM72XXz3/C/7l8juGC74r6oYbuSSgzy78LM/WPMt/vPofPHzVw/i9p7cFRX1tPWu2rBnTuTctvum07i0iIiKnTsWeHMPTnw5AaJSRveFiz5BlcuKUSuRYq1evHtN5Lftb+Jd7/gX+Oi2hC74Udwqrz13Np576FPdvvp/bl9/udCQRERGJIxV7cgxv33CxF/YPjnhMl+0g02ThMfoREud85rpbxnTe3Y98HTOzF8vh4YLvyTLsFfWYlGiMEzqvurCaj8/9OL/d9VtWVq7k7KKznY4kIiIicaI1e3IMb3864dQA1hM57vPW2uHmLJrCKZOcmdkLlx2GljR4rBw7mJjbMnxh8RcoySjhjrV30DfU53QcERERiRMVe3IMT1/6qFM4BxlgiKA6cUpCMDN6YWU9dKbAo+XYvsQbrU73pvOdC75DY38j92y4x+k4IiIiEicq9uRdPG4vnsE0wv6Ri73uaCcA2S5tpi6JwVT0w5VH9uF7pALblXh7R1YXVvOZBZ/h0f2P8mzNs07HERERkThQsSfvUpxTgbFmeI+9EQwXe2rOIonFTB2EVbUQNsMFX3Oq05Fi7nOLPseZ+Wey+tXVtAy0OB1HRERExpmKPXmXaXlVAKOP7NlOMkymmrNIwjEFQfhgDXij8Hg59lCG05Fiyuvy8p0LvkMwHOTOdXcStYnXkEZERET+Tq/W5V1K8qdjXVHCaYERj+mOdpLvKoxjKpH4MTkh7NU18FQp/KUEe34z5syuY44b67YPp3vu6arKruIrS7/CN9d/k/9687+4fv71jmURERGR8aViT95lWt50QukDI475Bm2AAINarycJzaRHsKtq4dkSeLl4uGnLsjaM+fsxp7Ptw1jPffv80/XPs/+Z9Y3r+cHmH7BwykLOKjrrtK8pIiIiE4+mccq7TMurIjxKJ86jzVm0Xk8SnPFaeF89zO2CLQXw/FTs8XcjmXSMMaw+dzUlGSV85W9foX2w3elIIiIiMg5U7MlRg71DZPvzR912odsOT2fTyJ4kA+MCLmyCpa2wNxv+XIYNJsafzcyUTO676D66h7r52stfIxJNkEpWREREjkqMVy0SE+0Nw5sth06w7UIa6aQYX7xiiTjKGDBnt8PFh6ExHR4tJ8df4HSsmJiTN4fbl9/O+sb1/GT7T5yOIyIiIjGmYk+Oam8Y3m4hPNq2C7ZTo3qSlMycHnh/HfR6+fIH/xeevjSnI8XE1TOvZtWMVTyw7QFerHvR6TgiIiISQyr25Kj2hj56BjqJpoSP+3zYhui3vSr2JGmZsgG4qhaXcVGwZT4pnVlORzptxhi+vuLrzM2fy1df+ir7Ovc5HUlERERiRMWeHNXe0MfhjgMjPn90vZ5RsSfJyxQE+e4jXyCSMkT+trmkNU/+KZ1pnjTuv/h+0r3pfOH5L9AVOHarCREREZl8VOwJANGopf1wPw3tIxd7PW934tTIniS5zr4W2s7a8f+zd+fxcVXn/cc/586+aJdlbZZsI+9mscHGOBiMMZgdmkBKmzQhCaRpaGjT7JQmJU2ALE3SEkIWQkiaX5MmEAgBYsBgVgMGbINtvNuStVmypZE0M5r13vP7QzKxkWRrvzPS83695uXR3Ll3vjZCmmfOOc8hmRuh4J1ZBOvKQdudamRKA6X84IIf0Nrdyuee/xwpK2V3JCGEEEKMkBR7AoCuwzHMlEVT+4EBn9OpQ7jx4GVirFUSYiS0y6TtjHeIlRwhd381eXtmZH3Bd/qU0/na8q+x8dBGvrXxW2id5X8hIYQQYpKTTdUFAEcaejpxNrbvB1b0+5wOK0SekY86dmdpISYzQxOavwfTkyRYX46RcNMxfw/aYdmdbNiuOuUq9ob28ovtv6AiWMHHFn7M7khCCCGEGCYZ2RMAtDVFUApaQnX9Hre0SVh3yXo9Id5LQVdNHZ2zDuA9UkDRlvmoVHZ/jvbPZ/4zl06/lO+9+T3+tO9PdscRQgghxDBJsScAaG+KklfiJ2Um+z0e1l1oLPKMwnFOJkR2iFYeIrRwN65IgOLN8zGSLrsjDZuhDL5x7jdYWrqUr778VTY0bbA7khBCCCGGQYo9AUCoOUphWWDA451Hm7Oo/PGKJETWiU9pp+3UHThiXoo3LcCIu+2ONGxuh5sfXPADZuTP4LPrP8uOth12RxJCCCHEEEmxJzDTFp2tMQpK/QM+p1OHcOAkoHLGMZkQ2SdZ2EX76Tswki6KNy/A0e21O9Kw5bhzuPfCe8n1xY9uswAAIABJREFU5PLpZz6Nqzh7RyuFEEKIyUiKPUFnawzL0hScZGQvT0lzFiEGI5kfpm3ROyjTkfUF39TAVH68+sckzSTTPzedWDpmdyQhhBBCDJIUe4LQoSjAgNM4tdZ06g7ZX0+IIUjlRGlbtB20yvqC75T8U7h71d24il38+cCfZQ8+IYQQIktIsSdob46CgvwBpnFGdQSTNLmyXk+IIUkHYrSd0VPwFW2ZjyPmsTvSsC2eupiGHzfQ0t3Curp1WDp7t5cQQgghJgsp9iaY6qpqlFJDuv3oe/dxpLMZt6f/dvFdugNARvaEGIZ0MEbbGe9gmAZFW+ZTECyxO9Kwdb3ZxYqKFdR21fJCwwuy6boQQgiR4bJ7MyjRx8H6g7z26L4hnbPtxUbcHievPbqPs686pc/xLqsDUOSo3FFKKcTkkg5203bGDoq2zOeWK/4TM9GE5cnOqZALixcSTUXZ1LoJv9PP0rKldkcSQgghxABkZG+S01oTj6TxBgfustelOwiqHBxKPhsQYrhSOVHaTt9Brr+AorfnoVIOuyMN29LSpcwrnMebrW+y9fBWu+MIIYQQYgBjVuwppe5XSrUqpbYd89i/K6UalVJbem+XHXPsK0qpvUqpXUqpNWOVSxwvEUujLY3vBMXe0U6cQoiRSeVG+NlTX8MZ9VG4dS6Y2fl5m1KK8yrPY0buDF5qeok9oT12RxJCCCFEP8byncYDwCX9PP59rfUZvbcnAJRS84HrgQW95/xIKZW9H3tnkXi4ZyqZL6f/Yi+lk8ToJteQYk+I0bCz4U1C8/bi7syh4J1ZkKV9TgxlsLp6NWWBMp6tf5b6cL3dkYQQQgjxHmNW7GmtXwDaB/n0q4Hfaq0TWusDwF5AFoKMg1ikp9gbaBrn0eYs0olTiNETn9pG56wD+I4Ukrd7JmRpnxOn4eTSGZdS4Clgbe1aWrtb7Y4khBBCiGPYMYfoH5VSb/dO8zza3rECOPZj4Ybex8QYi0dSuDwOnK7+B1I7LenEKcRY6K5sIVzdQKB5KsGD5XbHGTaPw8PlMy/H5/Tx+P7HCcVDdkcSQgghRK/x7rhxL/Af9HyO/R/AfwIfH8oFlFKfBD4JUFVVNdr5Jp1YJHnS5ixuPHjI3g2hhejP7bffbncEwjPqcca85O6vJu2PE58y2MkQmSXgCnDFzCt4eO/DPL7/ca6ZdQ1BV/D4Jxk9a/2Gq7Kqkvo6mSoqhBBCDMW4Fnta65aj95VSPwMe6/2yEZh2zFMrex/r7xo/BX4KcNZZZ2Xp5KfMoLUmFklRXBEc8DldVge5Rv6I3qQJkYluvP6WYZ97xyO3jU4IBaG5e3HEPeS/U0Pbou2kcqOjc+1xlu/J54oZV/DHfX/sKfhqrsHjOGYTeQvu2XzPsK9/86KbRyGlEEIIMbmM6zROpVTZMV/+FXC0U+ejwPVKKY9SagYwC9g4ntkmo1TCxEoP3InT0hZdulPW6wkxlhya9lN3YrlTFG6diyPutjvRsE3xT+GS6ZfQkejgydonMS3T7khCCCHEpDaoYk8p9b7BPPae478BXgHmKKUalFKfAL6tlNqqlHobuAD4LIDWejvwO+AdYC1ws9Za3iWMsVhvJ07vAJ04ozqMhUmedOIUYkxZ7jTtp+1EWQYFW+egsnRLBoDKnErOrzyfxkgjzzc8j9YyAUMIIYSwy2Cncd4NLB7EY+/SWv9NPw///ATP/ybwzUHmEaMg3tuJ0xfsfyRBOnEKMX7SgRiheXso3DqXvF0z6Zi3F7J09vTcwrmEk2HeaHmDPE8eZ0490+5IQgghxKR0wmJPKXUOsByYopT6l2MO5QKyD16Wi0WSOFwGTnf/owidVgcKgxyVO87JhJicEsUdhGfUk3ugilROlOi0ZrsjDdtZU8+iK9nFxkMbyXHn2B1HCCGEmJRONrLnBoK9zzv2t3UXcO1YhRLjIx5J4Qu6Bmy+0qU7yFG5GLK/vRDjJlLdiCscIHdfNalglGRBl92RhkUpxcrKlUSSEdbXr8d3is/uSEIIIcSkc8JiT2v9PPC8UuoBrXXdOGUS4yQWSZE/1T/g8S6rg2Jj6jgmEkKgoGPeXorfPJWC7bM5ctbbmN6k3amGxWE4WDN9DQ/teYiqz1QRSUX6bskghBBCiDEz2C4AHqXUT5VSTymlnj16G9NkYkylkibppDVgJ86EThAnRq40ZxFi3GmnRfupu1CWomD7bLCydPEe4HV6uXTGpRgegycPPEnaStsdSQghhJg0Blvs/R7YDNwGfOGYm8hSf2nO0n+x12WFAMiT5ixC2ML0x+mYuw93Vw65+6rtjjMihd5CGn7aQGusVTp0CiGEEONosN0401rre8c0iRhXf9l24SSdOGVkTwjbxEvaiXQ2E2woI5nXRbykHQwGXGd7MpVVldTX1Y9yysEJbw6zpHQJrx96nSJvEWeUnGFLDiGEEGIyGWyx9yel1KeBh4HE0Qe11u1jkkqMuXgkieFQuL39N1/ptDrw4MWjvOOcTAhxrK5T6nB3BsnfeQqHg91gwT2b7xnWtW5edPMopxuaM0vOpC3WxqvNr1LiL6E8WG5rHiGEEGKiG+w0zo/SM21zA/Bm7+2NsQolxl4sksJ7kk6cMqonRAYwNKGFu0FB4fbZOB39T73OBkopLph2AbnuXJ6ue5ruVLfdkYQQQogJbVDFntZ6Rj+3mWMdToydWO+2C/2xtEVEd8l6PSEyhOlNEpq3B1ckwNVn32R3nBFxO9ysmb6GhJlg3cF1WNqyO5IQQggxYQ1qGqdS6iP9Pa61/tXoxhHjwUxZpOLmgMVeRIexsMiRkT0hMkaiuINIRTMX8AH0wXpUVdTuSMNW5CtiRcUKnmt4jjda3mBp6VK7IwkhhBAT0mCncS455rYC+HfgqjHKJMZYLNrbnCXYf3OW8NHmLCpv3DIJIU6u65Q6mtr2w/oydHf/622zxbyiecwpmMObLW9SH7anaYwQQggx0Q12GudnjrndBCwGZGfcLBUP92zQ7MsZaNuFDhSKoModz1hCiJNxaH7xzDchafQUfFm+g8GKyhUUegt55uAzsn5PCCGEGAODHdl7rygwYzSDiPETi6RQBnh8/c/i7dKdBFUuDpXdIwdCTETNoVo4pxXqg7CtwO44I+IyXKyuWk3STPJcw3Oy/54QQggxyga7Zu9PwNHfwg5gHvC7sQolxlYsksIbcKGMATpxWh0UGsXjnEoIMWgLOnqKvVemoMu7UUWJk5+ToYp8RSwrW8bLTS+zvW07C4sX2h1JCCGEmDAGu8/ed4+5nwbqtNYNY5BHjIN4JIU/r//1eobPIEa3bLsgRAZTCvTKZvj9DFhXjv5ALcqZvaNipxafysHwQTY0baAiWEGBN7tHLIUQQohMMdg1e88DO4EcoABIjmUoMXYs0yLRnR6wE6e3omcT9VzZdkGIjKZ8JlzQBCEPvFJid5wRUUqxatoqXIaLdXXrMC3T7khCCCHEhDCoYk8p9UFgI3Ad8EHgNaXUtWMZTIyNeDQNgG+ATpzeaT3FXo4hnTiFyHRqWjec1g7bC9C1g+iZZfQUVsO9jSW/y88F0y7gSPwIrx96fUxfSwghhJgsBjuN81+BJVrrVgCl1BRgHfDgWAUTYyMW6RmU9Q4wsuep9ODEhQ//eMYSQgzX2Yeh0Q/PlaKvq0UF0gM/14J7Nt8z7Je6edHNwz53MKbnTWdu4Vy2HN7CjLwZTA1MHdPXE0IIISa6wRZ7xtFCr1cbw+/kKWwUD/fusRcYYBrnNC+5Km/MP8UXQgzf7bffftzXxY5ybiq6g4P3dfP/Ou7iL/20ss/y8uXUh+t5tv5Zrpt9HU5jsL+mhBBCCPFeg/0tulYp9STwm96v/xp4YmwiibEUi6TwBJwYjr7FnNYab6VXmrMIkeFuvP6WPo9FGxs5Zfdp/NO53yRaeajf8+545LaxjjZiHoeHlZUrefzA47x+6HXOKT/H7khCCCFE1jrh6JxSqkYp9T6t9ReAnwCn9d5eAX46DvnEKItFUgM2Z2mzDuPwO6Q5ixBZqLu8hXhRiNx91TijPrvjjEhVbhXzCufx1uG3OBTtv3AVQgghxMmdbCrmD4AuAK31H7TW/6K1/hfg4d5jIotYliYRTeEdoDlLo1kHICN7QmQjBR1z9mE5TAremQVWdk/FXl6+nIArwPr69aStE6xDFEIIIcSATlbsTdVab33vg72PTR+TRGLMJLpTaM2AI3sNZi0AOUo6cQqRjSxPis45+3BFAuQcmGZ3nBFxO9ysnLaSjkQHb7a8aXccIYQQIiudrNg70RBPds8TmoTikZ7mLAMXe3UkDydxqf6PCyEyX3xKiGhZC8GD5bg7cuyOMyLTcqYxu2A2W1q34Knw2B1HCCGEyDonK/beUErd9N4HlVI3AvJRa5aJHe3EeYKRvXhDfDwjCSHGQFdNLaYvQf6OGlTaYXecEVlevhy3w03FxyqwtGV3HCGEECKrnKzY+2fgY0qp55RS/9l7ex74BPBPYx9PjKZYJIXb58Dh7PufPaVTHDIbiddLsSdEttNOi9C8PTjiHvL2TLc7zoj4nD6Wly/HX+Pnd7t+Z3ccIYQQIqucsNjTWrdorZcDtwO1vbfbtdbnaK2lRVqWiUdSA47qNZv1WFgysifEBJHKixCpbsR/qARva6HdcUZkdsFsItsi/GDTD2iJttgdRwghhMgag9oYXWu9Xmt9d+/t2bEOJUaf1pp4JIVvgE6cDb2dOBP1ifGMJYQYQ+HpDSRzIuTvmomRyN61uEopmn7ZRNpKc+fGO+2OI4QQQmSNQRV7IvslY2ksSw/YnKXRrMOJi0SLFHtCTBiGJjR/D1hGz/o9snc7huThJP9w+j/wzMFneL7+ebvjCCGEEFlBir1JIhY5WXOWOsoclSD9D4SYUEx/nK6aOryhfM5bcLXdcUbkI/M/woy8Gdy18S4SpnwwJYQQQpyMFHuTxMm2XWhM11HpqB7PSEKIcdJd3kK8MMTVyz6Jbu9/Knc2cDlc3Hr2rTREGrh/2/12xxFCCCEynhR7k0QsnMLpMXC6+7Zhj1oRQrqNCin2hJiYFHTM3UciFYNny9Gm3YGGb1nZMi6uvpifb/05DeEGu+MIIYQQGU2KvUkiFknhC/T/iX5jb3MWGdkTYuKyPCl+88L34IgX3ii2O86IfGHJFzCUwbdf/7bdUYQQQoiMJsXeJNDTiTOJL2eg5iwHAahwVI1nLCHEOHu79mWY2wFbitDNPrvjDFtpoJRPnvZJ1tev58WGF+2OI4QQQmQsKfYmgVTCxEzrEzZn8akABUZ2f9ovhBiE5a2Qk4Jny9DJ7P0V8NH5H2V67nTu2ngXSTNpdxwhhBAiI2Xvb3oxaCdtzmLWUeGoQqnsbcsuhBgc5bZgVRNEXPDCVLS2O9HwuBwuvrL0KxwMH+SB7Q/YHUcIIYTISFLsTQIn2nZBa02jeVCmcAoxiajSOJx1BPbmwa48u+MM2/KK5ayuWs3P3v4ZTZEmu+MIIYQQGUeKvUkgFk7hcCpcnr6dODt0O906Is1ZhJhsFrVBeRRemoruyN7tGL645IsAfOf179icRAghhMg8UuxNAvFICm/Q3e80zcZ0TydO2XZBiMlFGcCqZnBa8HQ52szOadxlwTJuOu0m1h1cx4bGDXbHEUIIITKKFHuTQCySHHC9XoN5tNiTaZxCTDYqmIYLmqHNC69MsTvOsN2w4Aaqcqq4c+Od/TZrmVY9DaXUsG/TqqfZ8LcSQgghRs5pdwAxttJJk3TSOuG2C3mqgKCRO87JhBCZQFVH0ae2w9ZCdGU3anrE7khD5na4+fLSL/PpZz7Nr975FTeeeuNxxxsONnDP5nuGff2bF9080ohCCCGELWRkb4I7UXMW6OnEKev1hJjklh2G4jisL0NHsvMzwBWVK1g5bSU/ffuntERb7I4jhBBCZAQp9ia4E227YGmTJrOeCqcUe0JMZsqhYXUjmAqeKUdbdicani8u+SKmZfL9Td+3O4oQQgiREaTYm+BikRSGQ+H29f20vtU6RIqkrNcTQqDyU7DiEDT7YVOR3XGGZVrOND664KM8vv9xNrdutjuOEEIIYTsp9ia4WCSJN+DqvxOneRBApnEKIQBQc7pgVie8WYxu9tkdZ1huPPVGSvwl3PnanZiWaXccIYQQwlZS7E1wPdsuDNyJU6Eok5E9IcRRK1ogJwXrytGxvntzZjq/y8/nzvwcO9p38PDeh+2OI4QQQthKir0JzExbJGPmwJ0403VMMUrxKM84JxNCZCrltuDiRog7egq+LFy/d+mMS1lcspj/3vTfdCW77I4jhBBC2EaKvQnsRM1ZoKcTp6zXE0K8lypOwLkt0BiAN4vtjjNkSim+vPTLdCQ6uHfLvXbHEUIIIWwjxd4EdqJtF1I6SYvVLOv1hBD9UvM6YU5Hz/q9gwG74wzZvKJ5XDv7Wn6z8zd4ymX2ghBCiMlJir0JLBZJoRR4/H2LvWazAY1FhRR7QoiBrGiBonjPdgzh7Nt/7zOLPoPf5afsQ2Vore2OI4QQQow7KfYmsHgkiSfgwjD6duJsMOsAqHDKNE4hRP+UU/es39PAUxW4HG67Iw1JgbeAm8+4meCCIAe6DtgdRwghhBh3UuxNYLFI6gTr9Q7ixEmJUT7OqYQQ2UTlpWBVMxz28rfnfw7bBsiMnrV4Q719eOGHiTfE2dC0gbSVtim8EEIIYY/sm5cjBsUyLRLRNEVl/a+1aTTrKHVU4lTyLSCEODE1PYJeeoQlrIYtrbCoffxDWHDP5nuGdeoX//aLeL/k5a3Db3Hm1DNHOZgQQgiRuWRkb4KKR3s+wfYOtO2CWSfNWYQQg7eojTf3rofXpqDrsqthS3RHlJl5M9nUsolIMmJ3HCGEEGLcSLE3Qf1l24W+a2y6rQjt1hHZdkEIMWhKwf97/rtQlOhp2NKRXev3zik/B43mleZX7I4ihBBCjBsp9iaoWCQJgDfQd5pmo1kPQIVj+nhGEkJkuWQ6Dpc0gKHhzxXoRPb8Csl153JGyRns7dhLU6TJ7jhCCCHEuMie39RiSGKRFB6/E8PR9z9xo1kLICN7QoghUzlpWNMIYTesK0dbdicavEUliwi6grzU+BJWNgUXQgghhkmKvQkqfpJOnF58FBlTxjmVEGIiUGUxOLcF6oOwMXt+jrgMF8vLl9MWb2NH2w674wghhBBjToq9CUhbmng0hXeAYq/BrKPCWYVSffffE0KIwVDzO2B+CLYUoffk2h1n0GbmzaQ8UM5rh14jno7bHUcIIYQYU1LsTUCJ7jTaot+RPa01jeZBKqQTpxBipN7XAmXd8FwputVrd5pBUUpxbsW5JM0krx963e44QgghxJiSTdYmoFhvJ05vTt9ueZ06RFSHZdsFISap22+/fVTP96scPlH0DVy/L+D+9n+jwzo8ouuPhyJfEQuKFrC9bTvzi+ZT5CuyO5IQQggxJqTYm4COduL0BfqO7DWYdYA0ZxFisrrx+luGfe4dj9zW7/mpaBN5mxbyD9V3cWTxNrTL7PfcTLKkdAl7OvbwUuNLXHXKVTKtXQghxIQk0zgnoHgkhcvrwOHqpxNn+iCATOMUQoyadCBG+8JdOGNeCrfNASvzCyev08vS0qU0RZvY37nf7jhCCCHEmJBibwKKnbATZy25Kp8cI2+cUwkhJrJkQRcdc/fh6cgjf+cpoO1OdHLzi+ZT5C1iQ9MGUlbK7jhCCCHEqJNib4JRqJNuuyDr9YQQYyFWeoSuGQfxt0wh58A0u+OclKEMzq04l0gqwpbWLXbHEUIIIUadFHsTTH5wCpap8Qb7NmextEmTeVDW6wkhxkykupFoWQs5dZX4mjN/D77yYDk1+TVsbt1MOBm2O44QQggxqqRBywRTWtAzaufL6Tuyd9hqJUmSCucYjewpgzuuH0ETBiWfPQiR9RR0zj6AI+4hf9dMTE+SZGGn3alO6Jyyc6jtrGVD0wbWTF9jdxwhhBBi1EixN8GU5veM2vW3oXqjWQuMYXMWbXHF9+8b9umPffbGUQwjhLCNoQkt3E3xpgUUbpvNkcXb7U50QkF3kMVTF7Px0Ebqw/VMy8n8KahCCCHEYMhQygRTWlCN023gcjv6HGs0ezpxljvkjYwQYmxpp0nbaTvRDouit+eS5y+2O9IJnT7ldPLcebzY8CJpK213HCGEEGJUSLE3wZTmVw3YnKXBrGOKUYpX+cY5lRBiMrK8SdpO34FKO/j0ZXeiE5n7K8dpOFlRuYLOZCebWzfbHUcIIYQYFTKNcwLRWlNaUN3vFE6AxnSdNGeZgByKYa+VdGT+dmgiy6WD3bQv3MXUTbPhyST68gaUIzP3ZZiWM42a/Bo2tW5iVsEs8j35dkcSQgghRkSKvQkkFk4R8Obi66cTZ0qnaLGaWOw+x4ZkYiyZGtbdfcmwzl39mbWjnEaIvpKFXfz6ue9ww4W3wvoy9IVNqAz9oGF5+XIOdh3kxYYXuWLmFahMDSqEEEIMQubOqRFD1t4cBfpvznLIbMDCkpE9IYQt3tj7DJzdCntz4dXM3ZIh4AqwtGwpDZEG9nbstTuOEEIIMSJS7E0god5ir79tFxrMOmAMO3EKIcTJnNEOC9vhrSL0WwV2pxnQgqIFTPFNYUPTBhLphN1xhBBCiGGTYm8CCTVHiSWjuDx9O3E2mHU4cFLqqLAhmRBC0DN1c3krzOyCV6ai9+bYHalfhjI4v/J8YukYrzS/YnccIYQQYtik2JtA2g91cyhU1+8ak0azjjJHJU4lyzSFEPZRBrCqGcq64dkydKPf7kj9muKfwmlTTmNH+w78czIzoxBCCHEy8s5/Agk1R2npONjvsYZ0LXNcC8c50RApY9hdJXveQQohsoFyavQlDfBINTxZgb6mDlWYtDtWH0tKl3Cg8wAVH6sgYSbwODx2RxJCCCGGRIq9CSIeTdHdlaQ5VNfnWNSKENJtmb9eT1tc8f37hnXqY5+9cZTDCCHGkvJY6Mvq4eFqeHwa+q/qUMHM2szcZbg4r/I8Hks+xk/e+gm3LL7F7khCCCHEkMhwyAQROtQNQEuo78je0eYs0xzTxzOSEEKckMpJw+UNkDJ6Cr4M3HR9Ws40Qi+F+MW2X7CrfZfdcYQQQoghybzfrGJYQod6OnH2N7LXaNYCUOHM8JE9IcSko4oSsKYROt2wthKdzrx97Q795hC5nly+tuFrpK3MGn0UQgghTkSmcU4QoeYoDpdBe6Slz7EGsw6/ClCgimxIJoQQPW6//fYBjy3wnMMHrFvY/sMQD3X+N6DHL9hJmFGTW8++lc8//3ke2P4AN54q08aFEEJkByn2Joj25m7yp/rR2upzrMGspcJR3W+XTiGEGC83Xn/iNW+dB2tZsG8Z1TXVdNXUwjE/su54ZJjNm0bJmulreLL2SX605UesrFxJTUGNrXmEEEKIwZBpnBNE6FCUwrJAn8e11jSmD8p6PSFExotOayZS2USwoYxAfZndcfr417P/laAryG0v3ybTOYUQQmQFKfYmgFTCJNwWp6C0715QbVYrcWKZ34lTCCEUdNXUESs5Qt6+6fhaMmvqeZGviFuX3cr2tu08sP0Bu+MIIYQQJyXF3gTQ0dLTibO/kb2jnTgrpdgTQmQDBaG5e0nkd5K/owZ3KNfuRMe5ZPolXFR9ET/a8iP2hvbaHUcIIYQ4ISn2JoD25p5OnAWlfYu9xt5iTzpxCiGyhkPTvnAXaX+cwq1zcEb6zlqw09HpnLe+dCspM2V3HCGEEGJAUuxNAKHmKIahyCvx9TlWb9ZSZJTgU5n1ZkkIIU5Eu0zaTtuBdlgUbp1D0Jtvd6R3FfmK+Nryr7GjfQf3vnWv3XGEEEKIAY1ZN06l1P3AFUCr1nph72OFwP8B04Fa4INa65DqaRP5X8BlQDdwg9Z601hlm2jam6PklfhwOPvW7o1mnUzhFGNHGdxx/TC7JCr5rEmcmOVN0n7aToo2LeCTa76OTiuUMzO2ZLiw6kL+quav+Pm2n3Ne5XmcUXKG3ZGEEEKIPsZy64UHgB8CvzrmsS8Dz2it71JKfbn36y8BlwKzem9nA/f2/ikGIXSom8LyvlM4UzrFIbORRS75pxRjRFtc8f37hnXqY5+VvcrEyaVyonTM28tMawE834le1Uym7CLzpaVfYuOhjXzlxa/w4FUPEnD1/TkshBBC2GnMij2t9QtKqenvefhqYGXv/V8Cz9FT7F0N/EprrYFXlVL5SqkyrXXzWOWbKMy0RefhGKcsntLn2CGzAQtLOnFmupGMjvWeL8REFi9p59Ff/Jyr+ATrNz/Ji9GH7Y4EQMAV4I5z7+BjT36Mb7/+bW5fPvCm8UIIIYQdxntT9anHFHCHgKm99yuA+mOe19D7mBR7J9HR2o22dL+dOOvNWgAqZY+9zDaC0TGQETIxOTy1+X9ZfeVVXMAHOf3s04mXtA/63LHckH3x1MV8fOHHuW/rfZxXcR4XVl84Zq8lhBBCDNV4F3vv0lprpdSQF18opT4JfBKgqqpq1HNlm1Bzz7YLA3XidOBkqqN8vGMJMSgjGdF0ZMhUPjF+OubuwxnzkL+jhjbvdlK5UbsjAfDp0z/NK02v8G8b/o15RfMoD8rPXCGEEJlhvIu9lqPTM5VSZUBr7+ONwLRjnlfZ+1gfWuufAj8FOOusszJjpb6NQoeioCC/nw3VG8w6yh3TcCrbanohTmjd3ZcM+9zVn1k7iklEVjA07afuovjNUyncOpfDZ27F8ibtToXL4eI7532H6x67ji++8EV+cckvcBkuu2MJIYQQ4771wqPAR3vvfxT44zGPf0T1WAZ0ynq9wQk1R8kt8uJyO/oca0zXyXo9IcSEYrnTtJ+6E2UaFG6dgzIzY83qtNxp/Ps5/85bh9/ih5t/aHccIYQQAhjDYk8p9RvgFWCOUqpBKfUJ4C7gIqXUHmB179cATwD7gb2On1L5AAAgAElEQVTAz4BPj1Wuiaa9ubvfKZwRK0xIt8m2C0KICScdjBFasAdXJED+OzWQIXM8LplxCdfOvpb7t93Py40v2x1HCCGEGNNunH8zwKE+q9d7u3DePFZZJirL0nS0dDNtfmGfY41mHcCkLva0BlP3LOxSgFK6909bYwkhRkGiqIOumjry9k4nVVdBZHq/M//H3ZeWfIktrVu49aVbefDKB5ni79spWQghhBgvspgri3UdiWGmLQoGWK8HUOmcPs6pxpelIZ5yETztYt5oqKAz7iWWcpJMO0mYDizdd/Da5TDxONK4HSYeZ5qAO0nu2ddSG8on4E4SdCfxOtNSFAqR4aKVzbjCAXIOTCOVGyFR2Gl3JLxOL989/7tc/9j1fOXFr/CTi36Cw+g7zV4IIYQYD1LsZbHQoZ5OnP1tu9Bo1uFXQfJV31G/bJdMOwgnPHQlPESTbjSKoktvYUdjjNSRetLhw1jxCFYsjJWIgmWBYYBSKIcTwxPE8AYxfDk4/Pk480ooWHkDL9f+5TUMZb1b+OV4Er23nvsBdxKHkSHzxoSYzBR0ztnfO51zFkfOehszAxq2nJJ/CreefStf3fBVfrb1Z3zq9E/ZHUkIIcQkJcVeFgs197QdL+in2Gswa6l0VKMmyPCU1tCV8HAkGqA75QbA7UhT6O8m6E6y4a7P86nP/z2qCMAFFPTeTnhVIASE+OHt3+YffvoNokk3kaSbaNJFNOkhknBzOBogbf3lk3mF7ikEjxaAS66hoTOXHHcCvzuFy2GNyb+BEKIv7bBoX7iLKW+cSsG2ORxZtA0c9n8Yc03NNbx26DXufetezpp6FmeVnmV3JCGEEJOQFHtZLNQcJZDnxuM7/j+jpS0a0wdZ7rnApmSjx7IUbTEfbdEAKcuBy5GmNBgm1xvH4zTffV66s2VE0y51OsmPPvb5AY8b/jxcBWU488txFpQRKSijraAcV0E5hatu5Pn9x2ROdGNG2kiH2zEjbZjhI6TDbVjdHT0jjvEIViKKFY/2jDzaSRkj2utOiExg+uOE5u+laOtc8vbMoHPu/pOfNMaUUvzbsn9j25FtfOnFL/HglQ9S4D3ZB1BCCCHE6JJiL4u1N0f7HdVrsw4TJ5bV2y4cHclr7solZTkIuJKU5XaR60mMzVo6bXHF9+8bRs4If771Jn78jffTaubRYQZoN4N05AcImVPpMGfSafoxGXjNjpXoZmerp6eBjNIY8O59BRi99+EvjWaOvV+w6kb+r6Mah7IwsDDQvfc1Ru+fHpXCa6TwqiTe3vselcQRKODi27/d+xpD/2d77LM3Dv0kIcZAojhEuLqBnLpKUrkRustbT37SGAu4AnznvO/woSc+xG0v38bdq+7GUJmxVYQQQojJQYq9LKUtTfuhbuYvL+tzrNGsBaDSMX18Q42SeNpBc1cukaQHrzPFtPwOAu6U3bH6pRRY8TAzPS3MpKXf51gawpafLtNHt/YQs9x0W56em/bwP880U3TexVhaobXq+RPevZ8GrN6uohrg6PHe+8HTLuK5qBtLG1gorCHsqFJ584d5p7VnaqrTYeI0LFyGhcth4nKYuHtvHmcaI9NmBI9kVFLecE9I4Rn1uMJB8nbPIBXsJpUbsTsSFy++mO5Z3aT+LkXpVaUcfuzwkM6vrKqkvq5+jNIJIYSY6KTYy1LhUJx0wqSwvL/1ej2dOCucVeMda8SORP0cCueglKYsp4sif3dWdMV0W0k8VhK3lcJlpXr+1Gkc2sShTZzaxNAWCo2BRume0TmN4urKdgocW0jgIqHcxJWbuPIQVn7CKkCXCtCpArQZebQZ+SSU+7jXfuyzN7Lu7kve/VrrnuuaKCxtYGKQ0C4Slou4dhO3XMR1z/27/nc3C679KGnLePeWMB1Eku73dDLVvV1MTbyuFD5nGp8rRc/4ok2GORoLMiI5YSkIzd/Tu35vNkfOehvLnbY1UsPBBn74yA955uAzcC184oufYFrOtEGff/Mi2ZVICCHE8Emxl6XamwZozmLAL567B/9MPytvOHXoFx7hGq4XH3x2eCcaDoqv/DzN4VxyPHEqc7twZlijkwKriyrzEJVWC5VmKxXWYaZa7RT+aw7e9jf6PN8CTOXEVAZp5cBSRydZKqzekSWlNS4DfDpBno7gJYlHJ/HpBH4S/eYIKx+tRiHNRjHNRhFTznCRl+oi4vBjGk6U4t2iEtXzb+gnSX8zSSNvrWXKDdf2efzoHoUp00HSdBBPO0n03tqiAXRvkTftn37DgXYXAXeSgDuJz5XKvBFAMaloV5rQwt0Ub1pIwTuzaDtth92RUEpxfuX5tMXbeLruaa6dfS257ly7YwkhhJgEpNjLUu29nTj7bLtgwbRlVfhVkGt/+5EhX/eO628b0WjJRy6tGfJ5EcvL67EautIeYrVb6Gh4B7snLfl0nPnpA8xO1zHLbOAUs55CHX73eBQvjY4p7HNU8uvXGrn64rnEDQ9Jw0XKcJJULkzlGNQO7qsfWMsV3/+nPo87dZqg7iZXd5NnhSnSnRRbHRRbnZRY7Uw3m1iW2sp1V/ugcxsAMcNDxOkn7AjQ5cqhyxkkZbiG/PdXCpxK4zTS+Fxp8o4pPC0NibSTWMrFzs0b8C1ZTUskp+e83k6lOZ4EQU8Cj8PMipFZMbGkcqJ0zN5Pwc4acg4MfhRtLLkcLtZMX8NDux/iqdqnuKbmGpyG/AoWQggxtuQ3TZYKNUfx57nxBo5/I6+ciogOU2pU2pRsaNrSQV6LzcZQmtbffZWbbroaTh16wXj3yyPL4VCwMLWXM9M7WZjeyyyzAQcWJopGo4TNzjnsc1ZSa5TR4JhKSOW8W8g9tu4Fzry679rJkUorJx0qlw5ywVHa73MMbbLtPz7FQ7ctIZjuJpjuJseMUpwMoWI9z+k2PHS6cgm5culw5REzPIMqQgdiKPC5eorA9qfuYfmli0hbiu7ebSsiCQ/N4VwIg8uRJs+TINcbx+9KSeEnxk2s7DDuzhyCByuYU7HY7jgA5HvyWVW1irW1a3mp8SVWTltpdyQhhBATnBR7Waq9KdrvZuqecg8aTa6Rb0OqoTmczmFjbDZeleQc/y5+XvcWcPW4vb5Lp5mnaznV2sfXPh+kKPoj0hjsdlTxoGcVW5017HJWE1eeccs0VJZysD+kaXMX0uYufPdxh2WSY0bITUXIS4cpSoYoS/Q0hogbbtpdebS7Cyj0jU715TQ0ud4Eud4EECZpGkR6N75v6/ZzpDuA0zDJ88Yp8MXwuexdRyUmh65Ztbg7c/jIBV/mO1+/lW7dZXckZuTNYHHJYja1bqLEX8L8ovl2RxJCCDGBSbGXhU7UidNb6QUgV+WNd6whaU3nsTE2i4AR5xzfLrzGOHXb1Joq3cIyvZ3F1i78JIni4aE9aRrO+SSbXXMyurgbLNNw0GHk0eHq/T7QmoAZIz/VRUGqkynJEOWJwxz+QpDd4R/wqutUNrhOo9kxZVRe3+2wKPTHKPTHMC1FV8JDV9xLe7eftu4AXmeKnLOuJmwqchzxUXlNId5LOyxCC/aQ/8pc/nHaXbSftnNYPYXueGR096JcUrqE1u5WXmx8kWJfMSX+klG9vhBCCHGUFHtZ6ESdOD2VHgwMAirHhmSD05zK5414DblGjGW+XXiMsR/lceo0S/QOzje3UE4bSRy8pWbxmjGfvaqC/3rkm1xx/mljnmNMKIPVn1k7pFMMBUvKHaypcfDhCzU3xB/nhvjjHDDK2OA+jVdcp1JnlI1ouudRDkNT4ItT4IuTthSdcS+hmI/CC2/i880mZ/n3sCqwlZnuFpnmKUZdOtjNH165l+tX/DOB+jKiVc12R8JQBqurV/Pg7gd5svZJrp19LT6nb4An9zR4GS7ZukEIISY3Kfay0ICdOOkZ2Quq3IzduLctHeTNeA35RjfL/LtwKXNMX8+n45xrvc351hZy6aaeKfzWcSGb1OwJMYIHgLb4zDe+OqxTv37b19l0zWeZYoU4J/k2y1Nv8zfxp/hQ/EkajSlscJ3Ki65FHHBWjEpUp6Ep8sco8sd48ptf58O3fIoN3XN5tXsu1a5WLgpuYYl/L06VWZ1YRXZ76Z0/cfX7/47c/VUk87tI5UbtjoTP6WPN9DU8svcR1tWt4/KZl/f/c9uCezbfM+zXka0bhBBicpNiLwsN2IkT8E7zkqsyc71exPKyMTYbv5Hg7DEu9Fw6xQXWJlZbb+IlyQ5VzS+Ns9ijKkdltGq0DXvLilFy2CjgUe/5POo9n3yri2WpbSxPvc37E89xXeJZ9hvlrPMs5XnXYrqM4Ki8ZurIQT5U8AIfyHuFV7tn80zkNO4LXcwfus7h4uAWVgTeGb/pvWLC65izjyldp1OwfTaHl7yNdo7tB02DUeIvYUXFCp5reI6NhzayrGyZ3ZGEEEJMMFLsZaGBOnF2JjpxFbjINTJvvV7CcvJq92yU0pzt2417jAo9pS3O1u9wmfkK+UR5S53Cnx3LaFKjsxZtrAxny4qjRtqJ9L06jFzWepaz1rOcHCvKitRmVidf55OxR/h47FFed81nnXspbzrnjcrreY0UK4PbOT+wnbfj1awNL+a3nSt4tGsJF+dsYXXwLXxS9IkR0i6Tjvl7KNqygLxdM+mYv2dY6/dG27yiebR0t7C5dTNTfFM4Jf8UuyMJIYSYQKTYy0IDdeLcHdoNkHEje6ZWbIzNIq7dLPfvIGD0v1n4SE2zWvgb82kqOcIBVcYvjMs5YJQP+ny7R9cyUdgI8ITnXJ7wnEu12cSFyTdYmXyDc1Lb6FBB7r/Yw26zlUbHyBtMKAWn++o43VfHvsRUHg+fxSNdy3g6cgZrgptZFXxbij4xIsn8MOHp9eQeqCJR2EGs7LDdkQBYUbGC9ng7z9Y/S74nnyJfkd2RhBBCTBBS7GWZE3Xi3BPaA0BOBo3saQ1vxWcQsnI4y7uHQsfor5Vx6TTfWu3hc+Zv6SLA/Y7L2KJmDXm65nBH10Z7ZC1T1TnKud93Fb/0Xs7i9E5WJzfyj0vDuMN3sdk5mz95VvCmcx7WKKwXPcXTwi2ex6lNlvDHrqX8oescnoqcwRU5r3NBcJus6RPDFqluxBPKI2/3DFK5EdKBmN2RcBgO1kxfw4O7H2Rt7VqunXUtHucEWVMshBDCVlLsZZkTdeLc07GHdCSN1zNAVzcb1KeLaUgXM8fdQLkrNOrXn2E18SHzKUre5+FltYA/OlaMf+OVYXTDzGamcvC6awGvuxaw8c6b+O5tH+TSxAa+Gv05LUYhj7uXs859NmGj7/foUE13t/JPxY9xIFnCHzqX8dvO83gmcjrvz3tlFP4mYlJSEJq/hymvn07B9lkcPnMrOLTdqQi4AqyZvoY/7vsjTx98mstmXJaxjbaEEEJkDyn2ssyJOnHuDu0m0ZBAFWfAQhQgbHrZGq+m2NHJbHfTqF5bac0q602usF4mRA6rfhll4Y2rR/U1Bm0E3TAB7r7t66MYZny1RjW/817EQ55VnJ3axuXJl/h4/DE+FF/L8+7FPO4+l/3OyhG/zgx3K/9S/CjbE1X8vnM5P2m/hNK/m0F30oXfLVM7xdBYnhQdc/dStHUeufur6JpVZ3ckAEoDpZxbcS4vNLwgDVuEEEKMCin2ssxAnTgtbbE3tJd4QxzOsCPZ8UyteCNeg0NZLPbuH9UGmD4d58PmU5yq97NJzeK3jtWsr/0WC0fvJcQQmcrBBvfpbHCfTpXZzOWJl1mVfIOLkxvZ4pzFQ55VbHHOHlEnVKVgofcg8z31vNI9h5+GF7GvvYgCXzdTgxFcDpnaKQYvUdxBtOIQwYZy4kUhkoVddkcCYEHRAo50H3m3YYsQQggxEjJHJMsM1ImzIdxAd7qb+MG4TcmOty1RTdjys9i7f1Tb51fqVr6Y/l/m6Vp+b6zkAcdlE2e/vAnioKOMe/3XckPu17jfewXTzBb+I/oTfhD5HuclN2HokXViNZTmfYGdNN33KaYEInTEfOw+UszhqB9t/2w8kUW6Tqkj5Y9RsLMGlXLYHedd51acy1T/VJ6tfxZPpfx8E0IIMXwyspdlBurEuSu0C4DYQfubDTSlCqhLlVDjbqLE2Tlq151n1fJx83GiePkvx3XUGX2b1IjMETV8POxdxZ8857Ey+SbvT6znC92/5iPGEzzsOZ9n3SNf61iaE6HAF6M5nMOhcC4dMR8VuV0ytVMMinZYdMzbQ/GmheTtnknHgj12RwKOb9hSfUs18XQcr9NrdywhhBBZSIq9LHKiTpw723fiUA4SjWOzrcFgJbWTtxPTyTOizHU3jtp1z7a2c725jmaK+bHzGrrUyJt/iPGRVk7Wec7mGfcSlqTf4QPxZ/lU7GGu/Sc/rwXfx/PGGcManT261tHjNKnO76Ar4aGpK5d97YUU+rspDUZwGDLUJ04slRslPL2hZzuG4nZiU9vsjgT8pWHLQ/GHWFe3jstmSsMWIYQQQyfFXhY5USfOXe27mJ47nbdSb9mQ7C+2xatIaQfn+PZjqFF4o601l1ivcZn1KjtUFfc7riCh3CO/rnjXeO0vqJXBRtdCNroWMi99gHN3fY+r5rzCBdYm1huLh130Qc96vjxvgqD7CC2RIG3dfsJxLxV5neR4kqP8NxETTaSqEW9bAXm7Z5LIC2N5M+N7pjRQSvP/NGN8zGBj80aWlUvDFiGEEEMjxV4WOVEnzl2hXSwuWTzekY7jnbGYhnQxs92N5DlGYTqp1vyV9QIXWJt5Tc3jN47VWCpz1tVMFMPdXxCGv8fgDucMvvPbGN/6j49yifkql1s9Rd9zxiKeMxYNu+hzGJry3DD53jgNnXnUhgop8HVTlhOWUT4xMANC8/Yy5Y3TKNhRQ9sZ70BmNDUm9HyIiz5/EZsPb6bYX0xN/vD/fxVCCDH5SLGXRQbqxNmZ6ORQ9BBzC+faEQsA01IUrbmZoBFj1mhss6A1V1svcoG1meeMM/iDcf6IOjmKzNSgSrjPeRWVupVLzFe5zHqVldbmERd9fneKmuKeUb4j0QCRhIfKvE6CMsonBmD643TV1JK/6xQCDWVEpzX/5aABysafP+eWn0t7rJ319esp8BRQ5CuyLYsQQojsIsVeFgk19d+Jc1d7T3OWOQVz7IgFQEskiCPXx+menThGOn1Ta660XuZCaxMvGKdLoTcJHFv0rTFfe7foe8Y4k+eMRaSU6+QXeQ9DQVlOhFxPgobOPA6ECin0RykNRsbgbyAmgu6yVrxHCsjdX0WioIN0sHeGggX3bL5nWNe8edHNI87lMBxcPP1iHtr9EH+u/TPXzrpWGrYIIYQYFFntnUXamqIUVQT7PL6zfScAswtnj3ckALqTrp41Upsep8g5wjfSWnO59QoXWW/wknEqDxorpdCbRBpUCT93Xsm3nX/LPlXBldYGvpp+gPeZbw97y4aAO8Ws4iMU+aO0d/vZ21aEp3LBKCcXE4KCjrn7sRwmBTtmgZU5P3sCrgAXT7+YaCrKurp1WFr2lRRCCHFyUuxlCcvStDdHKeqvOUtoF8W+Yop9xeOeS2toCufgNCw6XvjViK+30trMGmsjG9RCfm+skkJvkmpQJfzMeRXfd3yQIyqPv7ae5db0/7DI2o0axmZ6hoLy3DAzC9sBmPq3d7K5qQxLlvGJ97DcKTrm7sMVCZBzYJrdcY5TGihlRcUK6iP1vNb8mt1xhBBCZAGZxpklOlu7MVNWvyN7u9p3MafQnimcHXEvsZSbyrwO9iVH1pTlmrlOrrFeYIuq4f8cq9BS6E16B4xy/ktdxwJ9gCvNl/mY+QQHVQl/Mt43rOsF3ClmFbXx6lObeUddQks4yLnT64a+lk8Z3HH9bcPKgLTPz3iJ4hDRshaCB8tJFIXsjnOc+UXzORI7wpbDW5jim0JNgTRsEUIIMTAp9rJEW2NPc5b3FnspM8W+zn28r2J4b35HwrQUh8I5+FxJ8r3xEV2r2mrmrvf7qFOl/I/jErS8IZ4Umg83DO55uHmWlZzrOshf+7Zzs36Y2R/286jZSK2jYkivaRia9id/yKXXzOG1+iqe2DmHpVX1TC/oGPxFtMUV379vSK971GOfvXFY54nx1VVTiyeUR/6OGrwuv91xjvO+8vfRFm9jff168r35tszqEEIIkR2k2MsSbU0RlIKC0uPfdOzv3E/aStvSifNwNEDaclCd3zGi2ZZFupNPmo/SHNb8vOQqUkq+LSeL8pz8IT1/LwXcpRdyrrWD88s2cGH4ezztXsqvvZfSYeQO6VrVBZ0UBXbycm01L9dO51BXG2dVNuJ0yFooAdppEZq/h+JNC3n/8k/bHec4DsPBmuo1PLjnQdbWrpWGLUIIIQYkwydZor0xSl6JH6f7+H3mjjZnGe9pnMm0gyPRAPneGH53atjX8eoEn0o/goHFZf/bTURl1ifoIvOYOHheL6Tm7gh/9JzHquQb/KTrTq6Lr8OthzYdM+hOcdGsvSyYeoh97YX8edds2rt9Y5RcZJtUXoRIdSPL516KPtB3Cr2d/C4/a6rXEE1FebruaWnYIoQQol9S7GWJtsbIgM1ZvA4v1TnV45qnOZyDUprSnPDwL6I1HzKfopgOfu64kt1t8mZFDF5HHO73Xc3NOV9ki2s2H4k/wb1d3+K85KaezkGDZCg4o/wQF9bsI2UZPLl7FrsOFw/lEmICC09voP7wHni+FN3tOPkJ42hqYCrnVZxHQ6SBV5tftTuOEEKIDCTz5bJAKmHSeSTG7LNL+xzb1b6LWQWzcBjj9yYkmnTRlfAyNRjGNYIpb6usTZyu9/GwcR57jcpRTCgmk2bHFO4MfIyFqb3cGP8jX+j+NVc6XuRnvmvY7Rz8hyClOREum7uLV+uqeKOhktZIgGVV9SP6HhcTgKH55fo7+fJ1P2bfz7r4v47v2p3oOPOK5nE4dpi3Dr/FFN8UZhXMsjuSEEKIDCLFXhZob46ChqKK40f2tNbsCu1iddXqccuiNRwK5+A0TIoD0WFfp8Zq4ErrJbaoGtYbi0YxoZistrlq+BfnZ7kg+QZ/F3+C/4z8F0+5l/Ir7+V0GjmDuobXaXL+zAO801rCW01ldMR8rJhRS75vZA2IRHY7FKojOquJOXvP5JbzvkF3eeugz73jkWF2bR2Cow1bnqt/jgJvgTRsEUII8S6ZxpkF2hp7NiovKj9+zUhLdwudic5xbc4STrrpTrkpCUYwhtmUJVdHuMF8giPk87+Oi2QvPTFqLGXwjGcp/5D7ZR70XMCq5Bv8uOtOrki8OOhN2ZWCBVNbubBmL0nTwdrdszjQXjDGyUWmi1Y2k8jvJHfvdBwxj91xjnO0YYvH6WFt7VriaflwQgghRA8p9rJAe2MUp8sgd8rxjSN2te8CGLdiT2toCefgdqQp8A1vTz2lLW4w/4yHJPc7LyeuMutNk5gYYsrLL31X8pmcL7DHWcXfxx7mB+HvsSC9b9DXmJoT5bK5uyj0xdhQV83G+kpMSz6YmLQUdMzbC0qTv6MGMmxNp9/lZ830NXSnunmq7inMQX64IYQQYmKTaZxZoK0pQmF5AOM9Q2lHO3GO1xqNzriXeNpFZV7HsEf1LrTepEY38mvHxTQrmWo0Ebz44LN2RxhQg2MqXw38PctSW7kp9kfuitzDetdi3sxxDH5TdGWQf/5H2XP2B9j2djuH/3jn2IYWGcv0JumcdYCCHbMIHiwnUt1kd6TjTPVP5bzK81hfv54NjRtYUbnC7khCCCFsJsVeFmhrjFB9at/CaFdoF1U5VQRcfbt0jjatoSUSxOtMDXsD9UrdymXWK2xWs9io5o1yQmGXj1xaM6zz7n55lIMMRCledZ/GZtdcro3/f/buOzyu6kz8+PdOb5JmRm0kjaoly7YsN9wNNtiO6cVAKMkCGyBLAoHd7CZsNrCbQEh+JCwhiQNJCBBIFhIIoZsOBjfcLVdZsnrvM5qm6ff3h0wRbuozks/nefzYvrr3nlf9vnPOed8PuDKwgYo7DPw96TJe155DeJB9HXv9DpqkQnJvf5KW5388tjELcasvvQtdl4WE2mz8VifhBF+sQxpgmnUaDr+Dss4yLDqx/FgQBOFMJ5ZxxjmfK0ifO3TCtgvl3eXj1l/P0acnGFGRbvIMa4udWg5zY/htvOh5XrlS7NMTxl1A0vCs/kLuSLibDXVhbva/zm/c/8vsUOWgrk/SBShM7katjJD21R/T7jGK9gxnIgmcxbVE1WEs5UUQib+fZYsyFpGbkMvm5s0Yp4/9i4GCIAhC/BLJXpzraTlWnCVrYHGW3kAvTZ4mZiTPGPMYosdm9QzqIAnawLDucWl0MzZ6eFa5Bp8kmlYLsdOmTOHyv/Vxn/FWlER4wPt7fuB9muSo87TXalURpiR34z30ER2eBOqdZrGP7wwkq8M4p1Wj9hpIrM2JdTjHUUgKVueuxqw1k31HNg2uhliHJAiCIMSIWMYZ57qb+9sbfDnZ+7Q4ywzr6Cd7X96DpbEVYii04dz3ES297UO+X3G0nnOjZXysmMMRxfg2fxcmt5HsF9ylnsE+VRFrAx9xjf995oYqeE53Pq9rzyEqnbxvpUKC7vW/pGjJfFrdCVR1J5NrdqJTh4cdizDxBJKdeDPbMDZm4E92ELS4Yh3SABqlhovyL+KZrmf4zoff4dmLniVBM7gWJIIgCMLkIZK9ONfd4kFnUqNPUA84Xt5TDsC05NGvxPnFPVhRWeID7yx0kptLlyQgSSd/WDjRHiytHOT6yPu0YeE1xdmjHqtwZhvpfsGQpOYF3Vf4SH0Wt/W9xK3+11gV3Mmjhq9Soco75T1SjD706hANTjNVPVbsiS7Rj+8M4yqsR+tIwlxeSOfCfciq+KqAmahNpOG3Deh+qOP7G7/PoysfRak4+QsZgiAIwuQjkr04193sJTnLiPSlPW6Huw9jM9qw6qxjOn5jKJk+WcssXd2wttldHN2KGTe/Ul5DaBzP7mIAACAASURBVJCFMARhPHx5VvAf5HNxqoGfFx/kf6O/4emmHH5cNQ1nWHPSexg1IQqTu2lwmmnsNeMLeclIcIstqWcIWRnFMb2KlL0zSTqah3P64Ft7jBdfhY97Ft/DfZ/cx8O7H+buBXfHOiRBEARhHImn7zgmR2V6Wr3MWJpx3NvKe8qZbh3bipZRGY4GMzErPKQpe4d8fV60leXRMjYpZlOnyByDCAVh+E48K1jEr+SFXBjdzg32PVxt7+QV5Tn91WO/kMF9cRZbrYxSYO2h1Z1At89IX0hNjtmJWhkd+3dCiLlQkgdPbhMJddn4kx3403piHdJxrp56NVXOKv5y+C8Umgu5sujKWIckCIIgjBNRoCWOubr7CAcix+3X84a81PXWMT15bJO95nAyPllHkaZlyDMVSjnCdZH36cXEG4plYxOgIIyBoKThVeU5PKT6Gh2SmX+KvMtdkRexyd0nvUaSIDPRTXaSk76QmqruZLxB9UnPFyYXd24zwQQ35oopKP0nnwmOpe/N/x5LM5fyk20/YWfbzliHIwiCIIwTkezFsU+Ls1izBpbOruipQEamJLlkzMaWj83qJSp82FSnr1L4Zauiu8ikmxeUKwlI8fnwIwin0iKl8mvlNTynXE2G3M1/hp/l0shm1HLopNeY9X6mJHejkGRqeqx0eQ2iPcOZQCHjKDkKMpgPF0Ecfs5VChUPrXiI7IRs/nXDv1LjrIl1SIIgCMI4EMleHOtq8oAE1oyByd6nxVnGchlnS9iKJ6pnqqZ5yLN6aXIPF0R3sFuayiFFwdgEKAjjQJYktilm8oDqJnZK0/hKdBc/DP+ZS6aefAW8Xh2mMLmbBG2AVnciTb1JROPw4V8YXRF9gN6ptWh7EzHV2WMdzgklahJ5bNVjaBQabv/gdrr6umIdkiAIgjDGRLIXx7oa3ZjTDGh0Ax8sD3cfJkWfQqohdUzG/XRWz6ToI0PlGPLFX41sIIiKl5QrxiQ+QRhvXknPc6o1/Er5VQJoeP16Az/0PkVq9MTfH0qFTK7ZSZrJjdOvo7o7mWBYVEGc7PpsXfjSO0mos6NxxmebA3uCnUdXPUqPv4c7PrgDX8gX65AEQRCEMSSSvTjW1eQhJdt03PGxLs7SGUnCFTUMa6/eHLmKYrmR9YqluCXj6S8QhAmkRpHFL1Rf4+73/MwNVfKY6+es9X+IUj6+5L4kQbrJS57FQTCi5Gh3Mq6AWNI82fVOrSWiC2A+XIQUis8EvySlhIeWP8SRniN8f+P3CUdFj0hBEITJSiR7cSrgC+Hu9pNiH5js+cN+apw1Y1qcpSpoQycFyVINraqcXgVXRDbSRApbFKVjFJ0gxFZUUvLQ1iC3J95NmaqIm/1v8Cv3w0wP157w/ARtkKLkbjTKCPUOC0nLrhf7+CYxWRXBUXIUZVCN+cjw+kCOhxXZK7hn0T1sbNrIfZ/chyy+KAVBECYlkezFqa4mDwAp2QOXAlU6KonIEWZYZ4zJuBpbIV2RJAo0bSikof3y/69ztFhx86LyPKKS+NISJrdOhZWfmm7hJ8abMcgBfuFZx52+v5EY9Rx3rkYVYUpyN2adH/PZX+ejmnwCYlnnpBVK9OCaUo++y8q5pfHb5uCa4mv49uxv80rVKzy862GR8AmCIExC4ok8TnU1Hkv2vjSzV97dX5xlRvLYJHuJC69ERZhcdceQrkuRnXx/qYad0jRqFFljEpsgxKMd6pncnng3L2rPY2VwF79z/5zVge1I8sA+ewoJ7Em99Lz3e1pdibxdMRVHny5GUQtjzWtvoy+lh7WLbkNuj9/P87dnf5vrp13PM4ef4cmDT8Y6HEEQBGGUiWQvTnU1udEnajAmaQccL+8px6w1YzPaRn1Md0CDoXgZeZoO1NLQGkKvjWwkFIFXlWePelyCEO8CkpZn9Jfyrwn/QaMijX/te56feR4jJ9I24DxJAveeN1hdVEUkquCdiqnU9lhiFLUwpiRwTqvC4e2E97KQ/fH561aSJH6w8AdcXHAxv97za16oeCHWIQmCIAij6OT1w4WY6mrykGo/vjjL4e7DTLdORxpq5ZRBKO9IhWiUAnX7kK6bFq2jVK7h7o0B/BcdH7MgTEabXvzwhMefYwZfz0zg/sJyfhV6iHX1BTxUO5W+6OfLNtNMXi6cVsGm2jy21ufS7TMwL6sZxeh/WwsxJKsjPPX+T7j76kdhQwbyBUNvZTMeFJKCnyz7Ce6gmwe2PYBWqeXywstjHZYgCIIwCkSyF4ci4Sg9LV5yVlsHHA9FQhx1HuXGGTeO+pj+kJKa7mQ8h95Dd/bgZxoUcpQrIpvoJIlfbXPxrYtGPTRBiEs3Xniq4htFPCQv4fLIJv49/zA35Xfyd+V5HFbks25L/xl6dZjVRVXsac6iojOVHp+ec/Lr0KtFZcTJpKGzApa2w2Yb7EmGs7pjHdIJqRVqHl7xMHd+eCf/veW/kSSJy6ZcFuuwBEEQhBESyV4c6mn1Eo3IpNgHFmepclYRjobHpBJnZVcqEVmBa8fLcPbNg75ukXyYTLp5UnkxoehfRz0uQZioPu3Ntz06g2siH/KtyKuURQt5KeHzqR2FBPPtzaQYvGxrzObNI8Wck19Hmsl70vsqJfjZdfcOKyZlHM4qnRFKnNCuh50pyCl+pNyTf35jSafS8ZuVv+HOD+/k3s33IiFx6ZRLYx2WIAiCMAIi2YtDnxVn+VKPvfKeY8VZRrkSZzgqUdmVQlZiL/U9TYO+TiMHuSiylRopg31S/JYYF4RYqlbY+YX0dc6L7uGC6HbK7zCxsXsn26xnfVa1Ns/qJEnvZ2NNPu8fLeQsezNTU7pOuOQvIsP76y4YViyr73x7JO+KMEySBPLyNnBo4YNM5CvrkMyhWId1QnqVnnUr13HnB3dyz+Z7AETCJwiCMIHF547xM1xXkxuVRkFSmmHA8UNdhzCpTdgT7KM6Xl2PhUBYxbS0ziFdtzK6hyR8vKI4h7jciCIIcSIiKXlfuYCfqW5gY32Yizs2cHvtn7H3tXx2jkXv58LiSjITXexqsrO1PodwVHxfTRaSWoY1zaCQ4R07cih+P7d6lZ51q9axwLaAezbfw/NHno91SIIgCMIwiWQvDnU1ekjOMqH4UrWGg90HKUkpQTGKPexkGY50pmLR95FuOr4/2Mkkyl5WRXexVyqiTpE5avEIwmTWIyVxyV/7eDbrCgwRH7fV/R+Xtb6LLuIH+vvxrSioZVZGK3UOC+9UTMXl157mrsJEISWGYHULODX9BVviuK2dXqXn0VWPssK+gge2P8Dv9/1e9OETBEGYgESyF2dkWaaryXNcM/VAJEBlTyWlKaWjOl6bO4Fev55pqR1Dmpy7KPIJSqK8rlw2qvEIwpngcOJUflVwC1ut81ng3Md3q59gdu8hkGUkCUpt7ZxbUIMvpObNiqlUdVnjOjEQBk+y+2BxB9Qkwo6UWIdzSjqVjl+e90sum3IZj5Y9ys93/pyoPLS2PIIgCEJsiWQvzri7/QT7wsc1Uz/Sc4SwHGZm8sxRHa+8IxWdKkSuxTnoa2xyN4vlQ2xSzKZLMo9qPIJwpggqtbyVvpLH8m/EoU7impb13NzwPCmB/mqNWUluLp5WQYrBx/bGHDbV5REIK09zV2FCmOWA6U7Ym4J8JCnW0ZySWqHmJ8t+wo0zbuTZ8mf5wcYf4A/7Yx2WIAiCMEiiQEuc6Wo6cXGWg10HAZiZMnrJXq9fS6s7kVkZrSgVg582uDiylSBq3lUsHLVYBOGMISmOq6b5rxJ8c56aB1fXc7vrCR7cEuT/bQoQiPSfn7hwLfI5N1DbYEObMys2cQujRpJAPrsNXGrYaENOCCFl+WId1kkpJAXfm/89kvXJ/Gr3r2jyNPHr835NqiE11qEJgiAIpyGSvTjT1ehGkiA56/hkL02fRroxfdTGOtKRikKKUpTcNehrcqJtzJareVOxGK+kH7VYBOGMIUe55JEnjjvcDNwZdXFz32v8aMUevnleFr/XX8VedTEAvpCTRqeF9Ose4B+9e7g8cQcqSSypm6gkJchrmuGVXHg3C3ltPZI5GOuwTkqSJG6eeTO5ibn816b/4vr117Nu5boxaQUkCIIgjB6R7MWZriYP5nQDas3A5VoHuw6O6qxeIKyktsdKvtWBTh0Z9HWXRrfgRs8GxbxRi0UQzjSbXvzwpG97nUyWWzT8ctoB7o/+gRfbMrmncgbtQR0oVKhMZt6ctYZyfzbfTH6XdFXvOEYujCZJG0W+sAleyoX1duQrGpCM4ViHdUqrclbxlwv/wnc+/A43vX0T9y+9nwvyh9cKRBAEQRh7ItmLM12NHmwFiQOOuYIu6lx1XDblslEb52hXMhFZwbTUwbdbmBptoFhu5CXFcgKSZtRiEYQzzY0Xnq4vZSGPyfNZHd3F5badXGzr4g3FMjYrSvnNfz/A98+N8IzjPO5rv5arkz7hXOMBFPFbyV84BSkxhHxRE7ye3Z/wXdaApIvvGdtiazF/vfivfHfDd/n+xu+zo20Hdy+4G51KF+vQBEEQhC8RyV4c8XtDuHv8zFyRNeD4oa5DwCD3651gP9CJzsn61hOEesp47OenOfdTsswl0S30kMBmRXzuGWrtHHxD+LG4XhBGU1hS8bZyMbsVxXw1soGvRjewUD7MJpuC+YZqCjTtPO1YybPOFezqm8I3LB+QqnLHOmxhGKQ0P/L5zfCmHd6yI1/S2N+XL46l6FN46oKnWLd3HX86+Cf2de7jf1f8L/lJ+bEOTRAEQfgCkezFka7G/ge1kxVnKUkpOf1NTrIf6It6/VoanBam5KhZuHjguW9899YTXlMqV5Mnt/OccjVhKT6/bDITRlYZdKTXC8JY6JQsPKZcyzy5kisjH7Pzm0ZaPLVUG7L5bsprbPLN4Hnn2fyo/XoxyzeBSXYf8uoWeC+rfw/fBU1IcV58Va1Q8+9n/Tvz0+dzz+Z7uPaNa7l7wd1cVXQV0lB6+QiCIAhjRrReiCMd9f3JXlrOwGWcB7oOkJeYR6Im8USXDVm314haGSZBGxjU+ZIc5eLIJ7RjYYc0Y1RiEARhCCSJPYpifqq6kd/tCmH3t7LEuZf0YBfLDYe4P/05CjVtPOtcwYOdV9EUssY6YmEYpAIPLG+DRhO8n4U8+O3UMbXcvpy/X/p3SlNKue+T+7jtvdto9bTGOixBEAQBkezFlY56F4kpOnQm9YDjo1mcxR9S4Q1pSDb4Bt1Efa58lEy6eUu5mKgkvmQEIVb6JB13vuVnZ1IpAUlDqbuS2a5y7FIX3015jVss79EeNnN/+7W81LuYoBznU0PCcaTpvbCsHWoT4L0slIr4XEnxZTajjT+u+SP3LrqXss4y1r62lr9X/l00YRcEQYixifFb5AzRUe8mPW/g7F27t53Ovs5RS/a6fQYkZCz6vkGdL8lRLohso4Vk9kpTRyUGYRKSZVSRCKpwGOkUW42iComQSkVUIV40GAm3OoFd5lnY/a0U+BpZ5CijzmBHYYhSqqvnhd5lrHfPZ7uviOvMm5mjqx30iztC7EmlDmRJhs02bl3zY+SIhKSM7z180N+P79pp13K2/Wx+tOVH3P/J/bxa9So/XPRDZiSLVSGCIAixIJK9ONHnDuLuPr44y6f79UpTSkc8RiQq4fDrMOv7UA2yifpZciU2HDylvBhZPC1OfrKMNhhE7w9g6POj9wfQ+wf+rQ2GUIXDaEIh1hZMwfK3l1GHwyjkwT+MRiWJrxcWwfoN+LUa/FoNfVoNXr0Ot0GPx6DDY9TTazQQVYrZqRORJYlGfSYdmmSKvHVM8TVgC3RyxFjALdYPWGY4wrPO5fy2+2Jmauu53rwp1iELQyDNdCJLUMoSeMeDvKYZSRX/CR9AlimLP675I69Wv8ojux/hujeu4+qpV3PX3Lsw68TeaEEQhPEkkr040dHQv18vPff4/XoqhYpia/GIx+jp0yPLCpINvkGdr5CjnB/ZTjMp7JNOVypeiGdSNIreH8DY14fR14fR58Pk6yPDlsFZ73/8hYQucMKkLSpJx5IyLQGthoBWg9tkYF9DI0WlMwipVATVKsIq1SlfFFBEo6jCEdThMIc3bWFhQQ76QBBdIICl143BH0D5hfGjgNtkwJFgwpFoosuSSJc5iV6TYSw+TBNSQKnlYGIxrUEHxZ4aznIdolWbitoY5EfprWzwlPKqayH/0349lpUmHvznB4n6PcMbTCzjHldSiZPnfvoc1y//d1if3V+0RTsxlkVKksQVhVewKmcVv9v3O54rf4536t7hltJb+Nq0r522TUN2bjZNDcOvkGzPsdNY3zjs6wVBECYLkezFiY46F0iQmpMw4PjBroNMtUxFq9SO6P6yDD0+AwZ1EL16cE17z5KPkI6DJ5SXiFm9OKaVJBJdbky+Y4ncZwld/x+Tz4ehz39cEheRJDIMBhShEG6jkY5kK306HX067YC/fXotAY0G+QRLL9e9+z53Lpg7rLjX/eNl7lo4e8AxSZYx9PlJ8PWR4O0jye3F4vZgdnnIbuv8LBEMqFTMsWej2neQ1tQU2lKTCavVJxrmjNGtsbDNMoc8XxO5fS2kBB1UGXP5iinCQkMlL/UuYdP8yzAvvIw0k4dkg2/IVTtPVq1XGDtbytdz/f3XwocZ8Eou8sWNSKb4brz+RQmaBO5ecDdXFl7JL3f/kkd2P8Kz5c9y++zbubzwclQn2ZPY1NDEo3sfHfa4d8y9Y9jXCoIgTCYi2YsTHfVuLOkGNPrPPyVROcqh7kNcXHDxiO/vDmgJRlSkm5yDOr9/Vm8HTaSyX5oy4vGF4ZGiUYx9fSR4fJi8PhK83v5/+3yfJXPfnloMr7414Dq/SolLq8Gt09BqNuGyWXHpNLi1Wnp1/ce9GjV//82fuPPmr8fovTueLEl4DXq8Bj1tKQPfpohGsfS6SXW6SHX0knDIw/T9h1EAEQlaE01UpZipTjbTaDadkfsCo5KSGmMu7dpUij01TPdUk+HvoMJUwDesH/LCukeZ+u1f0uZOpNtrJM3kwaLvE/v54pxU6EbWR+CdLHg5F/miJqTkwVVTRsGw2yAo1UoioeGXBP3i7FqhpZDHVj/GrrZdPLLnEX78yY956uBT3DzzZi6dcikapWbY4wiCIAgnJ5K9ONFR78I+zTLgWJ2rDk/IMyrFWbp9BlSKCEk6/6DOny+Xk4aTPyovRTwJjiFZxqZSkdXaTqLHi8nrJcHrO5bceTH5+o6bkfPptHiMBlwmI61pKby35RNmX7Qar1732Z+Q+uTf2hKQeOzPRBJVKOi2JNFtSeJIfja/efNd/uM/7yS920lGl4Osjm6WVzdxbnUTQZWKpvRk6jLTachIw689sx4kvSoDe5JKsAU6KfLWscC5j0Z9JpreevKtDjwBDW3uBJpdSXQeS/rMOr/4Vo9jUpYP+fIGWG+HV3OQV7cg5XhPf2GUYc+Q3TH3jlGfXZtvm8//Xfh/bGjcwB/2/4Eff/JjHtv3GDfNuImrp16NQS2WaAuCIIwmkezFAa8zgK83SNqX9+t1HgBgZvLIkr1AWIknqCXd5B7Uw9yns3qNpHFAKhjR2EI/VTiMpdeFudeN2fX5nySXm29PKYT3Pwb698Z5DXrcRgOtaal4jAbcJgNuo7H/30YDEdXAb9s/vr6eu/LssXi3Yi6kVtNkS6XJlspOQBMMkdnZTXZbJ7mtnRQ0txMF2lKsVGfbqLZn4NeNbEn0hCFJtOnS6NJYKPTWk9vXwuE7TPw5dJDt2plM0XTjDmhp95ho6jXT4QmTavRi1veJpuxxSkoOIK+th7ft8KYdeWEXzO0+7c/1++67b3wCHCRJkliZs5Lzss/jk5ZPeOLgEzy06yEeP/A4X5/+db427WuxDlEQBGHSEMleHGivcwEcl+yVdZaRoE6gwDyyhOvTdgvWQbZbmCdXkEqvmNUbJlUoRIrDSWq3g9QeB6ndDswu92czdFFJwm0y4kxMoNmWxisbNjLnhutwmYx4DfoT7o0TBieoUVOXZaMuy8YmWSbV0UteSwf5zW2cs/cwy8rKaUpLpi4xEXUwREgz+ff5hRVqjiQU0qpLQ9u+j3u9T7FNVcLjhrVIOisJ2gCugJYOj4lmVxLtHhMpBh9Wgw/lIKv2jgtJwc+uu3fY104WUkIY+Yp6+DgDdqRClxb5vFYk9ck/V7ded9ewxvrZK8P8eA+SJEkszVrK0qyllHWU8eSBJ3ms7DGePvg0tmtteEIeTGrTmMYgCIIw2YlkLw50NriRFBIp2QN/qZV1lDErbRaKETyoRKISjj49STo/KuXpq7hJwJrIDppJ4aCY1Ts9WSZPrWF6ZTUZnV2kdjuw9Lr4NEX26nV0Wi3U5NrpsphxJibQm2Aa0E7g+VdeI82WFpv4JzNJotNqptNqZufMqVh7XRQ2tFLU0MKDGZmEX3yNOnsGR/NyaciyTfoWD73qRC543Msffvo1vuZ/l8dcv+A53Rpe064gSRcgURvAG9TQ4TXS5kmgw2vEou8j2eBDqxr+vq1RI0e55JEnhnXpZCssI6ll5FUtkOKH7ang0PYv6xzsPr44NCdtDutWraPSUcmTB55k/fnrefbwsxRaCpmVMotUQ2qsQxQEQZiQRLIXBzrqXVgzjKg1nz9suoIuqp3VnJ93/oju7ejTEx1Cu4WrZqiw4eBPyotEBc6T0PsDZLd1kt3WSWZnD98qKIDtu/HpdHQkW6jOtdNptdKZbMFn0Mc63NNq7Rx+efPRuH689CQlsqM0kR0zp/Lxw7/jh2tWU1jXSGF9E36Nhsr8HMqLCuixTN4+YOEovKxbyWbNHG7zvczN/jdYGdzFY4avUq7Kx6QNYtIG8YVUdHuN9PgMdPuMmDSB/p8hk2iGbDyNxTJKSQLm9CCn+OGDTHgpF3lJB5Q442dBxgiKw6hT1Hz16a9S3lNOpaOSTGMms1JnkZeYN+x7CoIgnIlEshdjsizTUecmf87A0oP7O/cjIzM3bXhl7fvv3d9uQa8OYtCEBnXBvedoacdCmeir9zlZxupyU9DURk5rB2mO/mW3Xp2WlrRk/m/Hbopvu5neBNOEXPaamTCy5Gak1487SWKf38/mhfPYMn8O9rYOiqtrKTlaw6yKKtpTrBwuLKAqL3vStnPoVFh5wHQLi0IHuc33Er/wrONdzSKe0V2MS2HCoA5jMPdii7jp6dPT4zNQ77SQdfvT7G0OMSW5m0RdMNbvxoQx3GWUcPqllJLdh3xNLWzIgM02aDD1L+vUx8Fs7AiLwyzLWsZ823zKu8s50HWAt+veJkmTxKzUWRRbilErJ+f3pyAIwmgSyV6Mubv9+L2h4/br7e3Yi1JSUppSOux7e4IaAhEV9kG2W1gYPsRsm5K/KBcii1fwsfS6KWxspaCpFYvbSxRoT7awfeZUGmypdJsTQZL4+3sfcnWgFwK9wx5rosyOTTayQkFjpo3GTBs6f4CpNXVMr6rlvG27WLa7jKN5ORwuKqAr2RrrUMfEdvVM9iUW8TX/O1wa2MjS4D7+qjuf9dqziUhK1Moo6SYvaUYv7oCW8qNHKTct5HBHOikGL7kWJzlm5+BeTBLGjKSPIF/YBIfM8EkaPJ+PvKwDCl2xDm3EtEotc9LmMCt1FjXOGvZ17mNT8yZ2tO1gRvIMSlNKMaqNsQ5TEAQhbolkL8Y66t0ApOUObKa+r2MfUy1TR1SGekjtFmSZa/3vUeOIsju1eNhjTnSaUIgpDa1Mq2skvaeXKNCamsyBojxqs2z0naSS4xk3OzYJ+XVa9s8oZv/0qdg6u5hxtIbimnpKjtbQYbVwcFohVXk5RCbZ3j6/pOUp/WW8q1nEN/te4Zv+Vzk/uI0n9FewV93/s0CSIFEXoPOln/BPf3mQmh4r9Q4zu5uz2N2cSarx88RPr544Db8nE0kCZjqRM33wUUb/0s6jiVhMk2M/sEJSUGgpZIp5Cm2+NvZ17mNvx172de6j0FzI7NTZpOhTTn8jQRCEM4xI9mKso96FQiWRnPV5cZZwNMz+rv2sLVw77PsGw0rcAS2pRu+gyqjPDVcyNdLINzcH0F955s3qTdNqOXfnPgobWlFFo3Qnmtg6ezpHczJPmuAJk5Qk0ZaWSltaKpsXzGVqTT0lldWs3LqTJbv3U16Uz4uqyfejs0mZzo+M/8LC8CFu6XuV+71/YJuqhCf1l9Om/PwhWq8OU5LeQUl6By6/lnqnmXqHmV1NdnY3ZZFm8pBrcZKV6Br1Gb9NL344qvebjCRrsL9a5yELbE/l3muewt/QjtfeBvFUXXWYJEkiw5hBhjGD3kAvB7oOfL6vz5TJ7NTZ5CbkxjpMQRCEuDH5nlgmmI56FylZJpSqzxOsCkcFfeE+5qTNGfZ9u339M4KDKswiy1zrf5dOycwzZS6+deWwh51QJFkmt6WdWZV1fCsvn1BjGxV5WRzJz6bTkjQh998Joyuo0XBwWhEHiwvJautgZkUVcw5V8G7BFOo/3sKB4iJa0lMnz9eKJLFDPZM9qmlcHviYa/zv85j757yiXcHfdauPOz1RF6DU1k6prR1nn+6zxG9HYzYAFr2PrEQXWUmuUSnucuOFw9tLvG7LiIeeUCQFUOpAzvVw9P91MlO9GGNLOr1FdQSSB7esfyJI0iZxdtbZLLAt4HD3YQ50HeCt2rcwa81YVlgIRoJolJpYhykIghBTMUn2JEmqA9xABAjLsjxfkiQr8DyQB9QB18iy7IhFfONFjsp01ruZutA24HhZRxnAsIuzRKMSPcfaLagH0W5hZqSakkgtv9evJRR9ZlhjTiRSNEphYyvzyquwuL24DXp+0dFByje/TvAM6LsmDIMk0ZyRTnNGOgkeL/4n/sw/abUUNDTTk5TIgWmFVObnEVZPjtfPwpKKf+hW8aFmATf1vcFXAx+yOriTpHlqFHKU6AkSN7Pej1nfxixbG71+Hc2uRJp7EznUpAYt9AAAIABJREFUns7Bdhv27/yFP/xlN33VO+mr3YMc8MbgPTuzSIkhfv/2Pdz36G9JOppH8v7p+K0OXFPqCZsG13d1ItAqtcxNmztgX1/WN7K46KWL+MbMb3BV0VXoVLpYhykIghATsXwyOU+W5a4v/P8HwAeyLD8oSdIPjv3/P2MT2vhwtPkI+iOk5X2pmXpHGemGdGxG20muPDWnXzekdgvX+t+jR0rgPc0iYPIme1I0ytT6ZuaVV5Pk9dGVlMC7i+dSm5XO0z/7FXeJRE8YBLfJyLquTqLf+SZFdQ3MPFLFiu17WLznAEcK8zk4dQquxITT32gCcCgS+ZXxa6wPL+PWvtf446Vu2mv+xNvp51JpLDjhjKYkfZr4+SlJ7yAQVtLqSuCdN2tJKj0X08yVgIxBHcKkDZCgCaJXh047OTrZeuWNp0Cykw7LPozNNhLq7KTunE1fehfu/EYi+onbm+/LlJKSIksRheZC/vOG/2TeQ/N4cMeD/HH/H/nnkn/mmuJrRrQPXhAEYSKKp5ehLwfOPfbvZ4CPmOTJXltNf/XGjClJA46XdZaNqOVCt8+AThXCoD79fpnicB1zwkd5UncZQWmSLneRZfJaOlh04AgWt5dOcyJvLz2Lusy0ybP8Thh3TY42mpI0bFg4nRynm4X1bcw8cpTS8kqqUs1sz8mgKsV8wn6Vsay+Opx9b5uAp5jByu5DPP5PZm5q/AdVhlzeTj+XVl36Ka/VqiLkWZ10v/Ewi1dOxxdS4wlocQc1dHhMdCChkKKYNEEStAFM2gCaQaxIEIZIIePNbsVn68TUkImpKQN9RzK+jA48uc1EJlErDUmS8B7y8syFz7CzbSeP73+ch3c/zJMHn+TGGTdy3bTrSNBMjhdkBEEQTidWyZ4MvCtJkgz8QZblx4F0WZZbj729DTj1E8Qk0FrtRGdSk5T2eePtNm8bbd425pQMb7+eNnsm/rCarMTeQeUx1/rfxyUZeVu7ZFjjxbu0HieL95WT2eXAkWDk7aXzqMtMF0meMGJfrKAaTrSwNSeHsj4/M2oamF7TyA27y+k1GTg4JZeKPPuAJcKxrL463H1vAOvuDTNzyi0scJSxsmsrt9c+Q1nSTN5LPQeX+vQPz5IERk0IoyZEOhCOSngCWjxBDe6AFlegf6mdVhnun/XTBjBqgoMqMiUMjqwO457SgNfeSkK9HUNLGobWNPrSu/DktBA2Tp7lnQALbAtYYFtAWUcZj+9/nN/s/Q1/OvQnbppxEzfMuEHM9AmCMOnFKtk7W5blZkmS0oD3JEk68sU3yrIsH0sEjyNJ0r8A/wKQk5Mz9pGOobYaF7aCJKQvJB6f7tcbbnGWhLMuRSlFMetP/wt7SriJBeHD/Fl3EX5pclWc1AWCLNp/hOl1Tfi0GjbOK6E8PxtZceZVGhXGj0+vY1fJVPZML6SgqY2ZVXUs21fOwoOVHM3J5FDhKFQJlBSsvvPtkd9nmCKSkm3WsyhLKmFF9zaW9Oym1HWET6zz2GhdRJ9Kf/qbHKNSyJ8t+ZRlCIRVuIMaPAEtPT4D3T4jEjJGTZDEhWtxRfQkKPrEazWjIKoN0Tu1FndOM6bGDAwt6Rja0uhL6cGT00woyRPrELnvvvtG7V5z0ubw2OrHONx9mN/t+x2/Lfstf6v4G9+a9S2unHolaoVYxi8IwuQUk2RPluXmY393SJL0MrAQaJckKUOW5VZJkjKAjpNc+zjwOMD8+fMnbB3pPk8QZ7uP6UszBhzf27EXvUpPsWXove5anH0YihZjMfgG9Ur4NYH38Eh61muXDXmsuCXLTK9tZNGBCtShMHuLC9gzvZDQJCmcIUwMUYWCqpxMqnIySXH0UlJdT1FDMzNqGynIzsHR0EyNPYPocF58kKPc+cD/DCuudffeP6zrTsSv1PFO2rlsN89ldecmzu7ewUJHGZutC9liPYugcmgvIEkS6NRhdOowqUYfURm8x2b8PEENlvNu4SMfaKUgacpeUlW9pCpdaBWir99IRHVBXEX1uPOaMTbZMDXZ0HeVEjD3Mj17AbIcu4UQt15317Cv/dkr957w+IzkGaxbuY6yjjIe2f0ID2x/gD8f/jN3zr2TNXlrUIxC1VhBEIR4Mu5PwJIkGQGFLMvuY/9eA9wPvAbcBDx47O9Xxzu28dRW4wLAdoL9eqUppagUQ//UPLu9HoDkQczq5URaWRo6wF+1a/BJg38lPp6ZXR7O3bkfW4+TllQrm+aV4JgkhTKEiavLksTH82fxyazpTKtrImvnPs7avg9fWTnlBdkcLsjBa5i434NOTRIvZl3CxuRFrO7azOquzSxx7Oaj5MXssMwl/IWfZSPpk+fZ/w6X/PuP6Agn0Ra20BhOBWTMCi82lQObyilm/UZAVofx5DfhzWnB0JKGqTGTOy56EP7uR57pgCIXknrCvr56nDlpc3j6gqfZ1LyJR3Y/wvc3fp+nDj7Fv531byzNXBrr8ARBEEZNLKY70oGXjy1dVAHPybL8tiRJO4EXJEm6BagHrolBbOOmrdqJQimRlvN5MuIL+ajoqeCW0luGfD9/KMJfdzTSV7UDTWb+ac+/xv8+PrS8rj1nyGPFGyka5Warlavf20xIpeSDhbM5mpMp9uUJcSWoUbN/aj7ffOFlfn7rDZRU1zO3vJq55dXUZ6ZzsDCX5rTkCft126FL5Tn7Wux9razu3MTFHRtY1rOLDSlL2WOeCYxwv+CW58hRd5Gj7kKWwRk10hFOoiNs5kgwmyPBbAySnwyVA5vKgVXpmagfypiSlVG82W14s9r54McbueHqu2BjBmxPQ57mhBInUuLpi39NBJIksdy+nGWZy1hfu55H9z7Kbe/dxuKMxdy94G6KLEWxDlEQBGHExj3Zk2W5Bph9guPdwKrxjidWWqt7Sc1JQKVRfnZsf9d+InKEOalD36/3WlkLPd4g7t2vw/JTL33JjHRwdqiMl7Xn4lYYhzxWPDH3uli5dQfpqWnUZKSyad5M+nSTa/+hMDZiVRFTBppsqTTZUknw+phe08D0mibyW9pxJBgpL8ihMjcLv3ZiVsdt0mfwdM415Hvr+UrnJta2vcPy7u1Uz1QhyfIJq5MOlSSBRenFovRSrG3BH1XTFjbTFrZQE0qnOpSBRgqRrnSSqe6BYayUOOMpZLZXvsMNV18GbXo4aIH9VthnRc71wEwH2H3xm1ArGLAffjAklYR1pZUtl29hbdNauj/opuOVDqK+oVWHtefYaaxvHNI1giAIY0X8BoyBSDhKR72bmcuzBhzf0boDpaRkXvq8Id1PlmWe2lJLcXoC9Q37T3v+VwMfEEbFK9pzhzROXJFlph+tYdmuMsIqJf/R0kzR1RdO2FkRYfzFsiLmp9xGAztKp7F7RhEFTW2UVNezdF85iw5UUJOVTnlBNi2pE3O2r9aYy+OGHIo9NXylcxPPXWWgOfws65VLOCiduEffcOkUIfI0neRpOgnJCjrCZtrCZlqPLfe0f+fPbG8IkmtxkmbyiOqeQyBJQEYfZPQhe1RQbobDZlifAOZA/xLPqS4kTZy1y4jCo3sfHdald51zF1f99SoOrzmM/SI7i22LmWadNujk8Y65dwxrXEEQhLEgkr0Y6Gr0EAlFj+uvt7NtJyXJJRjVQ5tt21bTw5E2Nz+/qpR3T3NuWqSH84K7eUOzjF7FxNzPpg0EOHfbLgoammnMSOfDpQt562cPUTQBH4gFASCiVHI0N4ujuVlYet1Mr21kal0zRY2t9BoNHMm3U5Fnx6fXxTrUoZEkKhKmUGkq4OgD97Hu6jD/EnmdOimdtxRLKJdyRz2RVUtRstQ9ZKl7iMoSHZEkPjzUTJ1xOVXdKehUIXItTnLNDlKMcTwzFYckUxgWdCHP64bqhP7Zvs022J6KXOyCEgeSZeL364t4Iqywr2CGdQabmzfzUdNHHO45zNlZZ5NumPRdoQRBmGREshcDrdVOAGwFnyd7vpCPg10HuankpiHf709barEY1Fw+J+u0514d+IAoEi/pzhvyOPHA1tHJVzZtQ+8PsHXeLPbNKJ6Qsx6CcDKOpAS2zpnB9tJi8pvbmF7TyKKDlSw4dJT6jDQOmkwootHhVfKMEVmSeOFQmIzrbmShfJjzI9v5duQVaiUbbymWcETKGZPvY4UkY1P1N3S/5WtWWnoTqXNaqOpKpqIzFYM6SL61hynJPSRoJ36SMl4kpQxTXTDVhdyh60/6DifBQQuy3du/xDMn9q0bRirVkMoVhVdQ6ahkW+s2Xjr6EtOs01hkWyT68wmCMGGIZC8G2mp6SUjWYTR/vrdsb8dewnKYhbaFQ7pXY4+P98rbuf3cKejUylOemxx1sjq4g/c0i+hRxH4J24mcdB+VLLOkvpU1FXU49Tr+b9FMWpJM0NU8vgEKwjiJKJVU5WRRlZNFktvLtNpGiuua+G2Wnb4XX+dofg4VU/LospgnzAseUUnBNmkmO6XpLJIPsyayg9sjL1MjZfCWYjEVY5T0QX9PvxxLLzmWXkIRBU29SdT2WDjcns6hdhupRg9TknvIMTtRK+NsSWIck9L8sLIVeUlH//LOw2Z42w4JQVbNvgbZr0DSTdyPpyRJFFuLyU/KZ1f7Lg50HqC2t5almUspthQPeV+gIAjCeBPJ3jiTZZm26l4yp1oGHN/ZthOVQjXkZurPbK1DKUncsDjvtOeu9W9AQuYf2pVDGmM8nWgflSoc5txdByhsbKU2M50NC2eBWk1mDOIThFjoTTCyfdY0dsycyt5f/5HvLVtCSWU1s44cpducRMWUPCrzc+jTT4wWDhFJyVaplO3SDBZFD7EmupM7Ii9TLWXylmIxlVL2mCawamWUfKuDfKsDX1BNbY+Fmh4r2xpy2NmURY65lwJrD+kmUdFzsCR9BM7qRp7bDbX9SzzXLr4N/hJFLnLBTAdSSiDWYQ6bRqlhaeZSplun81HjR2xo3ECVo4rl9uUkahNjHZ4gCMJJiWRvnLl7/Hh7gyfcr1eaUjqkpSGeQJjndzVyYWkGtqRT7+UxR91cENzGBs18OpTWYcUeC0luD+dv2YPZ7WFbaTFlxaNb2EEQJhJZoWCj18vs5UvRBgIU1jVSXFPH0t37WLxnP40Z6VTl51CbnUVIrY51uKcVkZRsVc5ih2IGi6OH+Ep0J9+JvESVlMVbisUclexj/v1u0IQosXUwI72DLp+Bmm4r9Q4LtT1WTJoAhcndKAyxXQlx3333xXT8oZAUwBQ3THHz01UP88P/+QUcTYIjZmSbr3+JZ74b6dQLUeKWRWfhisIrONR9iG2t23i+8nkW2hZSmlIqGrILghCXRLI3ztpqeoGBzdS9IS+Hug9x88ybh3Svf+xuwu0P841leac994rAR6gI82Icz+p9mb29k698speopOCNFQtpSUuJdUiCEDcCWi2Higs5VFyIuddFcXUdhXUNrNqyg7BSSX1WBkfzc2jIyiCijO8n67CkYrNyNtsUJSw5lvTdGfkHVVIWbyoWU6XIHtkAkoKfXXfv4E9XaTFMXYJ/1ho8wVnYb/8Tm2s9FKZ0D322b4hjn+j6W687dTudU/nZKyMYe4RaemqQVrQjL+6EI0lwyALvZ4EhhFzihOlOJEMkZvENlyRJzEyZSW5iLhubNrK1ZStVzirOtZ9Lsj451uEJgiAMIJK9cdZW1YtaqyQ58/OKm7vbdxORIyzMGPx+vWhU5umtdczJNjMvx3LKcxOjHi4ObGGjei4tyrRhxz5uZJmZVf0l6B2JJt5edhZuo9gMLwgn40xKZPu8WWyfW0p6ZzdFdQ1MqW9kSkMTAbWa2uwsqvKy4/4HflhSsUk5m08UJSyJHuQr0Z3cFfkHR6N29uWOIGGVo1zyyBPDutQf7mT3W1toXXwJ9U4LiVo/hSndFFh70KoGkaiMYGyAN75767CvjReSNgqzHcilDmg09hd02ZkKu5P7l3jOciAlT7wlngmaBC7Kv4gqZxWbmzfzYuWLzE2fi6QSq08EQYgf8f67f9JpreklPT8RhfLz5R4723aiVqiH1Ez948pOaru8/Pq6019zReAjNIR4QfeVYcU8nhTRKMv2HqKkppHazDQ+XDiHkFp8mcYzv98f6xCET0kS7WkptKelsGX+HLLaOiisa6CgoZlpNXUsnVJIy/bdVOXl0JqWMu5LoofSyL6RFF5mDas1NazVVfDxPxupDL/Im4ol1ChOX3l4tOhUERwfPsE3b82nwWHmaHcKe5qzKGvJIM/ioDi1C6uhb9ziGW+juYRUUgC5Xsj1IjvVcMAKFUlQYe6v4jm7B+zeCbVSX5IkiixF2BPsbG3Zyu723RT8qABdto5A09ATWNGQXRCE0SaeoseR3xuiq8nDwkvyBxzf0baDWamz0KkG30PrqS21pCdquag045TnJUS9XBzYwib1HJqU8d0fyKhQcOHmXWS3d7G3uIDtpaKtwkSgUWtiHYJwArJCQVOmjaZMGxsXRchpbkN+/S2+Ul3HzMpqPAY9NTl2qnPttKWOT+I3nEb2+0jmsDyX8Nu/5/4Levi3yN+piGbzpmIJtYrxK9OkUsgUJDsoSHbg8Ok42p1CTY+Fmp5kUo0epqV1Yk/qnXQN28dqCalkDsE57cgLOvsreB60wPpssASQZ/WgUsb/ntMv0qv0rMpZxRTzFF7rfY3inxaz0LaQ2amzh7SXTzRkFwRhtIlkbxy1HHWCDFnFny+7dAVdHOk5wm2zbhv0fao63Gw62sX31kxFrTz1L5HLAhsxEIj7WT2Dr4+/ZOeQ1dHNhvmlVOSPcI+OIExiQ5kh+1SzHl5sbeG7/3IT+U0tFNY1MuNYRU+vXkdtdhY1OXZa0lOR46yHXwgVv9keRHnJN1gWPcDq6C6+G3mBI9Ec3lIsHtekD8Bi8LPQ0MTsjFZquq1UdKWwqTYfgzrI1NQuCpO7B7fEU+hvyzCvB3l2D1Qlwn4rfJzBT77+V+Q6N96sNmT1xPlY5iXmUXVvFRf9+SK2tW6j3lXPyuyVomKnIAgxI5K9cdRU4UClUZCe9/kP/T3te4jKURbYFgz6Pn/aUodGpeD6hTmnPM8Y9XFpYBOb1bNpUNqGHfdYszh7ufiDTSg0at48ez5NttRYhyQIcW04M2SfCqvVHM3P5Wh+LupQiNymVgoamig+NuPXp9VQdyzxa7KlEY2j4i4hSc1HynlsUZRydnQ/q6K7+W7kBcqjObylWEKd4tQrHUabVhVhenonxWmdNPcmUdGZQllLJgdabeRbeyhO7RrXeCYySQkUu5CnuqDFQMOjrZTULsLUkIkvsx1PditRbSjWYQ5KxB3h/NzzqXBUsLl5My9UvsCyzGVMs04TffkEQRh3ItkbR80VDjIKzShVn79qvqNtBxqFhlmpswZ1j15fiJf2NLN2ThbJJu0pz700uAkjfp7XrR5R3GMpo72TCzdsJqxSckNDA2tsl8Y6JEE4Y4TUaqryc6jKz0EVDpPd0kZBfRNT6huZXlVLQK2mLjuTmhw7jRnpI68sOVpxS2o2KM9ii2LWsaRvF/8eeZ7D0VzeUiymfpyTPoUE2eZess29OPp0VHSmUttjpao7hbRrf4rLryVBGxCr0gdBkoAsH79764f8zxO/xFSfhbExE2NzBj5bB56cFiL6+C/mIkkS06zTyDJl8WHDh3zU9BF1rjpW2FcMqcWSIAjCSIlkb5z4XEF6WrwULxo4w7azbSdz0uagVZ46cfvU33Y20BeK8I2z8055nkHu4/LARj5Rz6ROOX7FDIYit7GFNZs+wW008Mbq5RwpK2NNrIMShDNUWKWiNsdObY4dRSSCvbWdKQ1N5DW2UFxTT1Clwm6z0X3V1zmSk09oiHs1x6KqZFBS86HyLDYrZnFOdB+rorv5j8jzHIrm8ZZiMQ2K8V/RYNH7WZzTyNzMFqq6ktnlyqTeaUGjDJNs8GHV96FQyOMe10QUNvlwlhzFnd+IqTETQ2sahpZ0+tK68eQ2Ezb5Yh3iaSVoErhsymXs79rP9tbtPF/5PKuyV5GTeOqVOYIgCKNFJHvjpLnSAQzcr+fwO6joqeD2ObcP6h7hSJRnttaxpCCZabZTr/+/JLAZk9zH89r4TJ+Kq+s495OddFotvLnyHPy6wSW7giCMvahSSYM9kwZ7JopolMy2Dgoamljo95Py3usEFAp2mS1sTk5mm9WKO8YN3IOSmg+U89mkmMXy6D5WRnfzvcjfOBjN5y3F4pEPMJIZTUnBOT97hk6vkVZ3Ih0eE1aDj2SDD7UyOvLY4txoVPOMGPz0FtfgzjuW9LWkY+hIwW914MltJmh2j0KkY0eSJGanzsZusvN+w/usr13P7NTZLLItQqmIn2XSgiBMTiLZGydNFQ40ehWp2abPjm1t2YqMzLLMZYO6x7uH22np9fPjy0pOeZ5e9nN54GN2qGZQrbKPKO6xMOtwBct276PRls475y4lFOMHRUEQTi76haqej77xJt/51l2UNtVT0tTIsqOVRCSJmrR0DmXlcCgrG5fhxEvUWsYh1qCk4X3lAjYpZrM8WsbK6B6+H/kr067T80G4kWrVMAs/jaBX3hvfvZUkXYAkXQBvUE2X10in10iX14hZ30eK0YtuEhdzGW41zxNV8oxqQ7gK63HnNmFstmFsyiBl70wCSS48uc0ErE6I46Wyyfpkriq6iq0tW9nXuY8WTwurc1dj1g5/D64gCMLpiGRvnDQfcZBZZB7QX29z82YsWgslKadO3gBkWeaPm2rIsRpYNf3ULRQuDmwhUfbxtziswDlv/2EW7TtIdY6d989eFFfFHwThTDCcSp6figLPHOgCjGAoZrrGx4qgkxVdDq5ob+OKPTs4pDLwscbMxxozDUNoJzOaApKG95QLjyV9+1iWs4VLPY+wTVXC33Tnx+xFMKMmhFHjJBBW0uU14ujT4+gzkKD1k2r0YlCHxL6+QZDVETx5zXizWzG0pmFsyCR5/3SCJg+evGb8KT2xDvGkVAoVy+3LyU7IZkPjBl6sfJFzss6h2Foc69AEQZikRLI3Dtw9fno7+yg99/MHjKgcZWvLVpZmLR1UD55d9Q72Nji5//ISlKdo5GRU9zdR362axlFV7qjEPypkmYVlBznrYDkVBblsWLIg7sq7C8KZYCSVPAFuOm/gPrg+4G3A7PKQ39xGfnM7tztauN3XQk+CiVp7OrVZNv5ny4iG/f/snXd4HHed/1/Ttmul3dWqd0uybLn3nubghBRSSAgESIDcQYAEuKPkSMgBPwjlIOFIODjqcQcEkgBpTneau+NeZUuy2qr31fadnfn9sbJkY8lFllzn9TzzjDT1s21m3t9PGxMRwcxr0gI+8pNX+Nm3PsTN0bf5SeBRNsnT+JNlFXXyuclnNssJclP9ZKYM0B200x2ycThqwarE8NqDOI1iLqeELmkE89oI5rRjbU8npSEX997JxO0h5pVeia4NNnI/DylOLcZr9fJG4xu82fQmvoCP5bnLz7VZBgYGFyGG2DsLjJSvd6D7AD2RHpblLjulY/z3O4dx2RRum3viMKTPzDORqgd50nJucvVG9BroOqsONjC3voWteRm8UJaD3n02groMDAzOFn1OBzucpeyYUoo9FKaopZ3i5nZmVx1m7oFalpRMonvHfupyM2lNd53VwZ6BGDxluZoXzcu4MfouH4i8w08De9mgTOcv5qs5PMGevu17N59kCwnRkk/IOonGuAs9EUQL16JFGifUrosGUSec3Uk4sxNrZzqO+lzuvupB+HMMfU43lPUnWzucZzhMDm6cdCPb2rexrX0b7cF2LIXnxhtuYGBw8WKIvbNAc1UvFruCJ8c+tGxt81oEBJbkLDnp/jUdAd440M79V5VhNZ3gjhUL8ZUlJnbI5RyUi8bB8tPnOK+BrrNsx36m1bewp7SQrbOmkm0MWRsYXNQEbVb2lRaxr7QIczRGUWsH+pvrWXG4kek19YRNCg05mdTlZuLLTCdxlsK5Q4KVP1tW8YJpBTfEkqJvSXwP78lT+Ytl5YRdN+fmntpxdb0Wv+akU8gg5JiByTGVtOUBwnEZq6JOiG0XFSKEM7sIZ3Sx+uuv8E8f+Td4Oxu2pqPP6oaKfgT5/KqEKgoi87Pmk+vI5Y3GNyj5Rgm/3/d7Pjb1Y6cU9WNgYGBwMgyxN8Houo7vYC+5k9MQjgq/XNe8jkpPJW6L+7h98gvz8TUOe8jc19yHferlfOUD8/nXsH/Uc31hoYmfXGPhro3pbOp/c3xfyBgQdJ0V2/Yypa6JneXFbJpRgRGbZGBwaRE1mzhYlMdPW5r5l4/fSn5bJ8XN7RQ3t1FR7yMuSTRme6nLzaQxO4PYWSjYFBSTou958wquj67jA9F3+VHgp+yQy3nKspK9cum4nu/dNftOex8pJR1zbgXOxbfz150JAvvexL/lb6g9zad3oEtRMAiwq24d3FoPjXbYng7rsmC7B31mD0ztQ1DOL9GX48jh9vLb+cnvf8KP5B+xsXUj3136XTxWz7k2zcDA4ALHEHsTjL8rTKA3ytxrhkM4+6P97Onawz/P+OcR9/E1+vjZjp8BEAhL/PcLuUwvCfDAhu+Neh5ZjXHn6z/irapeypcsoHyM9j4+Tnk1gqZx+dY9TG5oZtuUUt6rLDOEnoHBJY4qy9TlZVOXl51s6dDRTXFzO0Ut7UzytZEQBJoz06nLzaQ+J5PwBLdkCQlWnrJczfPmFVwb3cDN0bf5XuC/2CuV8JTlanbI5eNy3frHPMdTp4b/+tl3mP6FnyDOfB8pM9+H0xwl3R7Eboqf0hEmor/hhYIgAIVB9IIgtNhgmwc2ZsIOD/qMXqjsRTCfP+0vLLKFpiea+I8D/8EP3/shtz5/K48sf+SUIoAMDAwMRsMQexOMr+r4fL0NLRvQdO2U8vW2H0ohocH8yaN79AAq6zZjjw7w729HmTVWpTdOiJrGVZt3MsnXxpbKcrZPHd9RcgMaxC8GAAAgAElEQVQDgwsfTRTxZXnxZXlZO6eSzO6+oQIvl23by4pte2nzuKjLTRZ4mUgigpm/W65gtXkpV8c2c2vkLb4d/G8OSgX8xXI1L07o2U+M2tuSLObiCNAdstEdsuGPWrANFnNJMYq5nBRBAHJDkBtCb7XCdg9s8cJON/r0Xpjeg2A5f0Tfhyo+xOzM2Xz1na/y6dc/zSemfYL7Zt+HIhptigwMDE4fQ+xNML6DvdhSTaRlDveeWte8jjRzGtM80064bzQusKPGQXleCFfK6PkaSjzKnOp3aMwoY23jNmaNm/Wnj6hprNy0g5LmdjbOqGDX5JJzaI2BgcEFgSDQnu6iPd3FphkVuP0Dg6Ge7SzZXcWS3VXMLSxiYNde6vLz6HalTkikQEwwsdq8nFdNi7ky9h63RdfwcPA33PhpO6/EdrJBmYF+jsIiZUkjMyWA1x6kJ2ylK2Sjoc+FWVJJtwdJs4Y5QaFmg0GE7DBc50PvNCfDO7elw24XemUfzOhBsJ0fPQ/LXeU8ef2T/Md7/8Hv9v6OrW1b+cHyH5DvHGOvSAMDg0sWQ+xNIFpCo2l/D8WzvAiDDyaarrG+eT2LcxYjiScuSrC71kE0LrFgyom9ejMOb8AaC7JlytXAtvEy/7SRgCu37KKkuZ31M6ewp7z4rJ4/Eomc1fMZGBhMAIJAT6qTnlQn26aWkRIMUdzcjmnjNubt3s/83fvpd9ipK8ilLj+PNq9n3IWfKsi8Zl7MGtMCVsS3c538Bx4I/S8+0cuz5st5yzSXmGAa13OeKqKok24P4bGF6I9Y6Azaafan0h5w4LEll0vi+ZWPdj4ieKOwqhm9x5T09O1ywx4X+pQ+0uzp59o8AKyylYcXP8yi7EV8c+M3ue3F2/jGom9wXcl159o0AwODC4hLMHP77NF2uJ9oSKVo2nCCdVVPFd2R7pP201ETsKXKSUFGhBxPbNTtzLEws6rfpS5rCh2uczfiJ2gaj2RnU9rUysYZFWdd6AGYFNOYJwMDg/OTAbuN3eXFfLypkd9/8EbeXjSXPmcK06tquPnVN7nrmee5fMMWiht9KPFTy2M7VRKCxFum+Uz9ryA/sH2MiGDm8+Gn+Y3/O9wReRWnFhjX850OggBp1gilnm6KXT1YZJX2QApVnV5a/CnEEsbt/VQQ3DGEla3wocNQ6of9Lr754T+gv5OJ7j8/wibfV/Q+nrnhGcpd5Tyw9gEeXPcgoXjoXJtlYGBwgWB49iaQ+t3diJJA/pThipvrmtcBsDhn8Qn33XPYQTAic/2i7hNuN7NmLWY1wpaKlWdu8FjRdS7ftJUKZyqbp5UboZsGBgYTQthq4UDZJA6UTUKJxSlsbqXI10JJUzNTautJiCItmV4acrOpz8thIMUxLufVdFhnms06ZRbT1Fpuib7NnZFX+WBkDWtMC3jOvIIWKWNcznW6CAI4zDEc5hjhuEzXYJP27pCNNEsExVt0Tuy60BDS4nBFG/rcbjY8uJcVyg1QlYZe5ofZ3Qiu0QddzwY5jhx+u+q3/GLXL/jl7l+yq3MXP1zxQ6Z6pp5TuwwMDM5/DLE3gdTv7SanLA2TdfhtXte8jqmeqaRbTxAmIkpsPpBKjidKQebooYmWaJAZh9dTkzOd7rSc8TT91NF1Ltu0lYraep7o6kSe8v5zY4eBgcElRdykUFNcQE1xAYKmkdXZRZGvlUJfC8u27mTZ1p30pDppyMtmg9WKqGloZ9rIXRDYq5SyVyklL9HOTdG3uTq2mWtiG9msVPKCaTl75NJzVnnYqqjkp/WTmRigK2inN2wl55NP8GaNn6mZHWQ6AkYxl5MgOOM8te6nrPheJex2w/40OOREnzQAc7oRPNFzZpssynx+9udZmL2QB9Y+wJ0v3cnnZn2OT1R+4qRpIQYGBpcuhtibIPo7w/S2BqlcNizCusJd7OzYyb0z7z3hvvapl+MPyayc23PCG/Ps6neQ1TjvTTlHXj1dZ/mW7UytqWPbtCn811+ruP/cWGJgYHAJo4sirZkZtGZmsHHuTJz+AQqbWyn0tTLjQDX/V1CI/1f/yUaTk02Kk80mJ71nWNnQJ2XyhO1D/MFyLddF1/P+2HoWx/fSKGay2ryUt0zzCAuWcXqFp4dJ0shxDpDpCLD+6VewXPUR1tSU4raGmJLZQUFan1HM5SQIDhWWdKDP7obdLtjrglonetFA0tN3goHYiWZ+1nz+esNf+famb/Of2/+Td5re4ZFljxjFWwwMDEbEEHsTRMPeLgAKpw/n673Z+CY6OisLRxdnCU0nddFtaJFG/vjrb466XY6i88mKKM/0iXzx0Z+Pm92njK6zdOtOph2qZcfUyWyZNQ3++vezb8dFglFcxsBg/PA7U9jjTGHPlHKUWJw9jz7BvSuWssLXyqpAsh1OpyuNppwsmnKyaPN60KTRPSMn6j/aJzr5o/VanrZcxfLYTq6Lrefe8N+4K7yat01zqfOeu9w5SdTxb3qae++fRV2Pi/0dGayvL2KnKcqUjE4muXuQpfOn5cD5iGBNwMIu9Fk9sMcFe9xQn4KeHYJZ3VAQHHFQ9lvf+taE2pVmSePHl/2Y1XWreWTTI9z6wq18ed6Xua38tqGCcAYGBgZgiL0Jo2FPN2mZNtIyhlsuvN7wOkXOIkrTRu879/LeVhRPHsuK6ihcPLqf7OaWl5H9+2mYfw/3LE0dWv7Isw+Nzws4EbrO4u27mFFVza6KMjbNmWE0TD9DjCIxBgYTQ9yk8HpggIolC0DXSe/ppaCljfyWdmbuP8icfVXEZZnmTC9NOVk05mThd6ac9nligok15gWsMS+gTG3guugGVsa2sPezDmr0F1mnT2G3XkSCsx9uJ4k6pek9TPL04OtPZX+Hl62+PHa3ZjHZ20V5eicW5fxoOXC+Ipg1mNedbMZ+IDUp+l7OB1cUfWYPlPkRpOEqqPfcMbY4l9O5hwuCwPUl1zMvcx4PrX+I/7fp//FW01t8e8m38dq8Yzq/gYHBxYch9iaAWETFd6iX6ZfnDS3ri/TxXtt7fGLaJ0YdddN1nSferCHe3UT+rP5Rj58R7WJO/142uOfSZ0oddbux0trpG32lrrOyupFZh5vZXJDF6sJ06GoedxsMDAwMxh1BoMvjpsvjZvv0qSixOLntHeS3tFHQ0kZRcysA/Q47vpwsGrOzaMk6/YfmarmQn8iF/Fa7gdy/fpWvrQxwt/gWQd3MVn0Sm/Vymjn75f0FAfLT+slP66cjYOdARwZ72rLY355BiaeHKRkdpJjPbSGS8x3BpMHMXvRpvVDrTLZseDsbtqQnG7RP7TuzE4iMzTMngPtKN2s/tJabu27moUUPcU3RNWdmi4GBwUWBIfYmAF9VL5qqUzR9+Gb+VtNbJPTECUM4X9vfTlXbAP0bn0ZceeWo272v4x1iosLbnhNX9BwrOSlpo66bt+8Q8w43s78knx1zppFjePQMDAwuUOImhfr8XOrzc0HXSR0IkN/SRn5LG+WHG6g8VIsmCMwuKKD7vfUcyiuiKSPrhCGfR+MXHfxpQ4zYlbdTTguLhEMsFaq4TNyPT/ewSS9nmz6JEGc/ty/DESTDUUd/xMyBjgxqu91Ud3nIT+unwtuJ1x486zZdSAgSUO5PVuv02WCnBzZnwHYPNy/6DGLEhGYZg3DW4Gc7fjZmu7507Ze49de38pV3vsKahjU8sOABPFbPyXc0MDC4aDHE3gTQsKcLk0Uiu3TY6/Z6w+vkOnKZ6h65TLKm6Tz62iFK0u007H8bGFnsFYaamBKo5VXvCsKydQKsH505+2uYt7+GA0V5vDtnmhG6aWBgcPEgCPQ7U+h3prC3ogwxkSCrs5u81naU97Zz1baNXL1tIxHFRE1uAdV5hVTnFdKV6jrptVBH5CB5HNTzsOkR5giHWSQc5IPiRm7SN7NHL+Q9vYwqcs96mGeqJcqigiZmZrdysNNLdZeHpr40XNYQ9ulX88id34LEGPsXChdur7+x5NxlyUUstl3H5dNvQdwkEM7oJlDQguo4ez3xYm0x/vfa/+XXe37NL3f/kg0tG/jyvC9zU+lNRi6fgcEliiH2xhld06nf203+VA+SlLzRDcQG2Ni6kTsr7hz1Yrt6TysH2wf4zztm8ZY+SsK8rrOq4x38sp2N7rkT9RJGZFZVLQv2HeJgYS7vzptuCD0DA4OLGk2SaMnKoCUrg8f/+ndu/97jTGpupNzXQJmvnmn1NQD0OlKoziviUF4hNXmFhCwnHoQLYWGdPpV1+lRy6GahcIh5Qg2zxTqCupldehHb9EnUkoXO2RNLVkVlVk4r0zLbqet1cbAznfT3f4HM6+7DbQvhsYVQTrOYy4tfumeCrJ14xppzB/DNez7K1772KLbWDGztXqKuPgJ5bUQ9vXAWbp2yKPOZmZ/hfYXv45sbv8nDGx5m9eHVfGPxNyh0Fk68AQYGBucVhtgbZzqbBgj1xyg6qgrn201vo2rqqCGcakLjsTcOUZ7p4IYZo/fLmxKooTDcwt+zVhE/w7Lhp8OMQ3Us2nOQ6vxs3p4/A90QegYGBueAE+YTTzDranazDsCTCu4Z5EQizO3rZW5vH7Or97Ogag8aUO1wsD0tja1pLvY7nSc8Zgse/q4v5nl9ARU0M0eoZa5QyxLxIP26jR16CRtzRND1szbAJksaZendlHq6efRrf6H87n+nM2inM2gn1RLBYwthU+LGeN8J6A104C+rZ6DIh70lA3tzFp49FajWMMG8NkJZHejyxFdBLUkr4X+u+R+eOfQMj217jJufu5m7K+/mnun3YFNsJz+AgYHBRYEh9saZhr3dIEBB5bDYe6PhDTKsGczwzhhxn+d2tnC4M8gvPjoHcZTmR5Ke4Jr2t+gwudmeNn1CbB+Jypp6luw6QG1eFm8umGkIPQMDg3PGifKJJ5q5uUXHLWsZnFZrGnk93ZS1t1LW1srtzc182OcjJklszs1DOnSYpkwvvU7HiKItgcQ+CtinF2DS41QKjcwRDrNM2M+Wf3LQOvAIG5QZbFSmc0gqQD8L4ZGCANHGPRS6+oipEt0hGz1hK/0RKxY5TrotRKo1bPTrOwG6ohIobCGQ34ql043Dl01qdTEph/MJ5XQQzG0jYZ3YJu2iIHL75Nu5Iv8KHtv2GL/a8yuer32eL8/7MquKVhmhnQYGlwCG2Btnimd6sToUbM5kKf1QPMT6lvXcUnYL4gg36HhC4z/XVFOZ42RVZdaox13Us430eB+/z/8g2lnKg5hS28jyHfupy8lkzcJZ6OKFm39hMDpGjz8DgzNDE0Ua0700pntZUzkDczxGSUc7ZW2t5O7bTcmuKqCKoMWMLzMdX2Y6zRkeQtbjC7PEUNihT2KHPgkrUdpf/CUfXSZyo+ttbo2+RWvUzOqOLF7szGJdrwdVP8l1WRB55I4za8ljkhNkDzZp74tY6ArZ8PlTaR1IwW0L4bKGMcsjtG44k3NfwPl+xyHqRDK7iWR2o/Q7sPuysfuysDdlE0nvJZTbRtTVPz4hniep5mkrsxH+WJivhL7CZ3/9WdqeaiNcEwYgryCPpoamcTDCwMDgfMIQe+NMep6D9DzH0P/vNr9LNBHl6sKrR9z+mW0+GntC/PbueaNeoG1qiCu7NnDQXsIhR8mE2P2PVNQ1cdn2vdRnZ/D64tlohtC7aDF6/BkYjC9RxcSB3HwO5Obzwurn+Pq/3Eteexd57V0UtHYwuSHZrqY7NWVI/LWmu1HlY4uzhDHzPzvjpHzwTl7TI1Tq9cyw1PCx/HruyW8ghJm9Qgm7xUkcFAqICsf/lh9fr3H9Y78e0+v4x5w7UdRx28K4rGGCMRPdIdtgiKcDhymK2xbCaY4OOy/18Tv3xUI8NUBfajX+qAl7cya2lkysXW5Ua4RgTht2y4lDf0/KKVTz1HSNAz0H2Cpvxf6QneLUYhZmLeShxWehT6+BgcFZxxB7E8xr9a/htriZkzHnuHWReILH11QzuyCNKyZnjHqMlZ3rULQ4L2deMZGmDlFe7+OyrXtozPLymiH0Lg50HUHTEHQdQdeHlnkcNuRYsjx48vlMBwR0AXRBQEcAQUj+feQJzgj7MTA4LQI2K1XF+VQV5ycbu/f5h8TftJoGZh6qQxVFmjM8NORk0JidQcB2bKGXsGBhq1DBVrECRVep0BuYodUwTa9jQeIAKiK1Qi77hSL2i8W0c/IqoWNFEMBhjuEwx4gnRHrCVnpDNhr7XMhiApc1jNsanpBznw5jqah5ttDMMQZKmhgo8mHtdGNrziK1tojv3PkX9DVhqOyDzPCEfISiIFLpqaQ8rZzdXbvZ0bGD+v56cv8pl8P9hylJHdugcmFBIY1NjWPatyC/gIbGhjHta2BgcGIMsTeB9Ef7ebvpbW4rvw1JPL6c9v9sqKelP8KPbps5qlcvM9LB/L5dbHLNodM88b1yrktxcvl7u2nO8PDqkjmn3E/KYILQdURNQ0wkkLREcp5IJJcNTrfPn0F6SzPCUcuGhN2R+SiH/7f3XwFNp36D1QFdENHE5PS5KxeT0dqMJorooogmSiSkI5OcnMsSmigZItHAAJKN3V2pdLlS2VkxCVlNkN3VQ0FrJwWtHRRu3wfsozs1hYZsL7OtVgRNOyaMPi7I7BEmsUechKhrFOstTNXrqdTquFlfy83aWrpwsl8spqZUxqzHRvT6jQeKpJHpCJJhDzIQM9F7lLcv4/Zv0xe24LREzklu31graj7y7Fn0cIk64cxuwpndyAErB35Xz2XWG6A6FTwR9Ip+KPUjWEcIkz1DFElhbuZcpnqmsqNjBzvm7eCmZ29iVdEq7pl+D5Pdk0/reI1NjWx+vnZMtiy8cdKY9jMwMDg5htibQF6qe4m4Fuem0puOW9cTjPGzN2u4qiKDJaXpI+wN6Drvb3+LiGTmTe+SCbYWShqa+OfsbFq9bl5ZOo+EIfQmBl1HUtXklFBZMqmQtJ5upIQ6KOaOFXejPSPpgC6KlGemI8dig2JLRJUVdFFAF8TB+eDfgjD0PwggwKvPvcLVN6ziiDcPQNCTRz/iBRR0PekZHJxEXUfQkoJTTWhIiQRKPD4kSkeyV4chAagqMqqssLysCOvAAKqikFCUpAfZEIQGlxiqLNGU5aUpy8v6WVNIGwgmRV9rBzMP1vHHgkIiTz9PY24WDbk5NOZkETMPCzdNEKkV8qgljxekZbh0P1O0eir1ehZq+3jpThvx/gepkorYqZSzSy6jWspHE8b3+i4I4DTHcB7l7Wtx59HUn4bo10izREizho1KnidAdYR5ev3jXPbYFKh2wn4XrM+EjRnoBQGY3A8FAcb5o8MqW1mSs4Q/3/Fnfvz2j3my6kleqX+FRdmLuHPKnSzPXT7igLWBgcGFgSH2JpDnap5jsmsyUzxTjlv30zXVhOIJ/u39FaPuPyVQQ2mogRcyVxKWJraBenGjj6vXbmJHOMyOZfOOyx0xODUETUOKx5HVOHJcPUbUHfn7HwXRLXOnoff2oEkSCVFCkyTiJoWoZBn6PyFJQ16z5HZJTxqCwBPf+wn3P/QvY7J3e0MzS1LGniPy3797is//2xeHFxztiUyoSIOeyOR7kJwrsTjWUIgPzK6E1pahXTVBQDWZUBWFuMmEqpiImxRUxZT0MBtPiAYXO4JAn9NBn9PB7sklmGJx9v78f/j0siUUNLdRXteIJgi0eT3U5+VQn59LvzPlmEP0Ck42SDPYwAxkXWXv737Ap++5nlnqIe6MvMLHeJkQZvbIpexSytgll9EkZo5rhc8j3r73fvFPXP7939MbttIbsdATtmGSVFzWZN7f6fbtu1QQFB2m9sPUfvRuMxxyJj199SlgTqCX+pPCzxsZ18tiYiDBF+Z8gbsr7+bpQ0/zZNWT3PfmfeQ58rij4g6uK7mOdOsog9MGBgbnLYbYmyCqe6vZ172Pr87/6nHrDncG+MOmBu6Yn09pRsoIe4OixbiubQ1t5nS2uGZOqK3FjT6ufncjHeluPn3wIPfIxtdiNARNQ1JV5HgcOR5Hise5c9FsMhsbkv8njg210Uk2Z07IMglZJmaxJL1bsowmJ+e/fOI33PXFz1wcYkYQ0AYFqcoJwsZ0nd8++l/ce/8/Jd9LNT70nirRKNZA4BhBrIkiccVE3GwibjITN5lw2axn1H/sUq1Ceqm+7guRmEnhlYEBypYuRNA0Mrp7KfS1UNjcwpLtu1myfTe9zhQa8nKoz8uhzes5JtxTFWReP5zAbL0BAKcWYLpaw0y1mplqNQvD+wDwCzYOSMXsl4vZJ5dQK+WhCuNxH9CHcvtyNIH+iIW+sJX2QArtAQd2U4w0S4RUSwRJ1I/b+0yqiEoX8OV0pFxDAZES03RmWlcwee88lH0uOtVm9kc2cSC6mQ51/KpopppTuWf6PdxVeRdvNr7JHw/8kR9t/RGPbXuM5bnLubH0Ri7LuwyTZBT3MjC4EDCe6ieIZ2ueRRZkriu57rh133+5CrMs8sWV5aPuf2XXBlyqn//O/ci4h9scTUmDj5Vrk0Jv9ZUrCK1bP2HnumDQ9aTwiMVQ4rHkPBZDjsWRE+qxmwJ5rlQ0USRsd6AqytB0ROCdTIwMRKIXh9A7HQSBUCxO3GIhbjm+/Pzxn0EcJRbDGgzi8PsBePD6K9FqqpPiz2wibjIN/Z2QlZO+p5dqFdJL9XVf6OiiSLvXQ7vXw5bZ03EEghT5Wij0tTC9qppZ+w8SNptozM2mPi+Hppws4opyzDH8ooP1plmsN80CwKv1MCNew9TEYaaqdSyMJMVfFJlqqYBFV5pR4/uplvLpF0cemDxVpMFKnm5bmJgq0Ru20hex0OxPpcXvxGGOkmaJ4DRHEQeF3xuPXzPm862875UzsvdccrJcwy51J9YOD872dFb03cJljltRrWHC3h5+ufH76Nr4dK5QRIVVRatYVbSKw32Hea72OV6sfZG3fW/jUBwsz1vOyoKVLMtdZjRpNzA4jzHE3gQQ1+K8ePhFVuStwG1xH7Nu8+FuXtvfzpffV443xTzi/pVekaXdW9maOp1GW96E2VnS4OPqtRtpHxR6cZNy8p0mmLPqddB1ZFVFiUW5akopnpYWlFgUJR4frlgJJEQR1WQiYrehKqZkzpmikJCTgu6n331szGGUBqNwJKTTZOIfvxFiIoESi/Lu31dzzcoVyLEYlqNEICRDQuMmE3Gz+SgxaD4l8W1wcdLa6TvXJowrAYedvRVl7K0oQ4nFyW9to6iphUJfK5MPN5AQRZqzMuhMS2PbX16hYxQvzN8AyAAySFeiLErrYVFaL4vTuvnaUhNyMNk6oUNwUS3nUy3lUy0VUCPnERLGll5gkhNkpgTIcAQIqzL9g8JvIGpBEDSc5ijW0gXEdRFFMEI9/xFdThDK6SCU04EYU7B0urB2enA05fCvNz0O/6uiFwShIAC5oXEp7lKSVsKX5n6J+2ffz6bWTbzW8BpvNb7Fy3UvYxJNzM+aj+caD01qHblS4Yh9hQ0MDM4NhtibANb51tET6TmuMEtC0/nuSwfIclr41LJRShtrGj+/zkJUMvFqxmUTZmNJQxNXr92UFHpXrThuBPhcMWFeB13HFItijkRQYjFMsSimWAxRSz5IXDt9MvFohLjJTMTuGPQSKagmE5pk/EzOJzRJImq1sflwEwszMoeWC4MiUIkmPbFKLHq8CBRF4iYTt82bjrOvl9igNzBh5ARe9OSkpJ1rEyaMuEnhcGE+hwvzETSNrM5uinzNFDW18HBmFvTupdOdRkNeDnV5OXS5T9yW4dDg9JtvfZv7H3mAUtVHWaKRskQTS+O7h7bziRnUSrnUSznUDU49gvOUf0uCADZFxaYMkJUyQDCu0B+20h+xkHHrw3yxJcZMSz1zrLVMszRiEeNn+E5dfGimOKHcDkK5HQhxiZe/+SqfvPd+aHDAoVQAdE8EckOQE0q2czgD8SeJEktzl7I0dynfWPQNdnTs4M3GN9nQsoHsO7L5pv+LOIVUSuWpTJInUypXUChPQpmgarAGBgYnx3iKnQCerXkWt8XNsrxlxyz/83uN7Pb189iHZmI1jRKauetPLC+U+VvG5YTkiQmLGBZ6HlZftfy8EXrjhq4jq3HMkQjmaBRTJII5Fh3y1mmiSMxkIuBIIWZO5n898fiv+fQDXzjHhl96jKcnV5ckYlYbMeuxvxsxkUCJRpMCMBpFiUWpzMnE0d01tE1iUATGBvMBk38bQt/g/OB0vZItImwvSId8D9t+8UceuvUminwtzNlzgHm79xOwWYfy/JqzMkatvByKw165lL1y6dCyFC1IaaKJskQTZWojU9R6LovvGFrfL9ipk3Kol3Jwz1Rwq000SRlEhZEjWY4gCOAwxXGY4uQ4/ax54ud88O7b2REuYXO4HAWVSksjc621zLTWYxejp/WeXAroSoLttW/zqatuQ9eATgs026DZDvvSYHcy0khPjUJWGDIikB4BTxRBPj5n8mTIosz8rPnMz5oPgOJW+OEvf87++C5q1Sq2xzcCICGRLeWRKxWSJxWSJxWRJxXiEtOH206JjNqC6lTIK8ijqWH88hYNDC4mjCeZcaYn0sO7vnf5yJSPoIjDIqo7EOWHrxxkYbGbm2bljrxzqAde+wbrG1W2V0yfEPuGhJ7Xw+orLxKhp+vku1Nx9vZiiYQwRyJIgx47TRCImc34nalELRZiZgvqCKF8UXX8exgZnJyzkT+mSRJRm42obVgE/vT3z3D/Vz6PKR4b8vQqsRj2wMDQdweS7SJig4VhYsqwEDQwOJuciVfymViMnZUV7KyswBKJUNjcRpGvmfLDDVQeqiUuyzRlZ1Kfn0NDbjaRkXJoj2JAtLNDrGCHMlxJ2q6FKNJaKU60UJxooSjRwrXR9dx0kxUCj6Eh0C66aRSzaJQyaZSyaBCz8UkZxIXj70GCAJHD27jb5eVjaW9TE8tmW3gS28Ml7IyUIBoM/eMAACAASURBVPUmmGxuZqalnhnWejJk/3HHuNQRRCAzkpzm9KCrQlL8tVmh3Zr0/B0c/F4JOnpajE+ufJjNzx8mLdNGWqYNV6YNk/XUHxPVXpWl5itZar4SgH6tj1q1ijr1EE2JeqrV/WyOvTu0vU2wkysVkisV4Fnp4VMPfgqX2YXD5DjtMNDPzf7caW1vYHApYYi9cWb14dWounpcCOcPXqkiGFX5zk3TRh+9eukrEPXzmdURblw1/iFlk+qbWLnuIhB6uo4pGsUSDmGJhLGEI3xh5TLo6SKuKITsjiFhFzOZjPA8gxHRZJmILBM52hOo68m+gYNhvkkhGMPh9yMelcf58A1XYWpqPKY9hGoyJX9TopGrYnB+ErFYODipiIOTipASCXLaOijytVDka6GkqRkdjmnrcKoERRv7xEnsk4cbY4u6RtV3P8NdX/08hVorBYl2ChJtzFUPIJMcUEkg0CamJwWgmJUUgVIWzWLG0HEkQWeyuYXJ5hY+nLqW+ngG2weF35P9K3iyfwXZcg8zLPXMsDRQam5FFpIVSsZczfMizDcTZB2yw8mJZCFjAjJ0WYamPM8ktr3SgK4NX+tsqSZcmTZSM2w4XGYcLgsp7uTc4TIjjxalBKSKacwxLWKOadHQspAWwJdowJdooDnRgC9Rz5bYWrI/ks1LdS8BIAkSqeZU0sxpx06WNMzSiT3EFxqFBYU0NjWOef+C/AIaGhvG0SKDixFD7I0zJaklfLjiw5S5yoaWba3v4amtPj69ooSyzFEqmh14AfY+A5d/nb3//gA3jrNdZYcbuHLDlgtW6EnxOLZQEGsohCUcGnrwjikmAikpPPfqW1x+x63JAhwGBmNFEIaqqEZs9uHluo6kqoMCMEr11h3MdLuwBgM4/MNeYR1IyMqQAIybTKimwQqtsiEEDc4fEpJEU242TbnZrF0wh/TePoqaksJv8Y49LN6xhyXFJTRseIv9hZOoz85DO43vryaIHOrW2GiawUZmDC2X9AQ5WicFiTYKE20UaG0UJNpYEN+PNCQCRQ591o7HX0VAshGUbQQlGyHJQrGpg2JTB7embqRDdbI7XMSuSBFrAjN5NTAHqxBlmqURe2WMZXd/dEy9/F780j2nvc+FhiAAKSqkBKA4AMC3v/451HiC/s4wfe0h+tpD9LaH6GsLUberk/DA8TmT1hRlSPjdtvTztNb2Y7JKmCwyZpuMYpaOGeC2iQ7KxUrKlcqhZbqus/Sjk3ngpQfoi/YNTd2Rbur669AZFp9W2TqiCOQCvbQ2NjWy+fnaMe+/8MZJJ9/I4JLHeDIeZ44kLh9BTWg89OxeslMt3H9V2cg7hXrgxX+BrOmw/F+AB8bVpsqDNSzfsp2WzAxeumIp6oUg9HQdSziMdVDgmeIxAOKyTCDFScRqJWKxog2Ku11NrSy/hIWe0TttghEEEopCWFEI2+08vXUP2ddcnVyVSKAMtomQ48NtOuwR/1ABoCOoksznrliMp7X1mDYdR1p1nKoX2vi8DcYVQaDL7aLL7WLrzErswRBFvhaEN99lyd6drNi9jZDZQlVBMVUFJVTnFRK0ji2nPCFINElZNElZHN3oR9ZVcrUOChLtFCZaMXW9zKrMEN5Yz1DPTQ2BkGQhOCgAvZKNIms7Vzt2EdJNHIjmsztcyO5IEenX/QtVnZAI9qH2tRHva0PtbwfNCNkfFRFkZXRPnSwqpDnScdkzcDkycDm8g/MMXHYv88uuxlfVe8w+aiJOX7CTnoEOegMd9ATa6R5oo6PfR0efj0Ckb2jbX/34V8efVADJLiE5JGSHTDglTL+jn2ZHM6J5WOFV/rKS8u+XE2mOEGmIEG4IE64Pk/Cf/PM28v0MLnYu3afjs8T/bmygqm2An985B7t5lLf75a9CuAc+9jeQzlyIHZ3Mv/ywjxWHGqnyunhqRglqX/tJ9z9XD5IWRcY+4McWDGANJb13OgIRq5Uep5OQzZ4UqudhWOa5fvg2eqedO3RJIiZJxP4x10nXk8Vh4jGk+HDTeE3XMYdD2AbUYxrH63CU8DtKBCoyqnysGDQ+b4OJJGi3sW9yKY//35+49f6vU+arZ2p9LVMaDzOn+gAAvvRMDuUXUp1XRH1WDolRihlt37t5DBZ4aHkqzBuPX4aoJ7AlItjVEI5ECHsiRIoaICPWfYwIDEpW5sn7CFps+G1Wbv55F1Pv/DKdUhrd9nLMuRUIaLilAF6pn3R5gDQxiCgcX5jk8Qu83exITdlPCQ1+tuNnYzxrjM/N/gAP/uF7SBETUtSMFElOtoiZlGgJkyJTEKMmhKOufJqsolrD7Ny2iYpFk1FtEVRbCNUWAfHERWNiepSAPkBA8/P6i6tZcOcCegp76J/XP7SNXbaTbksn3ZqO1+ol05Z5XE/AM833yy/Mx9c4xtYuF6hH0uDCwhB7E0hDd5D/ePUgl0/2cs20rJE3OvAi7HkaLv960rM3DuSkpIGus3DPQWYfaqQ6P5t3F8wk4xRDcM7mg6SYULEGAtgGAnzzxquRO9pRpaT3LmyzEbHa0C+A0Dfj4dvgOAQBTZaJyjIc1Y7s579+MtmX8Ujj+MFpSBCqcUzBAFLi2BHpI2Lw05ctxNPRjionRaCqyElxaPQQNJgAoiYTe0vK2VtSjqBp5Ha1U97UQLmvnst2beXKHVuIyTK1OfkcyiuiOr+IjjT30Hdxbm7RmM7bMjjXBImAbCcg2zl6qFLUE9gT4WERqIZIjQ+QFU1W2a3/GAT4Fk1SJrXksEGvZFdiEk2JdKpi+RADiQRuKYBH8uORBnBJI4u/C42TNWUfjUeeHWOO41HocgLVEUZ1hEfeQBOQoibkkBU5ZEEOJ+el2TNw1g230tFFjbgtRDwlhOoIEncEidtD6MrwddEkmHELZtxiOu3PtHPNg9cAEEvE6Ap30RnuTM5DnTT6G4fCQZ0mJ1n2LLLt2WTZs+AML5u+Rt+YRbJRWMbgbGCIvQlC03S+8sxuZFHge7dMH7koS6ATXvzSUeGb44Sus3zHPiprG9lXUsC6OZXo59FDoKiq2AID2AYCmMMhBCCuKKytrmPyVZcTM5uNh1aDi5+jGsePuFrTkNQ4clwdFoNqHEWSsIaCyCOIwcQRASjLg3mCw4JQlS+A8O3zmHfX7DvXJpwT1j7z5ihrMrCleZgTH2BBfIAFzS18oLEOgHZRYYviJJGSwp7Xd9Irjv93TxMkBmQHA7LjmOWSlsCeCPHrP27hrhumUaC3cS2buJ6NIEG/ZGcPxbylzWabVs5hLSsp/kiKP5cUIHXJHTz61aeIth5Cj48hasMoDjM6ok7CGiVhjRL1DC9+5PsP8eAfH0EKW1ACNpSAHTlgw9LlQmodLtijWiLEU4LEUwLEBudHC0AAk2Qix5FDjiNnaFk8Eac70k1bsI22YBuNA40c6j0EQMXjFXz2jc8yJ3MO87PmU+mpRBaNx2ODiwfj2zxB/H5jPVvqevjhB2eQnWo9fgNNg2fvhUg/fOzv4xK+CckP9MotuyhvbGHH5BI2T598XggnQdOwBgaw+/1YQoMCz2TC73YTcqQQN5tZ/cdnKb7umnNtqoHBeYEuiqgmM6rp2OpzT/zmz3z+374IuoasqkkxqMYH/04KQkskjBQYOC5M9KHrr8TS3DSiIEzI8gXhRT9X3HXFKNEZJ+GnF3hI4MevLT3pNl3AS0DKQIC81nbyW9t5X1s7N+TkQs8eelIctGS4afF6aMlwEzGfvKLiWN+3hCjhF1P45bY45ptXAqDoKrl6BwV6B/l6O6V6M8vEvSAmi8HspYjXtbls0aZQreWStvyjAAho5CtdlJraKDW3MsnUhkcaOOktdeV9r3D9Y78ek/2XQnGY0dAlDdURQnWECDPYB1UHMaagBOxDIlAZcGDtHFaKqjXM3Vc9iL7LDd4wpEcRTMfmSyuSQpY9K+nJI1kUpj/WT1uwjb/+4a805zaztnktAHbFztzMuSzMWsjC7IWUucpOuxWEgcH5hCH2JoC6riA/eKWKKyZ7uW1u3sgbbfoZ1LwO7/8RZE07dt0YRwVNgsBPcnIpb2xh87Rydkw5+U16QtF1LKEQdn8/1kAAUddRZQW/20MoJSnwDAwMxoggJls+jBZCrOtJAXjEO6jGqa7bxTS3G3MkjF1Vj4teUiUpKfyUozyER/IHL+ECSJcyp9PQvRU45DKDqwCxIp8Dv3qSL12/ipzOHibXNzOtNllivsfpoDnDQ4vXTavXQ8Q8MWHwR9veCGwkHUgHKnEIMcqlbibLXUyWu7lP+htmOekh2tNnpcY9j/f0CnYnilkfquDNYLKiaJoYoMTcTpHSQZGpg0JTB45xbvA+Zq8gIJ37sd3xRQDNHCdq7iPqGS7mIsRlTAN2lIGk+CvJrISNRzyAOrorBt7I4BRONo5XhkN0BUEYqubZ8rsWnvvtc/REeniv7T22tG5hS9sW3vUlewKmmdOYnzWfRdmLWJC1gEJn4Rk1gDcwONsYd+9xJqHpfOXpXZgkke/dMmPkC0LzdnjjW1BxPcwfYRRP1057VNASjfLxV5+ltKWJtbOnsq+0aGwvYBxQIhHsfj/2AT9SIkFCFAk6nYScTqIW63nhaTQwuOgRhKEiL0dyBv/y3m68K5MNj4+0kxgShEd5Bk3RKLZA4Dgx+OB1V2L1NRFXki0l4oP9Bc/XwkkGZ86ZNHR/KhJhZ8UkdlZMQtQ00nv7ye3oJqezh4o6H9Nrkv3BulJTaPV6aM5w05buOiXP33jY3kYGbcA7gKhp5NFNsdCO2ryWG92buVlfCyJEzArbxMlsYio7EqUciuWxPTxc8t4r9VNkSoq/IqUDwTS2SqVHeOPxsUe4rLzvlTM694WCrqhE3f1E3cliLI985yGeWP+LZOP4I5PPDodSkzsIOro7epQAjCQFoDQsAN0WN6uKVrGqaBUAbcE2trRtYXPrZja3bub1htcByLBlsDBrIYtyFrE4e/HZfeEGBmPAEHvjzO/W17G1oZdHb59JVqrl+A0ifnjmk+DIhBsfH5cHJGdggE++/Dcye7v5WmsLxbe9/4yPebqIiQRLSgvJqKvDEo+hAUGrDb/dTshqHc4ZjI7vCOh4cq4rahqcPsZndgYMtpNIKApRRgg1H2wwn/QMJqfa+t1M97iwDfiRjmorMVRJ1GQaFoImM3GTCU2SDCF4jjkd79xEoYkiHR4XHR4XO6YkxZW3p5/czm5yOrqpqGtkek09AD0pDtyZmVgPBojlWEikTPyjioZII14adS8/feY13lixAosWJTU+QJrqZ2b8MEsTe5OeJkXAZ8ngPWEyu/RSDiQKqItl8V442V6p4Es3c7BTxarEk5OsYpFV5DH0/DM4dQRbAgqDyYnBxvFB+VgBWO+AqsFBAFFH90T40LIvsH99C96CFNxZdiQlGbKZZc/ixkk3cuOkG9F1naaBJja1bmJL2xbWNa/jhcMvAFD6/0rZ0LKB/JR8su3ZRr6fwXmH8Y0cZ2bkpXHX4kJunp17/EpdTxZk6WuAu1eDzX3G58vs6eRTq/+GJRblt9fewguPfoex1eEaA7qOJRLG4U+2SyiYM42oINCd7iXoSEk+5AEXSlkIo6LmhYfxmU0gRzWYj1qSYvDPW3aR8b6rgOQAjxyLocRjyR6DsThKPIZ5sG3KERKiSNxs5oNzp5PS2zMkAk+nr6DBmXEm3rmJQhNF2tNdtKe72D6lFDGRwNvrJ7urh+zOHq5NcZLyejJvS02RiOVYiOVYiOZYSKSdhe+OIBCRLEQkC+14AZA1lVTVT1rcT1p8gJvVtdxKMtQvYLFRL2WxjXK+8bqAZcVthGIm+iPDAymymMAiq5gHxZ9ViWOWVUTjZzAhCALgUMEx3Dhe14EB5RgBOK/0St76v6qhfVIzbLiybLiz7bhz7Liy7bgybRQ4CyhwFnD75NvRdI1DvYdY37ye7+z/Dnu69rCrcxeSIJHjyCE/JZ88Rx5ui9sI+TQ45xhib5xZUOxmQfEoIm7jE7D3GbjyIShccsbnKmlu5K5XnyMuy/z8A3fQmp5x8p3GAUlVcQz4cfj9KGocTRQJpDj5zTMvcvO9nzwrNhgYnGvOxKt4MXgkNUkiZrUSs/6DV3AwPDTZXD6GEouixGJMz8vC3tk5vL8gEFMUoopCTFGIycl53BCBlySaJA2Jv50Vk3jiO4/y5sNXYG6JYmqJYG4MYzuY9NgkrOJR4s+M6jFxNhSTKsp0m9x0m5L3eFFP4FQDpMUHSIv7qYg3ME0/zF2XQYewiX32Yt4Tp7CNyTQmvEQSCpG4Qk/Ihj4UJK1jkpIi0KLEsZUvoTnuJkPuQxHOvidwzD36LhAEAXDGk9OkAQC+Ovvz9LQF6GwcoKc1SE9rkN7WIPV7utG1wYErAZweC6kZNlLTraRmWEn1erjZewef+fG9/GjzF2kJttA00IRvwMeGlg1Ass9fXkpeUvyl5GGVR4iiMDCYYAyxd7aoWQOvPwxTboRl/3rGh5tXtYdb3n2drlQXv33/rfSlOMfByBMw6MVL6e/HFkzm8oQtVvrcbkJ2B7oo0tznn1gbDAzOI87Eq3hReySPCg/Fbh9a/NPfP8r9X/08SiyGadAjqMRi2CMRUoPBoe10QSCuKMQGPYAxk4m4yXymrbAMLjA0QPWaUb1mgjOdyUGEPhVzSwRTSwRTSxRrbSi5rUkglm0hlmNOisCMs1P8SxMk+pRU+pRkXpig6zgSQf7w5BZuv7GAWWo1V+jbAfALNvZLxRwwF1MlFbCfYvoTNiJxhYgqE1Fl/FEz3pu/zsPtyUqgXtlPttxLptxHttJLttxLltxLijRxg0Vj7dEH49On71ygo5OWaSMt89hcy4Sq0dceGhJ/ve0h/J1hquv9REPq0HaPfuolxD8lKEgtocAZh9QYAUc3TaYafIl66v31HOw9CIDX6h0Sf1m2sVX4NTA4XQyxdzboroVnPgHeKXDTz+EMypsLmsb7N7/LZbu2cjCviD9efcO4JbOPhJhI4Bjwk+LvR4nHSYgi/tQ0Bpypo/YHMzAwuHg5E6+kJslErTJR67EPVUIigWlQ/B2ZLJEwjsDA0DbfvWUVekM9cbN5KBQ0bjYb4aCXCoJAwqUQcimEKlMAkAbUQeGXFH/OhmQjb10S+F1+PtLeQ7R63bR70k6vmqwgnnGhk7oP3wW6TrbWxVS1jsrEYSrVwyyKJPs1JhBoFLOolgs4ZCrgoFxAnZDNaz/4Pt/81+tojbtoU120qWnsj+QTP+pxzSGGyZJ7yZL7yFKSYjBL7sMr9xs9/sYRSRbx5Drw5DqOWxcJxunvCNPfGeIL936V6++8GfoVaHBAWMZBBlOYwhRAs8bo8tbR5KrCp1ezK7yLHR07kAWZgi8WsCbyIpXKLDLFXCPk02BCMMTeRBPxw5MfBkGCD/8JzMdfNE4VcyzGh9esZmpDLeunzeaFJVegTURfLF3HFI2S4u/DPtgyIWK20Jcx7MUzMDC4NJkIr6QuSUQl61Bu4BGOFoH7129i/qxpWIJBHP7hKAJNFJPCz2Qmbj4yN5MwCsNc9CRSZMKTHYQnJ++rYjgxJPzsm0NMPVCDeCAZMtzpctKa7qbVm6z4GT3RYKWucd93Hh6zXY8/9O3kH4JAq+SlVfKyhgUApGoDlCWaKFcbKU80sii+h/fFNgMQRWHL+8MUaK/ht6Tglx1ERDMaAj2JFFpVF21x19B8d6SQdaGpQ+cV0Mj551twZbgwyyomKYFZVjFLCRQpcdKfw6Xc4+90sdgVLMUKmcVOXtn+f9zwm0VD6/SYCH4F+k3gVxD7TWT488ioKWFuUCEmRWhxVtOUVkVj+X7+FPoVAGmJdKYwk+nmucxwzMYqnllVVwODIxhibyJJxOGvn4LuGvj4s+AqGvOh0vt6+Pirz+Ht6+Hvy65i47TZ42fnIIKmsbAkn2xfE+ZYFE0QCKakMOBMI2b0xDMwMDjLHC0Cn995gKLrrwWSEQdKNDroBYyiRKNYgwEc/sTQvpooJkNBzSaWlhZhDgWJm8xoRr/As8qYPcHj4F3716/dR2Z3HzmdPWR19TC9poFZh+oA6HY6aPW6aU130+Z1E7SOUD17AugXU9gqTmWrMijSBr1/5YlGytRGvOJb5EXakCKtAMQFmYBsY0CyE5DtDFjtBCUr+qAXLqSZ6FDTaFPTaI+n8eT+EIn0pfSGrWj68MCsgI5pUPj9oxCURO3SHhcROWOP2qnmOsoopEkZuP9/e/cdZ1dZJ3788z3l9uktk56QhOoSkKYIgiJFUXEFRVdxVVbXFXBXXde6q6zrWn5rAcUG2BWWtYGygAKistKkBwTSk8lMppd755ZTnt8f58zMnSQTUiaZTPJ9v17nddpzznnuvWfOvd95Wt8cGu02FmxcyUte/jKec55gXfYpHqr7I3/y78Tqt5hbPIxllRdwpKxkcXop2ZoUqYyDaI8+ajfpt96+Eobwy8vguTvg/C/DktP3+FRHrVvNG+++lcCyufb8i1gzb+E0ZhTccpnc4CDZkWEuOuGvqGDoa24hX1ODsexpvZZSSu2t0LYpZzKUM5P/823FHcNEgWAUDGZGRnjd8UfD5mj4gcC2d1gSONZ7sJpee1wSPA2la57rsnlOC5vnRL1p2kFAa//QeI+fKzZMDPQ+lM3Q2dJAZ3MjC1036rZxf0RAVaV/9yReyK+u/zV3XnU2uWCUWi9PTVAg5xeYV9qKTdRhS4hQsNNR8OdkmWdnyKeyeBmXq2+5jVNedgTGgB9aVAKbsu9QDmwqvkPZdxgpJ6s6iAFLQpKOT9P5H+CJzjZqUmVqk2VqkmXcQ2G4iBA+esOn9/jwz1z88T1u6/iZiz/OBz70Pk7mBQR+yMhIkWdGn+Sp8FGedZ7gd9mf8Tt+RrpSw4LNR7Bw+EiWecfQmGkkU5vgqAUnkR8ok61PaBVQNSUN9vYFY+A3n4DHb4AzPw4nvH2PTiNhyCse+j/Oevg+NrW08YOzXzt9HbEYQ2ZkhNzQIKliESPCaC7Hd35xO+f/3SVa/UkpNeuEjkPZcSYHgcbwnf+6hnde+mYSFY+k55HwPDLDQ9hVQ0T4lh31DJpwq3oJTYxXlT8YelA91AW2HZXmtTTCkdF3bPPgMO29A7T39LNoSzdHrO/gzKWHUfjpLXS2trClrZnO1hb66+v22/eiEYsRJ8eIU9XswxgyQYlcUKDGjwLABm+I9vJED7clK8Htb8mw5vGr+UshxzP5GtYUcgz52wbcgpXMYKVrsdI12OlayukaUvOP5vGuyb16+yN9+ANb8Po78Ac68Po78Pq34A92Qehvc1pt4rE3bMeiviHLyQ0nczInAzAY9vNE6WGekId5uuUxnm19kN8aoW10IfMHjuCv3/AuvvOR31MsFejoW0tH/1q2xPPOgfV4/s7HNrZdm8ALdppmKgsXLGTDxg17dCzAooWL2Lhp44xc+1Cjwd6+cO9XomEWTno3nP7BPTpFTSHPxXfdyvKOjTx4+DH8/LSzdq+B+RRszyM3NEhuaAg7CPBdl4HmZgq1dYSOw/q+AQ30lFIHDxFGSmX8mjp8YHRsuzHYgR/1DBpXB01UKqTjdspjfNvh0tNOpG1oKO4dNOocRtsuz37GsuhprKensZ7HVywBY2gYyfOX793Em045ibndPSzbsAmAUiJBV0sTnW0tdLY209PYsH9Lg0UYddKMOmm6k83jm93QI+fHAWAwSkOqyDvmbSaJN55miAxd0hRNNNIljXRJPQVJE/V7OggMcvXHr+SVX7w2KgGMSwQr6TTlxsMpLzqawFTf89GQEePVQR2fR771eQoVl4zr6c+IaVJvNXJa5ixOy5xFaALWB2t40nuYVe4jPJz9DX+efzsJUsyvLGXB0JGc2HEGueJfx0cbqKtAUzmaGuN5zcTn897j3sv9N6/Zo7yd/JrD9uq1bdy0ccaufajRYG+6PfIj+O2/wTEXwrmf3aPA6dRMln+66XskfI+bXnoODx5xzG6dZ7v/QBtDplSifmSEbCnqrayQTjOYq2E0lYrO7fvRpJRSB6hpLV0TIXBcio5LMTMxRATG4Pj+ePDnVirkUklqhocmBYGe44xXB42Gh0jguRoETqfOns3794IiDNTWcOPQIK2nnQLGUFMYpX1rD+3dPczt7mVxR9SWzrcsupsb6Wpppqu1ma6WJspVbdsffvL+/ZJlz3IZSNQzkKgH4Kxr13DFv7+fekZoN33MMf3MMX3MoZ+Tw6dIURk/dpgMWyUO/mhi1RKbFjNIn1NHyvWByaVCfiiTA8F4XiimMcai7eL/4BerwJaQulSJ+nSJulSR+nSJ+lSJ9D4OAg/2MQItsVnqrGCps4LXpC+mEOa54D9O4+xPnM2mkQ2sTTwFLT+lwW1koSxlwegK5vQux+nNwdoaGKu66waYOPB7yVGvZqS/RLomgePqs+tgpcHedMs2wxHn79EQC6ZSoeeqq/j2ggV0ZrL86Kzz6W5sfv4DtzHWRsIKfHLD8bAJvk9g2wzVNzBSWxeNgQXo4AlKqdliv4xPKILvuviuSzGOAb96/Q1c9uH34Xhe1RARcUng6OikMQB928ZzE/gJF89NcNTcVpxKGd9NaK2J3TS3pn5mMyDCSC7LSC7Ls4ctBiBdLNLe3cecnl7mdPdy7FPPcPyqvwDQX1dLV0sTm2rrqK9tpC9Xs9uf+ZZpyLYRYYBaBqSWp1hStcNQT545po/2qiDwhPAvpKlw0SVZGL6SIgm22C1stlrpsFrosFvpsFrpsFsoJZJkqkoN49Pihxa/u+pLvOZD/8BwKclgKcWW4RrW9jeOp0vY/nYBYH2qRMLZs2qE29rjdnOzdHzArJVj+MFhzlxwJsYY+kv9bBrZxMaRjTxReJjHnIdw2h3mLp/LguxCFvjLqBtsQwbS0JeE1bVcfNo/8pc/dQGQSDtkal3SNQkytQnSNQlSWUfbc38NhAAAIABJREFUAh4ENNibbivOiabdVHrmWbZ8+MOUn36aGwcHeOzS9+E77h5lIVksUjM8RDafRzCUUmkGmpoZzeb0x4ZSSu0JEfxEIhpftKogEGNwPS/qEMbzcLwKrueRyReww2He8ZITYf16DMRBZBwIJhL4blQiqGMFzh7FdJq1i+azdtF8ABzfp6Wvn/buXub09HHYxs18pr0dbv0lI8kUG5pb2NTUzKbGJjY3NlHeH/+wmIoIg9QwKDX8hcUT242hjgL3XvcVLvz7S5gXdDM/7OHwYAOneY9iMVGi3St1cfDXwuZ43mG10mM1UN60ipsue8+kS1rpWtzmRSRaFuG2LGI4XraSLeNp/OFuWi/8JNd8fTVe7wYqPevx+jZHPZrv8ms7tEulRISmdBNN6SZWtq7ECzy2FLawcXhjHAD+EfgjNYka5i+aT/tR7bRn5/K5l/8XN375LkZHKhRHPEaHKwx2Fxn7yMWCVM4lnUuQzrnRco1LMrNnv0+nzV72oDp/4Xw2xVW0DwUa7M0wEwT0XXc9PVdfjV1by/yvfZVPnXUW5+9moGf7Hm29m3j/2afRvmUzoWUxUlvLSF0dXkKHTVBKqX1CJK7Ouf2PeCsI+Pm13+dv3nIhbiUOBCsVkkOjk6qEmrHSRMcdL1Ucm9Kufk0fyHzHobOtlc62uGMTY/jff/8cb7voTSzu7WFRbw/HdEQ/KkOgp7aOTY3NbGpqYmNjM131DftmvNzdIcIQOe5eH5BNnjppl2s82sNe5gfdzAt7mBd0My/s5nTvEXKV4ni6Cg7PvDvN8Pzj6LBa2WI102k3s8VqZlAml3AaM4IXFih5LmXfoZSqobvYSP1hL6zqJdSQtANSrkfS8UnFU2KK8QJ1jMDJXNtlUe0iFtUuAmCoPDRe6rdmcA1P9z8NQOuVDdyU/QYr6o5mhXs0S615mNBQzHvjwV8p75EfKNG/pTB+fhH46EXXctu3nqSxPUNDe5bG9iz1rRns/VEdNISvPfK1PT78vce9dxozc+DTb5EZVHr2Wbo+8a8UH3uMmnPOYc4n/w2noQGAP/zPXbtwBsPCRMBx2QpHpjxcCzqMobellUKuRtuOKKXUDAptm039Q4zW1k3eYUw0VmClMh4AOp6H43kkS0WscKK7+39/3TmEq5+bHATGQWEQr+uz/gAiwupKhQcOW8EDh60AIFMuM7+/jwX9vSzo6+WIzs2csD7qmMK3LLrq6tlS38iWhgY2p9NIJcQkZuYz3flvDwuYE0+GZrfC8myeZZkCyzN5Fpef4YXtnZziPTk+TARAkQRdVtN48NdlNdNpNdOZaKY3WYcRi8c+dQWv+uK1lAObkudSioeJKHouQ6X0+LlEwvHAr3qataz909lIXbKOumQdxzQfQ2hC+kv9dBY6+fUDv+bp5se5r3IPADmpZbFzGAvdpSxsWcrC9sOYb7VhiUXgh5TyXhQI5is8tm4LvZtGWPNId1VJoFDXkqZhTobG9uxEEDgng5vQ4W1migZ7MyAslej9+jfou+467FyOuV/4ArXnv2pSkfQl5y2b8njL88gN9JHr78WtlAkti0J9M72NzXzpM//FZSecsD9ehlJKqT0hMjFMBJntdwfBePD3p//9LWee8eJovVIhVShMKhWEaEiBwHHwHZfAcQhcB99xCByXpmwmGvdVA8IZM5pM8mz7XJ5tnxttMIaG0QIL+nqZN9DHvIF+ju7YxEnrVnPBwkWYb20kqHfwmhN4TQn8RhevMUFQ58A+HlB7Z789dmY18L6PX8n5X/oItgloDftpD3vjqY/2oJcFwVZO9J7CZaKNXgWHrVYj77o4TVj6BZ1WE51OFAh2W40EYhOEEpUAVk3D5SQDxYm/nfmX/YA7n3OoS5eoTxfH2wM6B/o4gdMwxt/ussSiOd1Mc7qZb1zzDW465266w06e9VfxnPc0G4O13O79giD+nFKkWeAsYaG9lAXpxczJzqNt7ly+9aF/5Zu3fQK/EjDYPUp/Z4GBzrF5gfVP9GHCsSgQaptSkwLAhc0rCIMQy9Zn076mwd5+lv/DH+m68kq8TZuou+ACWv/lQ+OleTsVhqRHhskN9JIeHkKAUjbHUGs7o/X1Ovi5UkodJIxt49k2XirF759dx8o3vK5qZ1QqOBYMOp6H7Xs4vh+VDBZHsatKBj/yqjNh9XNVAWEUBFYHhIFjE9hOVEKobQeBfdwTqAgD2RwD2RyPL1wcbTOGuuIom79xNUckExw5kuLI7iTzq6oHe8awrlJmTbnCmni+0auw0fMohAdOUBOIPT5Q/LYsE9JkhmgPequCwV4W1nVyePm+Sb2FBlh0Ww10WU1stRqjKTWx3BvWUg6iUsDnHn8Qr+VUVvc1EoRjv4cMuUSF2lSZ2mRp0jzl+OO3ui17FjSNsWf5n4yI0GbPpc2ey2nJVwDgGY8twUY2BmvZ4K9hY7CWP5R/Q6U80UPrkdccyUW3XERLuoXWTCttmTZaj2il9fhWlmZaaUosJxxyGO4qM9BVGA8CNz7dT+gbPvT6r/PArevoHFjP5t7VbOp9jk29z9HRt4ayV5wqu2oPaLC3n5TXrqP7858n/7vfkVi0iIXf/S7ZU07e+UHGkBwtkB3oIzM0gB0EBI7DcEsb+YZm/FRq/2ReKaXUHtvbISN2eLwIJBLRtO2uMMQJApzA595b7uDcV55VFRD6JIvFSQHhmDAucRwLDF933NFktm7Fty0C28G37WjZsg/6oHC/9wQqwlAmy+/yIyx4zxU8BDwEuL5PU36U5pF8PBU4ZSTPuaPFSb3AjiZcBrIZHt6wkfbHnmQol2MklyWfzVBIp/bbmIC7N+SEBbQCrWy57l7qTn0NbYkySzMFlqRHWZIpsDQ9yuL0Vk5Ir6c5UZl0dN632VjKsLGY5lmzicOaN9Dn1rOeOaz32+gt1zBUSjFcTrJ1pHnSOIGuHYwHftmTL+Jjf93CHGeQFmeIhOxe76BnXX7bbqWfDVxxWeQcxiLnsPEAMDQBvWE3W4NOtoYdfPXOz9H2tja6R7tZ1beK/lL/Ds+VcTLUJGqoaamhZl4NOSdHymT47S13c8QRx5IsZUkWcxxdOZ6V4Sk4QRLbAdIBJuNDroLJeLiWg8Qd8exNcH4o0mBvHwsGB+m55hoGfvwTrGSS1g9+gIa3vhUrOXWnKc25LHVdW8gO9uFWKoQiFOvqydc3UaqpPei/ZJVS6mCyt0NG7OnxAfDQhg5e3NS03T4JQ2zfx/Y9bD+IlgM/3ha1Jzxu0TwyQ4PbHWuI2iP6tjNeKjgWIAa2TWDbhLZDJuFG/fLrd9ZuyWUmV+3N19aSB9ZXbbP9gIaRPHX5wqTp+HSG9ieentwBEDCaTlHIpMlnMuQzaQrZDKPpFMVkkmIqRTGVxJ2Gz+mF8xbv0XFbgEvOWz5p26Z4uideT5gKTQzTZIZoMsM0JYZocod5QW6YU2sT1HTfPen4UTvFgFtHf66evoY61llzWRPOY2PQwla/nsFymq6RHA0vfRvX9E0cV2cVaHGGaHGGabaHaXGGaXGGaLaHqbcL+7om7QHLEptWu51Wu50XcDwf+8E/8tXvf3V8vxd49BR76B7tpnu0m55iD8PlYYYrw+S9PPlKnpHKCH3lPkYq6/FWjPJs7mGC3C4E1x5ghGSYIkmKZR9bzm3rbyPtpEnbadJOmlwiFwWViRqStnZMWE2DvX0kyOfp/9736P/u9wgLBeovvJCWKy7Had7xuHlD3Vt57v57eea+P/LhV56B6e6klKthqG0uo7X1mP30XzmllFIHP2NZE0NJTOGqT3+Ry//lCuwgiAPBeB5UB4cBiXIZOwjY9jfwlRecjVm7Og7+qoLCqoBw8rqtnc3sosCx6W2oo7dhcuc/X//UZ3nj5X9L/WiZ+lKZ2lKZulKFulKZ2lKFuoF+5nZWSPnb/8C+ZMXhlG/4OcVUMp5SlJLRcimVpJxwKScSlBMuFTeeJ1wqrjse0P/+zlX77DVXJEEnzXTK9r+jrr7y0zSlDEsahCX1FksaLJbUV1hSn2dJwxZeVGdxhjNxh4bGsGXEsG4g5OnhFAuOP4G1Zi6bTAsdYRPdQT3PlOdxX3B4VQ+h4ODT7IzQbA/T7AzTYOfJHlOhayRH2vXIuB7ugd5OcJpYYu31GHwfveHTGBPiExDg4xsPHz9aDn1CH8yoDWUL3wupmDIlp0C2tYnB7hE63a2UrSKGye2YE3aC+mT9+NSYaqQl3ULWzR6S4wZqsDfNwtFR+n/0I/qvvY5gaIjcWS+n5fIrSB2+Yru0/Vs6eO7+e3n2/nvpXhf1zNW6+DBueexpjrv4DQQzOR6PUkqpQ56xLHzLwnefZziguC2hXTXde/udnHHmSyZtc7wKdhBs18nMmNCyCGyb977sRTRv6ZgICuMSxHAsMHScaMiCQ/CH2/Npq2uEOigSTVt3kMb1PNLlCulShXS5TLpc4dHb7ubM015MulQiXSpTN5JnTk8fqXJ5ys8LopLDSsLl/CVLSfmrqbguFdfGc5xoch0qjj2+HG23qYyv21xzn43t+wT2HlYRNiFv/ti/jq8WgafiCUCMoY48jWaYJuKSwboh2uuGWTy4ifnefZP+WRFaQslJMmRl2cAc1jGHTbSxOWymM2xka1DP+mIr+TBN86texJ2rq95bOyDjVsi4HplEFABm3AopN+491PVIOz62NfV7OhuEJuT+m9fs8fFjvZCKWLhYuLggEz2vYhFFKdUtlgKLxEiWP379/3jjG94OnWlCTyi5efLN3Yy0dDHS0MNwupehYICOfAfPDjw7fnjKTtGSaaHt9W3csf4Ojmo6inm5eQd9AKjB3jTr/vKXGfj+D8i+9HRaLr+C9DFHj+8zxtC7cT3PPfAnnnvg/+jduB6A9mWHc/rfvJ3lJ59KfdscLvn8VbQMds/QK1BKKaV2U9zeL3QcxobC/uNz61l54QU7Th6G4yWG2waJVuDjByFOpUIyCLB2UGoIE9VJxwNAa2L5pSuWkB0aGi8xHNuundBEPNfFc12Gc9nxbd/s7yNx4nHbJzaGZKVCouKR9LxoXqlMbKt4JCoV1j3wZ45Np0h6HjWFCq4fkPB9XM/HeZ4OZN68bDn85GcEAhXbpuzYVBybsh3Pq5fjecW2onS2zWnZLO1be6iMBZOui+c4+E4UPJqqAeXXMm/Sta/+yqdJWCEL6yyW1EtcKhiXDDYUOKy+h1Oyk0ss84Fh/WDIM4MOqwr1bJ47l01hE900MiSN9NsNDDiNiNsAdsN4W7NJb2swyty/+ybX3TxIMDpIWBggKAxGU3GIsJgnLA4TlvIExeEdDzK/t4PJ783QDzNRCG+HVOpH+O1jN/K6756OCcHqTZHpyJDZMofWp5eAF9eEayjD3FEq7QP0N26mL4yqlvaM9tB8XjMfuOcDAPh5n+KaIoVnChSeKVBaX8IEUwfis3FAdg32plnTO95B7XnnkTkuemCWR0fZ+OSjrHvkIdY99jD5vl4QYd7hR3Lm2/6OZSe9mNrm7Xus2u+Nw5VSSqn95PlKDL95/Y1c8fH3x4knSg2t8eqkURA4No2VGlqlaPnVK4+CrV3bXxcmgsPtAkWL0LL5q/lzSBYK0X7LmqheeqgGiSKUk0nKySQjO0l29c2/4opL37TjU4ThpODP9X1cP8D1fRKez72/uoNXnPVSEuP7fFwvSp/1feqLFRJxetcPtutg6I3zF8Add293XQMTpYlxADgREEbzTHMzR7/0JVXpHCqOy3rX4TnHwcNByiE1bpFau0Cj5GlKDdHUNswprcOcXe6hJjkw6boeFkOSYog0vWTYSCtbaKSLBnrCOgaopWAy/PnZHlqPPRk/sPFDa1InMtu9hxgcK8Qem8TQcf/vuOexelKJkHQyIJ0ISSVCEm5IwjEk3WjZmaol0F4M/XAgdJIiFtBaiqbj+jEh0JOCLZloeraWxKoG5rCUOY0lmDsK80b50BWX8c+/eT+9xV66R7vpaupi4NjoM3TEoS3bRnu2nbm5ubRmWnGtiefUbByQXYO9aea0tTFUKbHq5p+y7pGH6HjmKcIgIJFOs+gFx7H49W9i6fEnkmtonOmsKqWUUge+qlJD2IWOF4zhm5+7mksvezt2EGKHYRQQhiF2GMTb4nXPJxEvj4Vyl7z4hdAxeegFQxSgVgeFU82PbG9l/R8eo2yESkg8F6Khvw/sgHFfDTlhLItKwqKS2HFw/5PBQVqO2PUSJisMJwJGz+fn3/4Bb37n23A9n4TvTQSU43Nv0nq2WMIdibYvrKsj88RTu1xQ5YtQcWxGbJs+p46uHh/fDXAyAW42IJUNyWZDanJF6mpGWZ7r4+RcVUlQfAt0FwybWkO2lm5hS5ChO0yzNcjQETbQY2rpNTUMk8VIErESIAkCywVJgJVAJEFm2cnc91QWkZ3/nDehB2EJE5YgLGLCIoQlml/9Qe7fOB/XDnHtANcKcawA25oILLebS4hjGbBkvwwIv0Ni7VLQZVk2C5tXsHzusayYexyHbT2GxJONfPYtP8O6tULLvAJHzh2FpUWKkqer0MWWwhY6C508tPUh2Bq1TWzLtDE/N595NfNgFnahocHeNPv9j77DQ7f8DIDmhYt54asuYMlxJzB3xZHYjr7dSiml1D4lQtn3kXSWENil7jKMQYzBCgNu/Nb3ees7/wYrDLCCcIdziauZWkGAFYaT2rS987QTgcL2lyBqkxgFjTue3z+0gvrenrhE0cJIVRqxMJZE6xJtMyLTWuI4W2oVhZYVdxYTrT9eKvHS9rY9OtfVH7+SKz72TzhBMF7aGAWFccmjH+D4flQqGQeX4+l8n/zWXua3LSXhx1VaR32SWzzcMMQjajPZbRmcdBQMupmJwPDwTMAx3lacTIDtbl91MDRQ8lxG/SQFP8lImGY4iKZBK8u99/yeEy+9iL5UPT1uPUPkKAYJ/NDGCyy8sXlgx8s2XlCHHzbgBTZbK4NsHqrDC+ydliruyKJ/vgVbpggIx4JFmbzPseO5FfLrr93A5qHa8XXXCibt32mbRhNy/peu3a38bgA2hp3Ul5L0/PQOzn/tRfBEAzzWBGJIt5RYMm8hS+aOwpIiFatEV6GLjnwHHfkOHtz6IA9ufZAjvnQEQRhgz6LxrTX6mGYrTjmVhvZ5LFn5QmqadtzzplJKKaUOIHG7rsCy6BoaobzN8AfPKwyjoC8IuOm6H3Lx314cBYYmROJ9U859Pzo2DHnZEcuweravfjoVQzQ+YiBCKPDBc06nYcMGjCVxQDgWHE6sm/i1Itb4shHh6LltpAr5KF3V9rGA0ljWpPWDigi+4+A7DsVdKT2uctVd9/Dqd/z9dtutMCTpeyR8P2rr6PskfY9k1fram25h5ZlnkfA8MmGJWilSa49SY5fJ2qW4Y5cKqYRPS3KYualBnFSAxHHGxa93YeAXk67re1AqC8WSkC/BSBlGyoahsmGgZOgvh/SWQrpLIVtLhr5SEO2vWIyYNAWToCJJxE1iuUnEiZbFTSFO1bZECnESE9vG0yYmp3PHzhWlH9Pymn/mnrVTv68m8Am9EqZSxHglwlKBsJQnLOVpOOvdbB3JTarSOmluhTscIsNYMJApc9vDP+DV152C8QW2pqEjE02PNcEjzWAZEm1FFs6dx8K5R8KSImUpsiW/he/8x3ewL589gR5osDft2pcdTvuyw2c6G0oppZTaX+LSudBx2NQ/RDmTff5jdmBsuAsJxwJFMz4XE2LFczEmThPNo1LJkK3DHdS1z4nSBgHib3NMPO3I219yAnR07FI+DVQFgxYI/Mt5Z1C/fh2GsYCQiUARwQjjQTXx+tjya1YeSX1Pd9W+qvTE5xDG9yHE26P5irZmUvnhqmvLpHNRda1J+wVsS/bJeJChZVFMJCkmpg4ebxno59XHHLtrJzQGNwhIlis0eAVavGG6/ve/OfVV51Ln56k1BWrDUbIySkbKpKwKzZkK7bU+thNiOWC5IGKz87qIHmHoUwnLlI1DybiUSDBKgoIkKVgp8laKx++9j3lnvY6ihBQJo/n4ZChhKIphVMCXABjFGAiNEBrhzs98gpd++D8IjBCG1vj27acsockRhK0EoUVghGK+SHchy86qRVsyUUo4abIDMitezKbuJNlUQKalSHLuKHISGE+gM27v15GBh5vgz1Hwl2wusWTOQhaueQATGmQWDbiowd4+ILYD4S4MEqmUUkrtY6VSaaazcMjZm/fcjFXb3INjf3D9DVzx8jOe5wJxwFcV/IkJ+ck3v8/fXPqWSUGhmDBKF5rtto8vhwYwbFqzgWx7G5ioM5Hx64QGi7DqmgBVgaeBk5YsIDE4GG3bg9f9rpeeDGuf24Mj4XMXvhKee3Z8pDYjEi3HcxMHgdXz6v3/cOYpyGP3R4EJUdXLaL7NOkJoIIi3GYRzjlmBeeDPcZqJ9IEBHyEwEBjBr1r3x7YBA10+tzzq4JsGAtMwZbvQpAmpDX3qjEereMyxy7RaZWp7VjFn3mIaHI9626fW9sg5PlknIO0E2G5I2qmQdcu0OCGWY7Bcg+WEnPeiJFbp1l16jyuhkPcdCoHDSOCQ923e/8oBcpVvUyRJUZKMSoqClaYgE1NeUpPW/bht4q8+cyl1p74Zsd24dDERlywmJkobE6moNHKsVDGRQpw0IkLL6z7KT+6ayF/olQjyAwT5foL8+njej1UqMCfZwMJsG8vrl7J0yzIuPPWyA73p7XY02NsXwmC36xJX+9U/XTqNmVFKKXUoS+iYrfvdAf2ej5d6MWko6s6hESrp9I6P2QU/vvYnXPGKl+3RsVd9+otc9pF/jFbiksfqUshoHgeKYwGhMXFQCT/9/o1c+NY3TE43HlCy3bbxcxrDfffcy4tOf/HEOauvH59/fL7d+SEMDfPr3MnpxoPd6m3h5LwbMEcuw5LyHr1nAJz/ctimj9SJUtftq+MacTDigmQxIjztFfCWHMHG8Wq68TFWVKLqhCEJLyBR9kkWAhKeTzKe1v75MY475nCyYZm0XyJjyqRNhYR4k4LCaB5NOdeQSwRYiQC/Jok7sBrbjdryubaPJTsfe7CMS0FSdP1DlnTrnxklOR4sFrdZHiXJKKkokIz3BVhUjMt3r76eMy7/AJ5x8EMHL+lQyaTxWhbjhcuohA5hXPpZBJ6JJ4sAv6ePj82yaswHXLAnIucCXyEqY77WGPPZGc7SHvnD/9z1/ImUUkoptUMzWSI5m0tD9ybve/u6p+V927YEbRds6BugvIeB6l1Pr+GoC169R8cCfON7N3HZR160R8d+9T+/zLs+8B6AiVLNOMgcC1TFmKi6bvUU7//9bXdx1nlnTa6iu23a7aZwvOrv3PpakoXC5GtMkVcvngoO4FikX3IcfwEgM95brRELAdzQ4IQm6sgmCElUoqE33DDECULcICC/tYd0IksyDEmFASkTkBGfRMJguSF2IsR2DdakeYiVCJjnZrG6+rESBjthsN0Q2wmwnqePmXLoUA4TfPD1Bfzh/8eISTASJhgySQZNkiGSDJBk2CTpNxkGTB1DUsMINXiSJrSSDK5/fI8+65l0QAV7ElUk/hrwCmAz8KCI3GyMeWpmc7b7Ljlv2R4fe/W905gRpZRSahaaydKxA7pk7nnsTd739nXP5vdtpiR20qZvzFRVeh9ct5mT6/a8B9Wv/vDnEyWqEJeqmrid51hgGPdUW9V2VIzhnlt/y8tefV7cljQOIMOJ9qTR8rb7JpattrrtM2QMdmhw4qDQCcIoQPRC7HKIE0Z5w/NIWDZ2nD6aAhISkrR9krZPwvJIOAEJJ4gCxaqgsT2Rwh7NTwSViajkcWf8wMLzbPqWB8C39/g9nwkHVLAHnASsNsasBRCRG4DXArMu2BscGtyr42fzfxWVUkoppdQsE3diE+5CZ5P3rd3I8l0IVKfywy9cxUVXvB0rLlXcfgJ7in1rH1nFimOPwjImSsP2x44t22EUJNphiF00OIWo0yI73mfH+1wTkLJ8Uo5PUnySTkDC9knYUcDoOj6uExKGu1refOA40IK9eUDVyJNsBk6eobzsldzudtu8Df3vmFJKKaWUOlDtzW9dA7Rmd1C6t5P0QTz98tGbuOL8c3fjYmZSqeOPv/Fd3vqut050PDRe6hgFiGzTe61UlXQ+8cQqbtrN1zrTxEzRBe9MEJELgXONMZfG628FTjbGXFaV5l3Au+LVw4naTE6XZqB3Gs+nZie9DxTofaAieh8ovQcU6H2gIgfqfbDIGNOyox0HWsleB7Cgan1+vG2cMeZbwLf2xcVF5CFjzAn74txq9tD7QIHeByqi94HSe0CB3gcqMhvvg+fpt2a/exBYLiJLRCQBXAzcPMN5UkoppZRSSqlZ54Aq2TPG+CJyGXA70dAL1xtjVs1wtpRSSimllFJq1jmggj0AY8ytwK0zdPl9Uj1UzTp6HyjQ+0BF9D5Qeg8o0PtARWbdfXBAddCilFJKKaWUUmp6HGht9pRSSimllFJKTQMN9mIicq6IPCMiq0XkwzOdH7V/iMh6EXlCRB4VkYfibY0i8hsReS6eN8x0PtX0EpHrRaRbRJ6s2rbDz10iV8XPhsdF5PiZy7maTlPcB58UkY74mfCoiLyyat9H4vvgGRE5Z2ZyraabiCwQkbtF5CkRWSUi74u36zPhELKT+0CfCYcQEUmJyAMi8lh8H3wq3r5ERO6PP+8b444kEZFkvL463r94JvO/IxrsASJiA18DzgOOAt4kIkfNbK7UfnSmMWZlVVe6HwbuNMYsB+6M19XB5bvAtiOyTvW5nwcsj6d3AV/fT3lU+9532f4+APhS/ExYGbcjJ/5OuBg4Oj7mmvi7Q81+PvABY8xRwCnAe+PPW58Jh5ap7gPQZ8KhpAy8zBhzLLASOFdETgE+R3QfLAMGgHfG6d8JDMTbvxSnO6BosBc5CVhtjFlrjKkANwCvneE8qZnmuxINAAAHF0lEQVTzWuB78fL3gAtmMC9qHzDG/B7o32bzVJ/7a4Hvm8h9QL2ItO+fnKp9aYr7YCqvBW4wxpSNMeuA1UTfHWqWM8Z0GmMejpdHgKeBeegz4ZCyk/tgKvpMOAjFf9f5eNWNJwO8DPifePu2z4Ox58T/AC8XEdlP2d0lGuxF5gGbqtY3s/M/cHXwMMAdIvJnEXlXvK3NGNMZL3cBbTOTNbWfTfW56/Ph0HNZXD3v+qpq3HofHALiKljHAfejz4RD1jb3Aegz4ZAiIraIPAp0A78B1gCDxhg/TlL9WY/fB/H+IaBp/+Z45zTYU4e6lxhjjieqlvNeETm9eqeJuqvVLmsPMfq5H9K+DhxGVH2nE/ivmc2O2l9EJAf8FPhHY8xw9T59Jhw6dnAf6DPhEGOMCYwxK4H5RKW1R8xwlvaKBnuRDmBB1fr8eJs6yBljOuJ5N/Bzoj/qrWNVcuJ598zlUO1HU33u+nw4hBhjtsZf9CHwbSaqZel9cBATEZfoB/6PjDE/izfrM+EQs6P7QJ8Jhy5jzCBwN/AiouraY+OTV3/W4/dBvL8O6NvPWd0pDfYiDwLL4552EkQNbm+e4TypfUxEsiJSM7YMnA08SfTZvy1O9jbglzOTQ7WfTfW53wxcEvfAdwowVFW1Sx1ktml79TqiZwJE98HFcc9rS4g653hgf+dPTb+4fc11wNPGmC9W7dJnwiFkqvtAnwmHFhFpEZH6eDkNvIKo/ebdwIVxsm2fB2PPiQuBu8wBNoi58/xJDn7GGF9ELgNuB2zgemPMqhnOltr32oCfx+1oHeDHxpjbRORB4L9F5J3ABuANM5hHtQ+IyE+AM4BmEdkM/BvwWXb8ud8KvJKo8f0o8Pb9nmG1T0xxH5whIiuJquytB94NYIxZJSL/DTxF1Gvfe40xwUzkW027U4G3Ak/E7XQAPoo+Ew41U90Hb9JnwiGlHfhe3LOqBfy3MeZXIvIUcIOIfBp4hOgfA8TzH4jIaqIOvy6eiUzvjBxgwadSSimllFJKqWmg1TiVUkoppZRS6iCkwZ5SSimllFJKHYQ02FNKKaWUUkqpg5AGe0oppZRSSil1ENJgTymllFJKKaUOQhrsKaWUOuCJiBGRH1atOyLSIyK/msl87QoRye+Dc64UkVdWrX9SRD443ddRSik1u2mwp5RSajYoAMfEg9xCNNBtxwzmZ6atJBrrTSmllJqSBntKKaVmi1uBV8XLbwJ+MrZDRLIicr2IPCAij4jIa+PtR8fbHhWRx0VkeZz21yLymIg8KSJvjNP+q4g8GG/7lohIvP3E+NhHReQLIvJkvN2O1x+M97/7+V6AiPxzVfpPxdsWi8jTIvJtEVklIneMBbU7uraIJIArgTfG298Yn/4oEfmdiKwVkSum4w1XSik1u2mwp5RSara4AbhYRFLAXwH3V+37GHCXMeYk4EzgCyKSBf4e+IoxZiVwArAZOBfYYow51hhzDHBbfI6vGmNOjLelgfPj7d8B3h2fI6i65juBIWPMicCJwN+JyJKpMi8iZwPLgZOISuZeKCKnx7uXA18zxhwNDAKvn+raxpgK8K/AjcaYlcaYG+O0RwDnxOf/NxFxd/puKqWUOuhpsKeUUmpWMMY8DiwmKtW7dZvdZwMfFpFHgd8BKWAh8CfgoyLyL8AiY0wReAJ4hYh8TkROM8YMxec4U0TuF5EngJcBR4tIPVBjjPlTnObH21zzkvia9wNNREHbVM6Op0eAh4mCs7H064wxj8bLfwYWP8+1d+TXxpiyMaYX6Abanie9Ukqpg5wz0xlQSimldsPNwP8DziAKrsYI8HpjzDPbpH9aRO4nqv55q4i82xhzl4gcT9Tm7dMicifweeAa4ARjzCYR+SRRwLgzAlxujLl9F/MuwH8aY745aaPIYqBctSkgKlncXdueQ7/jlVLqEKcle0oppWaT64FPGWOe2Gb77cDlVe3sjovnS4G1xpirgF8CfyUic4FRY8wPgS8AxzMR2PWKSA64EMAYMwiMiMjJ8f6Lt7nme8aqS4rIirjq6FRuB94Rnx8RmScirVMlfp5rjwA1O7mWUkoppf/1U0opNXsYYzYDV+1g178DXwYeFxELWEfU5u4NwFtFxAO6gM8Qta/7goiEgAe8xxgzKCLfBp6M0z1Yde53At+O098DjFX7vJaoWunDcZDZA1ywk7zfISJHAn+KY9I88BYmtwPc1lTXvpuJaqv/uZPjlVJKHcLEGDPTeVBKKaUOWCKSM8bk4+UPA+3GmPcd7NdWSik1+2nJnlJKKbVzrxKRjxB9Z24A/vYQubZSSqlZTkv2lFJKKaWUUuogpB20KKWUUkoppdRBSIM9pZRSSimllDoIabCnlFJKKaWUUgchDfaUUkoppZRS6iCkwZ5SSimllFJKHYQ02FNKKaWUUkqpg9D/B7YuVc4jKWYLAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1080x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 624
        },
        "id": "JNdb2XDBrpqy",
        "outputId": "d54099e3-eac1-44e3-c161-bc5c1ad6e352"
      },
      "source": [
        "plt.figure(figsize=(15,10))\r\n",
        "sns.histplot(data=df_test, hue='Emotion', x='Message length', stat='count', multiple='layer', palette=palette,kde=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f5ce5e245c0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 624
        },
        "id": "UUmXazzmvEbt",
        "outputId": "d7e703a5-3f3e-4c23-f86d-ff70c97d150f"
      },
      "source": [
        "plt.figure(figsize=(15,10))\r\n",
        "sns.histplot(data=df_val, hue='Emotion', x='Message length', stat='count', multiple='layer', palette=palette,kde=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f5ce3cda400>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XvXlwCLZv2O7"
      },
      "source": [
        "## Text Pre-processing\r\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZIpU_9g9xEfd"
      },
      "source": [
        "The objective now is to convert the messages in our datasets into numerical features which will allow us to use machine learning algorithms to perform classification. Bag-of-words approach will be used through CountVectorizer method from Scikit learn library. But first, we need to clean our messages by removing words which are not important in a way they are too common and won't help our algorithms to make good predictions, and also by removing punctuation."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mMPbiuJn28My",
        "outputId": "9d6e8c2e-7245-4faf-e46d-b51653555b81"
      },
      "source": [
        "import string\r\n",
        "import nltk\r\n",
        "from nltk.corpus import stopwords\r\n",
        "\r\n",
        "nltk.download('stopwords')"
      ],
      "execution_count": 673,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 673
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rf2T-RabvNZV"
      },
      "source": [
        "def preprocess_message(message):\r\n",
        "  STOPWORDS = stopwords.words('english')\r\n",
        "  no_punc = [c for c in message if c not in string.punctuation]\r\n",
        "  no_punc = ''.join(no_punc)\r\n",
        "  words = [word for word in no_punc.split() if word not in STOPWORDS]\r\n",
        "  return ' '.join(words)"
      ],
      "execution_count": 642,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "peN-fpSq2qA_"
      },
      "source": [
        "df_train['Cleaned message'] = df_train['Message'].apply(lambda x: preprocess_message(x))\r\n",
        "df_test['Cleaned message'] = df_test['Message'].apply(lambda x: preprocess_message(x))\r\n",
        "df_val['Cleaned message'] = df_val['Message'].apply(lambda x: preprocess_message(x))"
      ],
      "execution_count": 643,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "id": "jF0yWJi03fDU",
        "outputId": "ec9af719-a08f-4df7-8ff7-6c04e43383d1"
      },
      "source": [
        "df_train.head()"
      ],
      "execution_count": 644,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Message</th>\n",
              "      <th>Emotion</th>\n",
              "      <th>Message length</th>\n",
              "      <th>Cleaned message</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>i didnt feel humiliated</td>\n",
              "      <td>sadness</td>\n",
              "      <td>23</td>\n",
              "      <td>didnt feel humiliated</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>i can go from feeling so hopeless to so damned...</td>\n",
              "      <td>sadness</td>\n",
              "      <td>108</td>\n",
              "      <td>go feeling hopeless damned hopeful around some...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>im grabbing a minute to post i feel greedy wrong</td>\n",
              "      <td>anger</td>\n",
              "      <td>48</td>\n",
              "      <td>im grabbing minute post feel greedy wrong</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>i am ever feeling nostalgic about the fireplac...</td>\n",
              "      <td>love</td>\n",
              "      <td>92</td>\n",
              "      <td>ever feeling nostalgic fireplace know still pr...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>i am feeling grouchy</td>\n",
              "      <td>anger</td>\n",
              "      <td>20</td>\n",
              "      <td>feeling grouchy</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                             Message  ...                                    Cleaned message\n",
              "0                            i didnt feel humiliated  ...                              didnt feel humiliated\n",
              "1  i can go from feeling so hopeless to so damned...  ...  go feeling hopeless damned hopeful around some...\n",
              "2   im grabbing a minute to post i feel greedy wrong  ...          im grabbing minute post feel greedy wrong\n",
              "3  i am ever feeling nostalgic about the fireplac...  ...  ever feeling nostalgic fireplace know still pr...\n",
              "4                               i am feeling grouchy  ...                                    feeling grouchy\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 644
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sMlFXbnu-OCL"
      },
      "source": [
        "We can have an idea of the words which are the most used in the messages depending on the emotion, using Counter method from collections library :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vGvzbCumyh8m",
        "outputId": "644f3fcc-2491-4f58-d3c9-b4e910b0ecc0"
      },
      "source": [
        "from collections import Counter\r\n",
        "\r\n",
        "words_counter = Counter()\r\n",
        "\r\n",
        "words = df_train.loc[df_train['Emotion']=='anger', 'Cleaned message'].apply(lambda x: [word.lower() for word in x.split()])\r\n",
        "\r\n",
        "for msg in words:\r\n",
        "  words_counter.update(msg)\r\n",
        "\r\n",
        "# Top 50 words observed in messages with Anger emotion\r\n",
        "print(words_counter.most_common(50))"
      ],
      "execution_count": 650,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[('feel', 1459), ('feeling', 721), ('like', 384), ('im', 342), ('really', 124), ('get', 121), ('time', 118), ('know', 115), ('little', 105), ('people', 105), ('want', 98), ('even', 95), ('would', 89), ('angry', 85), ('way', 85), ('think', 81), ('things', 80), ('one', 79), ('could', 79), ('bit', 77), ('dont', 70), ('irritable', 70), ('still', 70), ('cold', 68), ('life', 66), ('something', 65), ('back', 64), ('cant', 64), ('greedy', 63), ('offended', 62), ('ive', 62), ('going', 61), ('bothered', 61), ('fucked', 60), ('resentful', 60), ('selfish', 59), ('irritated', 57), ('much', 56), ('make', 56), ('say', 56), ('annoyed', 56), ('violent', 55), ('cranky', 55), ('insulted', 55), ('hated', 55), ('mad', 55), ('dissatisfied', 54), ('dangerous', 54), ('frustrated', 53), ('rude', 53)]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vvhvz-F5dN9E"
      },
      "source": [
        "Words which are the most used in all the messages :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L5LM2-o4dTkg",
        "outputId": "98b75817-af34-4103-f513-94583da787f2"
      },
      "source": [
        "words_counter = Counter()\r\n",
        "\r\n",
        "words = df_train['Cleaned message'].apply(lambda x: [word.lower() for word in x.split()])\r\n",
        "\r\n",
        "for msg in words:\r\n",
        "  words_counter.update(msg)\r\n",
        "\r\n",
        "# Top 50 words observed in messages with Anger emotion\r\n",
        "print(words_counter.most_common(50))"
      ],
      "execution_count": 651,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[('feel', 11183), ('feeling', 5112), ('like', 2908), ('im', 2430), ('really', 942), ('know', 853), ('time', 794), ('get', 740), ('little', 736), ('people', 654), ('would', 646), ('want', 644), ('think', 596), ('one', 596), ('still', 595), ('ive', 587), ('even', 568), ('life', 551), ('make', 523), ('bit', 516), ('something', 513), ('love', 500), ('much', 496), ('going', 487), ('dont', 482), ('things', 480), ('way', 470), ('could', 453), ('day', 428), ('back', 403), ('go', 393), ('good', 393), ('pretty', 392), ('need', 387), ('always', 355), ('see', 355), ('right', 346), ('feelings', 343), ('also', 337), ('say', 333), ('today', 327), ('feels', 323), ('work', 322), ('around', 315), ('cant', 301), ('made', 296), ('well', 275), ('though', 274), ('didnt', 272), ('help', 272)]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p5xyRImw-ssN"
      },
      "source": [
        "We can also generate a Word Cloud which is a nice visual tool :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        },
        "id": "KG8UjP3M7fWR",
        "outputId": "bfd17aad-6ed7-4c55-b931-6c5e7b7ae21a"
      },
      "source": [
        "from PIL import Image\r\n",
        "from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator\r\n",
        "\r\n",
        "# Concatenate all the messages:\r\n",
        "text = ' '.join(message for message in df_train.loc[df_train['Emotion']=='anger', 'Cleaned message'])\r\n",
        "\r\n",
        "# Setting stopwords\r\n",
        "stopwords = set(STOPWORDS)\r\n",
        "stopwords.update([\"feel\", \"feeling\", \"know\",\"like\",\"really\",\"thing\",\"make\",\"want\",\"even\",\"going\",\"one\",\"think\",\"time\",\"still\",\"way\",\"things\",\"people\"])\r\n",
        "\r\n",
        "# Create and generate a word cloud image:\r\n",
        "wordcloud = WordCloud(stopwords=stopwords).generate(text)\r\n",
        "\r\n",
        "# Display the generated image:\r\n",
        "plt.imshow(wordcloud, interpolation='bilinear')\r\n",
        "plt.axis(\"off\")\r\n",
        "plt.show()"
      ],
      "execution_count": 653,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QEhr_Uoh_Lls"
      },
      "source": [
        "Let's generate another Word Cloud but this time gathering all the emotions :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        },
        "id": "UQFbEqqf7-PH",
        "outputId": "41dfb4d6-88eb-4046-d2ed-ad70d2c296b5"
      },
      "source": [
        "text = ' '.join(message for message in df_train['Cleaned message'])\r\n",
        "\r\n",
        "stopwords = set(STOPWORDS)\r\n",
        "stopwords.update([\"feel\", \"feeling\", \"know\",\"like\",\"really\",\"thing\",\"make\",\"want\",\"even\",\"going\",\"one\",\"think\",\"time\",\"still\",\"way\",\"things\",\"people\"])\r\n",
        "\r\n",
        "wordcloud = WordCloud(stopwords=stopwords).generate(text)\r\n",
        "\r\n",
        "plt.imshow(wordcloud, interpolation='bilinear')\r\n",
        "plt.axis(\"off\")\r\n",
        "plt.show()"
      ],
      "execution_count": 656,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "asaIjXzof0jC"
      },
      "source": [
        "Looking at the distribution of messages containing the same specific word can help us to decide whether or not we should add this word into the stopwords list :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cvSatqK4_iaU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "51e9573e-7b32-43f8-c7aa-085f61bcdad8"
      },
      "source": [
        "df_train.loc[df_train['Cleaned message'].str.contains('feeling'), ['Emotion','Cleaned message']].groupby('Emotion').count()"
      ],
      "execution_count": 657,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Cleaned message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>737</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>753</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>1592</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>415</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>1557</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>216</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          Cleaned message\n",
              "Emotion                  \n",
              "anger                 737\n",
              "fear                  753\n",
              "joy                  1592\n",
              "love                  415\n",
              "sadness              1557\n",
              "surprise              216"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 657
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FZVy5tmMiD-q",
        "outputId": "c7479399-9e10-4237-bb67-bde8417282dc"
      },
      "source": [
        "for i in df_train.loc[df_train['Cleaned message'].str.contains('feeling'), ['Emotion','Cleaned message']].groupby('Emotion').count()['Cleaned message']:\r\n",
        "  print(i)"
      ],
      "execution_count": 658,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "737\n",
            "753\n",
            "1592\n",
            "415\n",
            "1557\n",
            "216\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "id": "TwupbejWfrCc",
        "outputId": "6115c146-f641-4cf8-aef9-c4923b09cd15"
      },
      "source": [
        "df_train.loc[df_train['Cleaned message'].str.contains('day'), ['Emotion','Cleaned message']].groupby('Emotion').count()"
      ],
      "execution_count": 659,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Cleaned message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>142</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>108</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>423</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>72</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>368</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>35</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          Cleaned message\n",
              "Emotion                  \n",
              "anger                 142\n",
              "fear                  108\n",
              "joy                   423\n",
              "love                   72\n",
              "sadness               368\n",
              "surprise               35"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 659
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "id": "fxgsgTfVf8t8",
        "outputId": "87300387-daef-4161-deae-5950cd731051"
      },
      "source": [
        "df_train.loc[df_train['Cleaned message'].str.contains('little'), ['Emotion','Cleaned message']].groupby('Emotion').count()"
      ],
      "execution_count": 660,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Cleaned message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>102</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>144</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>179</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>51</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>187</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>41</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          Cleaned message\n",
              "Emotion                  \n",
              "anger                 102\n",
              "fear                  144\n",
              "joy                   179\n",
              "love                   51\n",
              "sadness               187\n",
              "surprise               41"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 660
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "swnQ2MQhSad9"
      },
      "source": [
        "## Model Training & model selection"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z_ogbWuVgG7H"
      },
      "source": [
        "Now we can start building our model to perform classification.\r\n",
        "Using CountVectorizer, we are going to convert a collection of text documents (our cleaned messages) to a matrix of token counts."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1-fAqLX-utMr"
      },
      "source": [
        "X_train = df_train['Cleaned message']\r\n",
        "y_train = df_train['Emotion']\r\n",
        "\r\n",
        "X_test = df_test['Cleaned message']\r\n",
        "y_test = df_test['Emotion']\r\n",
        "\r\n",
        "X_val = df_val['Cleaned message']\r\n",
        "y_val = df_val['Emotion']"
      ],
      "execution_count": 661,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LNEu0SrZa6_t"
      },
      "source": [
        "For the moment, let's instantiate a CountVectorizer object without defining the additional words we don't want to include in our model. Later, we are going to see how much we can improve our model by tuning it upstream, and including more stopwords is part of the process."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MZ6SEFj4vWQj"
      },
      "source": [
        "from sklearn.feature_extraction.text import CountVectorizer\r\n",
        "\r\n",
        "vect = CountVectorizer()\r\n",
        "X_train_dtm = vect.fit_transform(X_train)\r\n",
        "X_test_dtm = vect.transform(X_test)\r\n",
        "X_val_dtm = vect.transform(X_val)"
      ],
      "execution_count": 662,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 243
        },
        "id": "1bbSSzs4wVtb",
        "outputId": "da1e6794-e8a3-4b99-d882-7bb897f69efe"
      },
      "source": [
        "pd.DataFrame(X_train_dtm.toarray(), columns=vect.get_feature_names()).head()"
      ],
      "execution_count": 663,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>aa</th>\n",
              "      <th>aaaaaaand</th>\n",
              "      <th>aaaaand</th>\n",
              "      <th>aaaand</th>\n",
              "      <th>aac</th>\n",
              "      <th>aahhh</th>\n",
              "      <th>aaron</th>\n",
              "      <th>ab</th>\n",
              "      <th>abandon</th>\n",
              "      <th>abandoned</th>\n",
              "      <th>abandoning</th>\n",
              "      <th>abandonment</th>\n",
              "      <th>abated</th>\n",
              "      <th>abbigail</th>\n",
              "      <th>abc</th>\n",
              "      <th>abdomen</th>\n",
              "      <th>abdominal</th>\n",
              "      <th>abducted</th>\n",
              "      <th>abelard</th>\n",
              "      <th>abhorrent</th>\n",
              "      <th>abide</th>\n",
              "      <th>abilities</th>\n",
              "      <th>ability</th>\n",
              "      <th>abit</th>\n",
              "      <th>able</th>\n",
              "      <th>ableness</th>\n",
              "      <th>abnormally</th>\n",
              "      <th>aboard</th>\n",
              "      <th>abominable</th>\n",
              "      <th>abortion</th>\n",
              "      <th>abou</th>\n",
              "      <th>abound</th>\n",
              "      <th>abraham</th>\n",
              "      <th>abroad</th>\n",
              "      <th>abruptly</th>\n",
              "      <th>absence</th>\n",
              "      <th>absolute</th>\n",
              "      <th>absolutely</th>\n",
              "      <th>absolutly</th>\n",
              "      <th>absorbed</th>\n",
              "      <th>...</th>\n",
              "      <th>yunhos</th>\n",
              "      <th>yup</th>\n",
              "      <th>yuuki</th>\n",
              "      <th>zach</th>\n",
              "      <th>zack</th>\n",
              "      <th>zackmdavis</th>\n",
              "      <th>zambia</th>\n",
              "      <th>zap</th>\n",
              "      <th>zara</th>\n",
              "      <th>zay</th>\n",
              "      <th>zayne</th>\n",
              "      <th>zealand</th>\n",
              "      <th>zen</th>\n",
              "      <th>zenden</th>\n",
              "      <th>zendikar</th>\n",
              "      <th>zenos</th>\n",
              "      <th>zero</th>\n",
              "      <th>zest</th>\n",
              "      <th>zhu</th>\n",
              "      <th>zico</th>\n",
              "      <th>zimbabwe</th>\n",
              "      <th>zimbabweans</th>\n",
              "      <th>zimmerman</th>\n",
              "      <th>zine</th>\n",
              "      <th>zipline</th>\n",
              "      <th>zoll</th>\n",
              "      <th>zombie</th>\n",
              "      <th>zombies</th>\n",
              "      <th>zone</th>\n",
              "      <th>zoned</th>\n",
              "      <th>zonisamide</th>\n",
              "      <th>zoo</th>\n",
              "      <th>zoom</th>\n",
              "      <th>zooming</th>\n",
              "      <th>zooms</th>\n",
              "      <th>zq</th>\n",
              "      <th>zucchini</th>\n",
              "      <th>zum</th>\n",
              "      <th>zumba</th>\n",
              "      <th>zz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 15046 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   aa  aaaaaaand  aaaaand  aaaand  aac  ...  zq  zucchini  zum  zumba  zz\n",
              "0   0          0        0       0    0  ...   0         0    0      0   0\n",
              "1   0          0        0       0    0  ...   0         0    0      0   0\n",
              "2   0          0        0       0    0  ...   0         0    0      0   0\n",
              "3   0          0        0       0    0  ...   0         0    0      0   0\n",
              "4   0          0        0       0    0  ...   0         0    0      0   0\n",
              "\n",
              "[5 rows x 15046 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 663
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 243
        },
        "id": "3sjVkVjrVqOh",
        "outputId": "e06de86e-2867-4c4d-fb9d-7169f8d44058"
      },
      "source": [
        "pd.DataFrame(X_test_dtm.toarray(), columns=vect.get_feature_names()).head()"
      ],
      "execution_count": 664,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>aa</th>\n",
              "      <th>aaaaaaand</th>\n",
              "      <th>aaaaand</th>\n",
              "      <th>aaaand</th>\n",
              "      <th>aac</th>\n",
              "      <th>aahhh</th>\n",
              "      <th>aaron</th>\n",
              "      <th>ab</th>\n",
              "      <th>abandon</th>\n",
              "      <th>abandoned</th>\n",
              "      <th>abandoning</th>\n",
              "      <th>abandonment</th>\n",
              "      <th>abated</th>\n",
              "      <th>abbigail</th>\n",
              "      <th>abc</th>\n",
              "      <th>abdomen</th>\n",
              "      <th>abdominal</th>\n",
              "      <th>abducted</th>\n",
              "      <th>abelard</th>\n",
              "      <th>abhorrent</th>\n",
              "      <th>abide</th>\n",
              "      <th>abilities</th>\n",
              "      <th>ability</th>\n",
              "      <th>abit</th>\n",
              "      <th>able</th>\n",
              "      <th>ableness</th>\n",
              "      <th>abnormally</th>\n",
              "      <th>aboard</th>\n",
              "      <th>abominable</th>\n",
              "      <th>abortion</th>\n",
              "      <th>abou</th>\n",
              "      <th>abound</th>\n",
              "      <th>abraham</th>\n",
              "      <th>abroad</th>\n",
              "      <th>abruptly</th>\n",
              "      <th>absence</th>\n",
              "      <th>absolute</th>\n",
              "      <th>absolutely</th>\n",
              "      <th>absolutly</th>\n",
              "      <th>absorbed</th>\n",
              "      <th>...</th>\n",
              "      <th>yunhos</th>\n",
              "      <th>yup</th>\n",
              "      <th>yuuki</th>\n",
              "      <th>zach</th>\n",
              "      <th>zack</th>\n",
              "      <th>zackmdavis</th>\n",
              "      <th>zambia</th>\n",
              "      <th>zap</th>\n",
              "      <th>zara</th>\n",
              "      <th>zay</th>\n",
              "      <th>zayne</th>\n",
              "      <th>zealand</th>\n",
              "      <th>zen</th>\n",
              "      <th>zenden</th>\n",
              "      <th>zendikar</th>\n",
              "      <th>zenos</th>\n",
              "      <th>zero</th>\n",
              "      <th>zest</th>\n",
              "      <th>zhu</th>\n",
              "      <th>zico</th>\n",
              "      <th>zimbabwe</th>\n",
              "      <th>zimbabweans</th>\n",
              "      <th>zimmerman</th>\n",
              "      <th>zine</th>\n",
              "      <th>zipline</th>\n",
              "      <th>zoll</th>\n",
              "      <th>zombie</th>\n",
              "      <th>zombies</th>\n",
              "      <th>zone</th>\n",
              "      <th>zoned</th>\n",
              "      <th>zonisamide</th>\n",
              "      <th>zoo</th>\n",
              "      <th>zoom</th>\n",
              "      <th>zooming</th>\n",
              "      <th>zooms</th>\n",
              "      <th>zq</th>\n",
              "      <th>zucchini</th>\n",
              "      <th>zum</th>\n",
              "      <th>zumba</th>\n",
              "      <th>zz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 15046 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "   aa  aaaaaaand  aaaaand  aaaand  aac  ...  zq  zucchini  zum  zumba  zz\n",
              "0   0          0        0       0    0  ...   0         0    0      0   0\n",
              "1   0          0        0       0    0  ...   0         0    0      0   0\n",
              "2   0          0        0       0    0  ...   0         0    0      0   0\n",
              "3   0          0        0       0    0  ...   0         0    0      0   0\n",
              "4   0          0        0       0    0  ...   0         0    0      0   0\n",
              "\n",
              "[5 rows x 15046 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 664
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3wpol4btXt1T"
      },
      "source": [
        "Now we import and instantiate a Multinomial Naive Bayes model, and we are going to feed it with the training document-term matrix we created one step before.\r\n",
        "\r\n",
        "According to Sklearn's documentation, the multinomial Naive Bayes classifier is suitable for classification with discrete features (e.g., word counts for text classification). The multinomial distribution normally requires integer feature counts. However, in practice, fractional counts such as tf-idf may also work."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sFI8JLgU6-Px"
      },
      "source": [
        "### Naive Bayes Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mqxTx-6IWPzA",
        "outputId": "a6ad062a-094f-4faa-b687-c837a45c7161"
      },
      "source": [
        "from sklearn.naive_bayes import MultinomialNB\r\n",
        "from sklearn.metrics import accuracy_score\r\n",
        "\r\n",
        "nb = MultinomialNB()\r\n",
        "nb.fit(X_train_dtm, y_train)\r\n",
        "\r\n",
        "y_train_pred = nb.predict(X_train_dtm)\r\n",
        "y_test_pred = nb.predict(X_test_dtm)\r\n",
        "y_val_pred = nb.predict(X_val_dtm)\r\n",
        "\r\n",
        "print('---- Accuracy Before Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))\r\n",
        "print('Validation accuracy', accuracy_score(y_val, y_val_pred))"
      ],
      "execution_count": 665,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy Before Tuning ----\n",
            "Training accuracy 0.906375\n",
            "Test accuracy 0.7895\n",
            "Validation accuracy 0.788\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_4pjv4TJhSAs"
      },
      "source": [
        "Let's see how the model perform with additional stopwords. To do this, we are going to write a script which will automatically add the words (from X most common words, eg. 200) which exceed a certain number of occurrences (count threshold, eg. 20) in a certain number of emotions (exceed threshold, eg. 3). \r\n",
        "\r\n",
        "Just for example, if the word \"like\" appear more than 20 times in 3 different emotions, the script can add it the stopword list."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TzesSJVRuHV3"
      },
      "source": [
        "max_score = 0.5\r\n",
        "val_accuracy_scores = []\r\n",
        "most_common_ls = [185, 195, 205, 215, 225, 235, 245, 255]\r\n",
        "exceed_threshold_ls = [3, 4]\r\n",
        "count_threshold_ls = [10,11,12,13,14,15,16,17,18,19,20]\r\n",
        "for mc in most_common_ls:\r\n",
        "  for exc in exceed_threshold_ls:\r\n",
        "    for cnt in count_threshold_ls:\r\n",
        "      additional_stopwords = []\r\n",
        "      for (i, j) in words_counter.most_common(mc):\r\n",
        "        count_exceeded = [False]*6\r\n",
        "        for id, count in enumerate(df_train.loc[df_train['Cleaned message'].str.contains(i), ['Emotion','Cleaned message']].groupby('Emotion').count()['Cleaned message']):\r\n",
        "          if count > cnt:\r\n",
        "            count_exceeded[id] = True\r\n",
        "          if id == 5:\r\n",
        "            if count_exceeded.count(True) >= exc:\r\n",
        "              additional_stopwords += [i]\r\n",
        "      vect = CountVectorizer(stop_words=additional_stopwords)\r\n",
        "      X_train_dtm = vect.fit_transform(X_train)\r\n",
        "      X_val_dtm = vect.transform(X_val)\r\n",
        "      nb = MultinomialNB()\r\n",
        "      nb.fit(X_train_dtm, y_train)\r\n",
        "      y_val_pred = nb.predict(X_val_dtm)\r\n",
        "      acc_score = accuracy_score(y_val, y_val_pred)\r\n",
        "      if acc_score > max_score:\r\n",
        "        max_score = acc_score\r\n",
        "        best_most_common = mc\r\n",
        "        best_exc = exc\r\n",
        "        best_cnt = cnt"
      ],
      "execution_count": 666,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FkhZxur0_Kfb",
        "outputId": "c0d07012-914a-4391-ff82-9894318ea2e1"
      },
      "source": [
        "print(max_score)\r\n",
        "print(best_most_common)\r\n",
        "print(best_exc)\r\n",
        "print(best_cnt)"
      ],
      "execution_count": 667,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.832\n",
            "215\n",
            "3\n",
            "16\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "j2uzPdM0S--F"
      },
      "source": [
        "Let's use the best hyperparameters we found:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DQFRU6TwhiaF"
      },
      "source": [
        "additional_stopwords = []\r\n",
        "for (i, j) in words_counter.most_common(225):\r\n",
        "  count_exceeded = [False]*6\r\n",
        "  for id, count in enumerate(df_train.loc[df_train['Cleaned message'].str.contains(i), ['Emotion','Cleaned message']].groupby('Emotion').count()['Cleaned message']):\r\n",
        "    if count > 12:\r\n",
        "      count_exceeded[id] = True\r\n",
        "    if id == 5:\r\n",
        "      if count_exceeded.count(True) >= 4:\r\n",
        "        additional_stopwords += [i]"
      ],
      "execution_count": 668,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bOs2kSq_YhMf",
        "outputId": "08eedad9-348d-4cd3-aebe-3f8bb6a74580"
      },
      "source": [
        "vect = CountVectorizer(stop_words=additional_stopwords)\r\n",
        "X_train_dtm = vect.fit_transform(X_train)\r\n",
        "X_test_dtm = vect.transform(X_test)\r\n",
        "X_val_dtm = vect.transform(X_val)\r\n",
        "\r\n",
        "nb = MultinomialNB()\r\n",
        "nb.fit(X_train_dtm, y_train)\r\n",
        "\r\n",
        "y_train_pred = nb.predict(X_train_dtm)\r\n",
        "y_test_pred = nb.predict(X_test_dtm)\r\n",
        "y_val_pred = nb.predict(X_val_dtm)\r\n",
        "\r\n",
        "print('---- Accuracy After Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))\r\n",
        "print('Validation accuracy', accuracy_score(y_val, y_val_pred))"
      ],
      "execution_count": 669,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy After Tuning ----\n",
            "Training accuracy 0.9351875\n",
            "Test accuracy 0.823\n",
            "Validation accuracy 0.832\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LDn2eZVopzzT"
      },
      "source": [
        "As we see, the script is working fine, we gained between 3 and 4% accuracy!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V8BocBWQ0saG"
      },
      "source": [
        "We can also do some improvments by tuning min_df :\r\n",
        "- min_df: float in range [0.0, 1.0] or int, default=1\r\n",
        "- When building the vocabulary, ignore terms that have a document frequency strictly lower than the given threshold. (This value is also called \"cut-off\" in the literature.)\r\n",
        "- If float, the parameter represents a proportion of documents.\r\n",
        "- If integer, the parameter represents an absolute count."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Utlh3AFdM_r7",
        "outputId": "37b6c4d7-7330-4857-db98-5914a9f88808"
      },
      "source": [
        "from sklearn.pipeline import Pipeline\r\n",
        "\r\n",
        "pipe = Pipeline([('bow', CountVectorizer(stop_words=additional_stopwords, min_df=10)),\r\n",
        "                 ('model', MultinomialNB())])\r\n",
        "pipe.fit(X_train, y_train)\r\n",
        "\r\n",
        "y_train_pred = pipe.predict(X_train)\r\n",
        "y_test_pred = pipe.predict(X_test)\r\n",
        "y_val_pred = pipe.predict(X_val)\r\n",
        "\r\n",
        "print('---- Accuracy After Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))\r\n",
        "print('Validation accuracy', accuracy_score(y_val, y_val_pred))"
      ],
      "execution_count": 670,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy After Tuning ----\n",
            "Training accuracy 0.9188125\n",
            "Test accuracy 0.866\n",
            "Validation accuracy 0.8695\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uOvPpqfV6Fzf"
      },
      "source": [
        "Trying different values, we see that it's working good. We increased test and validation accuracies by 3 to 4%."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mx9kLNol8Myk"
      },
      "source": [
        "### Logistic Regression Model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cM-Fwalw88T1"
      },
      "source": [
        "From Sklearn documentation, Logistic regression, despite its name, is a linear model for classification rather than regression. Logistic regression is also known in the literature as logit regression, maximum-entropy classification (MaxEnt) or the log-linear classifier. In this model, the probabilities describing the possible outcomes of a single trial are modeled using a logistic function."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6AWgddNrhbMM",
        "outputId": "d3e91059-8827-4e37-a6f5-ffb79a716de6"
      },
      "source": [
        "from sklearn.linear_model import LogisticRegression\r\n",
        "\r\n",
        "pipe = Pipeline([('bow', CountVectorizer(stop_words=stopwords.words('english'))), \r\n",
        "                 ('model', LogisticRegression(solver='newton-cg', C=6))])\r\n",
        "pipe.fit(X_train, y_train)\r\n",
        "\r\n",
        "y_train_pred = pipe.predict(X_train)\r\n",
        "y_test_pred = pipe.predict(X_test)\r\n",
        "y_val_pred = pipe.predict(X_val)\r\n",
        "\r\n",
        "print('---- Accuracy After Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))\r\n",
        "print('Validation accuracy', accuracy_score(y_val, y_val_pred))"
      ],
      "execution_count": 675,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy After Tuning ----\n",
            "Training accuracy 0.995375\n",
            "Test accuracy 0.8925\n",
            "Validation accuracy 0.9005\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "naRdTXAuFq0D"
      },
      "source": [
        "Finally, we see that Logistic Regression model performs the best. We get an increase of accuracies from 3 to 8%.\r\n",
        "\r\n",
        "We can save this model to be able to deploy it later."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o4sV7iB3SSpA"
      },
      "source": [
        "import pickle\r\n",
        "\r\n",
        "##saving the model\r\n",
        "with open(\"pipe_log_reg.pkl\", 'wb') as f_out:\r\n",
        "    pickle.dump(pipe, f_out)\r\n",
        "    f_out.close()"
      ],
      "execution_count": 706,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "exkL8LAiOSlh",
        "outputId": "d5ac17e0-668b-4252-9555-3012ddc0507e"
      },
      "source": [
        "pipe.predict_proba([\"I feel like i'm loving you\"])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.01187339, 0.00384238, 0.02680149, 0.94609178, 0.01031296,\n",
              "        0.001078  ]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ldO1hNN6qr2p",
        "outputId": "62ca8755-5d48-4826-8052-6b32d4e9cc99"
      },
      "source": [
        "pipe.predict_proba([\"love\"])"
      ],
      "execution_count": 699,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.14285806, 0.09008856, 0.07348533, 0.58355716, 0.09779101,\n",
              "        0.01221989]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 699
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o5RO7MzGtC_q"
      },
      "source": [
        "if we want to improve our predictions regarding the word \"love\", we can give more influence to love class :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2XcWIQgUrVaV"
      },
      "source": [
        "def reprocess_message(mess):\r\n",
        "  message = [c for c in mess.split(' ') if c.lower() != 'love']\r\n",
        "  message = ' '.join(message)\r\n",
        "  return message\r\n",
        "\r\n",
        "df_train.loc[df_train['Emotion'] != 'love','Cleaned message'] = df_train.loc[df_train['Emotion'] != 'love','Cleaned message'].apply(lambda x: reprocess_message(x))"
      ],
      "execution_count": 689,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PMBHiiRsrOJW"
      },
      "source": [
        "X_train = df_train['Cleaned message']"
      ],
      "execution_count": 690,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8Ton-CgKs4EO",
        "outputId": "c4e492c5-915a-42f8-a6bf-77e92573c296"
      },
      "source": [
        "pipe = Pipeline([('bow', CountVectorizer(stop_words=stopwords.words('english'))), \r\n",
        "                 ('model', LogisticRegression(solver='newton-cg', C=6))])\r\n",
        "pipe.fit(X_train, y_train)\r\n",
        "\r\n",
        "y_train_pred = pipe.predict(X_train)\r\n",
        "y_test_pred = pipe.predict(X_test)\r\n",
        "y_val_pred = pipe.predict(X_val)\r\n",
        "\r\n",
        "print('---- Accuracy After Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))\r\n",
        "print('Validation accuracy', accuracy_score(y_val, y_val_pred))"
      ],
      "execution_count": 691,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy After Tuning ----\n",
            "Training accuracy 0.995375\n",
            "Test accuracy 0.8905\n",
            "Validation accuracy 0.903\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I4SNYFLAtNU-",
        "outputId": "dce6575c-1259-452a-a341-b504c6f1a970"
      },
      "source": [
        "pipe.predict_proba([\"I love you\"])"
      ],
      "execution_count": 705,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.14285806, 0.09008856, 0.07348533, 0.58355716, 0.09779101,\n",
              "        0.01221989]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 705
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PFZsuItR3Ula"
      },
      "source": [
        "## Dealing with Imbalanced Data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NWsznm723jPO"
      },
      "source": [
        "We notice that some emotions are over represented in our datasets compare to others. For example, more than 50% of the messages are represented by Sadness and Joy. This kind of imbalance can be a problem for our models to make good predictions when it's about predicting classes that are under represented in our datasets : Here, it's the case for Surprise and Love which occupy less than 10% of the messages. Let's remind that the full dataset contains 20 000 records."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l1jgvSinW0F9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "754c0ad8-bcc5-462e-df52-4cca46d40d84"
      },
      "source": [
        "df = pd.concat([df_train, df_test, df_val])\r\n",
        "df.reset_index(inplace=True, drop=True)\r\n",
        "df.groupby('Emotion').describe()"
      ],
      "execution_count": 553,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>2709</td>\n",
              "      <td>2709</td>\n",
              "      <td>i feel complimented or insulted</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>2373</td>\n",
              "      <td>2373</td>\n",
              "      <td>i only have to think about a high school exper...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>6761</td>\n",
              "      <td>6760</td>\n",
              "      <td>i feel more adventurous willing to take risks ...</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>1641</td>\n",
              "      <td>1641</td>\n",
              "      <td>i tried to reconcile the two feelings into one...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>5797</td>\n",
              "      <td>5797</td>\n",
              "      <td>i hold it for a day my arm will feel numb and ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i admit to feeling bitterly surprised at how r...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Message                                                               \n",
              "           count unique                                                top freq\n",
              "Emotion                                                                        \n",
              "anger       2709   2709                    i feel complimented or insulted    1\n",
              "fear        2373   2373  i only have to think about a high school exper...    1\n",
              "joy         6761   6760  i feel more adventurous willing to take risks ...    2\n",
              "love        1641   1641  i tried to reconcile the two feelings into one...    1\n",
              "sadness     5797   5797  i hold it for a day my arm will feel numb and ...    1\n",
              "surprise     719    719  i admit to feeling bitterly surprised at how r...    1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 553
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xuW5ydWVC-J2",
        "outputId": "630e0ea9-61f7-42b8-d949-5cac7346c2af"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": 554,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(20000, 2)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 554
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FrCi0uJC5LFX"
      },
      "source": [
        "We are going to use the imblearn library which allow us to use some methods to perform Over sampling and Under sampling.\r\n",
        "\r\n",
        "Let's perform under-sampling based on RandomUnderSampler method:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xj2aICmK32di"
      },
      "source": [
        "from imblearn.under_sampling import RandomUnderSampler"
      ],
      "execution_count": 555,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YjSFTrb57C-A"
      },
      "source": [
        "X = df[['Message']]\r\n",
        "y = df['Emotion']"
      ],
      "execution_count": 556,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RfOKuY6d6K_F",
        "outputId": "2156f1c2-5f1e-41a6-955a-14b19b907376"
      },
      "source": [
        "rus = RandomUnderSampler()\r\n",
        "X_res, y_res = rus.fit_resample(X,y)\r\n",
        "print(X_res.shape, y_res.shape)"
      ],
      "execution_count": 557,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(4314, 1) (4314,)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function safe_indexing is deprecated; safe_indexing is deprecated in version 0.22 and will be removed in version 0.24.\n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f5m1we0D8LVG"
      },
      "source": [
        "res_df = pd.DataFrame(np.concatenate((X_res,y_res.reshape(-1,1)),axis=1), columns=['Message','Emotion']) # Resampled Dataframe"
      ],
      "execution_count": 558,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DMGeLeK-WjMW",
        "outputId": "70de4d53-3357-458a-882e-8670152e3465"
      },
      "source": [
        "print(res_df[(res_df['Emotion']=='sadness') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='anger') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='joy') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='fear') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='love') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='suprise') & (res_df['Message'].str.contains(r'\\bsad\\b'))].shape)\r\n",
        "\r\n",
        "print('-----')\r\n",
        "\r\n",
        "print(res_df[(res_df['Emotion']=='surprise') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='sadness') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='joy') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='love') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='fear') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='anger') & (res_df['Message'].str.contains(r'\\bsurprise\\b'))].shape)\r\n",
        "\r\n",
        "print('-----')\r\n",
        "\r\n",
        "print(res_df[(res_df['Emotion']=='love') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='sadness') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='anger') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='surprise') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='joy') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)\r\n",
        "print(res_df[(res_df['Emotion']=='fear') & (res_df['Message'].str.contains(r'\\blove\\b'))].shape)"
      ],
      "execution_count": 562,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(12, 2)\n",
            "(3, 2)\n",
            "(0, 2)\n",
            "(5, 2)\n",
            "(3, 2)\n",
            "(0, 2)\n",
            "-----\n",
            "(1, 2)\n",
            "(0, 2)\n",
            "(0, 2)\n",
            "(0, 2)\n",
            "(0, 2)\n",
            "(0, 2)\n",
            "-----\n",
            "(45, 2)\n",
            "(15, 2)\n",
            "(18, 2)\n",
            "(15, 2)\n",
            "(17, 2)\n",
            "(20, 2)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "vFFrKYUR_3Ow",
        "outputId": "bdcb51e8-ae56-4271-fbae-fbac2852d3a5"
      },
      "source": [
        "res_df.groupby('Emotion').describe()"
      ],
      "execution_count": 563,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Message</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>count</th>\n",
              "      <th>unique</th>\n",
              "      <th>top</th>\n",
              "      <th>freq</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Emotion</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>anger</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i found myself feeling so angry</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>fear</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i give you some tips on overcoming the feeling...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>joy</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i feel arsenal could be clearly superior</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>love</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i was just telling you how i feel about you an...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>sadness</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i have a massive identity of my own and always...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>surprise</th>\n",
              "      <td>719</td>\n",
              "      <td>719</td>\n",
              "      <td>i admit to feeling bitterly surprised at how r...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         Message                                                               \n",
              "           count unique                                                top freq\n",
              "Emotion                                                                        \n",
              "anger        719    719                    i found myself feeling so angry    1\n",
              "fear         719    719  i give you some tips on overcoming the feeling...    1\n",
              "joy          719    719           i feel arsenal could be clearly superior    1\n",
              "love         719    719  i was just telling you how i feel about you an...    1\n",
              "sadness      719    719  i have a massive identity of my own and always...    1\n",
              "surprise     719    719  i admit to feeling bitterly surprised at how r...    1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 563
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bMXauJMKDUzT"
      },
      "source": [
        "Here it is, all the emotions have now the same number of records, it means the dataset is perfectly balanced and logically no emotion will have much more influence than another when it will be time to make a prediction based on a new input message."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HDHPgSuKV-Ra",
        "outputId": "1266166d-4221-4f80-83ed-71292c07843c"
      },
      "source": [
        "X = res_df['Message']\r\n",
        "y = res_df['Emotion']\r\n",
        "\r\n",
        "print(X.shape, y.shape)"
      ],
      "execution_count": 564,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(4314,) (4314,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qRptcPPSEj6x"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\r\n",
        "\r\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=1)"
      ],
      "execution_count": 565,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Uc4miuWCWj4z",
        "outputId": "728cd45d-b1a7-466e-e6b7-ea6aac77cf8c"
      },
      "source": [
        "from sklearn.pipeline import Pipeline\r\n",
        "from sklearn.linear_model import LogisticRegression\r\n",
        "from sklearn.feature_extraction.text import CountVectorizer\r\n",
        "from sklearn.metrics import accuracy_score\r\n",
        "\r\n",
        "\r\n",
        "pipe = Pipeline([('bow', CountVectorizer(stop_words=stopwords.words('english'))), \r\n",
        "                 ('model', LogisticRegression(solver='newton-cg', C=6))])\r\n",
        "pipe.fit(X_train, y_train)\r\n",
        "\r\n",
        "y_train_pred = pipe.predict(X_train)\r\n",
        "y_test_pred = pipe.predict(X_test)\r\n",
        "\r\n",
        "print('---- Accuracy After Tuning ----')\r\n",
        "print('Training accuracy', accuracy_score(y_train, y_train_pred))\r\n",
        "print('Test accuracy', accuracy_score(y_test, y_test_pred))"
      ],
      "execution_count": 568,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "---- Accuracy After Tuning ----\n",
            "Training accuracy 0.9982613735149232\n",
            "Test accuracy 0.8621089223638471\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W2Z6cvivccnW",
        "outputId": "32e7cea5-d793-418b-fecd-964892dc281b"
      },
      "source": [
        "print(pipe.predict(['']))"
      ],
      "execution_count": 576,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "['love']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rytMx0cAXrUQ",
        "outputId": "1fcd4064-a943-4269-d2a5-9d80896707d9"
      },
      "source": [
        "from sklearn.model_selection import cross_validate\r\n",
        "\r\n",
        "cv_results = cross_validate(pipe, X, y, cv=5)\r\n",
        "cv_results['test_score']"
      ],
      "execution_count": 577,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.85168019, 0.85399768, 0.8400927 , 0.85283893, 0.84454756])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 577
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "RLy9UfQCZT--",
        "outputId": "b7692c8b-35d4-4ed3-a9df-c22e4c62521c"
      },
      "source": [
        "pd.DataFrame(pipe.named_steps['bow'].fit_transform(X_train).toarray(), columns=pipe.named_steps['bow'].get_feature_names())"
      ],
      "execution_count": 578,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>abandoned</th>\n",
              "      <th>abdominal</th>\n",
              "      <th>abhorrent</th>\n",
              "      <th>abilities</th>\n",
              "      <th>ability</th>\n",
              "      <th>able</th>\n",
              "      <th>abominable</th>\n",
              "      <th>abroad</th>\n",
              "      <th>absence</th>\n",
              "      <th>absolute</th>\n",
              "      <th>absolutely</th>\n",
              "      <th>abstract</th>\n",
              "      <th>abuse</th>\n",
              "      <th>abused</th>\n",
              "      <th>abuses</th>\n",
              "      <th>abusive</th>\n",
              "      <th>abyss</th>\n",
              "      <th>academic</th>\n",
              "      <th>academy</th>\n",
              "      <th>accept</th>\n",
              "      <th>acceptable</th>\n",
              "      <th>accepted</th>\n",
              "      <th>access</th>\n",
              "      <th>accessaries</th>\n",
              "      <th>accessories</th>\n",
              "      <th>accessory</th>\n",
              "      <th>accident</th>\n",
              "      <th>accidentally</th>\n",
              "      <th>accomplished</th>\n",
              "      <th>accomplishing</th>\n",
              "      <th>accomplishment</th>\n",
              "      <th>accomplishments</th>\n",
              "      <th>accord</th>\n",
              "      <th>according</th>\n",
              "      <th>account</th>\n",
              "      <th>acctually</th>\n",
              "      <th>accusation</th>\n",
              "      <th>accusing</th>\n",
              "      <th>accustomed</th>\n",
              "      <th>ache</th>\n",
              "      <th>...</th>\n",
              "      <th>yay</th>\n",
              "      <th>yeah</th>\n",
              "      <th>year</th>\n",
              "      <th>yearly</th>\n",
              "      <th>yearn</th>\n",
              "      <th>yearning</th>\n",
              "      <th>yearnings</th>\n",
              "      <th>yearns</th>\n",
              "      <th>years</th>\n",
              "      <th>yell</th>\n",
              "      <th>yelled</th>\n",
              "      <th>yes</th>\n",
              "      <th>yesterday</th>\n",
              "      <th>yesterdays</th>\n",
              "      <th>yet</th>\n",
              "      <th>yk</th>\n",
              "      <th>yo</th>\n",
              "      <th>yoga</th>\n",
              "      <th>york</th>\n",
              "      <th>youd</th>\n",
              "      <th>youll</th>\n",
              "      <th>young</th>\n",
              "      <th>younger</th>\n",
              "      <th>youre</th>\n",
              "      <th>youth</th>\n",
              "      <th>youtube</th>\n",
              "      <th>yummy</th>\n",
              "      <th>yunhos</th>\n",
              "      <th>yup</th>\n",
              "      <th>yuuki</th>\n",
              "      <th>zach</th>\n",
              "      <th>zealand</th>\n",
              "      <th>zero</th>\n",
              "      <th>zhuo</th>\n",
              "      <th>zombie</th>\n",
              "      <th>zombies</th>\n",
              "      <th>zone</th>\n",
              "      <th>zoo</th>\n",
              "      <th>zooms</th>\n",
              "      <th>zumba</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3446</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3447</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3448</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3449</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3450</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3451 rows × 6632 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "      abandoned  abdominal  abhorrent  abilities  ...  zone  zoo  zooms  zumba\n",
              "0             0          0          0          0  ...     0    0      0      0\n",
              "1             0          0          0          0  ...     0    0      0      0\n",
              "2             0          0          0          0  ...     0    0      0      0\n",
              "3             0          0          0          0  ...     0    0      0      0\n",
              "4             0          0          0          0  ...     0    0      0      0\n",
              "...         ...        ...        ...        ...  ...   ...  ...    ...    ...\n",
              "3446          0          0          0          0  ...     0    0      0      0\n",
              "3447          0          0          0          0  ...     0    0      0      0\n",
              "3448          0          0          0          0  ...     0    0      0      0\n",
              "3449          0          0          0          0  ...     0    0      0      0\n",
              "3450          0          0          0          0  ...     0    0      0      0\n",
              "\n",
              "[3451 rows x 6632 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 578
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bYmO2CjWaOMD",
        "outputId": "833ed3ea-4a05-47e6-e3a3-63d377a17915"
      },
      "source": [
        "from sklearn.model_selection import GridSearchCV\r\n",
        "parameters = {'model__C':[5],            \r\n",
        "              'model__solver':['liblinear'],\r\n",
        "              'model__max_iter': [1000],\r\n",
        "              'bow__min_df': np.arange(1,5,1),\r\n",
        "              'bow__ngram_range': [(1,1),(1,2)],\r\n",
        "              }\r\n",
        "model = GridSearchCV(pipe, param_grid=parameters, cv=5, scoring='accuracy')\r\n",
        "\r\n",
        "model.fit(X, y)\r\n",
        "print(\"The best parameters by CV:\", model.best_params_)\r\n",
        "print(\"The best score by CV = {:5.3f}\".format(model.best_score_))"
      ],
      "execution_count": 593,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The best parameters by CV: {'bow__min_df': 2, 'bow__ngram_range': (1, 2), 'model__C': 5, 'model__max_iter': 1000, 'model__solver': 'liblinear'}\n",
            "The best score by CV = 0.854\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XsU70boDcdBF",
        "outputId": "6305a6f6-6a84-445d-e2d3-b26e1ce92557"
      },
      "source": [
        "pipe.get_params().keys()"
      ],
      "execution_count": 170,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dict_keys(['memory', 'steps', 'verbose', 'bow', 'model', 'bow__analyzer', 'bow__binary', 'bow__decode_error', 'bow__dtype', 'bow__encoding', 'bow__input', 'bow__lowercase', 'bow__max_df', 'bow__max_features', 'bow__min_df', 'bow__ngram_range', 'bow__preprocessor', 'bow__stop_words', 'bow__strip_accents', 'bow__token_pattern', 'bow__tokenizer', 'bow__vocabulary', 'model__C', 'model__class_weight', 'model__dual', 'model__fit_intercept', 'model__intercept_scaling', 'model__l1_ratio', 'model__max_iter', 'model__multi_class', 'model__n_jobs', 'model__penalty', 'model__random_state', 'model__solver', 'model__tol', 'model__verbose', 'model__warm_start'])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 170
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VQAyShO8h1nu",
        "outputId": "4717794e-aaad-454e-b830-5d8e9ec029b4"
      },
      "source": [
        "print(pipe.predict_proba(['love']))\r\n",
        "print(model.predict_proba(['love']))"
      ],
      "execution_count": 601,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[0.25591268 0.21662775 0.23657052 0.14711438 0.11878784 0.02498683]]\n",
            "[[0.06960683 0.32915895 0.22877525 0.16212178 0.17764511 0.03269209]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jxop9ZFGWIlT"
      },
      "source": [
        "Finally, if we want to improve our predictions regarding the word \"love\", we can give more influence to love class :"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q3cbtJUKfz8L"
      },
      "source": [
        "def reprocess_message(mess):\r\n",
        "  message = [c for c in mess.split(' ') if c.lower() != 'love']\r\n",
        "  message = ' '.join(message)\r\n",
        "  return message\r\n",
        "\r\n",
        "res_df.loc[res_df['Emotion'] != 'love','Message'] = res_df.loc[res_df['Emotion'] != 'love','Message'].apply(lambda x: reprocess_message(x))"
      ],
      "execution_count": 606,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PKrv1Y2zDsJP"
      },
      "source": [
        "X = res_df['Message']\r\n",
        "y = res_df['Emotion']"
      ],
      "execution_count": 611,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C-ajEsBAGXuC",
        "outputId": "690540ca-1e81-40e8-cfb9-603cf241003b"
      },
      "source": [
        "model = GridSearchCV(pipe, param_grid=parameters, cv=5, scoring='accuracy')\r\n",
        "\r\n",
        "model.fit(X, y)\r\n",
        "print(\"The best parameters by CV:\", model.best_params_)\r\n",
        "print(\"The best score by CV = {:5.3f}\".format(model.best_score_))"
      ],
      "execution_count": 612,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The best parameters by CV: {'bow__min_df': 2, 'bow__ngram_range': (1, 2), 'model__C': 5, 'model__max_iter': 1000, 'model__solver': 'liblinear'}\n",
            "The best score by CV = 0.855\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EgVLiyXMGrPL",
        "outputId": "77a387be-e6b0-482a-d38b-e36bc51a0078"
      },
      "source": [
        "model.predict(['love'])"
      ],
      "execution_count": 684,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['love'], dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 684
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5kRdG03gGyDY"
      },
      "source": [
        "import pickle\r\n",
        "\r\n",
        "##saving the model\r\n",
        "with open(\"balanced_undersamp_pipe.pkl\", 'wb') as f_out:\r\n",
        "    pickle.dump(model, f_out)\r\n",
        "    f_out.close()"
      ],
      "execution_count": 624,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TJ0krWIKkabc"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}