{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "AO7EIAPjciKB"
      },
      "outputs": [],
      "source": [
        "### Ericsson_ML_Challenge_MaterialType_Prediction"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "###!mkdir ~/.kaggle"
      ],
      "metadata": {
        "id": "PW7LI6b4cnxf"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "###!cp /kaggle.json ~/.kaggle/"
      ],
      "metadata": {
        "id": "MDMSLwjmcn1P"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "####!chmod 600 ~/.kaggle/kaggle.json"
      ],
      "metadata": {
        "id": "wSrXLZjccn4X"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "####! pip install kaggle"
      ],
      "metadata": {
        "id": "h61s9qtIcn7Q"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "####!pip install keras-tuner"
      ],
      "metadata": {
        "id": "i9yUJOMjcn-b"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "###!kaggle datasets download -d saranyashalya/ericsson-ml-challenge-materialtype-prediction"
      ],
      "metadata": {
        "id": "eOmA0s-CcoBH"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "###! unzip /content/ericsson-ml-challenge-materialtype-prediction.zip"
      ],
      "metadata": {
        "id": "Q731Zv8FfZKi"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "###! pip install tensorflow"
      ],
      "metadata": {
        "id": "w-g98wUffZVP"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "####! pip install bayesian-optimization"
      ],
      "metadata": {
        "id": "ES8vmLlKfZkh"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from sklearn import preprocessing\n",
        "import matplotlib.pyplot as plt\n",
        "tf.random.set_seed(123)\n",
        "np.random.seed(123)"
      ],
      "metadata": {
        "id": "vATXQBAdfpGn"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "##!pip install bayesian-optimization"
      ],
      "metadata": {
        "id": "vNaWsWdpAhtv"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Import packages\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, BatchNormalization, Dropout\n",
        "#from tensorflow.keras.optimizers import Adam, SGD, RMSprop, Adadelta, Adagrad, Adamax, Nadam, Ftrl\n",
        "from keras.callbacks import EarlyStopping, ModelCheckpoint\n",
        "###from keras.wrappers.scikit_learn import KerasClassifier\n",
        "from math import floor\n",
        "from sklearn.metrics import make_scorer, accuracy_score\n",
        "from bayes_opt import BayesianOptimization\n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from keras.layers import LeakyReLU\n",
        "LeakyReLU = LeakyReLU(alpha=0.1)\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "pd.set_option(\"display.max_columns\", None)"
      ],
      "metadata": {
        "id": "s88RsMNifstL"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Import packages\n",
        "# Basic packages\n",
        "\n",
        "import pickle\n",
        "from math import floor\n",
        "from sklearn.model_selection import train_test_split, cross_val_score\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "# Evaluation and bayesian optimization\n",
        "from sklearn.metrics import make_scorer, mean_absolute_error\n",
        "from sklearn.metrics import mean_squared_error as MSE\n",
        "from hyperopt import hp, fmin, tpe\n",
        "from sklearn.model_selection import GridSearchCV, StratifiedKFold\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "pd.set_option(\"display.max_columns\", None)"
      ],
      "metadata": {
        "id": "5VENaHXffvkm"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = pd.read_csv(\"/content/train_file.csv\")"
      ],
      "metadata": {
        "id": "2fHwy3dlf3OP"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data = pd.read_csv(\"/content/test_file.csv\")"
      ],
      "metadata": {
        "id": "KWnFg3v7f9d2"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(train_data.shape, test_data.shape)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V0q6h0G9gq3t",
        "outputId": "1d393ce6-56e5-4ded-8d25-0cbf45e41945"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(31653, 12) (21102, 11)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u-ajVQqaguxv",
        "outputId": "b76e7e9f-68df-4257-b7f3-36facc679a52"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['ID', 'UsageClass', 'CheckoutType', 'CheckoutYear', 'CheckoutMonth',\n",
              "       'Checkouts', 'Title', 'Creator', 'Subjects', 'Publisher',\n",
              "       'PublicationYear', 'MaterialType'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 354
        },
        "id": "jYH3BBHfhhQq",
        "outputId": "84be46b1-4049-494e-f836-d41de0ec5ae7"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   ID UsageClass CheckoutType  CheckoutYear  CheckoutMonth  Checkouts  \\\n",
              "0   1   Physical      Horizon          2005              4          1   \n",
              "1   2   Physical      Horizon          2005              4          1   \n",
              "2   3   Physical      Horizon          2005              4          3   \n",
              "\n",
              "                                               Title               Creator  \\\n",
              "0                                         Tidal wave                   NaN   \n",
              "1                     London holiday / Richard Peck.  Peck, Richard, 1934-   \n",
              "2  Cinco de Mayo : celebrating Hispanic pride / C...      Gnojewski, Carol   \n",
              "\n",
              "                                            Subjects           Publisher  \\\n",
              "0             Tsunamis, Tsunamis Juvenile literature                 NaN   \n",
              "1                                                NaN             Viking,   \n",
              "2  Cinco de Mayo Mexican holiday History Juvenile...  Enslow Publishers,   \n",
              "\n",
              "  PublicationYear MaterialType  \n",
              "0             NaN         BOOK  \n",
              "1           1998.         BOOK  \n",
              "2          c2002.         BOOK  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-948279e6-6457-459b-ab5d-c60c542f8705\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ID</th>\n",
              "      <th>UsageClass</th>\n",
              "      <th>CheckoutType</th>\n",
              "      <th>CheckoutYear</th>\n",
              "      <th>CheckoutMonth</th>\n",
              "      <th>Checkouts</th>\n",
              "      <th>Title</th>\n",
              "      <th>Creator</th>\n",
              "      <th>Subjects</th>\n",
              "      <th>Publisher</th>\n",
              "      <th>PublicationYear</th>\n",
              "      <th>MaterialType</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>Tidal wave</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Tsunamis, Tsunamis Juvenile literature</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>BOOK</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>London holiday / Richard Peck.</td>\n",
              "      <td>Peck, Richard, 1934-</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Viking,</td>\n",
              "      <td>1998.</td>\n",
              "      <td>BOOK</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>Cinco de Mayo : celebrating Hispanic pride / C...</td>\n",
              "      <td>Gnojewski, Carol</td>\n",
              "      <td>Cinco de Mayo Mexican holiday History Juvenile...</td>\n",
              "      <td>Enslow Publishers,</td>\n",
              "      <td>c2002.</td>\n",
              "      <td>BOOK</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-948279e6-6457-459b-ab5d-c60c542f8705')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-948279e6-6457-459b-ab5d-c60c542f8705 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-948279e6-6457-459b-ab5d-c60c542f8705');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-5ea3d579-ba41-4503-adcf-005c5e0b9b72\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-5ea3d579-ba41-4503-adcf-005c5e0b9b72')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-5ea3d579-ba41-4503-adcf-005c5e0b9b72 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"MaterialType\"].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1F2CAv8dg1Uw",
        "outputId": "58112d41-7e3f-41b9-c0a2-2bbd73388261"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BOOK         21707\n",
              "SOUNDDISC     4149\n",
              "VIDEOCASS     2751\n",
              "VIDEODISC     1420\n",
              "SOUNDCASS     1020\n",
              "MIXED          347\n",
              "MUSIC          165\n",
              "CR              94\n",
              "Name: MaterialType, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data[\"MaterialType\"] = 0"
      ],
      "metadata": {
        "id": "Oz7vne3yg7cl"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from matplotlib_venn import venn2, venn2_circles, venn2_unweighted\n",
        "from matplotlib_venn import venn3, venn3_circles"
      ],
      "metadata": {
        "id": "GCxmtR_GhCum"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "set_numbers_train = set(train_data[['ID']].drop_duplicates().sort_values(by = 'ID')['ID'].tolist())\n",
        "set_numbers_test = set(test_data[['ID']].drop_duplicates().sort_values(by = 'ID')['ID'].tolist())\n",
        "venn2((set_numbers_train, set_numbers_test), set_labels = ('Train numbers', 'Test numbers'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 298
        },
        "id": "HJZFpA3nhGe9",
        "outputId": "8bdb6373-0fe9-4670-f111-f16f78b23525"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib_venn._common.VennDiagram at 0x786e69cb0b20>"
            ]
          },
          "metadata": {},
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### The above data explains the `size` of train and test data."
      ],
      "metadata": {
        "id": "wDGpYSs6hSeu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ce-2DXxthIJi",
        "outputId": "826f5298-c9dc-4448-8e53-e83a4edc2bbd"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['ID', 'UsageClass', 'CheckoutType', 'CheckoutYear', 'CheckoutMonth',\n",
              "       'Checkouts', 'Title', 'Creator', 'Subjects', 'Publisher',\n",
              "       'PublicationYear', 'MaterialType'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "num_var = [feature for feature in train_data.columns if train_data[feature].dtypes != 'O']\n",
        "discrete_var = [feature for feature in num_var if len(train_data[feature].unique()) <= 25]\n",
        "cont_var = [feature for feature in num_var if feature not in discrete_var]\n",
        "categ_var = [feature for feature in train_data.columns if feature not in num_var]"
      ],
      "metadata": {
        "id": "R-22uBJfhj81"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(\"The Numerical Variables are :\", num_var)\n",
        "print(\"The Discreate Variables are :\", discrete_var)\n",
        "print(\"The Continuous Variables are :\", cont_var)\n",
        "print(\"The Categorical Variables are :\", categ_var)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LbEgL7V2hpnu",
        "outputId": "53dcaee8-1989-4afe-9daa-78cfd600459d"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The Numerical Variables are : ['ID', 'CheckoutYear', 'CheckoutMonth', 'Checkouts']\n",
            "The Discreate Variables are : ['CheckoutYear', 'CheckoutMonth']\n",
            "The Continuous Variables are : ['ID', 'Checkouts']\n",
            "The Categorical Variables are : ['UsageClass', 'CheckoutType', 'Title', 'Creator', 'Subjects', 'Publisher', 'PublicationYear', 'MaterialType']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### CHECKING `NULL VALUES` OR NOT"
      ],
      "metadata": {
        "id": "5DIRfc7qkSKN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kb-rJ0dHkcod",
        "outputId": "9287a8f6-a8df-4338-b244-848630347148"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID                     0\n",
              "UsageClass             0\n",
              "CheckoutType           0\n",
              "CheckoutYear           0\n",
              "CheckoutMonth          0\n",
              "Checkouts              0\n",
              "Title                  0\n",
              "Creator            23137\n",
              "Subjects            1763\n",
              "Publisher          21916\n",
              "PublicationYear    21931\n",
              "MaterialType           0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = train_data.fillna(0)"
      ],
      "metadata": {
        "id": "QjoP__S3krV6"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T3ef48nWk3Ai",
        "outputId": "dc452f9d-5c91-484c-8f1f-e36ed12c5d67"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID                 0\n",
              "UsageClass         0\n",
              "CheckoutType       0\n",
              "CheckoutYear       0\n",
              "CheckoutMonth      0\n",
              "Checkouts          0\n",
              "Title              0\n",
              "Creator            0\n",
              "Subjects           0\n",
              "Publisher          0\n",
              "PublicationYear    0\n",
              "MaterialType       0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data = test_data.fillna(0)"
      ],
      "metadata": {
        "id": "p4g6ZjtfC7O0"
      },
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### CATEGORICAL   VARIABLES"
      ],
      "metadata": {
        "id": "l-lV7q2DiGIA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3IfNtha8h-pE",
        "outputId": "667cbc7d-9d8b-4334-d3ac-b844864e8445"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['ID', 'UsageClass', 'CheckoutType', 'CheckoutYear', 'CheckoutMonth',\n",
              "       'Checkouts', 'Title', 'Creator', 'Subjects', 'Publisher',\n",
              "       'PublicationYear', 'MaterialType'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"MaterialType\"].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ww0ENEz3i2Ws",
        "outputId": "93261c8a-b5a1-4d65-b211-a124343f8eba"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BOOK         21707\n",
              "SOUNDDISC     4149\n",
              "VIDEOCASS     2751\n",
              "VIDEODISC     1420\n",
              "SOUNDCASS     1020\n",
              "MIXED          347\n",
              "MUSIC          165\n",
              "CR              94\n",
              "Name: MaterialType, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### (0: 'BOOK', 1: 'CR', 2: 'MIXED', 3: 'MUSIC', 4: 'SOUNDCASS', 5: 'SOUNDDISC', 6: 'VIDEOCASS', 7: 'VIDEODISC'}"
      ],
      "metadata": {
        "id": "esw4E8B8BdNx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "c = train_data[\"MaterialType\"].astype('category')\n",
        "d = dict(enumerate(c.cat.categories))\n",
        "print(d)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z-Jn5LAFi7nM",
        "outputId": "d028adb5-2a43-4aaa-98f0-50ab7387fe12"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{0: 'BOOK', 1: 'CR', 2: 'MIXED', 3: 'MUSIC', 4: 'SOUNDCASS', 5: 'SOUNDDISC', 6: 'VIDEOCASS', 7: 'VIDEODISC'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data['MaterialType'] = train_data.MaterialType.astype('category').cat.codes"
      ],
      "metadata": {
        "id": "vruwCOqUjDu0"
      },
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Analysis For `Numerical` Variables"
      ],
      "metadata": {
        "id": "G6DLrIz_jUV-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#import seaborn as sns\n",
        "#import matplotlib.pyplot as plt\n",
        "\n",
        "for feature in num_var:\n",
        "  data=train_data.copy()\n",
        "  sns.distplot(train_data[feature])\n",
        "  plt.xlabel(feature)\n",
        "  plt.ylabel(\"Count\")\n",
        "  plt.title(feature)\n",
        "  plt.figure(figsize=(3,3))\n",
        "  plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "1wizFddojL_U",
        "outputId": "7fb7bff7-3c5f-4c08-e5c3-badf80360b25"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 300x300 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 300x300 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 300x300 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 300x300 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 354
        },
        "id": "MHXwPvlnkF7Q",
        "outputId": "8ad5e403-f5de-4ebd-d3be-11adb2a8c263"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   ID UsageClass CheckoutType  CheckoutYear  CheckoutMonth  Checkouts  \\\n",
              "0   1   Physical      Horizon          2005              4          1   \n",
              "1   2   Physical      Horizon          2005              4          1   \n",
              "2   3   Physical      Horizon          2005              4          3   \n",
              "\n",
              "                                               Title               Creator  \\\n",
              "0                                         Tidal wave                     0   \n",
              "1                     London holiday / Richard Peck.  Peck, Richard, 1934-   \n",
              "2  Cinco de Mayo : celebrating Hispanic pride / C...      Gnojewski, Carol   \n",
              "\n",
              "                                            Subjects           Publisher  \\\n",
              "0             Tsunamis, Tsunamis Juvenile literature                   0   \n",
              "1                                                  0             Viking,   \n",
              "2  Cinco de Mayo Mexican holiday History Juvenile...  Enslow Publishers,   \n",
              "\n",
              "  PublicationYear  MaterialType  \n",
              "0               0             0  \n",
              "1           1998.             0  \n",
              "2          c2002.             0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-be48054f-8776-400a-b2b4-ffab19c1334f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ID</th>\n",
              "      <th>UsageClass</th>\n",
              "      <th>CheckoutType</th>\n",
              "      <th>CheckoutYear</th>\n",
              "      <th>CheckoutMonth</th>\n",
              "      <th>Checkouts</th>\n",
              "      <th>Title</th>\n",
              "      <th>Creator</th>\n",
              "      <th>Subjects</th>\n",
              "      <th>Publisher</th>\n",
              "      <th>PublicationYear</th>\n",
              "      <th>MaterialType</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>Tidal wave</td>\n",
              "      <td>0</td>\n",
              "      <td>Tsunamis, Tsunamis Juvenile literature</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>London holiday / Richard Peck.</td>\n",
              "      <td>Peck, Richard, 1934-</td>\n",
              "      <td>0</td>\n",
              "      <td>Viking,</td>\n",
              "      <td>1998.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>Cinco de Mayo : celebrating Hispanic pride / C...</td>\n",
              "      <td>Gnojewski, Carol</td>\n",
              "      <td>Cinco de Mayo Mexican holiday History Juvenile...</td>\n",
              "      <td>Enslow Publishers,</td>\n",
              "      <td>c2002.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-be48054f-8776-400a-b2b4-ffab19c1334f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-be48054f-8776-400a-b2b4-ffab19c1334f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-be48054f-8776-400a-b2b4-ffab19c1334f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-64bb36af-7361-4a97-a466-d2f04eed278e\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-64bb36af-7361-4a97-a466-d2f04eed278e')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-64bb36af-7361-4a97-a466-d2f04eed278e button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "###! pip install klib"
      ],
      "metadata": {
        "id": "WJQs04h8lDAv"
      },
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#### Using KLIB Library"
      ],
      "metadata": {
        "id": "1YUThQhVlFG2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import klib"
      ],
      "metadata": {
        "id": "HWyQ_P7olDI_"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = klib.clean_column_names(train_data)"
      ],
      "metadata": {
        "id": "o9tbm6qvlS5y"
      },
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data = klib.clean_column_names(test_data)"
      ],
      "metadata": {
        "id": "zZaCLEKllvMs"
      },
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = klib.convert_datatypes(train_data)\n",
        "test_data = klib.convert_datatypes(test_data)"
      ],
      "metadata": {
        "id": "pp0K8dfumUfk"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = klib.mv_col_handling(train_data)\n",
        "test_data = klib.mv_col_handling(test_data)"
      ],
      "metadata": {
        "id": "ZQpyltvCmgYE"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.dtypes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tkqwblelmkP3",
        "outputId": "617f9b6f-cbe8-45ec-9ba3-f0db816e75f4"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id                     int16\n",
              "usage_class         category\n",
              "checkout_type       category\n",
              "checkout_year          int16\n",
              "checkout_month          int8\n",
              "checkouts               int8\n",
              "title                 string\n",
              "creator               object\n",
              "subjects              object\n",
              "publisher             object\n",
              "publication_year    category\n",
              "material_type           int8\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Data Conversion"
      ],
      "metadata": {
        "id": "BDLAhPSVmz93"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head(2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "id": "EX6JmgTAmzPM",
        "outputId": "e9cdac1b-0d1d-4e49-ee71-69c8ba80516f"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id usage_class checkout_type  checkout_year  checkout_month  checkouts  \\\n",
              "0   1    Physical       Horizon           2005               4          1   \n",
              "1   2    Physical       Horizon           2005               4          1   \n",
              "\n",
              "                            title               creator  \\\n",
              "0                      Tidal wave                     0   \n",
              "1  London holiday / Richard Peck.  Peck, Richard, 1934-   \n",
              "\n",
              "                                 subjects publisher publication_year  \\\n",
              "0  Tsunamis, Tsunamis Juvenile literature         0                0   \n",
              "1                                       0   Viking,            1998.   \n",
              "\n",
              "   material_type  \n",
              "0              0  \n",
              "1              0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-346bfda7-eaf9-4cd3-a821-6f63e774b014\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>usage_class</th>\n",
              "      <th>checkout_type</th>\n",
              "      <th>checkout_year</th>\n",
              "      <th>checkout_month</th>\n",
              "      <th>checkouts</th>\n",
              "      <th>title</th>\n",
              "      <th>creator</th>\n",
              "      <th>subjects</th>\n",
              "      <th>publisher</th>\n",
              "      <th>publication_year</th>\n",
              "      <th>material_type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>Tidal wave</td>\n",
              "      <td>0</td>\n",
              "      <td>Tsunamis, Tsunamis Juvenile literature</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>Physical</td>\n",
              "      <td>Horizon</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>London holiday / Richard Peck.</td>\n",
              "      <td>Peck, Richard, 1934-</td>\n",
              "      <td>0</td>\n",
              "      <td>Viking,</td>\n",
              "      <td>1998.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-346bfda7-eaf9-4cd3-a821-6f63e774b014')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-346bfda7-eaf9-4cd3-a821-6f63e774b014 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-346bfda7-eaf9-4cd3-a821-6f63e774b014');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-98322b3c-2ca9-47a8-99c5-56531870c9d0\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-98322b3c-2ca9-47a8-99c5-56531870c9d0')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-98322b3c-2ca9-47a8-99c5-56531870c9d0 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"usage_class\"] = train_data[\"usage_class\"].astype('category').cat.codes\n",
        "test_data[\"usage_class\"] = test_data[\"usage_class\"].astype('category').cat.codes\n",
        "train_data[\"checkout_type\"] = train_data[\"checkout_type\"].astype('category').cat.codes\n",
        "test_data[\"checkout_type\"] = test_data[\"checkout_type\"].astype('category').cat.codes\n"
      ],
      "metadata": {
        "id": "_EDPHnm6mzZk"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.dtypes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LcyOnuKSneSV",
        "outputId": "6813084d-7813-4408-fbe4-b86c466cfc6e"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id                     int16\n",
              "usage_class             int8\n",
              "checkout_type           int8\n",
              "checkout_year          int16\n",
              "checkout_month          int8\n",
              "checkouts               int8\n",
              "title                 string\n",
              "creator               object\n",
              "subjects              object\n",
              "publisher             object\n",
              "publication_year    category\n",
              "material_type           int8\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 354
        },
        "id": "_UaIZ8s4G9Jc",
        "outputId": "5a81c8e9-e48e-417e-cb75-4b33e551c255"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id  usage_class  checkout_type  checkout_year  checkout_month  checkouts  \\\n",
              "0   1            0              0           2005               4          1   \n",
              "1   2            0              0           2005               4          1   \n",
              "2   3            0              0           2005               4          3   \n",
              "\n",
              "                                               title               creator  \\\n",
              "0                                         Tidal wave                     0   \n",
              "1                     London holiday / Richard Peck.  Peck, Richard, 1934-   \n",
              "2  Cinco de Mayo : celebrating Hispanic pride / C...      Gnojewski, Carol   \n",
              "\n",
              "                                            subjects           publisher  \\\n",
              "0             Tsunamis, Tsunamis Juvenile literature                   0   \n",
              "1                                                  0             Viking,   \n",
              "2  Cinco de Mayo Mexican holiday History Juvenile...  Enslow Publishers,   \n",
              "\n",
              "  publication_year  material_type  \n",
              "0                0              0  \n",
              "1            1998.              0  \n",
              "2           c2002.              0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-25e44742-6511-4efd-8f80-8762c3c46828\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>usage_class</th>\n",
              "      <th>checkout_type</th>\n",
              "      <th>checkout_year</th>\n",
              "      <th>checkout_month</th>\n",
              "      <th>checkouts</th>\n",
              "      <th>title</th>\n",
              "      <th>creator</th>\n",
              "      <th>subjects</th>\n",
              "      <th>publisher</th>\n",
              "      <th>publication_year</th>\n",
              "      <th>material_type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>Tidal wave</td>\n",
              "      <td>0</td>\n",
              "      <td>Tsunamis, Tsunamis Juvenile literature</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>London holiday / Richard Peck.</td>\n",
              "      <td>Peck, Richard, 1934-</td>\n",
              "      <td>0</td>\n",
              "      <td>Viking,</td>\n",
              "      <td>1998.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>Cinco de Mayo : celebrating Hispanic pride / C...</td>\n",
              "      <td>Gnojewski, Carol</td>\n",
              "      <td>Cinco de Mayo Mexican holiday History Juvenile...</td>\n",
              "      <td>Enslow Publishers,</td>\n",
              "      <td>c2002.</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-25e44742-6511-4efd-8f80-8762c3c46828')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-25e44742-6511-4efd-8f80-8762c3c46828 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-25e44742-6511-4efd-8f80-8762c3c46828');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-be84351d-af7e-431c-a056-024c465fb741\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-be84351d-af7e-431c-a056-024c465fb741')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-be84351d-af7e-431c-a056-024c465fb741 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "def publication_year(s):\n",
        "    '''\n",
        "    '''\n",
        "    k = re.findall(r'\\d{4}', s)\n",
        "    if len(k) != 0:\n",
        "        try:\n",
        "            k = sorted(k)[0]\n",
        "            k = (pd.to_datetime('now').year - pd.to_datetime(k, format = '%Y').year)\n",
        "        except:\n",
        "            k = pd.period_range(start = k, end = '1678', freq = 'Y')\n",
        "            k = len(k)\n",
        "            k = (pd.to_datetime('now').year - pd.to_datetime('1678', format = '%Y').year + k)\n",
        "    else:\n",
        "        k = 0\n",
        "    return k###train_data['publication_year'][:100]"
      ],
      "metadata": {
        "id": "bhDL8oDG1taq"
      },
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"publication_year\"] = train_data[\"publication_year\"].astype(str)"
      ],
      "metadata": {
        "id": "RQAK24zgGgXD"
      },
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"publication_year\"] = train_data[\"publication_year\"].apply(publication_year)"
      ],
      "metadata": {
        "id": "bExo05Mgy7Ob"
      },
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data[\"publication_year\"] = test_data[\"publication_year\"].astype(str)\n",
        "test_data[\"publication_year\"] = test_data[\"publication_year\"].apply(publication_year)"
      ],
      "metadata": {
        "id": "M_iAO4DTHT9j"
      },
      "execution_count": 58,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.head(2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 254
        },
        "id": "r4hsvsL_HaqY",
        "outputId": "fa30d777-35cb-4954-dc38-ce425167c8f5"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id  usage_class  checkout_type  checkout_year  checkout_month  checkouts  \\\n",
              "0   1            0              0           2005               4          1   \n",
              "1   2            0              0           2005               4          1   \n",
              "\n",
              "                            title               creator  \\\n",
              "0                      Tidal wave                     0   \n",
              "1  London holiday / Richard Peck.  Peck, Richard, 1934-   \n",
              "\n",
              "                                 subjects publisher  publication_year  \\\n",
              "0  Tsunamis, Tsunamis Juvenile literature         0                 0   \n",
              "1                                       0   Viking,                25   \n",
              "\n",
              "   material_type  \n",
              "0              0  \n",
              "1              0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-beb8323b-894c-4e66-9a65-44abd8434e40\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>id</th>\n",
              "      <th>usage_class</th>\n",
              "      <th>checkout_type</th>\n",
              "      <th>checkout_year</th>\n",
              "      <th>checkout_month</th>\n",
              "      <th>checkouts</th>\n",
              "      <th>title</th>\n",
              "      <th>creator</th>\n",
              "      <th>subjects</th>\n",
              "      <th>publisher</th>\n",
              "      <th>publication_year</th>\n",
              "      <th>material_type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>Tidal wave</td>\n",
              "      <td>0</td>\n",
              "      <td>Tsunamis, Tsunamis Juvenile literature</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2005</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>London holiday / Richard Peck.</td>\n",
              "      <td>Peck, Richard, 1934-</td>\n",
              "      <td>0</td>\n",
              "      <td>Viking,</td>\n",
              "      <td>25</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-beb8323b-894c-4e66-9a65-44abd8434e40')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-beb8323b-894c-4e66-9a65-44abd8434e40 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-beb8323b-894c-4e66-9a65-44abd8434e40');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-4e682e53-5b4a-4cca-b327-aaf3cdbcd24d\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4e682e53-5b4a-4cca-b327-aaf3cdbcd24d')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-4e682e53-5b4a-4cca-b327-aaf3cdbcd24d button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "###!pip install textblob"
      ],
      "metadata": {
        "id": "OYYECAPoj150"
      },
      "execution_count": 62,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "###! pip install pickle"
      ],
      "metadata": {
        "id": "XCpFJm8eLgDa"
      },
      "execution_count": 64,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas, numpy, string, textblob\n",
        "import pickle\n",
        "from sklearn import model_selection, preprocessing, linear_model, naive_bayes, metrics, svm, decomposition, ensemble\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer, CountVectorizer\n",
        "import xgboost\n",
        "from keras import layers, models, optimizers\n",
        "from keras.preprocessing import text, sequence\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "nRctAEr7LgJw"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import string\n",
        "import re\n",
        "def clean_text(text):\n",
        "    '''Make text lowercase, remove text in square brackets,remove links,remove punctuation\n",
        "    and remove words containing numbers.'''\n",
        "    text = text.lower()\n",
        "    text = re.sub('\\[.*?\\]', '', text)\n",
        "    text = re.sub('https?://\\S+|www\\.\\S+', '', text)\n",
        "    text = re.sub('<.*?>+', '', text)\n",
        "    text = re.sub('[%s]' % re.escape(string.punctuation), '', text)\n",
        "    text = re.sub('\\n', '', text)\n",
        "    text = re.sub('\\d+', '', text)\n",
        "    text = re.sub('\\w*\\d\\w*', '', text)\n",
        "    return text"
      ],
      "metadata": {
        "id": "fx6OyqN3Lga-"
      },
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i7RPeRqkMrdc",
        "outputId": "b3fbcd0c-83a5-4736-9172-7a1715626379"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'creator', 'subjects', 'publisher',\n",
              "       'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xyy40V9ohYXt",
        "outputId": "55d57666-aebf-445f-8a03-29c5fe8f0a27"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'creator', 'subjects', 'publisher',\n",
              "       'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HQmwqwY9TTre",
        "outputId": "770d91a6-50b8-4a81-94f9-790d34ee112f"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id                  0\n",
              "usage_class         0\n",
              "checkout_type       0\n",
              "checkout_year       0\n",
              "checkout_month      0\n",
              "checkouts           0\n",
              "title               0\n",
              "creator             0\n",
              "subjects            0\n",
              "publisher           0\n",
              "publication_year    0\n",
              "material_type       0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"title\"] = train_data[\"title\"].astype('string')\n",
        "train_data[\"creator\"] = train_data[\"creator\"].astype('string')\n",
        "train_data[\"subjects\"] = train_data[\"subjects\"].astype('string')\n",
        "train_data[\"publisher\"] = train_data[\"publisher\"].astype('string')"
      ],
      "metadata": {
        "id": "iv9t_F0bU93t"
      },
      "execution_count": 70,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data[\"title\"] = test_data[\"title\"].astype('string')\n",
        "#test_data[\"creator\"] = test_data[\"creator\"].astype('string')\n",
        "test_data[\"subjects\"] = test_data[\"subjects\"].astype('string')\n",
        "#test_data[\"publisher\"] = test_data[\"publisher\"].astype('string')"
      ],
      "metadata": {
        "id": "Djk-M2Sfmi4H"
      },
      "execution_count": 71,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data[\"title\"] = train_data[\"title\"].apply(lambda x: clean_text(x))\n",
        "train_data[\"creator\"] = train_data[\"creator\"].apply(lambda x: clean_text(x))\n",
        "train_data[\"subjects\"] = train_data[\"subjects\"].apply(lambda x: clean_text(x))\n",
        "train_data[\"publisher\"] = train_data[\"publisher\"].apply(lambda x: clean_text(x))\n"
      ],
      "metadata": {
        "id": "Rywpo4fxUdP4"
      },
      "execution_count": 72,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_data[\"title\"] = test_data[\"title\"].apply(lambda x: clean_text(x))\n",
        "test_data[\"subjects\"] = test_data[\"subjects\"].apply(lambda x: clean_text(x))"
      ],
      "metadata": {
        "id": "x0UYBL_XMjnx"
      },
      "execution_count": 73,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.drop(columns =\"creator\", inplace=True)\n",
        "train_data.drop(columns = \"publisher\", inplace=True)"
      ],
      "metadata": {
        "id": "s7ILAnRtMVwe"
      },
      "execution_count": 74,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HMr45VFGMV1F",
        "outputId": "4b33229d-6b47-46e5-e945-87f91d1ffd4a"
      },
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'subjects', 'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 75
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Data Visualization"
      ],
      "metadata": {
        "id": "6p_gHwQYNmsz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "klib.cat_plot(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 977
        },
        "id": "liq47uKmNlFP",
        "outputId": "b00485d5-03e0-4456-d855-95580b39b12a"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSpec(6, 2)"
            ]
          },
          "metadata": {},
          "execution_count": 76
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1800 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "klib.corr_mat(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 320
        },
        "id": "phRebkLuNrlb",
        "outputId": "b8480ec7-0d9c-42e8-ab8b-fe585faed473"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<pandas.io.formats.style.Styler at 0x786e492c7100>"
            ],
            "text/html": [
              "<style type=\"text/css\">\n",
              "#T_f7299_row0_col0, #T_f7299_row0_col1, #T_f7299_row0_col2, #T_f7299_row0_col3, #T_f7299_row0_col4, #T_f7299_row0_col6, #T_f7299_row1_col0, #T_f7299_row1_col1, #T_f7299_row1_col2, #T_f7299_row1_col3, #T_f7299_row1_col4, #T_f7299_row1_col5, #T_f7299_row1_col6, #T_f7299_row1_col7, #T_f7299_row2_col0, #T_f7299_row2_col1, #T_f7299_row2_col2, #T_f7299_row2_col3, #T_f7299_row2_col4, #T_f7299_row2_col5, #T_f7299_row2_col6, #T_f7299_row2_col7, #T_f7299_row3_col0, #T_f7299_row3_col1, #T_f7299_row3_col2, #T_f7299_row3_col3, #T_f7299_row3_col4, #T_f7299_row3_col5, #T_f7299_row3_col6, #T_f7299_row3_col7, #T_f7299_row4_col0, #T_f7299_row4_col1, #T_f7299_row4_col2, #T_f7299_row4_col3, #T_f7299_row4_col4, #T_f7299_row4_col5, #T_f7299_row4_col6, #T_f7299_row4_col7, #T_f7299_row5_col1, #T_f7299_row5_col2, #T_f7299_row5_col3, #T_f7299_row5_col4, #T_f7299_row5_col5, #T_f7299_row5_col6, #T_f7299_row5_col7, #T_f7299_row6_col0, #T_f7299_row6_col1, #T_f7299_row6_col2, #T_f7299_row6_col3, #T_f7299_row6_col4, #T_f7299_row6_col5, #T_f7299_row6_col6, #T_f7299_row7_col1, #T_f7299_row7_col2, #T_f7299_row7_col3, #T_f7299_row7_col4, #T_f7299_row7_col5, #T_f7299_row7_col7 {\n",
              "  color: None;\n",
              "}\n",
              "#T_f7299_row0_col5, #T_f7299_row0_col7, #T_f7299_row5_col0, #T_f7299_row6_col7, #T_f7299_row7_col0, #T_f7299_row7_col6 {\n",
              "  color: #FF3344;\n",
              "}\n",
              "</style>\n",
              "<table id=\"T_f7299\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th class=\"blank level0\" >&nbsp;</th>\n",
              "      <th id=\"T_f7299_level0_col0\" class=\"col_heading level0 col0\" >id</th>\n",
              "      <th id=\"T_f7299_level0_col1\" class=\"col_heading level0 col1\" >usage_class</th>\n",
              "      <th id=\"T_f7299_level0_col2\" class=\"col_heading level0 col2\" >checkout_type</th>\n",
              "      <th id=\"T_f7299_level0_col3\" class=\"col_heading level0 col3\" >checkout_year</th>\n",
              "      <th id=\"T_f7299_level0_col4\" class=\"col_heading level0 col4\" >checkout_month</th>\n",
              "      <th id=\"T_f7299_level0_col5\" class=\"col_heading level0 col5\" >checkouts</th>\n",
              "      <th id=\"T_f7299_level0_col6\" class=\"col_heading level0 col6\" >publication_year</th>\n",
              "      <th id=\"T_f7299_level0_col7\" class=\"col_heading level0 col7\" >material_type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row0\" class=\"row_heading level0 row0\" >id</th>\n",
              "      <td id=\"T_f7299_row0_col0\" class=\"data row0 col0\" >1.00</td>\n",
              "      <td id=\"T_f7299_row0_col1\" class=\"data row0 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row0_col2\" class=\"data row0 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row0_col3\" class=\"data row0 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row0_col4\" class=\"data row0 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row0_col5\" class=\"data row0 col5\" >-0.01</td>\n",
              "      <td id=\"T_f7299_row0_col6\" class=\"data row0 col6\" >0.01</td>\n",
              "      <td id=\"T_f7299_row0_col7\" class=\"data row0 col7\" >-0.01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row1\" class=\"row_heading level0 row1\" >usage_class</th>\n",
              "      <td id=\"T_f7299_row1_col0\" class=\"data row1 col0\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col1\" class=\"data row1 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col2\" class=\"data row1 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col3\" class=\"data row1 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col4\" class=\"data row1 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col5\" class=\"data row1 col5\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col6\" class=\"data row1 col6\" >-</td>\n",
              "      <td id=\"T_f7299_row1_col7\" class=\"data row1 col7\" >-</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row2\" class=\"row_heading level0 row2\" >checkout_type</th>\n",
              "      <td id=\"T_f7299_row2_col0\" class=\"data row2 col0\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col1\" class=\"data row2 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col2\" class=\"data row2 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col3\" class=\"data row2 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col4\" class=\"data row2 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col5\" class=\"data row2 col5\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col6\" class=\"data row2 col6\" >-</td>\n",
              "      <td id=\"T_f7299_row2_col7\" class=\"data row2 col7\" >-</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row3\" class=\"row_heading level0 row3\" >checkout_year</th>\n",
              "      <td id=\"T_f7299_row3_col0\" class=\"data row3 col0\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col1\" class=\"data row3 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col2\" class=\"data row3 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col3\" class=\"data row3 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col4\" class=\"data row3 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col5\" class=\"data row3 col5\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col6\" class=\"data row3 col6\" >-</td>\n",
              "      <td id=\"T_f7299_row3_col7\" class=\"data row3 col7\" >-</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row4\" class=\"row_heading level0 row4\" >checkout_month</th>\n",
              "      <td id=\"T_f7299_row4_col0\" class=\"data row4 col0\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col1\" class=\"data row4 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col2\" class=\"data row4 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col3\" class=\"data row4 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col4\" class=\"data row4 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col5\" class=\"data row4 col5\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col6\" class=\"data row4 col6\" >-</td>\n",
              "      <td id=\"T_f7299_row4_col7\" class=\"data row4 col7\" >-</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row5\" class=\"row_heading level0 row5\" >checkouts</th>\n",
              "      <td id=\"T_f7299_row5_col0\" class=\"data row5 col0\" >-0.01</td>\n",
              "      <td id=\"T_f7299_row5_col1\" class=\"data row5 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row5_col2\" class=\"data row5 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row5_col3\" class=\"data row5 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row5_col4\" class=\"data row5 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row5_col5\" class=\"data row5 col5\" >1.00</td>\n",
              "      <td id=\"T_f7299_row5_col6\" class=\"data row5 col6\" >0.02</td>\n",
              "      <td id=\"T_f7299_row5_col7\" class=\"data row5 col7\" >0.24</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row6\" class=\"row_heading level0 row6\" >publication_year</th>\n",
              "      <td id=\"T_f7299_row6_col0\" class=\"data row6 col0\" >0.01</td>\n",
              "      <td id=\"T_f7299_row6_col1\" class=\"data row6 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row6_col2\" class=\"data row6 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row6_col3\" class=\"data row6 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row6_col4\" class=\"data row6 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row6_col5\" class=\"data row6 col5\" >0.02</td>\n",
              "      <td id=\"T_f7299_row6_col6\" class=\"data row6 col6\" >1.00</td>\n",
              "      <td id=\"T_f7299_row6_col7\" class=\"data row6 col7\" >-0.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th id=\"T_f7299_level0_row7\" class=\"row_heading level0 row7\" >material_type</th>\n",
              "      <td id=\"T_f7299_row7_col0\" class=\"data row7 col0\" >-0.01</td>\n",
              "      <td id=\"T_f7299_row7_col1\" class=\"data row7 col1\" >-</td>\n",
              "      <td id=\"T_f7299_row7_col2\" class=\"data row7 col2\" >-</td>\n",
              "      <td id=\"T_f7299_row7_col3\" class=\"data row7 col3\" >-</td>\n",
              "      <td id=\"T_f7299_row7_col4\" class=\"data row7 col4\" >-</td>\n",
              "      <td id=\"T_f7299_row7_col5\" class=\"data row7 col5\" >0.24</td>\n",
              "      <td id=\"T_f7299_row7_col6\" class=\"data row7 col6\" >-0.19</td>\n",
              "      <td id=\"T_f7299_row7_col7\" class=\"data row7 col7\" >1.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "klib.corr_plot(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 911
        },
        "id": "Rc7aQfyiNxP3",
        "outputId": "e1899fdb-0872-4462-a99b-16c7a0ddeb2a"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: title={'center': 'Feature-correlation (pearson)'}>"
            ]
          },
          "metadata": {},
          "execution_count": 78
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x1000 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "klib.dist_plot(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        },
        "id": "NVz5z-PTN4DP",
        "outputId": "f1e852a7-067c-47b6-fbf0-75864ccf5a17"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Large dataset detected, using 10000 random samples for the plots. Summary statistics are still based on the entire dataset.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: xlabel='id', ylabel='Density'>"
            ]
          },
          "metadata": {},
          "execution_count": 79
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x300 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABcQAAAEiCAYAAADXm9ECAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAADD+klEQVR4nOzdd3hUZdrH8e/09EpIQq8B6aCggAooICgoYEUFWXUtK6jriyuu3UUBsSsrdrCi0mxrQ1AUUBQEUakKhNAJ6cn08/4xyZCQ0ELCkOT3ua655sw5zznnnpAJz9znOfdjMgzDQERERERERERERESkljOHOgARERERERERERERkRNBCXERERERERERERERqROUEBcRERERERERERGROkEJcRERERERERERERGpE5QQFxEREREREREREZE6QQlxEREREREREREREakTlBAXERERERERERERkTpBCXERERERERERERERqROUEBcRERERERERERGROkEJcRERERERERERERGpE2pNQnzx4sUMHTqUBg0aYDKZmD9/frWe78EHH8RkMpV5tG3btlrPKSIiIiIiIiIiIiKVV2sS4gUFBXTu3Jlp06adsHO2b9+enTt3Bh/ff//9CTu3iIiIiIiIiIiIiBwba6gDqCqDBw9m8ODBh9zucrm45557ePfdd8nOzqZDhw5MmTKFvn37VvqcVquVlJSUSu8vIiIiIiIiIiIiIidOrRkhfiRjx45l2bJlzJo1i19//ZVLL72UQYMGsXHjxkofc+PGjTRo0IAWLVpw1VVXkZ6eXoURi4iIiIiIiIiIiEhVMhmGYYQ6iKpmMpmYN28ew4YNAyA9PZ0WLVqQnp5OgwYNgu369+9Pjx49ePTRR4/5HJ999hn5+fm0adOGnTt38tBDD7F9+3Z+++03oqOjq+qtiIiIiIiIiIiIiEgVqTUlUw5nzZo1+Hw+0tLSyqx3uVwkJiYCsG7dOk455ZTDHueuu+5i8uTJAGXKs3Tq1InTTz+dpk2b8v7773PddddV8TsQERERERERERERkeNVJxLi+fn5WCwWVqxYgcViKbMtKioKgBYtWrB27drDHqckeV6RuLg40tLS2LRp0/EHLCIiIiIiIiIiIiJVrk4kxLt27YrP52PPnj2cddZZFbax2+20bdu20ufIz8/nzz//ZNSoUZU+hoiIiIiIiIiIiIhUn1qTEM/Pzy8zOnvz5s2sWrWKhIQE0tLSuOqqqxg9ejRPPPEEXbt2Ze/evXz99dd06tSJCy644JjPN378eIYOHUrTpk3ZsWMHDzzwABaLhZEjR1bl2xIRERERERERERGRKlJrJtX85ptv6NevX7n111xzDTNmzMDj8TBx4kTeeOMNtm/fTr169TjjjDN46KGH6Nix4zGf74orrmDx4sVkZmaSlJTEmWeeySOPPELLli2r4u2IiIiIiIiIiIiISBWrNQlxEREREREREREREZHDMYc6ABERERERERERERGRE0EJcRERERERERERERGpE2p0QtwwDHJzc1HVFxERERERERERERE5khqdEM/LyyM2Npa8vLxQhyIiIiIiIiIiIiIiJ7kanRAXERERERERERERETlaSoiLiIiIiIiIiIiISJ2ghLiIiIiIiIiIiIiI1AlKiIuIiIiIiIiIiIhInaCEuIiIiIiIiIiIiIjUCdZQByAiIiIiIiIiIiJ1k8/nw+PxhDoMqQFsNhsWi+W4j6OEuIiIiIiIiIiIiJxQhmGwa9cusrOzQx2K1CBxcXGkpKRgMpkqfQwlxEVEREREREREROSEKkmG169fn4iIiONKcErtZxgGhYWF7NmzB4DU1NRKH6vWJ8QnT57MY489RmpqKmvWrMFsLls2fdCgQSxfvpyRI0cybdq0EEV59ObOncv8+fNZsWIFO3fu5KGHHmLcuHFl2qSnp9OlS5dy+5566ql89dVXZda9+eabvPjii2zZsoW4uDjOPfdc7r33XpKSkoDAH6cXXniBRYsWsWXLFqKjo+nVqxf3338/jRs3Dh7nlltu4d13360w5vvvv5/bb7/9sO/rzTff5NlnnyUjI4NWrVpx7733ct5555Vpk5ubyz333MOnn36K1+ulX79+TJkyhZSUlMMeW0RERERERERETh4+ny+YDE9MTAx1OFJDhIeHA7Bnzx7q169f6fIptT4hDoH6MpmZmSxdupQzzzwzuH7btm389NNPREVFhTC6Y/PRRx+xdetWzjvvPGbMmHHYtvfdd1+Z93vw+5w1axa33XYb48aN49xzz2Xbtm1MnDiR9evX8/nnnwOwevVqPvnkE6666ipOO+00MjMzefzxx+nfvz9LliyhXr16AIwfP54xY8aUOf68efOYPn06/fv3P2ycc+bM4fbbb+eOO+7g7LPPZt68eYwaNYpPP/2U7t27B9tde+21rFu3jieeeIKwsDAmTpzIZZddxsKFC7Fa68SvsoiIiIiIiIhIjVdSMzwiIiLEkUhNU/I74/F4lBA/HLvdTp8+fZgzZ06ZBPHcuXNp27ZtlRRjP1Fee+214Cj3IyXEW7RoUSahfLDZs2fTu3dvHnrooTLrx40bR0ZGBo0aNeKMM87gxx9/LJNw7tGjB506dWLWrFmMHTsWgObNm9O8efMyx3n44Ydp06YNHTp0OGycU6ZMYcSIEdxzzz0AnHXWWfz+++9MnTqV999/H4Dly5ezcOFCZs+ezTnnnANAq1atOOOMM/j4448ZPnz4Yc8hIiIiIiIiIiInF5VJkWNVFb8z5iM3qR1GjBjBRx99VGbW2tmzZ3PJJZdU2H79+vVcddVVNG3alEaNGnH55ZezefPmMm2ef/55zj33XJo2bUpaWhpXXHEFmzZtKtPmlltuoVevXnz//ff06dOHRo0a0b9/f1atWlWp93FwyZfj4fV6iYmJKbPu4NexsbHlRl83bNiQevXqsWvXrkMee8eOHSxbtoxLL730sDFs2bKFTZs2MWzYsDLrR4wYweLFi3G5XAAsWLCA2NhY+vXrF2zTunVrOnbsyIIFCw57DhERERERERERERGoQwnxQYMG4Xa7WbRoEQDr1q3j999/Z8SIEeXabtmyhUGDBpGVlcW0adN46aWXyMzMZNiwYcEELQSSvtdffz1vv/02zzzzDH6/P7hfaXv27GHChAmMGzeO1157DafTyahRo8ok52+55RYSEhKq9D2PHz+eevXqkZaWxm233VYurquvvpqvv/6aDz/8kLy8PNauXcuTTz7JoEGDaNSo0SGPu2nTJvbu3UtaWtoh28yZMwe/38/FF1982Bg3btwIBJLbpaWlpeF2u9m6dWuwXatWrcpdBUpLS2PDhg2HPYeIiIiIiIiIiNQMPj94TuDD5w/1O5YTrc4kxCMiIhg8eDBz584FAgnb7t2707Rp03JtH3vsMeLj45k7dy5Dhgzh/PPPZ9asWWRlZfHWW28F2z366KOMHDmSM888k4EDBzJz5kzcbjcffvhhmeNlZWXx8ssvc8kllzBw4EAmT57M9u3bWbFiRbCN2WyustItdruda6+9lmeeeYYPP/yQsWPHMn/+fIYNG1YmCX/JJZcwZcoUbrjhBpo2bUrv3r2Ji4vjlVdeOeSxDcNgwoQJpKamHjbZfbifb2nZ2dlAYCR6aXFxcQDBJH52dna5NiXtSo4hIiIiIiIiIiI1l88PGUWwteDEPTKKji0pPmnSJLp37050dDT169dn2LBhrF+//rD7zJgxA5PJVOYRFhZWps3jjz9O/fr1qV+/Pk888USZbT/++COnnnoqXq/36AM9jJUrVzJgwADi4uJITEzkhhtuID8//7Dxljz27NlzyOM2a9asXPvJkycHt2/ZsoWzzz6byMhIzj77bLZs2VJm/yFDhjBnzpwqeY+HU2cS4gAXX3wx//vf/ygqKmLevHmHTOguWrSIQYMGYbVa8Xq9eL1e4uLi6NixIytXrgy2++mnnxg+fDgtW7YkKSmJhg0bkp+fz59//lnmeCkpKZxyyinB123atAFg+/btwXXPPfcce/furZL3mZKSwuOPP875559P7969ufXWW3n55ZdZs2YNn3zySbDdxx9/zH333cf48eP5+OOPeeGFF/jrr7+49tprMQyjwmNPnjyZxYsX89///pfIyMgK22zYsIFff/31kOVoREREREREREREDuYnMGrbbAK7ufofZlPgfMcySPzbb7/llltu4YcffuCrr77C4/EwcOBACgoKDrtfTEwMO3fuDD5KqiIA/Prrr9x///3MmjWLd999l3vvvZc1a9YAgZLHN910E9OnTy9X1rgyduzYQf/+/WnVqhU//vgjn3/+Ob///jtjxowJtrn88svLxLpz507OO+88+vTpQ/369Q97/IcffrjMfuPGjQtu+7//+z8aNmzIqlWrSE1NZfz48cFt7733Hmaz+YjVJqpCnZhUs8Q555yDzWZj0qRJbN26tVzd6hKZmZlMnz6d6dOnl9tmt9sByMjI4OKLL6Zr1648+eSTpKSkYLfbueKKK8qUVYHyo59LjnFwu+o0YMAAIiMjWb16NcOHD8cwDP7v//6P0aNHc+eddwbbNW3alPPPP59FixYFJ68sMXPmTKZOncqzzz5Lnz59DnmuDz74AKvVelQTXZaMBM/NzSU5OTm4vmTUd3x8fLBd6QsIpduVHENERERERERERGo+qwmsJ2IYrx/cFY8JPaTPP/+8zOsZM2ZQv359VqxYwdlnn33I/UwmEykpKRVuW7duHZ06dQrm4jp16sS6devo2LEjU6dO5eyzz6Z79+7HFughfPLJJ9hsNqZNmxacq3D69Ol06tSJTZs20apVK8LDwwkPDw/us3fvXhYuXMirr756xONHR0cf8n2WlGtu3bo1Y8aMCSbEs7Ozuffee1m4cGEVvMMjq1MJcZvNxtChQ/nvf//L2WeffcgrGvHx8QwYMIDrrruu3LaoqCggMMljQUEBb7zxRjDh7fV6y9XpPlnt27ePffv20bFjxzLrO3XqBFDuloVPPvmE8ePHc/fdd3P11Vcf9thz586lT58+1KtX74hxlNQO37hxY5k64hs3bsRut9OsWbNgu2+//RbDMMrUEd+4cSPt2rU74nlERERERERERESqWk5ODsAR5wbMz8+nadOm+P1+unXrxqOPPkr79u0B6NixIxs2bCA9PR3DMNiwYQMdOnTgzz//5PXXXy9Tdvl4uVwu7HZ7MBkOBJPf33//Pa1atSq3zxtvvEFERMRRVYOYPHky//nPf2jSpAlXXnkl//znP4Mj2zt37syCBQsYOHAgX375ZTAPeeedd3LLLbfQuHHjqniLR1SnSqYAjBo1ikGDBnHTTTcdsk2fPn2CV2a6du1a5lGStHU6nZhMpjK3KsyfP7/KavlUtS+++IKCggK6du0KQL169YiIiGD16tVl2pW8Lv0L+P333/P3v/+93Gjyivz8889s3rz5qMulNGvWjFatWpWruz5v3jzOPvvs4Gj6/v37k52dzbfffhtss2nTJn799Vf69+9/VOcSERERERERERGpKn6/n9tvv53evXvToUOHQ7Zr06YNr732Gh9++CFvvfUWfr+fXr16kZGRAcApp5zCo48+yoABAxg4cCCTJk3ilFNO4cYbb+Sxxx7jiy++oEOHDnTt2pXFixcfV8znnHMOu3btYurUqbjdbrKyspgwYQIAO3furHCfV199lSuvvLLMqPGK3HrrrcyaNYtFixZx44038uijj/Kvf/0ruP3xxx9n3bp1NGvWjI0bN/L444+zePFiVq1axejRo7nsssto0aIFN910E263+7je5+HUqRHiAKeeemqZiTErMmHCBM4991wuueQSRo8eTf369dm9ezdLly6lZ8+eXHzxxcFbIMaOHcuYMWNYt24d06ZNq3Dix6Mxbtw4Zs2adcQ64uvWrStTqH/t2rV8+OGHREREMGDAAADuvfdezGYzp512GrGxsaxcuZKnnnqKrl27csEFFwCB2zRGjx7Nq6++SnR0NL1792bbtm1MmTKFtm3bBt/f+vXrufrqq2nRogWXXXYZP/30U/Dc9erVo3nz5mXimz17NuHh4cHzHM37vOuuu7jhhhto1qwZZ511FvPmzWPFihVl6p336NGDc845h3HjxvGf//yHsLAwJk6cSPv27Rk6dOjR/IhFRERERERERESqzC233MJvv/3G999/f9h2PXv2pGfPnsHXvXr14pRTTuHFF1/kP//5DwA33XRTmQG8M2fOJDo6mp49e9KmTRt++uknMjIyuOKKK9i8eTMOh6PceUoqWwBcffXVFZaDbt++PTNnzuSOO+7g7rvvxmKxcOutt5KcnFxm1HiJZcuWsXbtWt58880j/jzuuOOO4HKnTp2w2+3ceOONTJo0CYfDQcOGDcvk+1wuF+eddx4zZ85k4sSJREdHs379egYNGsSLL75Ypv54VapzCfGj0aJFCxYsWMAjjzzCnXfeSUFBAcnJyfTq1StYnqNdu3ZMmzaNKVOmMHLkSDp06MCMGTP429/+Vqlz+v1+fD7fEdvNnz+fxx57LPh61qxZzJo1i8aNGwdHd5dcdZo5cyZFRUWkpqZy9dVXM2HChDIj2h944AHq1avHe++9x/PPP09CQgJnnXUW99xzT/BDtWLFCnJzc8nNzWXw4MFlYhk5ciTTpk0Lvvb5fHz44Yecd955ZT6AR3qfF198MYWFhTzzzDM888wztGrVijfffJMePXqUaffaa69xzz338M9//hOfz0ffvn2ZMmVKlUwoICIiIiIiIiIicrTGjh3LJ598wuLFi2nUqNEx7Wuz2ejatSubNm2qcPu+fft46KGHWLx4MT/++CNpaWm0bt2a1q1b4/F42LBhQ7kyyACrVq0KLsfExBzy/FdeeSVXXnklu3fvJjIyEpPJxJNPPkmLFi3KtX3llVfo0qULp5566jG9R4DTTz8dr9fLli1baNOmTbntjz76KAMHDuTUU0/l73//OxMnTsRmszFixAgWLlxYbQlxk2EYx1g6/uSRm5tLbGwsOTk5h/1HFhERERERERERkZOD0+lk8+bNNG/enLCwsOB6jx+2FoDdfGIm1fT6we2HppFgO8rzGYbBuHHjmDdvHt98802ZOfGOls/no3379px//vk8+eST5baPGjWKHj16BM/z8MMP88svvwCBuQ8XLVpEly5djvm8h/Laa68xbtw4tm/fTlxcXHB9fn4+qampTJo0ibFjxx7zcd9++21Gjx7Nvn37iI+PL7Nt7dq1XHjhhaxatYrIyEg6d+7Mww8/zEUXXcTTTz/NN998w/z588sd81C/O8dCQ2tFREREREREREREjsItt9zCO++8w4cffkh0dDS7du0CIDY2Nlhje/To0TRs2JBJkyYB8PDDD3PGGWfQqlUrsrOzmTp1Klu3buX6668vd/yvvvqKDRs2MHPmTAC6d+/OunXr+Oyzz9i2bRsWi6XC0dbH4vnnn6dXr15ERUXx1VdfceeddzJ58uQyyXCA9957D6/Xy9VXX13uGMuXL2f06NF8/fXXNGzYkGXLlvHjjz/Sr18/oqOjWbZsGf/85z+5+uqryyXDDcPghhtu4KmnniIyMhKA3r178/LLL5OWlsYbb7zByJEjj+s9Ho4S4iIiIiIiIiIiInLS8BqA/wSd5xi98MILAPTt27fM+tdff50xY8YAkJ6eXqYed1ZWFn//+9/ZtWsX8fHxnHrqqSxdujRYmrlEUVERY8eO5b333gvu36hRI5577jn+9re/4XA4mDlz5hEntzyS5cuX88ADD5Cfn0/btm158cUXGTVqVLl2r776KiNGjCiXKAcoLCxk/fr1eDweABwOB7NmzeLBBx/E5XLRvHlz/vnPf5apK17ipZdeIjk5mSFDhgTXPfjgg1x55ZWcfvrpDBo0iFtuueW43uPhqGSKiIiIiIiIiIiInDCHKnvh80NGUaB0yoliM0OjcLCcgBItcvxUMkVERERERERERERqBUtxcvoE5sMxo2R4XaOEuIiIiIiIiIiIiJwULGawhDoIqdV0/UNERERERERERERE6gQlxEVERERERERERESkTlBCXERERERERERERETqBCXERURERERERERERKROUEJcREREREREREREROoEJcRFREREREREREREpE5QQlxERERERERERERE6gRrqAMQERERqUsMw6DI4yPf5aXQ5aPQ7aPQ7aXA7aPI7cXl9ePxGXh9fjw+P26fgcfnL3MMU8mzCaxmMzarGYfFjN1qxmYxY7OYiLBbiXRYiHRYibBbiHJYibBbsVs1HkLkeBmGgcdn4PMbGBj4DfAbBoYBfr+B2WzCYjZhLX5YzCZMJtORDywiIiIi1U4JcREREZEq4vT42JfvIjPfHXzOLnKTU+Qhu9BDdvGztzjBbRTvZ5Q6hmGUPeZBLw/p4FRb6dybqdRzmM1CXISN+Ag7seE2YsNtgdeRdpKiHNSLcpAQaVfiXOoEl9dHTqGH/YVusgo85Dk95Dm9FLi95Dm95Lu85Du9FHl8uL1+nB4fLq8fl9ePcfCHtRSD8p9Ja/HFqnCbhTCbhXC7pXjZTITdSkyYlegwG9FhVmLCi5+LX0c5rEqoi4iIiFQRJcRFREREjpJhGOQ6vezKcbIzp4hdOU525TrZnetkb56bPKenTJLbMIqfSy2XbLNazITZrITbA8mxMLuVMJsFm9WC1WzCZjFjsZixmU1YLeZAgrtcstzAWzyC3OPz4/H6g8tOtw+nx0tRqWcwcLp85Lh8bM1yYgLMprIjzk0EHrHhNpKiHSRFO0iJCSM5NozU2DBSYsKIDbcpOScnPcMwyCnysCfPxd7Sj3wX+wvcZBW6yXd6A20p9fksXqjosxs89mHOa8KEUVELvx88kO30lWpb/Fzqs1fyuvR2s8lUJkEeG24jvvhCVnxE8SMycKErwm7R51NERETkMJQQFxERETmIYRhkF3rIyCpiW1Yh2/YXkpFVREZWIYVuX7lEmb9U4tthtRAX6Sh+2IkJtxMTbgs+YsPsxIRbcVgtwWR0deeuSso5FLp95Do95BZ6yCnykFvkIbfITW6Rh5wiN1kFLrIL3Hh9PpwFHvYUeDDtyi8TZ8ko8waxYTROiKBRfDiN4iNonBBOUpRDiTg54ZweHzuyi9iZ42R7dlFgOdvJzhwnLm+pzyulLlKV+szaLBaiw21Eh9uJdFiJDLMRYQ+UGIoKsxLpCFy4slvM2K0WHFYzDpsZR/HFq8DnwoS5JIldfPXK6w+UVPH6DHzGgQtXLk9gpHmRx4fL48PpCZROynd6KHR5yXd5KHB6KXB5yHd6cHl8mEwGRQUe9hZ4Aueg1MWsg5LpDmvgLpDESDtxETYSIu3ERdgDCfQIO/GRdhIi7ITZzPq8ioiIVELfvn3p2LEjFouFmTNnYrfbmThxIldeeSVjx45l9uzZJCcn89xzzzF48GAAfvvtN+68806+++47IiMjGThwIE899RT16tUD4PPPP2fixIn89ttvWCwWevbsyTPPPEPLli0B2LJlC82bN2fOnDk899xz/Pjjj7Ru3Zrp06fTs2fPkP0saiqTcbh7/U5yubm5xMbGkpOTQ0xMTKjDERERkRoo3+Vl2/5A0ntbVlHxcyH5Tm8wYeY3Agm0QKETEzERdupFh1EvJox60eHUj3FQLyqMhCg7UXYrFlP1J7mri89vkO/ykpnvZm++i8x8F3tznezLCzyyC1yAEUzAmU0HEnMOq4XGCeE0Lk6QN4oPJMwTI+1KvMlxMQyDzAI3O4oT3tuznezMLmJ7dhH7C9yHvEhV8nmNj3SQEOUgPspBQmTgERdhIy7STqTNgsV8IKF9MjEM8Pj95BV5yXN5yS0KJMlznR5yCgMXsnKL3OQWesgrcuN0e4MJcgAzpUaflxp1bgIcNgsJEQeS5ocacR5ms4TkvYuISO3mdDrZvHkzzZs3JywsLNThHJO+ffuycuVK/vWvf3H55Zfz3nvv8eCDDzJw4ECGDx9O3759eeqpp3j//fdJT0/H7XaTlpbG9ddfz+jRoykqKuKuu+7C6/WycOFCAObMmYPJZKJTp07k5+dz//33s2XLFlatWoXZbA4mxNu2bcvjjz9O69atueeee/jpp5/YtGkTVmvdGfNcFb87SoiLiIhInWAYBnvzXPy1r4At+wrYnFlAemZhmWTawYnvxOgwkmPDSYmLIDU+ggax4STHhhFmNZ+UybMTwe31syfPxc7sInZkF7Iju4jd2YXszXXiN/wVJsrD7RaaxEfQIimSZvUiaVEvkoZx4VgtqlMu5RW6vaTvLyQ9szDwXHzB6uC7M/wcSHxHOmwkxYRTPzacpOgwkuPCSYkJo160A4elbnxeDQOcxTXRswrdgXkLCt1kF7rJKQyUdMorcpNb6MblDZRtKfm8li7ZcnDiPNxuIT7CHkyaxx2UME8oHonusCpxLiIiR++wSc21T8K6JwPLvd6C5L4HtuVvhq/OCiw3Hg6nPVd2328vhP0rA8vDM8pu+2sGrL43sHzas9B4RKVi79u3Lz6fj++++w4An89HbGwsI0aM4I033gBg165dpKamsmzZMhYsWMB3333HF198ETxGRkYGjRs3Zv369aSlpZU7x759+0hKSmLNmjV06NAhmBB/5ZVXuO666wD4448/aN++PWvXrqVt27aVei81UVUkxOvO5QMRERGpM/x+gx05RWwuSX7vK2TLvgIK3N4yo0d9xc/xkYEEWv2YCBoUlwBJjQuv04nvQ7FbzcVlUsKBhOB6j8/P7lIlK3ZkFbI7p4h9uUW4nD5ydubx+6684IhVu8VMk4QIWtSLpHm9QKK8SUKEJvOsQ7w+PztznMGE99biJPi+fFeFiW+z2UxiVODOjKSYwMWplNhwUmPDiQmz1vnPqskE4TYL4bEWUmIP/eXQb0CR20dWkZuc4qR5TpGH7IKS0ebuYDklj8+H0+kj21nElv1Fh02cR9itFYw0DyTQSyfT9RmXUPP7A2WMfH4Df/Fz8GEY+P0UPxuYTBTfQWIKXOw1Fz+biteZS5YDz3aLGXNd/2MkUhU8uVC0PbDsc5XdZvgObHNnld/XuffA9oN5Cw5s8xYeV4idOnUKLlssFhITE+nYsWNwXXJyMgB79uxh9erVLFq0iKioqHLH+fPPP0lLS2Pjxo3cf//9/Pjjj+zbtw+/PzBEJz09nQ4dOlR43tTU1OA56lJCvCooIS4iIiI1msfnJyOriM378oOJ762Zhbi8PvyUT6ilxEXQICGShvGBBGzDhAii7JY6n0w7XjaLmUYJETRKiCiz3u0NJD237S8kPTOfjP2FbN+fT77Hx9rd+azfXVyj3AQWk4nGCeG0qBdF83oRtEiKomlihEae1nAlNfm3lhrtnZ5ZSEZ2ER6fv0ypk5KLVLERDlLiIkiJi6BhfOCRGhdWZ0Z7VyezCSIdFiId4TSKC6+wTcm/R6HbR3ZhYALSnEIP2UWB0eY5xSVaSkaee30+nE4vWU4vpszDjziPclgDdcyLk+YJpUaal9Q3jw23KalYB/n9Bi5voMa+0+vD6fEHJob2Bmrtl9TfL9nm8vjw+AzcPj8urx938cTS7uJlt6/sc0ld/4Nvkj/eW+YP/k21WszYLCbsVjN2S2DeAbvVVDwPQeBhK14Os1pw2ALPYTYLYTZz8NlhLf18oJ3NYlIZMqn9bDEQ3jCwbHGU3WayHNhmjy+/b1jSge0Hs0Ye2GaNqLjN0YZos5UNy2Qqs67kc+r3+8nPz2fo0KFMmTKl3HFKktpDhw6ladOmvPzyyzRo0AC/30+HDh1wu92HPG/pc8ixUUJcRESCSr6IuIsn/vL6DkwC5vUbgcRi8ZcIvxF4DYERMVZLYKSMxWzCYjJhNgdGkjpKJiCzavIuOX5Oj4/0/YX8tbeALZmB0d/bssom1UqS3zaLhdT4SBokRNA4MZKmiVGkxoURZjHX2PreNZHdaqZpYgRNEyOgdWDSIL/fYE+eiy37CkjfX8C2zAJ2ZBVQ6PKwcW8hm/YWBkuumE0mmhQnyVskBcqtNFGS/KTl9PjIyCoifX9BccmTIrbuLyhXk7/kLg271UJKbAQp8RE0iI+gYVzgoopGfIdWSUI7ymEhylFyR0jFfP7AhL2ly7TkFI8+zy1ezisece7z+3EWecks8mLaV3jIxLnFZArWNj+QOD+QMI+PDGwLt1nUtwgxv9+g0OMrnhDWS4HLR4HbS4HLi/OgBLazZCLZkqS358D6Io8fd3Epn4MT1MbBy0b5bSXrjQr2qeg4FTGbzMH5BAIXZAIT5Pr9RqmyasaBPrBhlM+GB34o4AETPsBTZlNJ89K/76WfS88BUGZ9qWWzyRRIjtsshNsOSppbyyfUK0qwO4r76LbiRL3DFki02y3qr8tJ4pQ7Ao+KRDUvXwqltD4fHXpbizGBxwnWrVs35syZQ7NmzSqs9Z2Zmcn69et5+eWXOeusQDmY77///kSHWacoIS4iUsv4/AZ5Tg85RQceBS4v+S4f+U4PBW4f+S4vhS4vRYf4IgJHN1qmdJvDdZ1LttmKE+MRditRDgsRdiuRdgsRjsBzdJiN2AgbceGB26rjwm1Eh1lVZ7iOynd52bIvkPjevDdQ83tnthOfYRxIfhcnwMPsVhrER9IwIZJGiZE0TYgkNTYMq9mk5PdJyGw2kRIbRkpsGGe0TAQOTJq4ZV8B6ZmBx/b9BeQ73WzcW8ifewsxrys7krxlUhQt6kXSIilK5VZOMMMw2J3rCtb4Ts8sIH1/IbtzXWU/owQ+p2AqLnMSTmpCYMR3o/gIkqIc+pzWcBaziegwK9FhVpokHLqdz29Q4AqUaskucBcn0AOJ85zCA+Va8pwewKAoz83OPHdwToLDTQwaSJzbSiXMDyTS4yNs6kccgdfnL5PIznd5KXT7KHB5KXAH+pCFLm/xa9+BxHdxX9JfasT1oZLXpRPVhnGItsVMJhN2a8nIakvgUZLItVmCAy5sJSOtLWVHXVtLRmJbSm23mbGaTVjMZixmigdvBAZyBP4GBX65juVPUcn7Dg4a8RvBQSVunx9PqZHq3oNGrpceye7y+gLPHh8ur694v8CIeLfXFxjh7vHhDY4ANSjy+cB5YC6AwM+Nsq/L/EzLxl7R+yxZZ7dasJeMcC/uu5eMdi+zzmImzBbo10eW9OuLn6PDrMRH2Imw66KVCMAtt9zCyy+/zMiRI/nXv/5FQkICmzZtYtasWbzyyivEx8eTmJjISy+9RGpqKunp6UyYMCHUYddqSoiLiNQgXp+fvfkuMvPd7Ct+3l8QWN5fEKgBmuv0YhQnI6DUF5BSXz6O5suI2WzGZjZhsZixmotHzBQP1zMDFNdShMAXgsAjcLtWSW1Gj8/A6zuQZHe6/eS5/ZgKvUDZDryJQ3foY8JsxAUT5TZiw23ERthJjLSTGGWnXpSD+Ag7Fg0nrJH8foNduU62ZhaSvj9Q7mTr/kL25R2oI1x6RGlUuP1A8jshkqaJkSRF2ZVUq+FMJhP1ohzUi3JwWrMDWbXMfDeb9+WzNbOAbfsKyNifT0HxSPI/9xUGa5JbzSaaJETQMimS5sWjyZskRGBTIuy45bu8xaO9A0nvrZmFbNtfhMvrq/AzGhlmIyUukpS4cBrER9IoIYIGceGEqyZ/nWYxm4gJtxITbqVpwqFvU/f4DHKLPOwvKCnVcuA5t+hA8tzp8eJ0+ch1FbE1K1DfPNA/OahPAcSG2w4aYV6+VEu0w1pjE3eGEbjDr6DUCO3CUons/FLJ7EJ3oE1+qRHd7uLPMhx+xLVhHLrfaLUERiuH262EO6yE2wLJa4fVgs1i4Fn7DRaTieTu5xHusOOwlRrRbLXgsB8oG2KzHKjZfTL/k5iLg7OW+sMWSdXevVTyMzcIXFQqKSlTeqR9IJEeSKy7PGUT7CXJdLc30MZd3Nbj9eP1+YJ3gpa+qOF0l020lyiTaDeVT7yX/rcqWXRYLcQX3+0RF2EjKdpBg7hwGhY/Ih1KSUnd0KBBA5YsWcJdd93FwIEDcblcNG3alEGDBmE2B+7MmDVrFrfeeisdOnSgTZs2PPvss/Tt2zfUoddaJuPgAlo1SG5uLrGxseTk5BATExPqcEREqoRhGOwvcJORVcTOnCJ25TjZWfzYm+cK3KZJ2S8mBy+bMBHhsBIVbiPKYSPCYSXcbiXCHniOdARGZYdZLYTZA19Wwqzm4mVzmQmBDtfZLR978XOpdX7DwO314ywZFeP1Uej2UVQ88qjI7Q3ealvg8pJXFBgdllfkocDpoSS1H7xNlAMJ9NJfes0mEwmRdpKiHNSLspMU7SCxeLkkyRZuV4mFUCty+9hanPROzyxkS2bBIRNrfgKTXTZIiKRhQgSNi5PfCZF2LCfxF2SpXn5/4G/k5n0FbMnMZ1tmAdszCyh0e4IjSUv+TtjMZpokhNOsXiTNEiNomhhIkusLeMXyXV62ZwUmRd2eXRQsfZKZ765wkkurxUz9mECd75TiEd+NEiKIC7dhMZ3ciSyp2fwGFHmKy7QUBEq1ZBUEyrPkFtc2zy10k+v04Pf7y5RlqahECwSSmlFhVqIdgbvTYsKsRIfZiHRYibBbCLdbiCh+hNushNnKjkIOjFY2YTMHynJVlFw3iktteEtN5uj1GQdqZAdLiZQqKVI6AerxU+Q5MHK7ZJR2gduH1xcYPXw0AyIqSmobRuBuq3C7lXB7cWLbZilObluJcFiILN2PtFuL7/AL/HzsVvMhE9get5sZT08CYMztd2Oz24/531yqXunfBY/fj8drFI9g95UdzV6qDntwufi1q/j3tMjtDdR7L74YU+jy4vSUHQBTug9f8hwXbqNRfARNEsJpmxrDKakxxIbbygcrtZLT6WTz5s00b96csLBDTwYtcrCq+N1RQlxEJIR8foMd2UWBWsiZhWzdF3jOc3rKJR9KOq1Wi4W4SDuxEQ5iI+zERzqIL74tOLZ4IqroMGtgtCw1OyHh8xvkubyB26mLPME6pDnFI8TyijxkFbjIKXTjNwJfBA8kyMt2uAEi7VaSogPJ8pIkeb0oB0nRDpKiHMSE19zRYScTwzDILfKyPTuQWNuRU1S87GRPnvPAxZuD6n3Xjw0vriMcSeP4CBonRhDlsCr5LUfk9wfKrWzeV8DWffmkZxaQkZmP0+MN/D2A4MSdJiApykGzehE0Lk7iNowPp0FcWJ2oS24YBnvzXcWJb2cw+b0ju4icIk+5pFnpi1PJceE0iIsktTjxnRITht1i0qhvOSkZBvgMgxyntzhpfmBy0JLa5iVlWgpcB+o8H3zHWkV3sFFq3aFeB9oH+mIlgwxKktblYj3S60OUHTk4+W02mQi3WwmzWwmzBZLWJaO1wx2BgRGBpL6VSEcgsR1ZPGAiwmHBYqqeO638Ph+b/lgDQKt2HTFbav/f2rrOMMDp9QXnFSh53pfnZE9uEbtzisgrdAcHt5RczDaboGFcOO2Kk+PtG8QQH6kLKLWVEuJSWUqIKyEuIjWM1+fnr30F/LEjl7U7c1m/Ky9Qe5EDX2x8RnHpgOhw6sWEUS86jKSYMFJiwqkfE0Z8hC044Y8EeP0GOYUeMgsCZWQy813sL3CRXeAqrlPqwuk+MErlUAlzu9VMvUgH9WPKjiwPjDa3kxjpUFmWUrw+P3vyXKUSaoHk2o7sIvJd3goTawYQHW4nJT6SBnERNEyIoFFxvW+bfq+lCvn9gcTvln0FZGQVkrG/gJ1ZheQUuKDUF++SvwNmk4nkaAeNE8JJiQkjOSZQ4zwlJozEqJr12ff6/GQWuNmT62JPnpPdxc+7cpzsyHaWuSOj9P89BhAX4aBe8f839WMjaJwQSH5H2i26OCW1kmGA2+cn1+kl3+kh3+Ulz+mlwOkhr/hOtiK3LzgS1uXxFj/78foDtaE9Xj9GuTT20Supk+2wmrHbLIFazTZzmTraJbWzw6zFc684LMG7/0rmZXFYAyXuavqACKkbCtw+dmQXsSu7iC378vlzTy67sgrLJMfNJmiXGsNZretxeotEonSHV62ihLhUVo1PiL/wwgu88MILbNmyBYD27dtz//33M3jw4KPaXwlxEakJdmQX8dOW/fyakcPG3fk4ixMRfiOQgLBbLaTERdIgIYKG8ZE0TgxMNqZaq1XHMKDQ4wvUWs93sS8vUHN9f34gaZ5V4ApO4lVRwpzgOhOJkXbqRdkDk35G2IpH5tuCE3fFRdiJCasdI81dXh9ZBZ5gnfo9eS725gUSa3vzXOzLdwfr1Vc0gV58lIOkmHCSYsKoHxNOamw4DeLDiS0upyASCnlOL+n7AwnyHdlF7MkpYld2YXA0eUUXzCxmE8nRYdSLspMQZQ9O4pcYZSch0hGcUNBuMVfrZ9/r8weTdblOD9mFB+osZ+YXPxcElg/+bJa+KHWRYxaJYS6sjng2JP0jMNFlXDipsWFEKPEtclQOHqHt9RmB5LjfX+pOKCP4ucMwDkzoaDZjMQVGjpdM7li6RF0t6EKIVEq+08v6Xbls2J3Hn7tzycjMx2wCiwnsFjPdmsRxZqt6dGsaVyfu6qrtlBCXyqrxCfGPP/4Yi8VC69atMQyDmTNnMnXqVH755Rfat29/xP2VEBeRk5FhGGzJLGT55kyWb84iI6swmCT0GRDhsNE0KZqWyTG0To6hSWIEdk0GGFKGAW6/Pzi6PDPfxf58N/sLXGTlu8qUZTlUHUQ4MNLcYjKVSZLHhtuICrMS5QgkzaIcNiIdFqJKTToVXjyJVHUl0/z+QI3SIrePIk9gMq3cIm+gXrsz8Jzr9JJd6GZ/gYesQjcFrsCo+orq1PuLew82iyWY9E6KCSclLpzk2HBSYsMIs5qDPyeRk5nfb5BT5GV7diE7s4vYneskM99JZp6L/flOfIepQVw6iWW1mIO1hyMd1uLJ4czBuRlsFhN2qwWTqeTzZBxIqhkGbp8RrCPsLqkbXDxRXq7TQ1HxRGcV1QgunfAGsFjMJESGER/lICHKQUJUGPVjHCRHO6j3WRcs7j00So5m3wUZJ+AnLCJ1md/vJ2vvHgDik+pjNmuyYzk6mfkufvgrk5//2sfOrALMJrCaINJh5YIOKZzfKVWjxmswJcSlsmp8QrwiCQkJTJ06leuuu+6IbZUQF5GTye5cJwvW7mbZn5mByS8pKX9ipnn9GNo1iqdNagyN4sID9b2VJKxRfH6D7EIPe/NdwYm7coo85BZ6yCtyk+t0k1/kodDlxSgeaQ7lE2hQPoFesmwxmwi3WwizWrBaTIFJuoqfrebAMsF9y/4C+fx+PD4Dr9+Pz28Eln2BybeK3H5cXl+w7dFOuGUQSHjHhNuIi3QEE2v1osKoF+0gMdJBfIRNv89Sq/mKJ/Hck+sMjsbOLvkbUBh4FLg8lHSpS3/2ofzn/XAflcN9NkueTZgId5TU/7UF5o4ofiREOoiLtJEQ4ThkeS2X08lt11wGwCv/7oK388PH+iMRETkmmlRTqkJGViE/bNrHz5v3kVPowgJEOSwM7pDCBZ1SiQnTZJw1jRLiUllV8btz0lxK8/l8fPDBBxQUFNCzZ88K27hcLlwuV/B1bm7uiQpPRKRCfr/BL9uy+PKP3azelo2veBS4xWyhdYM4OjROoEvjOGLDrEoY1nAWsylQRzzq0F/i/MV1SPOcXnIK3WQXT/6ZW+Sh0O2l0OWl0O2lqNSzy+OhzZ5vAfg98UwKvAYmvGWOe7hfnZJtR5yQq/jZbDLhsFmIdNgIt1uJDLMS4bARWbwcG16qHEyknQibRTXrpU6zmE2BiXejHRVuNwzwGwZFHj/5Lg/5Ti8FLi/5Li9urx+3z4+nuMaw2xeoN2wYpe+cOPD5slnM2KxmHNbiWsHFNYMj7FaiHVaiwq1E2q2BzySVu/vCZDIRG58AQGHbe1BaSkSqm8lkIiIqOrgsUhmN4iO4pHsTLj6tMSu27OfT1RnszCrkgxXb+XTNLga3T2ZI5wbEhisxLiJHFvIR4mvWrKFnz544nU6ioqJ45513OP/88yts++CDD/LQQw+VW3+kEeIffPABL774Ihs3bsQwDFJTUzn99NO57777SEpKAgL1zFu1asWAAQOOGPOnn37KqFGjWLVqFU2aNDnKd3poZ599Nh07dmTatGnHfazKcLvdPProo7z//vtkZ2fTrl077rvvPvr06VOm3dSpU1m6dCm//PILubm5fP3113Tt2rVMm6+++opnnnmG9evXk5eXR2pqKueffz533XVXmX+jhISEQ8bzxx9/kJKSUuG29PR0unTpUm79qaeeyldffVVm3ZtvvsmLL77Ili1biIuL49xzz+Xee+8N/puLHI+cIg+L1u1hwdrd7M1zBRLhmGidEkePVvXp1DiWSJtFSXA5rAOT2Rk43YHJuoqKJ+ry+P2BeqR+Px5v4NlbXKfkwP/cB15bLWasZhNWiwmLyYzVasJmNmO3mgNlWezFZVnMB2qVKsktIiIiIjWRYRj8sjWLT1dnsH1/AVYTRNgtjOzemMEdUjCro3vS0whxqaxaMUK8TZs2rFq1ipycHGbPns0111zDt99+S7t27cq1vfvuu7njjjuCr3Nzc2ncuPFhj//ss8/y0EMPcfPNNzNhwgQMw2Dt2rXMnj2bXbt2BZOj06dP57zzzjuqhHht8+9//5v33nuPe+65h1atWvHOO+9w+eWX88UXX9C5c+dguxkzZtC8eXP69OnDxx9/XOGxsrKyOPXUU7nhhhtISEhg7dq1TJkyhbVr1zJ37txguy+++KLcvv/4xz+IiIg4ZDK8tPvuu48zzzwz+DoqKqrM9lmzZnHbbbcxbtw4zj33XLZt28bEiRNZv349n3/++RGPL3IoBS4vH63ewWdrduH0+vAa4LBZ6d6iPme2SaZRXJiSjHLUSkqpmE0mbMWT8omIiIiIyOGZTCa6NUuga9N4Vm/L5pNVGWRk5vPq0i0s/SuTf/RtScO48FCHKSInqZB/87bb7bRq1QoIjPL96aefeOaZZ3jxxRfLtXU4HDgcFd+ueigvvvgiI0eOZOLEicF1AwYM4NZbb8Xv9x9f8LXAjh07mDlzJo888gg33HADAOeeey5nnXUWjz32GG+//Xaw7Zo1azCbzXz//feHTIhfdtllZV6feeaZ2O12/vnPf7Jz505SU1MB6N69e5l26enp/Pnnnzz44INHFXeLFi3KHaO02bNn07t373J3FIwbN46MjAwaNWp0VOcRKeHy+vji9918uGo7eU4vXj+kxEfRq20KPZonaDS4iIiIiIjICWYymejSJJ7OjeNYvH4Pc3/eyu878xj/wa9ccVojhnRugEUjlkTkICfd9M5+v79MnfDjlZOTc8gRxyWzW3fu3Jlt27bxyiuvkJCQQEJCAu+88w4AHo+Hu+++mxYtWtC0aVPGjRtHQUFBpeP58ccf6devH6mpqfTq1atcmY8Sy5cv56KLLqJRo0Y0bdqUv//97+zduze4vUuXLvzrX/8qt999991H+/btjzrZ//vvv+Pz+ejXr19wnclkol+/fixcuBC32x1cX9nZwEvKo3g8nkO2mT17NiaTiYsvvrhS5ziY1+stV0ZHE69KZfj8Bl+v3c3ts1bx1g9bySryEhcdwdV92nDXkA6c0yaJKLuS4VJ5Xq+XBR9+wIIPP8Dr9R55BxGR4+Bxu5nx70HMvKsv0V+feeQdRESOk/o6ciKYTCb6tE3mgWGdSUuNo8DjZ+aP6dwz7ze27S8MdXgildK3b19uv/324OtmzZrx9NNPhyye2iSkCfG7776bxYsXs2XLFtasWcPdd9/NN998w1VXXVVl5+jcuTOvv/46b7zxBrt3766wzZtvvklycjIXXnghX3zxBV988QUDBw4E4OGHH+a1115j3LhxvPbaa/j9/grrmB+N3bt3c+mll2K324PHHD9+PDt37izTbvny5Vx44YXExMTw6quv8tRTT/HLL7+U+bmMGDGCjz76CJ/PF1xnGAbz5s1j+PDhmM1mbrnllsPW6gaCFx8OHnnvcDhwuVxs3bq1Uu/V5/PhdDpZvXo1U6dOZfDgwYettz5nzhx69epFw4YNj+r448ePp169eqSlpXHbbbeRlZVVZvvVV1/N119/zYcffkheXh5r167lySefZNCgQRodLkdtw+487py9mpcW/8WefDcR4Q5GnN6Kuy/sxBnNE7BZlAWX42f4/Wxe/web1/+BoTuXRKSa+f1+fvjTzrItMZgL0kMdjojUAerryImUEOXgtoFtGdW7JVarlXV78rlzzhoW/FFxPkikJvnpp5+C1R3k+IS0ZMqePXsYPXo0O3fuJDY2lk6dOvHFF19UaR3vqVOnMnr06OAVlaZNm3Leeefxj3/8I5ig7dSpE3a7nfr165cpw5GVlcVrr73Gbbfdxj//+U8gUE5kyJAh5ZLYR2P69OlAYJLPktHKDRs2ZNiwYWXaPfzww3Tp0oU33ngjOAt3u3bt6N27N1999RUDBgzg4osv5umnn2bx4sXB0d3Lli1jx44dwVHWZrMZi8Vy2JhatGgBwIoVK8okrH/++efgz6AyOnXqFPwZnXvuubz00kuHbPv7778HE9ZHYrfbufbaaznnnHOIjY1lxYoVPPHEE6xatYoFCxZgswVmlL7kkksoKCjghhtuCI5M79OnD6+88kql3o/ULS6vjw9+zuDTX3fi8RvY7Tb6d2hIv7bJRNhOuhtrpIazWCz06j84uCwiUp2sVivX987E5MnBHB4X6nBEpA5QX0dONJPJRO+0+rRvGMdbS//it4ws/rv4L7ZnFzHqjKaacFNqrJJ5EOX4hTSz8+qrr7JlyxZcLhd79uxhwYIFVT6pZbt27Vi6dCnvvfceN954IzExMbz00kuceeaZrFmz5rD7/vHHHxQVFTFkyJAy64cOHVqpWFasWMFZZ51VpnTH2WefTXx8fPB1YWEhP/74IxdddBE+nw+v14vX66VVq1Y0bNiQlStXAtC+fXvatGlTZqLKuXPn0rJlS7p27QrAc889V6bMSkXatWtHz549eeihh1i+fDn79+/nueeeY8mSJQDBhPyxev/99/n88895+umn2bBhA1deeWWZ0eylffDBB9hsNi666KIjHjclJYXHH3+c888/n969e3Prrbfy8ssvs2bNGj755JNgu48//pj77ruP8ePH8/HHH/PCCy/w119/ce2112IYRqXek9QNm/bkcffcNXy0egdOn0HHZkn8+6IuXNAxVclwqRZmi4X23XrQvlsPzPqSKCLVzGK1ctqtSzj1/34ja8CvoQ5HROoA9XUkVOIi7dzSvw1DujTG64f5q3cy+Yv1FLkrzk2IHK2+ffsybtw4br/9duLj40lOTubll1+moKCAv/3tb0RHR9OqVSs+++yz4D6//fYbgwcPJioqiuTkZEaNGsW+ffuC2wsKChg9ejRRUVGkpqbyxBNPlDvvwSVTnnzySTp27EhkZCSNGzfmH//4B/n5+cHtM2bMIC4uji+++IJTTjmFqKgoBg0aVKlBvrVNncju2O12BgwYwKRJk/j222/54IMPKCoqYurUqYfdb9euXQDUq1evzPrKXpHZvXt3uWMdfPzs7Gx8Ph/33HMP9evXL/PIyMhg+/btwbYXX3wxn3zyCW63G6/Xy0cffcSIESOOOa7//ve/JCYmMmjQIFq1asUrr7zCnXfeCXDI+utH0r59e3r06MHo0aN5++23+e6778okrEsYhsHcuXPp379/mQsDx2LAgAFERkayevXq4DH/7//+j9GjR3PnnXfSu3dvLr/8cl588UW+/PJLFi1aVKnzSO3m8fl5d3k6983/nfT9RTgcdkb1acu1Z7ciISLk8w+LiIiIiIjIMTKZTAzp2ojr+6ZhMptZviWLez78jb15VTd3nVQ9p9OJ0+ksM6DR4/HgdDrLzU9X0rb0XHperxen01lmXrzDta2MmTNnUq9ePZYvX864ceO4+eabufTSS+nVqxcrV65k4MCBjBo1isLCQrKzsznnnHPo2rUrP//8M59//jm7d+/msssuCx7vzjvv5Ntvv+XDDz/kyy+/5JtvvgkOij0Us9nMs88+y++//87MmTNZuHBhufkGCwsLefzxx3nzzTdZvHgx6enpjB8/vlLvuTapk1mec889lw4dOrB+/frDtitJBu/bt48GDRoE1x9p1PWhJCcnl7n6U6L0utjYWEwmE3fccQfnn39+ubaJiYnB5REjRvDoo4/y9ddfY7fb2bdvX6UmpWzatClff/016enpFBYW0rp1a6ZNm0ZKSgqNGzc+5uMdrH379thsNjZv3lxu2w8//EBGRgYPPvjgcZ+nxL59+9i3bx8dO3Yss75Tp04AbNmypcrOJbXDzpwinvxyA1v3F+L1Q8dmSVzao5kS4XJCGIZBbtZ+AGLiEyp9Z46IyNHw+/3s3R0Y9JGUnFLpSdNFRI6W+jpyMjiteSL1ohw8v2Adf+4r5K65a7h7UBtaJ0eHOjSpQEmi+M033yQ2NhaAefPm8dZbbzFw4EDGjh0bbDtq1ChcLhcvv/wyycnJAHz66ae8+uqr9OnTh//7v/8Ltr3++uvJzc3l+eefD5YN/vrrrznvvPOOOcbOnTtz7733AoE5EidPnky9evX4+9//DsD999/PCy+8wK+//sqCBQvo2rUrjz76aHD/1157jcaNG7NhwwYaNGjAq6++yltvvcW5554LBBLuR5oD7+AJNydOnMhNN93Ef//73+B6j8fD9OnTadmyJQBjx47l4YcfPub3W9vU+mzPnj17qF+/fpl1RUVFbN++nbZt2wbX2e12nE5nmXbt2rUjPDycTz75JJhMhUA5jsro1q0br732Grm5ucGyKYsXLy5TpzsyMpLu3buzfv167rnnnsMer0WLFnTr1o05c+bgcDjo2LEjaWlplYoNCP4xKCoq4u233+bqq6+u9LFK+/nnn/F4PDRt2rTcttmzZxMVFcXgwYMrffwvvviCgoKCYKmYevXqERERwerVq7n88suD7UpGkFdFkl9qjzUZOTy1YAN5Li92u43LerTg9OYJaL5MOVG8Hg/vv/I8AGNuvxub3R7iiESkNvO43Txw+00APDPzfRxhYSGOSERqO/V15GTRLCmKf1/YkWlfrWdHVgH3ffQHdw5I49RmlbtbXeq20nlCi8VCYmJimYGZJcn5PXv2sHr1ahYtWkRUVFS54/z5558UFRXhdrs5/fTTg+sTEhJo06bNYWNYsGABkyZNYt26deTm5gZHxhcWFhIREQFAREREMBkOkJqayp49eyr3pmuRWp8QP/PMMznvvPM455xzSElJYceOHbzyyitkZmZy4403BtulpaXx3XffsWjRIuLi4mjatCkJCQmMGTOGZ555hvDwcDp16sTcuXMrHOl86qmn0rhxY+bPn3/IWG666SZeffVVLr30Um6//Xays7OZPHkyCQkJZdo99NBDDBs2jGuvvZYRI0YQFxfHjh07+Oabb7jyyis588wzg21HjBjBpEmTsFgs3HHHHWWOM27cOGbNmnXEEe0vv/wyMTExNGzYkPT0dP773//icDi47bbbyrRbsmQJ+/btY926dQB89913pKen06RJk2AyevTo0XTp0oX27dsTFhbGb7/9xvPPP0/79u254IILyhyvpMzL+eefT3h4eIWxJSUlccUVV/Dcc88BcO+992I2mznttNOIjY1l5cqVPPXUU3Tt2jV4fJPJxOjRo3n11VeJjo6md+/ebNu2jSlTptC2bVvOPvvsw/48pO748vddvL50C26fQYOEaP7WN42UaHXQ5cSzOxyhDkFE6pCIMAsYEPHnc/ja3xnqcESkDlBfR04WCZEO/nVBe175diNrtmUx5csN3DO4DZ0bx4U6NCnl/fffB8BR6m/H8OHDufDCC8tNzvvmm28CgYGuJS644ALOO++8cnfCvfLKK+XalozIPlY2m63Ma5PJVGZdyd0wfr+f/Px8hg4dypQpU8odJzU1lU2bNh3z+bds2cKQIUO4+eabeeSRR0hISOD777/nuuuuw+12BxPiFcWpufXqQEL8rrvu4vPPP+fee+8lMzOTxMRE2rVrx/z58znrrLOC7e69917Gjx/PNddcQ35+Ps8//zxXXnklDzzwAD6fj2effRa/38+QIUN44IEHuOmmm8qcx+v1HnLSyBIpKSm8//77TJgwgb/97W80a9aMqVOnMnHixDLtTj/9dP73v/8xefJkxo4di8fjoUGDBpx99tk0b968TNvhw4dz//33YxhGufrhfr//iDEBuFwupkyZwo4dO0hISGDIkCH8+9//JjIysky7yZMnByfbBIJlTkaOHMm0adOAwCj4efPm8fTTT2MYBo0bN2bUqFGMHTu2zB8cgIULF5KZmXnYMi8+n69Mbac2bdrw2muvMXPmTIqKikhNTeXqq69mwoQJWK0Hfp0feOAB6tWrx3vvvcfzzz9PQkICZ511Fvfcc0+ZP6hSN3l9fmYu28qXv+/CY0CnZklc1asFkZo0U0LAZrdzzW0TQh2GiNQRjrAw3r7iJyzO7fjSf2K3EuIiUs3U15GTjcNm4cZz0njlm42s2rqfyZ9v4L4L2tKuQUyoQ5NiYRXcwWaz2coldw/V1mq1lskRHaltdSup7tCsWbMKz9eyZUtsNhs//vhjsHpDVlYWGzZsoE+fPhUec8WKFfj9fp544olg4r/kQoIcmcmowZcFcnNziY2NJScnJ1iCRETkcPJdXp76agO/bc/BY5gY0Kkx53dqgE01UkREpI5I/qxRICEe1pDdgzNCHY6IiEhIeH1+Xli4gd8zsoh2WHhwyCmqKX4COZ1ONm/eTPPmzStMVJ/M+vbtS5cuXXj66aeD65o1a8btt99epq63yWRi3rx59OjRgy5dutCnTx/+9a9/kZCQwKZNm5g1axavvPIKFouFm2++mc8++4zXXnuN+vXrc88997Bw4UKuu+664HlKn2P16tXBGIYOHcqSJUu4++672b59O1lZWcTFxTFjxoxghYoS8+fPZ/jw4TV6lHhV/O5oOKSI1Bl78pzcO28Na7bnYJgtXHVWGkO7NFQyXERE6pSs094is9fnZJ32VqhDERERCRmrxcyN/VrTJjWWPJePhz9dx197C0IdltRCDRo0YMmSJfh8PgYOHEjHjh25/fbbiYuLC47unjp1KmeddRZDhw6lf//+nHnmmZx66qmHPGbnzp158sknmTJlCh06dODtt99m0qRJJ+ot1XgaIS4idcKePCcPf/wHu/NcRIY7+FvftrSqF4EmuZdQ83m9fPflJwCcNXAIlhNwy56I1F0ej4e3Xw6Uurvq77dUeOuxiEhVUl9HTnYuj49nvlzLX3vySIiw8Z8L29E4ISLUYdV6NXmEuISWRoiLiByF0snwmMhwxp3XgdZJSobLycHv97Pxt9Vs/G11mTkTRESqg9/n44dvF/LDtwvxH8VcMyIix0t9HTnZOWwWxg1oS5PEKPYXenjwk7XsynGGOiwRqUa6NCsitdrePBf/+WRtMBn+jwHtSI2xH3lHkRPEbLHQo0//4LKISHWyWC2MuGpMcFlEpLqpryM1Qbjdym3nteXJz/5gR3Yhj36+ninDOxBu1++sSG2kkikiUmvtzXPx8Cd/sCvXSUxkGP8Y0F7JcBERqfMsBZvB8IHJgi+yeajDEREROWnkFLp59OM15Ba56dk8gbvOS8OkW4urhUqmSGWpZIqIyCEoGS4iIlKxeovPIvmr1tRbfFaoQxERETmpxEbYufmcNMwmMz9s3s8HK7aHOiQRqQZKiItIrbO/wM1/SiXDb+6vMily8jIMg4K8XArycqnBN22JSA3h9/vJLLCRWRiGX39yROQEUF9HappmSdFc1bM5fgNm/bSNn7bsD3VIIlLFlBAXkVrF6fEx9Yt17Mx1Eh0RSIY3iHWEOiyRQ/J6PLzzwlO888JTeD2eUIcjIrWcx+3m+g868rePBpCTdFGowxGROkB9HamJeqXVp98pKXgNePrrTWRkFYU6JBGpQkqIi0itYRgG0xZt4s+9BdjtNm449xQlw6VGMJnNmMz6L1lETgyzxYLZYiGv4xOhDkVE6gj1daQmurhHU1qnxJDn8jHp8/UUur2hDklEqogm1RSRWuPtH7fy0aod+Exm/n5uOzo2iA51SCIiIiIiIlJD5Tk9TPp4DVkFLro3jeeewW00yWYV0aSaUlmaVFNEpNiidXv4aNUO3H4YcXpLOqQqGS4iIiIiIiKVFx1m46Zz0rCazfy8NUuTbIrUEkqIi0iN99v2HF757i88BvTr0IizW9dDF+1FRERERETkeDVJjOKqXi3wG/Dezxls3J0f6pBE5DgpIS4iNdrOnCKeWrABp8+gfZN6XNS1EWYlw6UG8Xm9LPnqU5Z89Sk+r+oSikj18ng8zH3kQuY9PIDo7y4MdTgiUgeoryO1wRmtkujeoh4ev8HTX2/C6fGFOiQROQ5KiItIjZXv8jLls3XkOr00SIhm9JktsVmUDZeaxe/388cvP/PHLz/j9/tDHY6I1HJ+n48vf4Uvfg/HnLUq1OGISB2gvo7UFlec0Zy4SDsZOUXMXJYe6nCkhhszZgwPPvhgqMOos5QQF5EayTAMXvhmEztynESGO7iuXxsibfqTJjWP2WKhW68+dOvVB7PFEupwRKSWs1gtXN55J1e0X4/FbIQ6HBGpA9TXkdoi0mHlmjNbYhjw+e+7WLE1K9QhSYg0a9aMp59+usy6Ll26VGuCu7CwkBtvvJH69etjNpsxmUzBhxLrx84a6gBERCrj89928fOWLHyYGX12G+pH2UIdkkilWCwWTj2zb6jDEJE6wmq10e/fPwGwP8SxiEjdoL6O1CanNIjjnHYpLPxjF9O++YtnLu9EdJi+i1Y1t9sNgM1mw1Q8QZjP58Pn82E2m7FarVXa1lIDLtY98sgjvP/++7z88st06tSJt956i//85z888MADjBgxItTh1TgaTikiNc5fe/N568d0PAYM7taUNvUjQx2SiIiIiIiI1AHDTm1CalwEmYVu/vvtZgxDd1xVtUmTJjFp0iQKCwuD65YsWcKkSZP43//+V6bt448/zqRJk8jJyQmu++mnn5g0aRIfffRRmbbPPPMMkyZNYu/evcF1q1atqp43cZC3336bqKio4OPtt9/m0UcfLbPuu+++O+T+06dPZ8KECVxyySWkpaXx8MMP06NHD7KysujUqdMJeQ+1SaVGiP/111+0aNGiqmMRETmiIrePZ7/eiMvrp23DBAackoxJZcOlBjMMA7fLBYDd4QiOVBARqQ6GYVBUWABAeESk/uaISLVTX0dqG7vVwrVnt2LSJ2tY9lcm32yIp1+bpFCHJSe5Cy+8kNNPPz34+q677qJhw4bceuutwXUNGzascN+srCz2799Pr169yqzv3bs3v/zyS/UEXMtVKiHeqlUr+vTpw3XXXccll1xCWFhYVcclIlKOYRi88t1fwbrhV/bSJJpS83k9Ht54dgoAY26/G5vdHuKIRKQ2c7tc3HHtlQA8M/N9HOrHi0g1U19HaqPGiZFc2LUx81ek88r3m2mfGk39GP2fWlXuvvtuIFDapETv3r0544wzMJvLFrsYP358ubbdu3enW7du5dredttt5dp26dKlSmL2+XyH3R4dHU10dHSZ1wkJCbRq1eqIxy6J9+Bz1JRyLyejSpVMWblyJZ06deKOO+4gJSWFG2+8keXLl1d1bCIiZXyzYS/fb9qH1zAx8szWJERoGgQREZHKCt/2VqhDEBERqbEGdGhAy/rR5Ll8Kp1Sxex2O3a7vcwdJRaLBbvdXqYmeFW1rYzdu3cHlz0eD9u2bavUcY5GVFQUTZo0YcmSJWXWL126lFNOOaXazlubmYzj+MR6vV4++ugjZsyYweeff05aWhrXXnsto0aNIimp+m8Xyc3NJTY2lpycHGJiYqr9fCISOhlZhfx77m/ke3yc27EJF3VtiFmDw6UWMAwDw+8HwFQ8W7iISHUxDIN6/2uMxbkDwhuw5/yMUIckIrWc+jpSm+3NdfLQ/NWYDD939G9N71b1Qh1SjeF0Otm8eTPNmzevcZUnmjVrRlFREW+//TZNmzblmWeeYdq0aVx99dU8/vjjJCcnl9unqKioTJ3ziiQkJGA/xF0006ZN49///jcvvfQSXbp04fXXX+fJJ59k7dq1tGzZskreV01RFb87xzWpptVqZcSIEXzwwQdMmTKFTZs2MX78eBo3bszo0aPZuXPn8RxeRAQAl9fHMws2Uuj10Tw5jiGdGygZLrWGyWTCbLFgtlj0BVFEqp3JZMJqBqvZ0BwcInJCqK8jtVlSTBjndmiIAbyxbAsFLm+oQ5ITZOjQodx666107NiR/fv3M3HiRObOncuCBQsqbP/ee++Rmpp62MfSpUsPeb5//OMf/Otf/+L//u//6NixI1988QWffPJJnUuGV5XjGiH+888/89prrzFr1iwiIyO55ppruO6668jIyOChhx4iNze3WkupaIS4SN0wc+kWPl2zE7vdxvghnakfZTvyTiIiIlKhsO1zMfkKMSwROBuOCHU4IiIiNVqW08+0z39lf14R/U9J5vqzWoQ6pBqhpo8Qv/3227n99ttDHUqdFLIR4k8++SQdO3akV69e7NixgzfeeIOtW7cyceJEmjdvzllnncWMGTNYuXJlpYISESmxflcen/+2C78Bl/RspWS41Do+n48fF33Jj4u+POJELCIix8vr9fD2ohzeWuwhP3loqMMRkTpAfR2p7awWMyN7BpLgC9buZsPuvBBHJCJHUqmE+AsvvMCVV17J1q1bmT9/PkOGDCk3c2v9+vV59dVXqyRIEambXF4f07/9EwOD3q3r06ZBXKhDEqlyfp+PX39axq8/LcOvL4kiUs18Xh9ffTyPrz6eh8+rvzkiUv3U15G6IC0lhj5p9QF45bu/8Pr8IY5IRA7HeuQm5X311Vc0adKkXBLcMAy2bdtGkyZNsNvtXHPNNVUSpIjUTR/8nMHOnCLiI+xc3qMpOZq0W2ohs8VCp+49g8siItXJYrUwYOjw4LKISHVTX0fqiqvOaMKKrVmk7y/kf7/t4sLODUIdklSTLVu2hDoEOU6VSoi3bNmSnTt3Ur9+/TLr9+/fT/PmzXUblIgct0178vj018DEvNef1YIIh5UcZ4iDEqkGFouF0/sNDHUYIlJHWK02Lrn8EsAAw4mBSpGJSPVSX0fqipgwG1ed0YQXv/2T2Ssy6NkikaRoR6jDEpEKVKpkyqHm4czPz69xhfBF5OTj9vp54Zu/MDA4q3USpzaND3VIIiIitUb9BaeQ+kks9RecEupQREREapW+aUm0TYnB7fXx+pLNh8yfyQH6GcmxqorfmWMaIX7HHXcAYDKZuP/++4mIiAhu8/l8/Pjjj3Tp0uW4gxKRum3uygy2ZxcSG27jml7NQh2OSLUyDAPDH6gxaDKbMZlMIY5IRGozwzDw+sHwm0DfP0XkBFBfR+oSk8nE9Wc1Z8KcNaxMz+KnLVn0aJ4Q6rBOSjZb4C61wsJCwsPDQxyN1CSFhYXAgd+hyjimhPgvv/wCBP5DW7NmDXa7PbjNbrfTuXNnxo8fX+lgqssHH3zAiy++yMaNGzEMg9TUVE4//XTuu+8+kpKSAOjcuTPnnXcejz32WIijrVhubi733HMPn376KV6vl379+jFlyhRSUlKOuO/nn3/OI488wqZNm2jUqBG33347V111VZk2brebiRMn8v7775Ofn0/37t157LHHaN26dXW9JZEK/bU3n49W7wDgujNbEOWoVGUnkRrD6/Ew4+lJAIy5/W5spf5vFRGpam6Xi0vf6AZ0Y8atUaEOR0TqAPV1pK5pFB/B0M6pzPtlO28s20LnxrE4NG9HORaLhbi4OPbs2QNARESELpjJYRmGQWFhIXv27CEuLg7LccxLcUyZpkWLFgHwt7/9jWeeeYaYmJhKn/hEefbZZ3nooYe4+eabmTBhAoZhsHbtWmbPns2uXbuCCfGT3bXXXsu6det44oknCAsLY+LEiVx22WUsXLgQq/XQ/4w//PADo0aNYtSoUTz66KMsXryYW2+9laioKC666KJguwkTJjB37lwmTpxIamoqTz75JMOGDWPZsmU14t9Zagevz8/0b//Ebxj0bFFPV9JFRESqUc6pr6HKpiIiIlVvWNeGLN64j335Lv63ZifDuzYKdUgnpZJBniVJcZGjERcXd1QDhA/HZNTgYj25ubnExsaSk5NzyKRt+/bt6devH88//3y5bX6/H7M5UEb9ZB4hvnz5cgYNGsTs2bM555xzANi4cSNnnHEGr7zyCsOHDz/kvhdffDEFBQV8/vnnwXV///vfWbNmDT/88AMA27dvp0uXLkydOpUxY8YAkJWVRadOnbjzzju59dZbq+/NiZQyd2UG7/+8jZgwG1Mv7Uxs+IHbX7LdsNMJsZr7S2oZwzBwu1wA2B0OjYoQkWplGAZFhQUAhEdE6m+OiFQ79XWktsvzQJwdkg+aUm/Jpn08t3AjDquFpy/vQnyk7o44FJ/Ph8fjCXUYUgPYbLbjGhle4qhHiI8YMYIZM2YQExPDiBEjDtt27ty5xx1YVcnJyTnkVYOSZHhF9u/fz6WXXorf72f27NkkJiayfv16Hn74Yb7//nt8Ph+9e/dm8uTJNG/eHIBx48bx559/8r///Q+AzMxM0tLS6NKlC19//TUQmHi0RYsWvPTSSwwbNuyo3sOCBQuIjY2lX79+wXWtW7emY8eOLFiw4JAJcZfLxffff8+DDz5YZv2IESOYM2cO6enpNGnShEWLFuH3+8vEEx8fT79+/fjqq6+UEJcTYk+uk3m/bAfgml7NyiTDRWozk8mEQxNSi8gJYjKZiIhUqRQROXHU15G6qlfLRD7/bRcb9+Tx7k/p/KNvq1CHdNKyWCxVkuQUOVqHzggfJDY2NnglNzY29rCPk0nnzp15/fXXeeONN9i9e/dR7bN7926GDh2K3W7nww8/JDExkS1btjBo0CCysrKYNm0aL730EpmZmQwbNgxX8dXuXr168csvv+B0OgFYunQpDoeDX3/9lby8PCAw2tvr9dKrVy8AEhISuOWWWw4bz8aNG2nVqlW5K+lpaWls2LDhkPtt3rwZj8dTrg54WloaQHDfjRs3kpSURFxcXLl2GzduPGxsIlXBMAxeX7oFj89Px4ax9GqZGOqQRERERERERCrNZDJxTa9mACzesJc/9+aHNiARCTrqEeKvv/56hcsnu6lTpzJ69Ghuv/12AJo2bcp5553HP/7xD5o0aVKufUZGBsOGDaNx48a89dZbREZGAvDYY48RHx/P3LlzCSu+ut2jRw+6devGW2+9xXXXXUevXr1wuVysWLGC3r17s3TpUi644AIWLVrEjz/+SP/+/Vm2bBmtWrWifv36QOAq2OFGqgNkZ2dXeKEhLi6O7Ozsw+4HlNu3JPFdsv1Qx4+NjSUrK+uwsYlUhRVbs/glPQur2czfejfXbZRSp/h8PlYt+w6ALj3P0sgIEalWXq+HRdOuxuQrYtiZCeT3mBHqkESkllNfR+qyVvWjOKt1Et9t3MsbS7fw4IXt9X1X5CRw1CPESysqKqKwsDD4euvWrTz99NN8+eWXVRZYVWnXrh1Lly7lvffe48YbbyQmJoaXXnqJM888kzVr1pRpu3nzZs4//3zatGnDrFmzgslwCEwoOmjQIKxWK16vF6/XS1xcHB07dmTlypVAINneoEEDli5dCsCyZcvo3bs3PXv2DK5bunQpPXv2DB537969PPfcc9X9YxA5aTk9PmYs3QLAkE6pNIgLD21AIieY3+dj5dJvWbn0W/w+X6jDEZFazuf1MWdpEbN/BMuuhaEOR0TqAPV1pK67ontj7FYL63fnseyvzFCHIyJUMiF+0UUX8cYbbwCB0cU9evTgiSee4KKLLuKFF1446uNMmjSJ7t27Ex0dTf369Rk2bBjr16+vTEiHZbfbGTBgAJMmTeLbb7/lgw8+oKioiKlTp5Zpt3LlSjIyMrjqqqtwOBxltmVmZjJ9+nTq169f5rFs2TJ27NgRbFcyMjw3N5fffvuNXr16BRPiLpeLlStXBsulHK24uDhyc3PLrc/Ozi5X5uTg/YBy+5aMDC/Zfqjj5+TkEB8ff0yxihyr+b9sZ1++i3pRDoZ3axjqcEROOLPZTLuup9Gu62lHvGNIROR4mS0WBrXdy/mtNmMxG6EOR0TqAPV1pK5LjHJwUecGALz9Yzoury4MiYTaUZdMKW3lypU89dRTAMyePZuUlBR++eUX5syZw/3338/NN998VMf59ttvueWWW+jevTter5d///vfDBw4kD/++KPM6Oyqdu6559KhQ4dyyfeLL74Yi8XC9ddfz7vvvkufPn2C2+Lj4xkwYADXXXddueNFRR2YmKhnz57cd999LFmyhMTERNLS0igsLOShhx7iu+++w+VylRkhfjRat27Nt99+i2EYZW6t2bhxI+3atTvkfs2bN8dms7Fx40bOPffcMvvBgVrirVu3Zs+ePeUS7Bs2bChXf1ykKm3PLuKTX3cCMKZXMxxW3T4pdY/FaqX3gAtCHYaI1BE2m43h//4EE36yKjc2RkTkmKivIwJDOqeycP0eMvNdfPrrTkZ0axTqkETqtEr1ggsLC4mOjgbgyy+/ZMSIEZjNZs444wy2bt161Mf5/PPPGTNmDO3bt6dz587MmDGD9PR0VqxYUZmwKrRnz55y64qKiti+fTvJycnltk2aNIkrrriCq6++mh9++CG4vk+fPqxbt45OnTrRtWvXMo/SSeNevXpRUFDAf//732Diu2PHjoSFhfHMM8/QsGHDCmuXH07//v3Jzs7m22+/Da7btGkTv/76K/379z/kfg6HgzPPPJOPPvqozPp58+aRlpYWjKNfv36YzeYy7bKzs/nmm28YMGDAMcUqcrQMw+D17zfj9fvp2iSeU5vqbgQREZETwXAk4nckYTg0ibWIiMiJ4LBauLJHIAczf9UOsgrcIY5IpG6rVEK8VatWzJ8/n23btvHFF18wcOBAIJB8jomJqXQwOTk5ACQkJFT6GAc788wzGTduHPPmzWPZsmXMmTOHESNGkJmZyY033ljhPk888QRDhw7l8ssvD9YHnzBhAn/++SeXXHIJ8+bNY8mSJcydO5fx48czZ86c4L5paWkkJSWxZMmSYGkUi8XC6aefXmZdiaSkJMaNG3fY99CjRw/OOeccxo0bx/z588tcSBg6dGiw3WOPPUZSUhLbtm0Lrrvzzjv56aefGD9+PN9//z2TJk1i9uzZTJgwIdimYcOGjBo1igceeIC3336bhQsXMmrUKKKjoxkzZszR/aBFjtGyPzP5bUcONouZv/VqpolFREREREREpNbq1TKRtORo3F4f7/6UHupwROq0SiXE77//fsaPH0+zZs04/fTTgyOhv/zyS7p27VqpQPx+P7fffju9e/emQ4cOFbZxuVzk5uaWeRzJXXfdxa5du7j33nsZPnw4999/P1FRUcyfP58LLqj4ti2TycRzzz1H//79ueSSS/j9999p0aIFCxYsID4+njvvvJNLLrmEhx9+mMLCwnJlS0p+HqWT37179y6zrYTP58Pv9x/xfbz22mv07duXf/7zn9xwww20aNGC9957D6v1QNUbv9+Pz+fDMA7UgzzjjDN44403+OGHH7jkkkuYM2cOzzzzDMOGDStz/EmTJnH11Vfz8MMPM2rUKKxWK/PmzTuuCxwih1Lk9vHmD4G7SYZ1aUj9mLAQRyQSOh63m1ce/w+vPP4fPG6NFBGR6uVyOvnHlcP5x5XDcTmdoQ5HROoA9XVEAkwmE6N7NgPguw37SM8sDG1AInWYySidPT0Gu3btYufOnXTu3Dk4Mcby5cuJiYmhbdu2x3y8m2++mc8++4zvv/+eRo0qrqX04IMP8tBDD5Vbn5OTo8StSA3y5g9b+fTXHaTEhPHYJZ2xW498bS7bDTudEGs7AQGKnEAet5sZT08CYMztd2Oz20MckYjUZi6nk9uuuQyA6VP+Bs2GhzgiEant1NeR2i7PA3F2SD7KcV5PL9jAD39l0qVxPBMGH3v+TESOX6UT4lVp7NixfPjhhyxevJjmzZsfsp3L5cLlcgVf5+bm0rhxYyXERWqQnTlF3PnBr3j9fu4a1JauTY6udrgS4lJbGYZBYX4eABFR0SofJCLVyu/3Y5/bEotrFzHxiew9PyPUIYlILae+jtR2x5oQ35Xj5P8+WIXPb3DvBe3o0DC2egMUkXKsR25SXkFBAZMnT+brr79mz5495Up+/PXXX0d1HMMwgvW9v/nmm8MmwyEwSaTD4ahMyCJyknjrh614/X66NI4/6mS4SG1mMpmIjNZFXRE5McxmM4mRHiwWJz7lpETkBFBfR6SslNgwzm2bzJd/7OKd5ek8MqyDLhSJnGCVSohff/31fPvtt4waNYrU1NRKf3BvueUW3nnnHT788EOio6PZtWsXALGxsYSHh1fqmCJy8vo1I5sVW7OwmE2M6tk01OGIiIjUSXlt78fkzcewRoU6FBERkTrp4lMbsXjjXv7am88Pf+2nZ8vEUIckUqdUqmRKXFwcn376aXCiyEqf/BCJ9Ndff50xY8Yccf/c3FxiY2NVMkWkBvD5De6a8ysZWYUM7pDKNb2aHdP+KpkitZXP5+O3n38AoMNpZ2CxWEIckYjUZl6vh4X/+xiAc84fitWq/1hFpHqpryO13bGWTCkxe0UGs1dsIzkmjCcu7YzVcuS5tUSkalRqhHh8fDwJCQnHffKToHy5iJwgX6/dTUZWIVEOKxefWvHEuSJ1kd/nY/m3CwBo17W7viSKSLXyeX3MfXsGAH0Gnq+EuIhUO/V1RCo2pFMqX/2xi925Thas3cOgDimhDkmkzqjU5af//Oc/3H///RQWFlZ1PCJSC+W7vLz/c2DSrktPa0yUo1LX4kRqJbPZTOsOnWndoTNms0aFiEj1MlssnNHnHM7ocw5mJaVE5ARQX0ekYmE2Cxd3CwwWm7sygyK3L8QRidQdlSqZ0rVrV/78808Mw6BZs2bYbGVHlqxcubLKAjwclUwRqRneWLaF/63ZScO4CB67pBMW87HPO6CSKSIiIiIiInKyqWzJFACvz8+ds39lZ04RI7o14rLTGld9gCJSTqWGaQ4bNqyKwxCR2mp7dhFf/B6YMPeaXk0rlQwXERGRqlP/ixZYnNvxhTVkz3l/hTocERGROstqMXN598Y8vWADn/66k4HtkomLsIc6LJFar1IJ8QceeKCq4xCRWurNZVvx+Q26NYmnU6O4UIcjIiJS55n87uBDREREQuv05gm0qh/Fpj35zF6RwfVntQh1SCK1XqULeGVnZ/PKK69w9913s3//fiBQKmX79u1VFpyI1Gyrt2WzalsWFrOJUT2bhjockZOSx+1m5jOTmfnMZDxuJadEpHq5nE6ueqcrV8y9gLzwDqEOR0TqAPV1RA7PZDJx1emB78uL1u9hV44zxBGJ1H6VSoj/+uuvpKWlMWXKFB5//HGys7MBmDt3LnfffXdVxiciNZTPb/DGsq0ADGqfQmpseIgjEjl5uV0u3C5XqMMQkTqi0G2i0G0m64z5oQ5FROoI9XVEDu+U1Bi6NI7H5zf44OdtoQ5HpNar1KSa/fv3p1u3bjz22GNER0ezevVqWrRowdKlS7nyyivZsmVLNYRanibVFDl5fb12Ny9/9xdRDitPX9GVKEelKjQFaVJNqa0MwyA3K3CnVUx8AiaT6uyLSPXx+/3s3R2Y2yMpOQWzudI3jIqIHBX1daS2O55JNUvbsq+ACXN/BWDyiE40qxdZBdGJSEUq1QP+6aefuPHGG8utb9iwIbt27TruoESkZnN6fLxffFX74m6NjjsZLlKbmUwmYhMSiU1I1BdEEal2ZrOZ5NQGJKc2UDJcRE4I9XVEjk6zepH0alkPgFk/aZS4SHWqVC/Y4XCQm5tbbv2GDRtISko67qBEpGb7aPUOcoo8JMeEMaBdcqjDERERERERETnpXXZaYyxmE6u2ZbF2Z/m8m4hUjUolxC+88EIefvhhPB4PELjim56ezl133cXFF19cpQGKSM2yv8DNJ7/uBODKHk2wWjT6TORw/D4fv69czu8rl+P3+UIdjojUcj6vlx9ev54fX7mSiFXjQx2OiNQB6uuIHL2U2DD6takPwKzl6VSiyrGIHIVKZaqeeOIJ8vPzSUpKoqioiD59+tCqVSuio6N55JFHqjpGEalB3v95G26vj7TkaHo0Twh1OCInPZ/Px9IFn7F0wWf49CVRRKqZ1+tlxud7eP2rfGzp74c6HBGpA9TXETk2F3drhN1qYf3uPFamZ4c6HJFaqVKFfWNjY/nqq69YsmQJq1evJj8/n27dutG/f/+qjk9EapD0zEK+Xb8XgFFnNFWNQJGjYDKbad6mXXBZRKQ6mc1mejbNwuQvwmzSqDMRqX7q64gcm/hIO4Pap/DR6u2891M63ZrE6bu1SBU75oS43+9nxowZzJ07ly1btmAymWjevDkpKSkYhqEPqUgdZRgGb/24FQODM1ok0jo5OtQhidQIVquV/hddGuowRKSOsNntXHvfK5gML7kmTXotItVPfR2RY3dhlwYsWLub9P2FLNmUyZmt64U6JJFa5ZguzxqGwYUXXsj111/P9u3b6dixI+3bt2fr1q2MGTOG4cOHV1ecInKSW52Rw68Z2VjNZq7s0STU4YiIiMgh+KLb4I1pjy+6TahDERERkQpEOaxc2LkBEChL6vX5QxyRSO1yTMNCZsyYweLFi/n666/p169fmW0LFy5k2LBhvPHGG4wePbpKgxSRk5vPb/DWD1sBOK99MvVjwkIckYiIiIiIiEjNNahDCp//tos9eU4WrtvDwPYpoQ5JpNY4phHi7777Lv/+97/LJcMBzjnnHCZMmMDbb79dZcGJSM3w7YY9ZGQVEuWwMrxbo1CHI1KjeD0e3v7vk7z93yfxejyhDkdEajm3y8VdN43hrpvG4Ha5Qh2OiNQB6uuIVE6YzcKIbg0BmLMyA6dHk9KKVJVjSoj/+uuvDBo06JDbBw8ezOrVq487KBGpOZweH+//lAHAiG6NiHKoHqnIsTAMg8L8PArz8zAMTXAnItXLMAxysvaTk7Ufa+YPoQ5HROoA9XVEKu+ctvWpHx1GTpGHz3/bFepwRGqNY8pc7d+/n+Tk5ENuT05OJisr67iDEpGa4+PVO8gucpMcE8bAdof++yAiFbNYrYy45sbgsohIdbLZbTw5dC1m9x7q/7KSfQ0yQh2SiNRy6uuIVJ7VYuay0xrz/KKNfLR6B/3bJWsQmkgVOKYR4j6fD+th/gOzWCx4vd7jDkpEaoasAjcf/7oTgJE9mmC1HNOfFBEBzGYzickpJCanYDbrMyQi1ctsttA8sYgW8bnov20RORHU1xE5Pr1bJdIkIYJCt5cPV20PdTgitcIxXVYyDIMxY8bgcDgq3O5SHUKROuX9n7fh9vpIS47m9OYJoQ5HREREjkJBi1sweXMxrDGhDkVERESOwGQycUWPJjz2+To+/20XgzukkhBpD3VYIjXaMSXEr7nmmiO2GT16dKWDEZGaIz2zkG/W7wXg6jOaYjKZQhyRSM3k9/nY9McaAFq164jZYglxRCJSm/m8Xr7c2QOA08/sg/7iiEh1U19H5Ph1bRxHm+Ro1u/OY+7KDK4/q0WoQxKp0Y4pIf76669XVxwiUsO89eNWDAzOaJFIWnJ0qMMRqbF8Ph/ffvYhAM3btNOXRBGpVl6vlzdeeAaAU8/orXq+IlLt1NcROX4mk4mRpzfhwY9+Z+G6PVzQKZXU2PBQhyVSY6kHLCLHbPW2bH7NyMZqNnNljyahDkekRjOZzTRu0Tq4LCJSncxmMx26nhZcFhGpburriFSNtikxdG0Szy/pWXzwcwa3nts61CGJ1FgmwzCMUAdRWbm5ucTGxpKTk0NMjGogipwIfr/BXXN+ZVtWIRd0TGVUz2Yn5LzZbtjphFjbCTmdiIiIiIiIyBHleSDODslh1X+urZkF3DXnVwAmjehE83qR1X9SkVqo1l+enTx5Mo0bNy63/t577yUxMZE333zzuM+xZs0aJk+eTGFh4XEfqyIJCQk899xz1XLsikydOpXhw4fTrFkzEhIS+OWXX45qvxkzZjBixAjatm1LkyZNGDBgAP/73//KtTMMg6effpqOHTvSoEEDBg4cyE8//VTVb0Oqybcb9rItq5Aoh5Xh3RqFOhwRERE5RvUWnUbyZ42ot+i0UIciIiIix6BpYiS9W9YD4L2ftoU4GpGaq9YnxCvy4IMP8sILL/DEE08watSo4z7emjVreOyxxygqKqqC6Mr74osvuPTSS6vl2BWZMWMGHo+HPn36HNN+TzzxBI0bN+bxxx9n5syZtG/fnquvvpp33323TLtnnnmGyZMnc/PNN/Puu++SkpLCxRdfzJYtW6rwXUh1cHp8vP9z4D/d4V0bEeVQ1SUREZGaxuLchcW5HYtzV6hDERERkWN0WffGWMwmVm3LYu3O3FCHI1Ij1bls1iOPPMKzzz7L448/zpgxY47rWD6fD7/fXzWBHUb37t2r/RylrVmzBrPZzPfff8/HH3981Pt98803JCYmBl/369eP9PR0nn/+eUaOHAmA0+nkqaee4pZbbuEf//gHAD179qR79+48//zzPP7441X7ZqRKffLrTrIK3dSPDmNg++RQhyNSK3g9Hua8Ph2Ai/92E1ab6gKJSPVxu1zcPLcTJqMdj1+WFepwRKQOUF9HpGolx4RxTttkvvpjF+8uT+ehC9tjMplCHZZIjVKnRohPnjyZJ554gilTpnDttdeWWV9RWZVmzZoxefLk4OuhQ4dyxRVX8O6779KjRw9SUlKYPn06Y8eOBaB169YkJCTQuXPn4D5//PEHF198MY0aNaJp06Zcc801ZGRklDnPW2+9Rc+ePWnQoAEtW7Zk8ODBrFy5Mrj94JIpP/zwAxdccAFNmzalSZMm9O7du9wo7ONR2QmWSifDS3Tq1Ilduw6MPlq+fDl5eXkMGzYsuM5utzNkyBC++uqrSp1XTozMfBcfrd4BwMgeTbBZ6tSfD5FqYxgGudn7yc3eTw2e1kNEagjDMNiVa2VnXhj7zv4u1OGISB2gvo5I1RvRtSF2q4UNu/P4aYsucIscqzozQvyJJ57gscce45FHHuHvf/97pY+zatUq0tPTmTBhAnFxcXTo0IHc3FyeeOIJPvjgA2JiYnA4HABkZGRwwQUX0Lx5c6ZPn47L5WLixIkMGTKE7777jujoaJYuXcqtt97K2LFjGTBgAEVFRaxYsYKcnJwKz5+bm8sVV1zBGWecwSuvvILdbmf9+vVl2r/zzjuMHTuWjz76iDPPPLPS77Uq/PDDD6SlpQVfb9y4EQhcPCgtLS2NjIwMioqKCA8PP6ExytGZ9dM23F4fbZKjOaNFQqjDEak1LFYrQ6/8W3BZRKQ62ew2xj80ObgsIlLd1NcRqXrxkXaGdEpl7soM3l2eTrcmcVg1aE3kqNWJ/40KCgp45JFHGDVqFDfffPNxHSsrK4sFCxbQqNGByQSbN28OQJcuXcqMkn7hhRfwer3MmTOH+Ph4ADp27EjPnj159913ueGGG1ixYgXx8fE8/PDDwf0GDhx4yPP/+eef5Obmcv/999OuXTuAcrW+zWYzFosl5LfMzJ49m+XLl5eZuDQ7OxuHw0FYWNnpl+Pi4jAMg+zsbCXET0Kb9uTz3ca9AIzq2Szkv1sitYnZbCalUZNQhyEidYTZbKFV23ahDkNE6hD1dUSqx9BODVjwx2525hSxcN0eBrZPCXVIIjVGnbh8FB4eTq9evZgzZw4//PDDcR2rffv2ZZLhh/PDDz9w1llnBZPhEBgJ3aFDh2AcnTt3Jisri1tuuYVFixZRWFh42GM2b96c6Oho/u///o958+axb9++cm2uuOIK9u7dS+/evQ95HMMw8Hq9wUdV10L//fffueOOO7jyyiu54IILqvTYcmIZhsGby7YAcFbrJFrVjwptQCIiIiIiIiJ1XLjdwiWnBvJTs1dkUOT2hTgikZqjTiTEzWYz77zzDi1btmTkyJH88ccflT5WUlLSUbfNzs6usH1SUhJZWYEaT2effTbTp09n3bp1XHLJJbRu3Zqbb745uP1gcXFxzJ07l6ioKG6++Wbatm3L0KFDj/k9vfvuu9SvXz/4GDdu3DHtfzjbtm3jsssu49RTT+Wpp54qF7/L5cLpdJZZn52djclkIi4ursrikKrxw1/7Wb87D7vVwsgeGtkhUtX8fj9/rfudv9b9fkImahaRus3n8/H7nPH88cE/CP/j0VCHIyJ1gPo6ItXnnLb1SY0NJ9fp4ePiOb9E5MjqREIcICYmhtmzZ5OYmMgll1zC1q1bg9vCwsLweDxl2ns8HgoKCsod51hKRcTHx1c4gnvv3r1lRo1fdtllfP3112zYsIHJkyfz6aefcv/99x/yuKeeeioffPABmzdv5t1332Xv3r1cffXVRx0XwKBBg/j666+Dj7vuuuuY9j+UzMxMLrnkEurVq8fMmTOxHTSDeEnt8E2bNpVZv3HjRho1aqRyKScZl9fH2z8GPisXdW5AQqQ9xBGJ1D4+r5evP5rN1x/Nxuf1hjocEanlvB4Pz72/gWdnZ2DfND3U4YhIHaC+jkj1sVrMwYFrn6zZyf4Cd4gjEqkZ6kxCHAIjs+fOnYvJZGLEiBHs3r0bgAYNGuB2u9m8eXOw7eLFi/H5ju52E7s9kCQ8eNTz6aefzuLFi8nOzg6u27hxI7///jtnnHFGueMkJiYyatQo+vbty4YNG4543vDwcAYMGMC1117L1q1by53/cBISEujatWvw0aTJ8Y/8zc/P57LLLsPtdvPee+8RExNTrk2PHj2Ijo7mww8/DK7zeDx88sknDBgw4LhjkKr12Zpd7Mt3kRjpYEjn1FCHI1IrmUwmUhs3JbVxU9XnF5FqZzKbaJ+cR4ekfZhMRqjDEZE6QH0dkerVvVk8bZKjcXt9fPDztlCHI1Ij1IlJNUtr0qQJH3zwAUOGDOHSSy/lk08+oX///kRGRnL77bdz6623smPHDl588cVyEz8eSlpaGgCvvvoq559/PhEREbRr146bb76Zd955h4svvpg77rgDp9PJo48+SqNGjRg5ciQAkyZNIisri969e5OUlMQff/zBwoULDzn555dffslbb73FBRdcQKNGjdi9ezcvv/wyp59+ejDeWbNmMW7cOObPn3/YOuKHsmTJEvbt28e6desA+O6770hPT6dJkyZ07dr1kOcYPXo0a9as4bnnnmPbtm1s23bgD3H37t2BwGj8f/7zn0yZMoXExETatWvHq6++yv79+xk7duwxxyrVJ6vAzfxV2wEY2aMJDqslxBGJ1E5Wm40hI8eEOgwRqSPsdgd3PPAoJr+LArMj1OGISB2gvo5I9TKZTFx1RlPu//A3vlm/l/M7ptI4ISLUYYmc1OpcQhygXbt2zJo1ixEjRnDFFVcwZ84cZsyYwX333ceoUaPo0KEDL7zwAkOHDj2q43Xq1Im77rqLN998k2effZaGDRuyevVqGjVqxCeffMJ9993HjTfeiNlspm/fvjzyyCNER0cD0K1bN6ZPn878+fPJy8ujQYMGjB07lvHjx1d4rubNm2MymZg4cSL79u0jISGBvn37limx4vf78fl8GEblRv1MnjyZJUuWBF8/+OCDAIwcOZJp06Yd8hzffPMNQIXJ/P379weXb7vtNgzD4PnnnyczM5MOHTowe/ZsmjVrVql4pXq89/M2nB4frepH0btVYqjDERERkSriSewZ6hBERESkCqUlR9OjeSLLN2fy9o/pTBjcNtQhiZzUTEZls6YngdzcXGJjY8nJyamwPIeIVM7mfQX8e+4aDAwevqgDacnRoQ6JbDfsdEKs7chtRURERERERE6EPA/E2SH56IoMVJudOUWM/2A1Pr/BvRe0o0PD2NAGJHISq1M1xEXkyAzDYObSLRgY9G5Z76RIhovUZl6PhzkzpjNnxnS8B03wLCJS1dxuFxP/dRsT/3Ubbrcr1OGISB2gvo7IiZEaG07/U5IBePvHrZWuGiBSFyghLiJlLNmUybpduditFkaefvyTrYrI4RmGwf49u9m/Z7c6rSJS7Qy/QcbWzWRs3Yw598iTuIuIHC/1dUROnBHdGhFus7B5XwHfbNgb6nBETlp1soa4iFSsyO3jrR+2AjCia0PqRWmyLZHqZrFaGXzZ1cFlEZHqZLPbeGDgRszufaT8OJh9F2SEOiQRqeXU1xE5cWLDbVx8aiPe+mEr7/6YzunNE4iw63MncjCNEBeRoNkrtpFd5CY1NpwLOqWGOhyROsFsNtOoWUsaNWuJ2az/lkWkepnNFro0yKNryj4s+pMjIieA+joiJ9ag9ik0iA0n1+lhzgpd+BapiP43EhEAtu0v5LPfdgFwTa9m2PQtWUREpFYqajSSgqbXUdRoZKhDERERkSpmtZgZ3asZAJ//vovt2UWhDUjkJBTSjNfixYsZOnQoDRo0wGQyMX/+/FCGI1JnGYbBa0s24zcMejRLoEvjuFCHJFJn+P1+0v/cQPqfG/D7/aEOR0RqOZ/PxxLPZXzPjWS1mxzqcESkDlBfR+TE69I4jm5N4vH5Dd5YtkX1+0UOEtKEeEFBAZ07d2batGmhDEOkzlv6ZyZrdwYm0hzVs1mowxGpU3xeL1/MeZcv5ryLz+sNdTgiUst5PR6mTfkP06b8B6/HE+pwRKQOUF9HJDRG9WyK1Wxm9bZsVqZnhzockZNKSCvrDx48mMGDB4cyBJE6r/REmsO7NiApWhNpipxIJpOJeikNgssiItXJZDbRtEWr4LKISHVTX0ckNFJjwxncMYWPV+/gzWVb6NQoVqVRRYppqlmROm72ygyyCt2kxIQxpFODUIcjUudYbTaGj/57qMMQkTrCbndw96QnQx2GiNQh6uuIhM6Iro34bsM+duU6+ey3XVzYWd/5RaCGTarpcrnIzc0t8xCRytu2v5DPf9sJwJjezXW1WEREpA5IWDKIegu7kbBkUKhDERERkWoUbrdw5elNAJi7MoOsAneIIxI5OdSo7NekSZOIjY0NPho3bhzqkERqrJKJNH1+g+6aSFNERKTOsOX+hj3nF2y5v4U6FBEREalmZ7WuR6v6UTg9Pt5Znh7qcEROCjUqIX733XeTk5MTfGzbti3UIYnUWAvX7Sk1kWbTUIcjUmd5PR4+evs1Pnr7NU1wJyLVzu12MeF/bfjXgjNx+jVviIhUP/V1RELLZDIxpldzAL7buJc/dqjagkiNSog7HA5iYmLKPETk2GXmu4ITaV7RvTH1o8NCHJFI3WUYBru3b2P39m0YhhHqcESkljP8Buv3RLJuXzx7zv091OGISB2gvo5I6LWqH0X/U5IBePm7v3B7/SGOSCS0QjqpZn5+Pps2bQq+3rx5M6tWrSIhIYEmTZqEMDKR2sswDF79fjNFHh+t60czqH1KqEMSqdMsVisDhl8eXBYRqU5Wm42bxv87uCwiUt3U1xE5OVx5ehNWbM1iZ04R837J4PLuyrtJ3WUyQniJ9ptvvqFfv37l1l9zzTXMmDHjiPvn5uYSGxtLTk6ORouLHKUlm/bx3MKNWM1mJl/ckUbxEaEO6ahku2GnE2L13V1EREREREROEnkeiLNDcg248Xr55v08+dV6LGYTk4Z3oklizcgHiFS1kJZM6du3L4ZhlHscTTJcRI5dTpGHGUu3ADC8a8MakwwXERERERERkePTo3kC3Zsl4PMbvPTdn/j9KmMkdVONqiEuIsfnjaVbyHN6aJIQwUVdGoQ6HBEB/H4/O9K3sCN9C36/avmJSPXy+31sXfAo6V/cR9ifL4Y6HBGpA9TXETm5/K13c8JtFjbtyefLP3aFOhyRkFBCXKSOWLE1iyV/7sOEiRv7tMRq0cdf5GTg83r5dNZMPp01E5/XG+pwRKSW87g9THr5Bx59bTWO3x8JdTgiUgeoryNyckmItDOyR6B++KyftrEv3xXiiEROPGXEROqAQreXV7//C4AhnVNpmRQV4ohEpITJZCIuMYm4xCRMJlOowxGRWs5kMtE4rojGMXmYTLpNWkSqn/o6IiefAe2SaZMcjdPj47XvNxPC6QVFQiKkk2oeL02qKXJ0Xl78F1+v201KTBhTLumEw2oJdUjHTJNqioiIVA3Hzk8w+YswzOG4UoeEOhwREZEarSZNqllaRlYhE+aswev3c9u5afRsmRjqkEROGI0QF6nlft6yn6/X7Qbg72e3qJHJcBEREak6rtQhOBteqmS4iIhIHdYo/sDcYq8v2UxOkSfEEYmcOEqIi9Ri+wvcTP/2TwAu6NSA9g1iQxyRiIiIiIiIiJwMhnVtSMO4CHKdHl5a/KdKp0idoYS4SC3l9xtMW7SJfJeX5vUiGdm9cahDEpEKeD0e/vfem/zvvTfxejQqQ0Sql9vt4umJ9/H0xPtwuzWJlohUP/V1RE5eNouZW89thdVsZsXWLL76Y3eoQxI5IayhDkBEqsdHq3fw+44cHFYLt57bGqtF179ETkaGYbB961/BZRGR6mT4DdatWR144cwEe4PQBiQitZ76OiInt6aJkVx5ehPeWLaFN3/YSrsGMTSKjwh1WCLVSglxkVpo4+483v95GwBjejcjNTY8xBGJyKFYrFb6XTA8uCwiUp2sNhu3n7UZsyeLBt/1YN8FGaEOSURqOfV1RE5+gzuksHpbNqszsnnu601MHN4BmwbVSS2m326RWqbQ7eW5hZvwGwY9W9Sjb1pSqEMSkcMwm820at+J/2/vzsPsKMuE/3+f2s7We5LOBoFAWCTKKiAuIIIEBASXUdTf6z64gI7LgDoojrO8IiqDC+qo4zKKGyOoA4LyIgYUBEGWACEaEkggSac7vZy1qp7t90c1PWZYBExokr4/19VXd9d5Tp27truec586Ty1Zuj9BIKdlIcT2FYYhR+05xot3fwh5nyuEeDpIX0eIZz6lFO968Z70lGMeGG3xg5vXTXdIQmxXcjYSYifivec/rl/L5kbKnO4yb3/RYpRS0x2WEEIIIZ5BssGXks47mWzwpdMdihBCCCGeIfqqCe988Z4AXLFiI7evH5/egITYjqQgLsRO5Po/jfDb+0YIlOI9L1lCrSRfSRTimc45x/DGhxje+BDOuekORwixk3POcnvvP/CHOZ9l9KCvT3c4QogZQPo6Quw4Dl7Uz7Kl8wD48q9XM9GRG+GKnZMUxIXYSdw/0uLrv1kLwKsP2YW953ZPc0RCiCfCGsNPvvN1fvKdr2ONme5whBA7OZ1rzjvng5x3zgfRubzJFUJsf9LXEWLH8vrDF7Frf5WJjuYry++Tm+GKnZIUxIXYCUy0NZ/+5SpyY9l/lz5OPXDhdIckhHiClFJ09fTS1dMrQxwJIbY7pRQDcwYZmDMoOUcI8bSQvo4QO5ZSFPKel+xFHAbctm6MS//w0HSHJMQ2p/wO/FFPvV6nt7eXiYkJenp6HrftlVdeyde+9jVuv/12Wq0W8+fP5+ijj+aMM85gyZIlT1PEW1uxYgVXXHEF733ve6lWq0/6+ffffz8f/vCHWbFiBaOjo/T393PYYYfx0Y9+dKtlWrNmDV/84he55ZZbWLlyJXvttRc33HDDo85zYmKCT37yk/zsZz9jbGyM+fPn89a3vpUzzzxzqs3AwMAjnjc4OMi999479f/VV1/N5z73OVatWkWj0WD+/Pm87GUv40Mf+tBf3FbiydHW8c+X38MfhxrM763wz6c+m66dcKiU8Rw2ptAbT3ckQgghhBBCCCFEoaGhL4G55emOZNu6dtVm/n35fQC879i9ed4es6Y5IiG2nZ2vavYoPvGJT/C5z32Ol7/85Vx44YXMnj2btWvXcvHFF/O2t72N5cuXT0tcK1as4Pzzz+dv//Zvn1JBvNVqMTg4yLnnnsvChQsZGhri3/7t3zjllFO47rrrmDWrSFb33nsvv/zlLznkkENwzj3muG2tVouTTz6ZKIr4l3/5FwYHB1m9ejWNRuMRbU8//XRe9apXTf2fJMlWj4+NjXHIIYdw+umnMzAwwMqVK/nUpz7FypUrufTSS5/0sopH573nq9et4Y9DDWpJxNnH77NTFsOFEEIIse30/f4NBPkILpnN+KEXT3c4QgghhHgGOnqfQR4cbXPFio186df3MbenzOLZtekOS4htYqevnD18pfLf//3f8w//8A9T05///Ofzhje8gV/84hfTGN1fZ+nSpXz+85/fatqBBx7IoYceyrXXXsurX/1qAI4//nhe9rKXAXDGGWdw2223Per8LrzwQprNJtdffz21WpHkXvjCFz5q24ULF3LooYc+Zmyvec1rtvr/hS98IUmS8P73v5+NGzcyf/78J7aQ4nFdfudGrv/TMIFS/N2xezG/tzLdIQkhhBDiGa40spwwfQhbliHWhBBCCPHYXn/4bjw43uGO9eN85her+NdXPJu+avKXnyjEM9xOP4b4RRddxODgIGedddajPr5s2bKpv9M05ZxzzmG//fZj/vz5HHnkkVx++eVbtT/55JM57bTTtpq2YsUKBgYG+M1vfjM1bWBggM9//vOcd9557LPPPixZsoQzzjiDVqsFwPe+972pYUj22msvBgYGOOCAA/7q5X14OJM8z6emBcET28zf/e53ecMb3jBVDN/WHo5Na7mB07bwh3VjfO+mdQC88Yjd2H+XvukNSAjxlBhj+OWlP+CXl/4AIzeaEkJsZzrP+eQ1e/Cv1x9KbmQsXyHE9id9HSF2XGGg+Ltj9mJBX4UtrYzP/vKP5ObRRx0QYkeyUxfEjTHcdNNNHHnkkcTxXx54+PTTT+fb3/4273nPe/jOd77DPvvsw5ve9CauvPLKp/T6X/va11izZg0XXXQRZ511Fj/+8Y/5zGc+A8Bxxx3HBz/4QQAuueQSfvGLX/Cd73xn6rkDAwOcccYZT+h1nHNorVm3bh1nn302Cxcu5KSTTnpSsa5bt46hoSFmzZrF61//eubNm8cee+zB3/3d39FsNh/R/sILL2RwcJDdd9+dt771rTz44IOPOl9rLWmacscdd/DpT3+aE044gUWLFj2p2MQjrR9t84Vr/oTHc8y+c1m2dN50hySEeIq8czywehUPrF6Ff4whrYQQYltxznHz+j5uemgeQy/+w3SHI4SYAaSvI8SOrZpEnLWsGJ71T5sbfO36NezAtyMUAtjJh0wZHR0lyzJ22WWXv9j27rvv5vLLL+eCCy7gzW9+MwDHHnss69at4/zzz+eEE0540q8/d+5cvvrVr07N68477+RnP/sZH//4x5k9ezaLFy8GimFOHh7v+2FhGD7hK7vf9a53cckllwCwePFiLrvssid948qhoSEAzj33XE466SR++MMfsmbNGj7xiU/QarX4+te/PtX2tNNO47jjjmNwcJCVK1fymc98hhNOOIHrr7+evr6+rea7//77s3HjRgCOOeaYqfUhnrotzYzzf7GKjrY8a34Pb3nB7nK3diF2YGEY8qJlJ039LYQQ21MURbzh9OKii7DSP83RCCFmAunrCLHjm99b4X3H7s3//flKrv/TMLsOVHn5AQumOywhnrKduiD+sCdSLLzxxhsBOOWUU7aa/opXvIJzzjmHVqv1pIcSOfroo7f6f5999nnCN5QcHh5+wq/zD//wD7zjHe/gwQcf5Ctf+QqveMUruPLKK5/QBwEPe/hGm3vuuSdf+tKXADjqqKMIw5D3ve99fPSjH2X33XcHmHocirHYn/e85/HiF7+Y//zP/+S9733vVvP90Y9+RKvV4t577+Wzn/0sr3/967n00kulI/QUjbVy/uWKlQw3Uub2lHn/S/cmCnfqL3oIsdMLwpB9DzhkusMQQswQYRTxomOW/eWGQgixjUhfR4idw7MX9vLm5+/ON367lu/ftI6+SsyRe8+Z7rCEeEp26krawMAA5XL5MYfz+HPj4+PEcUx//9ZXysyZMwfvPRMTE0/69f/3VdpxHJNl2ZOez1+y2267cfDBB/Pyl7+c//qv/8I594ibbf4lD1/Z/b9vonnUUUcBcO+99z7mc5cuXcqSJUu4/fbbH/Wxww47jDe+8Y1cfPHFXH/99Y8Yl108MRNtzT9fcQ8bJzrM6S7zsZP2o6f8l4cCEkIIIYQQQgghhPhrHbd0HscvnYfH8+Vf38dvV49Md0hCPCU79RXiURRx+OGHc91112GMIYoee3H7+/vRWjM+Pr7VsB/Dw8Mopejt7QWgVCptdcNKKIrpzxTVapW9996bNWvWPKnnLV68mFKp9JiPp2n614bG0qVLieOYtWvX/tXzmmnqqeZfrriHDeMdZtVKfOzEZzG767G3lxBix+G9Z2xL8a2g/llzZAgkIcR25Zxj7LZvgk2ZP38e+a6vmu6QhBA7OenrCLFzedPzd0dbzzX3DvHFX61GKXj+nrOnOywhnpSd+gpxgHe/+90MDQ3x2c9+9lEfv/rqqwF43vOeB8BPfvKTrR7/6U9/yv777z81XMqCBQtYvXr1VjcQuPbaa59SbEmSANum2Pywer3O3XffPTW8yZOJ5eijj+a6667bavrDy3bAAQc85nNXrFjB6tWrOeiggx73NW655Ra01uy2225PKraZrpkZ/u8VK1k/1qa/mvCxk/ZjsKc83WEJIbYRozU//saX+fE3vozRerrDEULs5HSec875P+Wcz/6Cym3vn+5whBAzgPR1hNi5KKV4+4sW8+J9BvF4vvir1dy0Zst0hyXEk7JTXyEO8NKXvpT3vve9fOpTn2LVqlW88pWvZNasWTzwwANcfPHF1Ot1XvrSl7J06VJOOukkPvaxj5GmKUuWLOGSSy7h5ptv5uKLL56a38tf/nK++93v8qEPfYiXvexl3HzzzfzsZz97SrHtvffeAPzHf/wHL3vZy6hWq+y3335AMVTLaaedxhe+8IXHfP55551HvV7n8MMPZ/bs2axbt46vfvWrZFnGO9/5zql27XZ7qvC/fv16Go0GP/3pTwF4wQtewOzZxSd5Z599Nscffzynn346p512GmvWrOGf/umf+Ju/+ZupG4B+4Qtf4P777+cFL3gBc+bMYeXKlVxwwQUsXLiQN77xjVOv+cY3vpEDDzyQpUuXUi6Xueuuu/jiF7/I0qVLOfHEE5/S+pqJWpnhX69Yyf1bWvRWYj560n7M65ViuBA7m3KlOt0hCCFmkJ6SBtx0hyGEmEGkryPEzkUpxTuO3APrPNf/aZjP/+pPvD9QPHf3gekOTYgnRPk/v9R5B1Ov1+nt7WViYuIR43X/bz//+c/52te+xu2330673Wb+/Pm85CUv4cwzz2SPPfYAoNPp8M///M9cdtlljI2Nsddee3H22Wdz8sknbzWvz3/+83zta19jfHycY489lje/+c284hWv4Gc/+9nUGNwDAwN84hOf4D3vec/U87785S9zzjnnMDo6OjXtU5/6FN/5znfYtGkTCxcu5I477ph6/ute9zouuuiix1ymK6+8ki996UusXLmSVqvF/PnzOeKIIzjrrLO2ukJ83bp1HHjggY86jz+PGWD58uX80z/9E/fccw99fX28+tWv5qMf/ejUcCpXXXUVF1xwAatXr6bZbDJ79myOOeYYzjnnHObNmzc1nwsvvJDLLruMtWvX4r1n11135aSTTuLMM8/8i9tKFCY6mvOvupf7hpt0l2POPWk/dh2YuR3J8Rw2ptArw6YLIYQQf5XKA99C2RY+rNHZ7c3THY4QQgixQ2to6Etg7gy8ds05z0XXrua3940QBQHvf+neHLJb/19+ohDTbMYUxIXYkTw41ub8q1axuZHSVYr42En7sdus2nSHNa2kIC6EEEIIIYQQ4plmJhfEAawrhk25cU1RFH/HUXvwor3mTHdYQjyunX7IFCF2NHc+OM6/Xf1HOtoy2F3mQyfsy8K+ynSHJYQQQgghhBBCCLGVMFCccfSeeDy/W7OFi65dzYNjHU47dFe5ia54xpKCuBDPIL+8exPfuuF+nPfsO6+HDxy3Nz1luSRaiJ2ZMYbrrizuRXHkCS8niuTULITYfnSe859fKe5R88Z3vod48ibvQgixvUhfR4idXxQG/N0xezG3p8xPb3+In97+EA+NdTjzJUsox+F0hyfEIwTTHYAQoviK0bd+u5Zv/HYtzntetNcczjnxWVIMF2IG8M5x38oV3LdyBd7JTe6EENuXc47f/3Y5v//tcpzkHCHE00D6OkLMDEopXnfYIt794iVEQcAtD4xy7k/vYriRTXdoQjyCfDQrxDSb6Gi+9OvV3LF+HIDTDl3EKQcukK8WCTFDhGHI816ybOpvIYTYnqIo4q2HrkeZCRZcs4QtJz443SEJIXZy0tcRYmY5cu85zOst85lfrGLdaJtzLlvB3y/bh73ndk93aEJMkZtqCjGNblqzha//Zi2NVBOHAWccvYTn7TFrusN60h4ca/Dt6+5nxaYxtDYoBeUoopVpSklEqEA72KWvgiKgrxox1MiopzkoqEYx3aWIidxQCkLiwLOpkbGgNyFUMY00Y7ilSSoxu/V0M9ZJWbelSU85RCuFcpZGS2O9Z/HcXhyesWZOiEc7R24sgYrYdXaVcilGa0sQBnjtyJ1jtJmhs5yWcQx0Vdl3YS9rhprUOx26KiWMMbS0BQMqsFTjMmGsSJKIdifHe6h3UspBRE9XQiu31KoRvUmZ8Syj0UyJgpBSKcB5iHxApRKRasuG4Qb9vQmRj1Alj9WKdpox2N/NwbvMZrSRsqHRphIHpLnlgeEmUQRRBN2lCi6wjNZzyonCaEDl1Dshh+0xi462rBmawCuIIsUes3oYaWeMTnRQeJJKTF+lTDvLGG0aBntiKkmJzBjqLU1fLSY1njzXLJhdw+iAB7aMETlQYcSsnjJxEmAtxIFi41iTTBu6qyV26eumkWuarZRGZtDOkgQR82dV2DTaoVxSjDZzcJ7+WplaNWHTaJueWsRArcpwo0UlidHGMdFqE8cJtVJEEocY5xmrtxnoruLwVKOYWjVE59DIc/LM0kg7QMjc3iq9tQSDZ7yV0mgaFs2u0dGeViclUiFtb6lGirFmivUh5dhTTmKWzO1nVq3KQ+MNHhpv0Wx16KqVqZUTvHeMN3NqlYjuJGG0kRJFijAMmNdTJcstdz04QhRG9FcjrFJ0lUvsMbtGbhV/2jhKlml6+spY41m3uQHOUY5DumoJIQFREmKtp5GmYBXVUsTs/hqzKiUeGG3SV02wzjMy0WaPOX1sSVM2j7epxRE+UMztqVBLQjbXMzrW0BMnBLFnSz0nihVxGBIqTyuztLOcchihQlXsG2GMcQanFKFX7DKrm4cmWjTbGbv012hbizdACMY4wiDAWk8UesbrGXWt6a0kzO2t8dBokyQulj+JQh7a0qKSKOb0Vlk/3MA4UMrTW4kYaVqqCXgHs3traGvJtcMHjlZqqQQBNvBEKqS7lNDxmtCG1MohTnnGmhm5tlRKMaVSRHcYkpQDrAmIlOVPmxtgLeVKib5yzFArxWpLHAT0VsuUKiHKBdQ7KXlm6O4tk7VyRjuaaqQYaWT01cr09sUkYYlmM2W82aGnu4TWvlh3UUitViLXhnpDE0fgFJTCgCAIGW+1qcYhw/WcRtuwcHaNubMrbBlPUd5CEINzqDigFEAn93SXY3zgaXcsvd0JeWrZ3OiQeEtULtGdxPgA6q0c4xy9XSWWzOoBr7hn4yilckRfucy6kVG0D6hECu88QRRTKwXU2xmNtiUJDc0MauWQOd1VtrQMg30xjY4F5xkcqNHJNK3cUYtg43jKvN4qhJ5ERcSJYvN4h/5ambFGm9RYAhWg8xwbRMzqSZhf62Jjs4VCsaC/wkOjHXaf00UjtWxptkCFxMrRyi3VMCZMFI2WZk5/hVaq2bBlgt5KBe8VXd0xNodMZ7S0RaGIgCgKqFViOpkhyy0LZvUwr69KmhtWbxqjVi4xv6fCaDsjdw6lPD5X9PfGbB7PiBNFEkTkTtNqG0pJQBiGtFNNkoTsO7eX+4ZboBzGeiqlEGM8xnnGGxneGRbM7iHNLdUkRIUBEZ6h8ZTZ/WV27emiYSz1ZsbwRIMkKRHHikopYbfeGr9/YDPaGPqqFeb2lsktdJci7h+t44wiCDylOCKJA+IgQnnPcLNDO9P0dsUYAyFgPCQlBTZkdk/CA5vrNDspYZDw3cHXMqCG2Wxm8cnkl9w/3EBjGahWCOKAVlOzaLDKcN3QytvgAypxTCmJMM5QUiFb2hlzu8sMT6S4wGG0o7s75sGhJt3lMlEMxjniKKK3XKJajthS74CCOApRoaLe6KC8olyO6a4m4KHRyUlzQ7UcoggJlSKOFaELihxvLFEUsPtAL0ONDrnRlKKI8XZG6C2b25ZZtRDlQ6qVCG0smfb01hKqScTG8QYTLU1PpUSlEhW5LHC0OoauJKbRSXHeEcUJWSeDJKDdzpnT102SBMX+2FclDkImmi3auSP0DpeEOO2pJIq5fd2kmaHZ0QxPNKhnjv6umAX93Qz0lNHa8tB4iyiA7qRMx1p6qwkTrQ7jjRyvFP09EQ8NZ3SVA/q6inO11orZAyUaDYOzjlo1opM7slzTyTTd3RW64ojUORIUC2Z1Ya2inCgmWppGJ2egu0zgHEPNFOfBWoOn2FZp6ohLoFxAb3eJIAjJTU6WebrKMaU4opXnZKkmTELGGykDtYQkKrHbnBq59gy1O+hU08oNBJ7AKNrG0l0NCVWEwVNSIbnLGR7L6euJGOztoZXlDHaViYOQ9RNNvLUEQcSsWhmnPKPNNptGU7A5vT3ddJcD1m1psvucHga7ajxYb6IcaKtpdRzlBHKniFVAFHo6ORBYlAtYNKeLOIwYbqS08py95vcxPNahYTKSICae7Lf2d5foKUesHm7QXYoZa7SJwxDvAyqVgDy37NLXw5zuEvdumqC3EqFURBA4Noy36EoSrAJnLH1dCdaDyRzDzRa1coXBrphVmyYgVChtiUshfdUy9XZOqRST5znaKcpxQE9XQrtl6a4mhFHA2o1jRGHAnJ4qYaDoGEMQBEw02mgDSRIwv6+LVGuGxlNQnnIcEIWKekszq7dMqZTQaWusgu5KxMRERjvXxKWQtGOY3VulXApptg0drUniAGMsY42cWhLQ110jUJ5GmmOdx2lHrSshCBSVOKGnHNPUmp6khHGOoXqLVpYThQGhCthrfj+95YQ7HxzF54YgiXBYeiolJtoZeFBAGMK6oRb91ZCe7hrzemrU4pDVw+Ok2jDezilHiq5KmSQOmVut8lCzSbtdLGcljmhmhvVDE1hgfn+JNFd0VxJUQHG8lUt0jGG42SJwAZUkQinHRKoJVUA5DNg41mbRYBf9tSojzRbt3JIoRTu1NLM2XiUs6IsJoph2x7NwVolSHLN+pEkcB1jnMdphnQECumoxlSim3k5ptA1JrNDWEgUhXbWYQEW0OynWKxyWgWqVWjliy0SHXHlqcUw5CUhzRxhCOzOEFB96luKQchQx3u4w3OhQjkP6amW6koSRdofRiRbeB/TVEhbMqjHayGnnltAbglJEs6mpliLm9FYZa2YM1MqMtTNCYKTRIVLQ31+hr1QmdZZGI6NUiujkhjgKqSQhtSjmgS11jDGEYUB3VxVjDU47OtqANSSVMtoajPb0VUtoZ8gMLOqvMq4tXSoiU4Z2xxFFjnZmCX3A/IEK60Za1MqKDaMpURgw0FPGGEOaW5xXDHSX6S0nlEshJvdsaNQxBrorCaUgZLjdot7I6eSG/kpCqZzgvadWiSnFIQ+NdkgCQ6lco7crIHIxrTxFWzDWMFirYrynbRytrIMxCpwlVIpaKcR5RU93RE9cIg4CNk402TyeUa0q8AGlKKSnUiKJFLn11IKA0VSz9+wu7h1uEseQqBAi6I1KzO5KSJ3nqL1nc8jusygnO8a1rSPNjE9ftYoHRltEQcDfPHcXTtp/AWEgF/+J6bdjHEVC7GSameGbv1nLb+8bAWDRQJUzjl7CbrNq0xzZUzPRcty8fpQ1Q02sBeUhKUEnhVIMSoFxsLme4VH0lOOiOKktoYUoiaiVQlppThxFxIFipJmzcTxBKUUrt7QyQzlRjPVaWu0OQ42cSggW8B5SDQ5o5xM4LM2OAzzeFW9qwhDq7YyoFGOMJwxAG4/1jmY7R+eggaGGRjvPupE67dRRKhcFM+3B6KLIkCQZYVQUuFq5RQHtDOIgo2sipWUstTiku9Kh0c5opo4wUpRChVWewBdvTnOj2NIy9DSLN79R6DE+oJM5Nrc0cRAwNJ6yud6hlECuPcP1nBAIFVTLGd5ZGtoTe7AKlIEOUClN0MktG7e0cQFEChodT72VUu8UX1VN4pSuUodMG5o5jE1AUCrhs4y6ga5KiDGQZ5aRlsZ52DiWEQOE0NvoEIYhzntCpRhr5OQWyqWM0Zal1clpp5rcFdtfkTHWydhcN1QiaObF/rOxrukqwWgLKuPQU02ZaGlKSYA2nk7mCUNNKYIkDDBK0WhZuhqGwFqiUkw1DtDW084t2lg6OQQBbGnmVCshzoY00w7tHBqZpq0dnbYmDhUpUA08Ex1QQbFu47BDK4PB7g7rRluMNFKaHUutktOdBBjvaaSOSgy1uMRYJyWKQkIUm3szssxy/5acKCz2U68U1VJIo6MxzrFmqEmuLb2NDtbClpbDAQGWrolif0gihXWOyfeExAH011P6u6psGG3QUy1hrGGspWnlMN5qM9Y2JCEQBAyNd6gkCVuaKbnOqZbLRArGOxlxCEEQEyhDqhVZZkliRfHFFE8Uxhiriw+OLIy1NCP1No3MMt42dHKN84oYR8cFlJUj9QFx4Gi2LNpBFGUMNzKG6zlxAJUkJYwUo42cOITNE5pN42nxbtdDJYamhiQE52BjQ4OzpAZC70gtxb4XAApqpZhMawIVUC5FOOeopwZroJwExJGiFBVFQ+sVIbBxvI01UCrn9CQB420zWZCHrnJGOfS4sMhFeQ5d9ZQsM6R5sf6bFra0WlTrEIch7czRyDy1Ro6x//PGvVKOMNrSTj1RANZDKSq2S7PjSEJo22K7tje32NLJmGgY8BDGKVp7ogiSAFIDtSTEek9mHF3jIdbAWNvigXKsqcQB3jvaxSqjNp5S71gCD/dtblAqBfSWOmwczVBhsQq9hyjsECUBWceRTX6D3QJhwzHcqNPKPGNNRTP34GFzK8Vknra1lAPFWMezpZWD90RhSBIHjNVzemrFMexskSsyCxGGLRMpI92GkWabIAgYbWo2jrdo5pZGK2OskxMXq4m2hiRUxHFAJ7VsbnboaMdYy1GNWligexxyF5DlDuOLc08ABCGUoyLvZBrGO4aheodMW9YPtymV2myp54y3O+ROoZTDGU9PI2asrokjTxhFmLx4Yx5HijDwpBqisMjF9480AXDWUyopnC6OnWbmsQ7q+TjtzBXbJlAkKEbaObMaKeN9OeNZTrvtGKlnlMsZSahIkqIQvWpDA2+LHL+pXiY3nq6KYsOWDqgQZQ1xOaYUUhTfnGMizUm1p6uk0F4VBVpC4shifMTseokHR1poA2C4p3sPan6AEdvL3SN1HtzSwnsYKhf7SDvzjGUZ401DmhoIIIlDSrFCG08pCqh3NFsmcoabafHBt4HaOIy0oRK1UP5/clc5aVMqx9SbKUGgUATEkaPZ8Thf9BWqlRjlAlq5xhg3macCAg9BFBIF0E412he5YKJj2TKRYbQmLsU0O5rAF8fW2ESxL1eSgMw4vIdyHFIrhWxp5LQNVCJNOQFjFKHytHOoliIaqcEBSaxJdXEcZg7G0wZRAK28eIMfhgH1lsa6ImdFIeQOygo21y2Z1nRySyMv1sNERzPanqB3LMU6x0gjRVnorqSkxtJdjqmnOa20OAY2j8F4DlEAtbGUzBTLtKXdod4uVm4lUeTGk9sixq5mk3IAeZEmGe7kOBOQlBTNZk4zt/RV2ygfsKXZJlBg3WRfyhbHTBwUv7smUkLl0dYX27YUEQeetvVkqSMqQbsDIxM5KkqZyDVpZplotUlN0ccKFDhf7BuVCIIwwHpXFEK0Z8LCSFMz2tS0Us9QV0YUBgyNNotzYhjSW2ljFbRaGRNp0derdRpUIxjuQJrV2diVMVJv41WA1Y7cFOvQKUgmzzGG4sNWr6CeaaIgYqydoXNHmju2NDIybYmCABX4or86FlOtBqwf6dAVKyZaniQCpxTl2JMa2NLQDHZX+dOmcboqEYGKICj2zWopwuJwDnrKCcYajCs+SKhWMjbWYtaPdAjDYh2VIqiVc1ptTVwJ0LlDUQxB0FsJaGpFNSn6BxtGc8IANjczFAptHCpUtJoGpya3f0NjjWa0NTmfoNgO9QyGGjnlOKCVF8m/K4F6x9POPeUEshyGGzlJKaDTsWTWEwdgLbRsUUTobWi8UrQ7liAolqE6kREoTxxHVJOE3GRUSgnGOcbqGZmBcPI83jKOWeUS924YwxhPKQkw1tNTiZlo5wSBwjlPKVQMtz1DdUOtXhT4q0nEAyN1jC3OkyFQKedEgWJjrcNYM6eVWvrqGUkc0zGa4YZBAeNNQ05xbg2VxzpHV61CZi31eo5SEMUByjlSXay3JITxDFr5OD1VzWizRW6L41Nr6HgIyRlvFn2cjlWMdGJKUcjmsZQ49FgPzgZo7QhiKMcQxzGdtiazxXqxk79LCagoIusYUMX0rmpOKQmZaGbFxUdhQBxF5NYWH4rkRdskiggiRSUKaLRyGp2iT9FdTqkmMePtjHrqUUBlImNLK2e8nWOswhpHKYFOzmRfLWW8Y+nr6jDezom9YqxjiRV0NTXd5ZTMFB/+leKQ3BjCICAJI6qlkE0THYyGIILuSoY2DmcdxoExUC7laOuL5Su3MaY4VsdbmlaqKcUR3htaGkIcmSni3tJO2TiRU4k8Y53i3F/0dyG3RX9mqJFRiyLiksKZgC2NNsZBrRIRK0W9rWnbIqdsbuWUouKNSTmCUiliS8MQOEhKmmpJEQURrSzHO4V1nuGqwTpLbi3NzIEv+lZKFcey9VAtQVe5RKhCRhpt6imUgiInRQl0JSlh4Mmso5bE1DuaZpqzerhDHHpUGBIA1ThiTk+Zdm6Z31Nm/0UDlJ/oG/VpNrurxCdOWcqXrl3NzfeP8v2b13Hz2lHeedSe7DpQne7wxAwnBXEhnma3rRvjq9etYaydo1CccuACXnXILsShDOkvxIzkPTWVFcVRn1C8jRVCiO3Eez6y/gNsaTjacRd7z53ugIQQOz3vqZIRASnS1xFiJinHIe9/6d4s/+Mw/3njA9w33OQjl67gVYcs5OT9FxBJHURMEymIC/E0mWhrvv/7dfx61WYAFvRWePfRe7JkUMbREmImC3G8qrwCgO91DqK4zkkIIbYP5Q39t32XfmDFvm+d7nCEEDNAgOP4cAWE8GN9EFb6OkLMKEopXrzPIPvv0sfXr1/DH9aN8cPfr+fmtaO866glLJolV4uLp58UxIXYzlqZ4fI7N3DFik3kphjj9GXPmcdrDt2VUiSdQSEEaC9XRgghnj4+iNhx7yIkhNgRGenrCDHjDdQSzlq2D79ZPcK3b7iftSMtPnLZnbxk37m84qCFDNSS6Q5RzCBSEBdiO8mM5Rd3D/Gz2x+imRkAlgx28f89bzf2nSc3gRVCFKwK+X568OQY4kIIsX35IGb08NPZPFGMQy+EENubUyE/cwf/zxjiQogZSynFi/aaw7MX9PKN367l9/ePcvU9m/j1qs0c+6y5nHLgAvqqUhgX258UxIXYxjJjue6PI1z6hwcZaxc351jYV+W0w3blubv1o5SMmSeEEEKI6fP+/i8R9zQYs118z3xwusMRQgghxAzTX0v44HH7cM+GOj+6ZT33bqpz5V0buebezSxbOpeTD1hATzme7jDFTkwK4kJsI+tH2/y/lUNc/6cR2nlxRfisrhKvee6uvGjJbIJACuFCCCGEmH5HVm9gMNrCxnwW3xud7miEEEIIMVPtt6CHj5+8HysemuBHt6xn9eYm/33HBq6+Z4gX7Dmbo/edw55zuuTCQrHNSUFciL9CZiw3rRnl/60c4o9Djanpg91lTnj2PI7dby6xfC9QCPE4Au84Ir4fD9ysFyEDpwghtiflLN++dU8qaiEvf86G6Q5HCDEDKO84SN2PjeB2twgnfR0hxJ9RSrH/Ln08Z2Evt60f55Jb1rN2pMU19w5xzb1D7NJf5eh9BnnhXrPprchV42LbkIK4EE9SI9XcsX6CP6wb4/b141NXgwdK8dzd+jl2v7k8Z2GvfIIphHhCFJ69ohEAbtG7TnM0Qoidn+P6tXMB+EH13VRnT3M4QoidnsKzOCj6Onc46esIIR6dUoqDF/Vz0K593L2hzvI/DvO7NVt4cKzNd353P9+/eR0HL+rjubsPcMCufVIcF38VKYgL8Rd471k/2uG29WPctm6cVZsa+D+7DdWc7jLH7DvIUXvPoV/uiiyEeJIcitv0Avzk30IIsT15FdDe9TCaqWfIzmev6Q5ICLHT8yjudgswrvhbCCEej1KKZy/s5dkLe3nz83fnhvu2cO2qzawZbnLz/aPcfH8x3tvi2V0ctGsfB+zax5LBLkIZplY8Cc+IgvhFF13Epz/9aTZt2sQBBxzAF77wBQ477LDpDkvMUGOtnPuGm6zeXPzcN9yko+1WbRYNVDl4UT8H79bPkjldMj64EOIp8ypghVmAQwZLEUI8DVRIZ5fnsmXC/tnH+0IIsf14FbDKLyCzIKNJCiGejFop4qX7zeWl+81l3ZY2N9w3wh0PjrN2pMXakSZrR5pcetuDVJOIJYNdLJ5dY/HsGnvMqTGnqyTf3BePadoL4j/84Q/5wAc+wFe+8hUOP/xwLrzwQpYtW8aqVasYHByc7vDETso6z0gzY8N4h00TKRsmUjZNdHhovMNoK39E+yQK2W9+Dwcv6uPg3fqZ3VWahqiFEEIIIYQQQgghZp5Fs6osmrWI0w5bxHg7584HJ7hj/Th3PDhOMzPc+eA4dz44PtW+qxSxx5wuFvSWmddbZrCnzLyeMnO6S3KvNzH9BfELLriAv/3bv+Utb3kLAF/5yle44oor+MY3vsGHP/zhaY5O7Ci892TG0coMrczSyg2tzNDIDBNtzWgrZ7ydM9rOGWtrxts51j36dVEKxa4DFfac08WSweJnl/6qfP1GCLF9eE8JjQM0EchXiYUQ25P37K9upV1KScMKYzx/uiMSQuzsvCdBA2CkryOE2Ab6qglH7j2HI/eeg3OeB0bbrBlusma4xZqRJutHO39WJN/6uQrFrK6EgVpCXyWmt1r87qsWP12lmGoSUk1CaqWIUhTIleY7oWktiOd5zq233spHPvKRqWlBEHDsscdy4403TmNkYjpds3KIDeMdMuPIrSM3Dj35O5+a5v9n2uRv49yTep04DJjXU2Z+b5n5fRXm9ZZZ0Fth0UCVShJup6UTQoithTheW7kDgO91DgIk/wghth/lDQ/dcBMAn3/57zi9LgVxIcT2FeA4KbwDQvixPggrfR0hxDYUBGpqqJRjnlVM09axfrTN2pEWm+opQ/WMoYmUTfWUzFhGmhkjzeyJzV8pqklIKQ5JwoA4VMRhQBwGJNH/+v/hx6OAWhJx6kELt+OSi7/GtBbER0ZGsNYyd+7crabPnTuXe++99xHtsywjy/5nh52YmACgXq9v30DF0+rXdz3APRue2jYNAzX5SV5ErRRSLUXFJ32VmL5qQm8lpr8W01cpPg185Kd8Hp220OlfvxwzSbNRR3da2KyNtaA8WMBmYB0oVfw2cYRHoYkxaYrVDhwoF2J8iE01KgpRgcJmGhNplFLY3BY/TpFHITrtYFKNjorXwYE14AGdapy36Myj8HgLxoGLIIsivI3JnSdWYKxHO4fONCYHBxgV0OmEmHYbnTtCApx2GMDq4jVCp/ARBKHCaFcsb14spyHAaIexAZnX6DTHZA5nFEGocHhAEVrQVmEzW9xgKFCgPMYHWO0wUUjaKpO1M3TaIbBgco9NNShwCgwh3hUfECkHJgBlinWSdQKy3KKzDh7wAeQdj04zbFZ8eGQsZC7EGIvVkDuInMHrHGvA+qDYbrkj6xiUB5flGICwWFcuDLHeEyuFzjTWgvGQJaA7OSYzWAfOAQp0GmJTi46LdQaT+4YHlxXbSgcROjMENsBaj8l98XoWUAE2AJs5rAqwrti42gRo5zGZxTqHy8GrIkatQqwN0WmK0ZC3DdY6dGpQkcJ60IHHZkWMBMVP2glpBYa808amKW7yNXOrsHhM5rEWMmuwaYYPAzyKPLHo3GKzFBWADcEFCu0D0naA8Q6TtbC5QwcBWEvqi6Rjsja5LY4BZRTOO8zkaS8IQMcRWeAxnTZaGYw1mNSQdUJ0p41JLWFUjNWZBZrQafI0xRqN8RYfgO3kEEEYRCgsxhTbOLAKpSgO4CDCWoMLA5yFrA263cYah+44tNZ4r7DeYXyAVsVvFTh05jCu2BfzSGMzXRwbLsJHxbGtAjChRacZKgDvINbgLJjJbW3DCGcd1oH3DmMnrydT4AKIbYSxBqUCQh/irMNkFmcoYtGg4ghlA7RRRCHoTgfnIFTFNJPZImcp0D4kiDwEISbTWA02CDG5xeTFPj8VnwOigFx7XOaxvog5UMWxZnyIMQ6beQjAegg0qDAo9qMQ7OStKWwAOgyxmcU5sE7hco+KQIdgDGQuAA/GTq5vBy4rsol2EJsAjytidqC9Im8rUMU+FdoA7TQmyyCA0Be5zEdFADp12MkvTT380bJRCpt5jFKYyRyjwxCjPcY6dDD5eBCgvEcFAdoE2EyjVYTJDN6Cn1xvluI415HHph1cEJBFrtjn2wFZJ8NkRZINgiKnm0ChrMJkjjwMixyYOawt1qm1oH2AzR3OF/uHp3hNY4rtZAzknZAscmTGYrIOkVdkMcW5xBW511lPFkWY1BCEHh+FGF0cH0Fc5G5rQAWQtkNsp13si86jncIb8BT5yrriNY32aKPwoQIUNtXkQUSn7cnyHJ06TJZjVECgFN4EdEoBNmvjLGgCdGTJjSd3CpOmRQDWAhEqABVEhM5hMl1sG6fQqGJ/8AFEDutD8shgs3axPE6jdXGlZqPjyDotdNpGAbkPCAMw2pN3DDo16NRCAIEJCHKFdR4dBZjUkCuLzTK8Am0gp8jl1kxuDIpcHLsA7WNMlhEWiQYVePKsuL2ecUWOQAXY3GCsQ4UQqIDAQxAGxXlPG8zkfPOOQqcZxhjwETYzuMnzjbGT53QXYKzDezA2QNuQPNXFsWyL48dYhVMekxfnVZMVY6tb+/B+ODlPV+xbNgMTBNggQGcG7yb7OpPtsgB0arFaY7SbOtc5wIQBWhm0c9gsAwtahRjrSIkwqcZmRY4zFlwOVhV/GzP59+Sx5z1op9DWF20daDt5/KiH11HROXBWYdIck1syFRP6AJN2iphN0ad4eBmnYlVB0S9xxb5vfIgP/NQ5Q3kwk+dtwpC07cm1xXQ6xfbUngBwfvI8byCIApxz+EDhtS9eM4S8HaJzTxYanAowWRvrIQwDNFHxDao0x2UPb1fITbGv6SBARwaTdWCqDwahKvKOCorcoH2xT3oFphPhwwiT5ujckace08mwxqHCAKU8BkXmI0IVYLIUYxUm8yhb3BTS6OJY15EjVR6TttAqQhGhQotOc3IXAsU5MfUa7wzahpg0I1chOogxWVrkSTOZt5jsA6mgWM+AChTaqyIGo1CRwmS6OO5USKAUxjhUWJzX3OT216nBGIPJivkEiqLvnIFWAaFRWO3xQG7BZB6Xe4wr+rxaFTuTzR3W+iL9mGJfeXgfcarok7nJx7RXBIEHHZJbTaYzQqdxRpOmRV9H6zY+CMk7kHmD6bQxzhPYAOc8mY+wqcYWPXkIVHGuBawKyEJLZCNM2iZ3xbrzgEaBUujAkKdFjs3CEGyM1RqXWdTksW0A7QIcFH0Z5dDWYjNdpC4boHxxXlUB6KA4HrWCXIFJ29jJ40Vbps5BerK/a70i70QEUYhNMwiL/d35Yh8NXZEzlS3OO9b9z3HoFIQOgiDE5havJs+jQYixISbNCYJi/fs4xFiLUwqrbXE/migkCNVkntbYyf3WuABtYmyW4yZzr3agOwbT0RinsHYythxUCDrQmMyShTG2o4v1l7nivV0QknlNbg0mM0Vfw1hQAToKyXxYHDsawskcYp3HueIYNbbYZsb6oq9JsS2tgiy26FQT2AjnDEZT5GlbxJ2nBptqTORxk/1krUDryZzki9yQhxHeK5wO0FkHayEnxCuFzgzWTp2qivcZDnQMgS3er1gPIcWx56IQnWmUUxjvyZTFOouxDpsXO4P3k+97H+6neDAYvAqxWac4PwZFfMpP9o0DX7zfcRE2NeiOw2ZpcRyFAQFgbISOi/cXnVaDer2Oz2PEXza7BLMXVmBhZWqa9556xzDcTBlvacZTTb2tmUhzJtqGiVTTygwdbenkdmpkgaz95F+/uxzxkj27t9XiiCepu7v7ca/sV977abufzoYNG1i4cCE33HADRxxxxNT0s88+m+XLl3PTTTdt1f4f//Ef+cQnPvF0hymEEEIIIYQQQgghhBBiBzAxMUFPT89jPj6tV4jPnj2bMAwZGhraavrQ0BDz5s17RPuPfOQjfOADH5j63znH6Ogos2bNkvF8hJhG9XqdXXfdlfXr1z9uwhFCiB2N5DchxM5K8psQYmcmOU6Ima27+/Gvzp/WgniSJBxyyCFcc801nHrqqUBR5L7mmms488wzH9G+VCpRKpW2mtbX1/c0RCqEeCJ6enqksyGE2ClJfhNC7KwkvwkhdmaS44QQj2ZaC+IAH/jAB3jTm97Ec5/7XA477DAuvPBCWq0Wb3nLW6Y7NCGEEEIIIYQQQgghhBA7kWkviL/2ta9leHiYc889l02bNnHggQdy1VVXPeJGm0IIIYQQQgghhBBCCCHEX2PaC+IAZ5555qMOkSKE2DGUSiU+/vGPP2JIIyGE2NFJfhNC7KwkvwkhdmaS44QQj0d57/10ByGEEEIIIYQQQgghhBBCbG/BdAcghBBCCCGEEEIIIYQQQjwdpCAuhBBCCCGEEEIIIYQQYkaQgrgQQgghhBBCCCGEEEKIGUEK4kKIv8pFF13E7rvvTrlc5vDDD+fmm2+e7pCEEDPcddddx8knn8yCBQtQSvGTn/xkq8e995x77rnMnz+fSqXCsccey5/+9Ket2oyOjvKGN7yBnp4e+vr6eNvb3kaz2dyqzZ133smLXvQiyuUyu+66K+eff/4jYrnkkkvYd999KZfLPOc5z+HnP//5Nl9eIcTM8clPfpJDDz2U7u5uBgcHOfXUU1m1atVWbdI05YwzzmDWrFl0dXXxqle9iqGhoa3arFu3jhNPPJFqtcrg4CBnnXUWxpit2vz617/m4IMPplQqsWTJEr71rW89Ih7pBwohtpUvf/nL7L///vT09NDT08MRRxzBlVdeOfW45DYhxLYkBXEhxFP2wx/+kA984AN8/OMf5w9/+AMHHHAAy5YtY/PmzdMdmhBiBmu1WhxwwAFcdNFFj/r4+eefz+c//3m+8pWvcNNNN1Gr1Vi2bBlpmk61ecMb3sDdd9/N1VdfzeWXX851113H6aefPvV4vV7nuOOOY7fdduPWW2/l05/+NP/4j//IV7/61ak2N9xwA6973et429vexm233capp57Kqaeeyl133bX9Fl4IsVNbvnw5Z5xxBr/73e+4+uqr0Vpz3HHH0Wq1ptq8//3v57//+7+55JJLWL58ORs2bOCVr3zl1OPWWk488UTyPOeGG27g29/+Nt/61rc499xzp9qsXbuWE088kaOPPprbb7+d973vfbz97W/nF7/4xVQb6QcKIbalXXbZhfPOO49bb72VW265hZe85CWccsop3H333YDkNiHENuaFEOIpOuyww/wZZ5wx9b+11i9YsMB/8pOfnMaohBDifwD+sssum/rfOefnzZvnP/3pT09NGx8f96VSyX//+9/33nt/zz33eMD//ve/n2pz5ZVXeqWUf+ihh7z33n/pS1/y/f39PsuyqTYf+tCH/D777DP1/2te8xp/4oknbhXP4Ycf7t/xjnds02UUQsxcmzdv9oBfvny5977IZ3Ec+0suuWSqzcqVKz3gb7zxRu+99z//+c99EAR+06ZNU22+/OUv+56enqmcdvbZZ/ulS5du9Vqvfe1r/bJly6b+l36gEGJ76+/v91//+tcltwkhtjm5QlwI8ZTkec6tt97KscceOzUtCAKOPfZYbrzxxmmMTAghHtvatWvZtGnTVrmrt7eXww8/fCp33XjjjfT19fHc5z53qs2xxx5LEATcdNNNU22OPPJIkiSZarNs2TJWrVrF2NjYVJs/f52H20iOFEJsKxMTEwAMDAwAcOutt6K13ir37LvvvixatGirHPec5zyHuXPnTrVZtmwZ9Xp96krMv5S/pB8ohNierLX84Ac/oNVqccQRR0huE0Jsc1IQF0I8JSMjI1hrt+pwAMydO5dNmzZNU1RCCPH4Hs5Pj5e7Nm3axODg4FaPR1HEwMDAVm0ebR5//hqP1UZypBBiW3DO8b73vY8XvOAFPPvZzwaKvJMkCX19fVu1/d857qnmr3q9TqfTkX6gEGK7WLFiBV1dXZRKJd75zndy2WWXsd9++0luE0Jsc9F0ByCEEEIIIYQQ4sk544wzuOuuu/jNb34z3aEIIcQ2sc8++3D77bczMTHBf/3Xf/GmN72J5cuXT3dYQoidkFwhLoR4SmbPnk0Yho+4s/fQ0BDz5s2bpqiEEOLxPZyfHi93zZs37xE3TjLGMDo6ulWbR5vHn7/GY7WRHCmE+GudeeaZXH755Vx77bXssssuU9PnzZtHnueMj49v1f5/57inmr96enqoVCrSDxRCbBdJkrBkyRIOOeQQPvnJT3LAAQfwuc99TnKbEGKbk4K4EOIpSZKEQw45hGuuuWZqmnOOa665hiOOOGIaIxNCiMe2ePFi5s2bt1Xuqtfr3HTTTVO564gjjmB8fJxbb711qs2vfvUrnHMcfvjhU22uu+46tNZTba6++mr22Wcf+vv7p9r8+es83EZypBDiqfLec+aZZ3LZZZfxq1/9isWLF2/1+CGHHEIcx1vlnlWrVrFu3bqtctyKFSu2+uDv6quvpqenh/3222+qzePlL+kHCiGeDs45siyT3CaE2Pam+66eQogd1w9+8ANfKpX8t771LX/PPff4008/3ff19W11Z28hhHi6NRoNf9ttt/nbbrvNA/6CCy7wt912m3/ggQe8996fd955vq+vz//0pz/1d955pz/llFP84sWLfafTmZrH8ccf7w866CB/0003+d/85jd+r7328q973eumHh8fH/dz5871/+f//B9/1113+R/84Ae+Wq36f//3f59q89vf/tZHUeQ/85nP+JUrV/qPf/zjPo5jv2LFiqdvZQghdirvete7fG9vr//1r3/tN27cOPXTbren2rzzne/0ixYt8r/61a/8Lbfc4o844gh/xBFHTD1ujPHPfvaz/XHHHedvv/12f9VVV/k5c+b4j3zkI1Nt1qxZ46vVqj/rrLP8ypUr/UUXXeTDMPRXXXXVVBvpBwohtqUPf/jDfvny5X7t2rX+zjvv9B/+8Ie9Usr/8pe/9N5LbhNCbFtSEBdC/FW+8IUv+EWLFvkkSfxhhx3mf/e73013SEKIGe7aa6/1wCN+3vSmN3nvvXfO+Y997GN+7ty5vlQq+WOOOcavWrVqq3ls2bLFv+51r/NdXV2+p6fHv+Utb/GNRmOrNnfccYd/4Qtf6Eulkl+4cKE/77zzHhHLj370I7/33nv7JEn80qVL/RVXXLHdllsIsfN7tNwG+G9+85tTbTqdjn/3u9/t+/v7fbVa9a94xSv8xo0bt5rP/fff70844QRfqVT87Nmz/Qc/+EGvtd6qzbXXXusPPPBAnySJ32OPPbZ6jYdJP1AIsa289a1v9bvttptPksTPmTPHH3PMMVPFcO8ltwkhti3lvffTc226EEIIIYQQQgghhBBCCPH0kTHEhRBCCCGEEEIIIYQQQswIUhAXQgghhBBCCCGEEEIIMSNIQVwIIYQQQgghhBBCCCHEjCAFcSGEEEIIIYQQQgghhBAzghTEhRBCCCGEEEIIIYQQQswIUhAXQgghhBBCCCGEEEIIMSNIQVwIIYQQQgghhBBCCCHEjCAFcSGEEEIIIYQQQgghhBAzghTEhRBCCCGE2IG8+MUv5n3ve99jPr777rtz4YUXPm3xCCGEEEIIsSOJpjsAIYQQQgghxBN36aWXEsfxdIchhBBCCCHEDkkK4kIIIYQQQuxABgYGpjsEIYQQQgghdlgyZIoQQgghhBA7kD8fMmXz5s2cfPLJVCoVFi9ezMUXXzy9wQkhhBBCCPEMJ1eICyGEEEIIsYN685vfzIYNG7j22muJ45j3vve9bN68ebrDEkIIIYQQ4hlLCuJCCCGEEELsgP74xz9y5ZVXcvPNN3PooYcC8B//8R8861nPmubIhBBCCCGEeOaSIVOEEEIIIYTYAa1cuZIoijjkkEOmpu2777709fVNX1BCCCGEEEI8w0lBXAghhBBCCCGEEEIIIcSMIAVxIYQQQgghdkD77rsvxhhuvfXWqWmrVq1ifHx8+oISQgghhBDiGU4K4kIIIYQQQuyA9tlnH44//nje8Y53cNNNN3Hrrbfy9re/nUqlMt2hCSGEEEII8YwlBXEhhBBCCCF2UN/85jdZsGABRx11FK985Ss5/fTTGRwcnO6whBBCCCGEeMZS3ns/3UEIIYQQQgghhBBCCCGEENubXCEuhBBCCCGEEEIIIYQQYkaQgrgQQgghhBBCCCGEEEKIGUEK4kIIIYQQQgghhBBCCCFmBCmICyGEEEIIIYQQQgghhJgRpCAuhBBCCCGEEEIIIYQQYkaQgrgQQgghhBBCCCGEEEKIGUEK4kIIIYQQQgghhBBCCCFmBCmICyGEEEIIIYQQQgghhJgRpCAuhBBCCCGEEEIIIYQQYkaQgrgQQgghhBBCCCGEEEKIGUEK4kIIIYQQQgghhBBCCCFmBCmICyGEEEIIIYQQQgghhJgR/n9Lyq4PNHvbNAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "klib.missingval_plot(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U0O1W3d2N8TE",
        "outputId": "e2d2f73b-101c-4661-e253-6a2cfe8a1494"
      },
      "execution_count": 80,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "No missing values found in the dataset.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uJnaqcdoMV4x",
        "outputId": "4c3c9239-63c3-4501-80c0-ef4cb41195d7"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'creator', 'subjects', 'publisher',\n",
              "       'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "  ### Vectorization Of the Column Using `TF-IDF Vectorizer`"
      ],
      "metadata": {
        "id": "otPlP174s1Ae"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_extraction.text import TfidfVectorizer"
      ],
      "metadata": {
        "id": "R6qX96WTr76N"
      },
      "execution_count": 82,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "MAX_WORDS = 200000"
      ],
      "metadata": {
        "id": "AWC5hiJzLW8_"
      },
      "execution_count": 86,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers import TextVectorization"
      ],
      "metadata": {
        "id": "1FYH9vHGV6Rc"
      },
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vectorizer = TextVectorization(max_tokens = MAX_WORDS,\n",
        "                               output_sequence_length = 1800,\n",
        "                               output_mode = 'int')"
      ],
      "metadata": {
        "id": "ix9loA53V6U6"
      },
      "execution_count": 88,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IbZ5GMZEWD0K",
        "outputId": "b8d5e3f7-85c2-4fc8-ec1e-9790117545bf"
      },
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'subjects', 'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "vectorizer.adapt(train_data[\"subjects\"].values)"
      ],
      "metadata": {
        "id": "FsKptB20V6YN"
      },
      "execution_count": 90,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vectorizer.get_vocabulary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lg3Uet10WJvV",
        "outputId": "8365a798-adb6-4421-9d4a-24c2611adbff"
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['',\n",
              " '[UNK]',\n",
              " 'fiction',\n",
              " 'juvenile',\n",
              " 'and',\n",
              " 'literature',\n",
              " 'music',\n",
              " 'states',\n",
              " 'united',\n",
              " 'films',\n",
              " 'history',\n",
              " 'drama',\n",
              " 'recordings',\n",
              " 'for',\n",
              " 'the',\n",
              " 'biography',\n",
              " 'stories',\n",
              " 'video',\n",
              " 'fictitious',\n",
              " 'women',\n",
              " 'character',\n",
              " 'hearing',\n",
              " 'impaired',\n",
              " 'popular',\n",
              " 'life',\n",
              " 'books',\n",
              " 'mystery',\n",
              " 'etc',\n",
              " 'feature',\n",
              " 'new',\n",
              " 'century',\n",
              " 'of',\n",
              " 'th',\n",
              " 'american',\n",
              " 'england',\n",
              " 'childrens',\n",
              " 'rock',\n",
              " 'social',\n",
              " 'in',\n",
              " 'english',\n",
              " 'language',\n",
              " 'york',\n",
              " 'songs',\n",
              " 'jazz',\n",
              " 'war',\n",
              " 'travel',\n",
              " 'guidebooks',\n",
              " 'state',\n",
              " 'americans',\n",
              " 'comic',\n",
              " 'african',\n",
              " 'strips',\n",
              " 'aspects',\n",
              " 'poetry',\n",
              " 'works',\n",
              " 'love',\n",
              " 'great',\n",
              " 'customs',\n",
              " 'science',\n",
              " 'animals',\n",
              " 'britain',\n",
              " 'type',\n",
              " 'large',\n",
              " 'world',\n",
              " 'manuals',\n",
              " 'france',\n",
              " 'psychological',\n",
              " 'washington',\n",
              " 'america',\n",
              " 'relations',\n",
              " 'historical',\n",
              " 'description',\n",
              " 'china',\n",
              " 'children',\n",
              " 'fantasy',\n",
              " 'sound',\n",
              " 'television',\n",
              " 'psychology',\n",
              " 'n',\n",
              " 'family',\n",
              " 'art',\n",
              " 'indians',\n",
              " 'folklore',\n",
              " 'motion',\n",
              " 'japan',\n",
              " 'domestic',\n",
              " 'north',\n",
              " 'california',\n",
              " 'friendship',\n",
              " 'animated',\n",
              " 'government',\n",
              " 'pictorial',\n",
              " 'schools',\n",
              " 'y',\n",
              " 'handbooks',\n",
              " 'picture',\n",
              " 'politics',\n",
              " 'dogs',\n",
              " 'humorous',\n",
              " 'cookery',\n",
              " 'with',\n",
              " 'relationships',\n",
              " 'criticism',\n",
              " 'piano',\n",
              " 'mexico',\n",
              " 's',\n",
              " 'police',\n",
              " 'novels',\n",
              " 'authors',\n",
              " 'italy',\n",
              " 'detective',\n",
              " 'folk',\n",
              " 'self',\n",
              " 'graphic',\n",
              " 'comedy',\n",
              " 'programs',\n",
              " 'sisters',\n",
              " 'study',\n",
              " 'c',\n",
              " 'west',\n",
              " 'rhyme',\n",
              " 'spanish',\n",
              " 'child',\n",
              " 'computer',\n",
              " 'foreign',\n",
              " 'country',\n",
              " 'religious',\n",
              " 'u',\n",
              " 'civilization',\n",
              " 'london',\n",
              " 'men',\n",
              " 'chinese',\n",
              " 'ireland',\n",
              " 'conditions',\n",
              " 'people',\n",
              " 'tales',\n",
              " 'man',\n",
              " 'investigators',\n",
              " 'suspense',\n",
              " 'private',\n",
              " 'woman',\n",
              " 'christian',\n",
              " 'management',\n",
              " 'brothers',\n",
              " 'to',\n",
              " 'calif',\n",
              " 'human',\n",
              " 'pacific',\n",
              " 'behavior',\n",
              " 'africa',\n",
              " 'mothers',\n",
              " 'orchestra',\n",
              " 'personal',\n",
              " 'business',\n",
              " 'humor',\n",
              " 'young',\n",
              " 'decoration',\n",
              " 'design',\n",
              " 'miscellanea',\n",
              " 'health',\n",
              " 'seattle',\n",
              " 'murder',\n",
              " 'civil',\n",
              " 'teaching',\n",
              " 'john',\n",
              " 'fathers',\n",
              " 'studies',\n",
              " 'los',\n",
              " 'cats',\n",
              " 'angeles',\n",
              " 'materials',\n",
              " 'germany',\n",
              " 'region',\n",
              " 'adventure',\n",
              " 'japanese',\n",
              " 'd',\n",
              " 'blues',\n",
              " 'dance',\n",
              " 'political',\n",
              " 'death',\n",
              " 'speakers',\n",
              " 'exhibitions',\n",
              " 'europe',\n",
              " 'architecture',\n",
              " 'dictionaries',\n",
              " 'daughters',\n",
              " 'national',\n",
              " 'india',\n",
              " 'gardening',\n",
              " 'excerpts',\n",
              " 'cooking',\n",
              " 'b',\n",
              " 'wash',\n",
              " 'south',\n",
              " 'case',\n",
              " 'musical',\n",
              " 'jewish',\n",
              " 'horror',\n",
              " 'painting',\n",
              " 'short',\n",
              " 'interpersonal',\n",
              " 'girls',\n",
              " 'education',\n",
              " 'musicians',\n",
              " 'christmas',\n",
              " 'northwest',\n",
              " 'middle',\n",
              " 'magic',\n",
              " 'interior',\n",
              " 'french',\n",
              " 'detectives',\n",
              " 'city',\n",
              " 'russia',\n",
              " 'photography',\n",
              " 'industry',\n",
              " 'fairy',\n",
              " 'time',\n",
              " 'egypt',\n",
              " 'families',\n",
              " 'instrumental',\n",
              " 'high',\n",
              " 'vietnamese',\n",
              " 'amateurs',\n",
              " 'school',\n",
              " 'imaginary',\n",
              " 'technique',\n",
              " 'san',\n",
              " 'river',\n",
              " 'authorship',\n",
              " 'sex',\n",
              " 'economic',\n",
              " 'scotland',\n",
              " 'instruction',\n",
              " 'king',\n",
              " 'bears',\n",
              " 'pictures',\n",
              " 'construction',\n",
              " 'bildungsromans',\n",
              " 'anecdotes',\n",
              " 'western',\n",
              " 'race',\n",
              " 'guitar',\n",
              " 'church',\n",
              " 'southern',\n",
              " 'philosophy',\n",
              " 'crimes',\n",
              " 'experiments',\n",
              " 'ecology',\n",
              " 'teenage',\n",
              " 'rap',\n",
              " 'string',\n",
              " 'jews',\n",
              " 'canada',\n",
              " 'operas',\n",
              " 'animal',\n",
              " 'married',\n",
              " 'conduct',\n",
              " 'baseball',\n",
              " 'arranged',\n",
              " 'teenagers',\n",
              " 'rights',\n",
              " 'presidents',\n",
              " 'finance',\n",
              " 'reading',\n",
              " 'mythology',\n",
              " 'william',\n",
              " 'persons',\n",
              " 'paris',\n",
              " 'latin',\n",
              " 'culture',\n",
              " 'parent',\n",
              " 'interpretation',\n",
              " 'characters',\n",
              " 'texas',\n",
              " 'journalists',\n",
              " 'spain',\n",
              " 'sons',\n",
              " 'hawaii',\n",
              " 'george',\n",
              " 'buddhism',\n",
              " 'artists',\n",
              " 'place',\n",
              " 'francisco',\n",
              " 'arts',\n",
              " 'east',\n",
              " 'vietnam',\n",
              " 'plants',\n",
              " 'gardens',\n",
              " 'on',\n",
              " 'massachusetts',\n",
              " 'mass',\n",
              " 'christianity',\n",
              " 'alphabet',\n",
              " 'spiritual',\n",
              " 'vocals',\n",
              " 'translations',\n",
              " 'conflict',\n",
              " 'british',\n",
              " 'patients',\n",
              " 'natural',\n",
              " 'marriage',\n",
              " 'into',\n",
              " 'birds',\n",
              " 'antiquities',\n",
              " 'house',\n",
              " 'collections',\n",
              " 'alaska',\n",
              " 'violin',\n",
              " 'medicine',\n",
              " 'food',\n",
              " 'exercises',\n",
              " 'college',\n",
              " 'identity',\n",
              " 'diet',\n",
              " 'care',\n",
              " 'australia',\n",
              " 'against',\n",
              " 'development',\n",
              " 'mice',\n",
              " 'communication',\n",
              " 'space',\n",
              " 'modern',\n",
              " 'legal',\n",
              " 'interviews',\n",
              " 'county',\n",
              " 'military',\n",
              " 'law',\n",
              " 'island',\n",
              " 'florida',\n",
              " 'chicago',\n",
              " 'patterns',\n",
              " 'serial',\n",
              " 'homes',\n",
              " 'diaries',\n",
              " 'training',\n",
              " 'islands',\n",
              " 'terrorism',\n",
              " 'teachers',\n",
              " 'saint',\n",
              " 'prevention',\n",
              " 'investigation',\n",
              " 'youth',\n",
              " 'public',\n",
              " 'policy',\n",
              " 'work',\n",
              " 'students',\n",
              " 'nature',\n",
              " 'ill',\n",
              " 'concertos',\n",
              " 'age',\n",
              " 'actors',\n",
              " 'vocational',\n",
              " 'orphans',\n",
              " 'handicraft',\n",
              " 'counting',\n",
              " 'trade',\n",
              " 'musicals',\n",
              " 'exercise',\n",
              " 'basketball',\n",
              " 'romantic',\n",
              " 'indian',\n",
              " 'holocaust',\n",
              " 'religion',\n",
              " 'pets',\n",
              " 'games',\n",
              " 'bible',\n",
              " 'single',\n",
              " 'kings',\n",
              " 'italian',\n",
              " 'gay',\n",
              " 'medical',\n",
              " 'james',\n",
              " 'greece',\n",
              " 'environmental',\n",
              " 'day',\n",
              " 'carolina',\n",
              " 'problems',\n",
              " 'mental',\n",
              " 'literary',\n",
              " 'rome',\n",
              " 'physical',\n",
              " 'guidance',\n",
              " 'wit',\n",
              " 'valley',\n",
              " 'disabilities',\n",
              " 'revolution',\n",
              " 'mississippi',\n",
              " 'home',\n",
              " 'actresses',\n",
              " 'therapy',\n",
              " 'success',\n",
              " 'sea',\n",
              " 'color',\n",
              " 'service',\n",
              " 'oregon',\n",
              " 'mexican',\n",
              " 'trials',\n",
              " 'treatment',\n",
              " 'rulers',\n",
              " 'identification',\n",
              " 'dinosaurs',\n",
              " 'aged',\n",
              " 'virginia',\n",
              " 'pennsylvania',\n",
              " 'voice',\n",
              " 'union',\n",
              " 'novelists',\n",
              " 'intelligence',\n",
              " 'horses',\n",
              " 'dwellings',\n",
              " 'diseases',\n",
              " 'cuba',\n",
              " 'childhood',\n",
              " 'park',\n",
              " 'missing',\n",
              " 'guides',\n",
              " 'town',\n",
              " 'planning',\n",
              " 'microsoft',\n",
              " 'i',\n",
              " 'greek',\n",
              " 'w',\n",
              " 'players',\n",
              " 'pioneer',\n",
              " 'narratives',\n",
              " 'mary',\n",
              " 'j',\n",
              " 'henry',\n",
              " 'frontier',\n",
              " 'forest',\n",
              " 'coast',\n",
              " 'a',\n",
              " 'wars',\n",
              " 'victims',\n",
              " 'israel',\n",
              " 'fitness',\n",
              " 'collection',\n",
              " 'catholic',\n",
              " 'boys',\n",
              " 'jersey',\n",
              " 'farm',\n",
              " 'exploration',\n",
              " 'software',\n",
              " 'grandmothers',\n",
              " 'crime',\n",
              " 'web',\n",
              " 'spy',\n",
              " 'prisoners',\n",
              " 'hygiene',\n",
              " 'haunts',\n",
              " 'female',\n",
              " 'encyclopedias',\n",
              " 'central',\n",
              " 'boston',\n",
              " 'abuse',\n",
              " 'vampires',\n",
              " 'multimedia',\n",
              " 'louisiana',\n",
              " 'thomas',\n",
              " 'program',\n",
              " 'periodicals',\n",
              " 'de',\n",
              " 'cartoons',\n",
              " 'biology',\n",
              " 'use',\n",
              " 'making',\n",
              " 'landscape',\n",
              " 'drawing',\n",
              " 'cancer',\n",
              " 'brazil',\n",
              " 'air',\n",
              " 'soviet',\n",
              " 'sexual',\n",
              " 'richard',\n",
              " 'edward',\n",
              " 'spouses',\n",
              " 'sports',\n",
              " 'operations',\n",
              " 'knitting',\n",
              " 'interactive',\n",
              " 'body',\n",
              " 'ballet',\n",
              " 'babies',\n",
              " 'yoga',\n",
              " 'recipes',\n",
              " 'dragons',\n",
              " 'textbooks',\n",
              " 'revenge',\n",
              " 'toy',\n",
              " 'symphonies',\n",
              " 'singers',\n",
              " 'queens',\n",
              " 'legendary',\n",
              " 'directories',\n",
              " 'russian',\n",
              " 'queen',\n",
              " 'marine',\n",
              " 'international',\n",
              " 'infants',\n",
              " 'ghosts',\n",
              " 'discovery',\n",
              " 'pigs',\n",
              " 'murderers',\n",
              " 'maintenance',\n",
              " 'federation',\n",
              " 'enterprises',\n",
              " 'charles',\n",
              " 'biographical',\n",
              " 'twins',\n",
              " 'repair',\n",
              " 'netherlands',\n",
              " 'naval',\n",
              " 'mountains',\n",
              " 'legends',\n",
              " 'illinois',\n",
              " 'heroes',\n",
              " 'comedies',\n",
              " 'adult',\n",
              " 'workers',\n",
              " 'mount',\n",
              " 'georgia',\n",
              " 'electronic',\n",
              " 'christ',\n",
              " 'arthur',\n",
              " 'writing',\n",
              " 'shakespeare',\n",
              " 'movements',\n",
              " 'judaism',\n",
              " 'jesus',\n",
              " 'clothing',\n",
              " 'catalogs',\n",
              " 'series',\n",
              " 'parenting',\n",
              " 'german',\n",
              " 'first',\n",
              " 'countries',\n",
              " 'clergy',\n",
              " 'witches',\n",
              " 'planets',\n",
              " 'nutrition',\n",
              " 'investments',\n",
              " 'ii',\n",
              " 'file',\n",
              " 'criminals',\n",
              " 'automobile',\n",
              " 'sonatas',\n",
              " 'saxophone',\n",
              " 'physicians',\n",
              " 'period',\n",
              " 'murders',\n",
              " 'lewis',\n",
              " 'fla',\n",
              " 'employment',\n",
              " 'early',\n",
              " 'conservation',\n",
              " 'campaigns',\n",
              " 'st',\n",
              " 'sites',\n",
              " 'philippines',\n",
              " 'loss',\n",
              " 'islam',\n",
              " 'friends',\n",
              " 'railroads',\n",
              " 'poets',\n",
              " 'orchestral',\n",
              " 'officers',\n",
              " 'lawyers',\n",
              " 'immigration',\n",
              " 'evolution',\n",
              " 'dancing',\n",
              " 'asia',\n",
              " 'adaptations',\n",
              " 'water',\n",
              " 'voyages',\n",
              " 'ocean',\n",
              " 'immigrants',\n",
              " 'etiquette',\n",
              " 'elizabeth',\n",
              " 'black',\n",
              " 'theater',\n",
              " 'systems',\n",
              " 'racing',\n",
              " 'rabbits',\n",
              " 'other',\n",
              " 'movable',\n",
              " 'houses',\n",
              " 'hiking',\n",
              " 'furniture',\n",
              " 'fire',\n",
              " 'ex',\n",
              " 'alternative',\n",
              " 'therapeutic',\n",
              " 'survival',\n",
              " 'republic',\n",
              " 'rearing',\n",
              " 'mathematics',\n",
              " 'low',\n",
              " 'kidnapping',\n",
              " 'examinations',\n",
              " 'drug',\n",
              " 'area',\n",
              " 'actualization',\n",
              " 'underground',\n",
              " 'ships',\n",
              " 'scores',\n",
              " 'remodeling',\n",
              " 'radio',\n",
              " 'questions',\n",
              " 'publishing',\n",
              " 'princesses',\n",
              " 'perception',\n",
              " 'mixed',\n",
              " 'maine',\n",
              " 'emigration',\n",
              " 'disease',\n",
              " 'correspondence',\n",
              " 'buildings',\n",
              " 'big',\n",
              " 'toys',\n",
              " 't',\n",
              " 'suites',\n",
              " 'paul',\n",
              " 'orleans',\n",
              " 'monsters',\n",
              " 'methods',\n",
              " 'medieval',\n",
              " 'marketing',\n",
              " 'irish',\n",
              " 'horse',\n",
              " 'franklin',\n",
              " 'columbia',\n",
              " 'robert',\n",
              " 'regions',\n",
              " 'internet',\n",
              " 'creative',\n",
              " 'beauty',\n",
              " 'sources',\n",
              " 'philadelphia',\n",
              " 'ornament',\n",
              " 'ensemble',\n",
              " 'elementary',\n",
              " 'dating',\n",
              " 'insects',\n",
              " 'bilingual',\n",
              " 'appreciation',\n",
              " 'wine',\n",
              " 'vocabulary',\n",
              " 'technology',\n",
              " 'supernatural',\n",
              " 'old',\n",
              " 'iran',\n",
              " 'ethnic',\n",
              " 'birthdays',\n",
              " 'vocal',\n",
              " 'regency',\n",
              " 'participation',\n",
              " 'mother',\n",
              " 'moral',\n",
              " 'm',\n",
              " 'legislation',\n",
              " 'justice',\n",
              " 'infancy',\n",
              " 'grandfathers',\n",
              " 'costume',\n",
              " 'battle',\n",
              " 'arizona',\n",
              " 'policewomen',\n",
              " 'peter',\n",
              " 'parties',\n",
              " 'gospel',\n",
              " 'employees',\n",
              " 'clark',\n",
              " 'beings',\n",
              " 'army',\n",
              " 'arab',\n",
              " 'violoncello',\n",
              " 'soul',\n",
              " 'scientists',\n",
              " 'resources',\n",
              " 'rain',\n",
              " 'movement',\n",
              " 'montana',\n",
              " 'halloween',\n",
              " 'doctrines',\n",
              " 'control',\n",
              " 'colorado',\n",
              " 'class',\n",
              " 'bedtime',\n",
              " 'activity',\n",
              " 'wizards',\n",
              " 'stephen',\n",
              " 'small',\n",
              " 'rhymes',\n",
              " 'pa',\n",
              " 'nursery',\n",
              " 'northern',\n",
              " 'martial',\n",
              " 'collecting',\n",
              " 'change',\n",
              " 'brown',\n",
              " 'band',\n",
              " 'adventurers',\n",
              " 'trumpet',\n",
              " 'sir',\n",
              " 'quartets',\n",
              " 'puzzles',\n",
              " 'jewelry',\n",
              " 'explorers',\n",
              " 'emperor',\n",
              " 'dog',\n",
              " 'disorders',\n",
              " 'celtic',\n",
              " 'caricatures',\n",
              " 'soldiers',\n",
              " 'restaurants',\n",
              " 'railroad',\n",
              " 'producers',\n",
              " 'louis',\n",
              " 'fashion',\n",
              " 'erotic',\n",
              " 'directors',\n",
              " 'archaeology',\n",
              " 'accidents',\n",
              " 'va',\n",
              " 'trains',\n",
              " 'settings',\n",
              " 'sacred',\n",
              " 'recreation',\n",
              " 'real',\n",
              " 'poland',\n",
              " 'minnesota',\n",
              " 'iraq',\n",
              " 'hatha',\n",
              " 'geography',\n",
              " 'estate',\n",
              " 'congresses',\n",
              " 'collectors',\n",
              " 'ca',\n",
              " 'biblical',\n",
              " 'as',\n",
              " 'widows',\n",
              " 'weight',\n",
              " 'weddings',\n",
              " 'videos',\n",
              " 'traffic',\n",
              " 'tibet',\n",
              " 'thrillers',\n",
              " 'textile',\n",
              " 'sleep',\n",
              " 'role',\n",
              " 'prehistoric',\n",
              " 'grief',\n",
              " 'film',\n",
              " 'automobiles',\n",
              " 'venice',\n",
              " 'trees',\n",
              " 'services',\n",
              " 'outdoor',\n",
              " 'ornamental',\n",
              " 'labor',\n",
              " 'expedition',\n",
              " 'ethics',\n",
              " 'equipment',\n",
              " 'choruses',\n",
              " 'wilderness',\n",
              " 'voices',\n",
              " 'veterans',\n",
              " 'triangles',\n",
              " 'system',\n",
              " 'robots',\n",
              " 'play',\n",
              " 'parents',\n",
              " 'korea',\n",
              " 'influence',\n",
              " 'hunting',\n",
              " 'divorced',\n",
              " 'circus',\n",
              " 'building',\n",
              " 'urban',\n",
              " 'spies',\n",
              " 'personnel',\n",
              " 'mind',\n",
              " 'joseph',\n",
              " 'industrial',\n",
              " 'harry',\n",
              " 'garden',\n",
              " 'football',\n",
              " 'fear',\n",
              " 'educational',\n",
              " 'conversation',\n",
              " 'bill',\n",
              " 'bear',\n",
              " 'arctic',\n",
              " 'advertising',\n",
              " 'adoption',\n",
              " 'wildlife',\n",
              " 'visual',\n",
              " 'vegetarian',\n",
              " 'utah',\n",
              " 'themes',\n",
              " 'southwest',\n",
              " 'security',\n",
              " 'problem',\n",
              " 'pregnancy',\n",
              " 'physics',\n",
              " 'owners',\n",
              " 'nev',\n",
              " 'jack',\n",
              " 'flute',\n",
              " 'esteem',\n",
              " 'digital',\n",
              " 'contemporary',\n",
              " 'caribbean',\n",
              " 'bibliography',\n",
              " 'yorkshire',\n",
              " 'style',\n",
              " 'soccer',\n",
              " 'slavery',\n",
              " 'parks',\n",
              " 'painters',\n",
              " 'metal',\n",
              " 'la',\n",
              " 'fishes',\n",
              " 'elephants',\n",
              " 'divorce',\n",
              " 'community',\n",
              " 'characteristics',\n",
              " 'anatomy',\n",
              " 'ability',\n",
              " 'working',\n",
              " 'techniques',\n",
              " 'structures',\n",
              " 'skills',\n",
              " 'rites',\n",
              " 'plays',\n",
              " 'phrase',\n",
              " 'korean',\n",
              " 'household',\n",
              " 'grammar',\n",
              " 'disorder',\n",
              " 'associates',\n",
              " 'artistic',\n",
              " 'adolescence',\n",
              " 'zealand',\n",
              " 'trail',\n",
              " 'theory',\n",
              " 'student',\n",
              " 'snow',\n",
              " 'shipwrecks',\n",
              " 'prisons',\n",
              " 'opinion',\n",
              " 'motives',\n",
              " 'michael',\n",
              " 'media',\n",
              " 'intellectual',\n",
              " 'imagination',\n",
              " 'f',\n",
              " 'electric',\n",
              " 'earth',\n",
              " 'dress',\n",
              " 'david',\n",
              " 'comics',\n",
              " 'wolves',\n",
              " 'witchcraft',\n",
              " 'wild',\n",
              " 'size',\n",
              " 'santa',\n",
              " 'roosevelt',\n",
              " 'reggae',\n",
              " 'reducing',\n",
              " 'ranch',\n",
              " 'puget',\n",
              " 'photographers',\n",
              " 'monkeys',\n",
              " 'inheritance',\n",
              " 'hollywood',\n",
              " 'group',\n",
              " 'frogs',\n",
              " 'flowers',\n",
              " 'ethical',\n",
              " 'deaf',\n",
              " 'computers',\n",
              " 'company',\n",
              " 'bluegrass',\n",
              " 'assassination',\n",
              " 'ancient',\n",
              " 'ages',\n",
              " 'wales',\n",
              " 'vegas',\n",
              " 'tennessee',\n",
              " 'taiwan',\n",
              " 'suicide',\n",
              " 'succession',\n",
              " 'species',\n",
              " 'sciences',\n",
              " 'portuguese',\n",
              " 'occupations',\n",
              " 'nicholas',\n",
              " 'method',\n",
              " 'las',\n",
              " 'knights',\n",
              " 'illness',\n",
              " 'ducks',\n",
              " 'bush',\n",
              " 'alabama',\n",
              " 'wisconsin',\n",
              " 'violence',\n",
              " 'travels',\n",
              " 'supplies',\n",
              " 'stress',\n",
              " 'specimens',\n",
              " 'rich',\n",
              " 'production',\n",
              " 'processing',\n",
              " 'pilots',\n",
              " 'missouri',\n",
              " 'lullabies',\n",
              " 'kennedy',\n",
              " 'job',\n",
              " 'historic',\n",
              " 'habits',\n",
              " 'h',\n",
              " 'ghost',\n",
              " 'game',\n",
              " 'endangered',\n",
              " 'emotions',\n",
              " 'camps',\n",
              " 'book',\n",
              " 'beadwork',\n",
              " 'bass',\n",
              " 'asian',\n",
              " 'argentina',\n",
              " 'airplanes',\n",
              " 'adultery',\n",
              " 'warfare',\n",
              " 'shows',\n",
              " 'r',\n",
              " 'provence',\n",
              " 'practices',\n",
              " 'power',\n",
              " 'peoples',\n",
              " 'peninsula',\n",
              " 'navajo',\n",
              " 'moving',\n",
              " 'libraries',\n",
              " 'knighthood',\n",
              " 'inspector',\n",
              " 'influences',\n",
              " 'hong',\n",
              " 'foods',\n",
              " 'flight',\n",
              " 'economics',\n",
              " 'desert',\n",
              " 'brain',\n",
              " 'baltimore',\n",
              " 'atlantic',\n",
              " 'aeronautics',\n",
              " 'walker',\n",
              " 'organized',\n",
              " 'nevada',\n",
              " 'network',\n",
              " 'maryland',\n",
              " 'lesbians',\n",
              " 'jane',\n",
              " 'inventors',\n",
              " 'herbs',\n",
              " 'healing',\n",
              " 'forecasting',\n",
              " 'cultural',\n",
              " 'cross',\n",
              " 'criminal',\n",
              " 'composers',\n",
              " 'anime',\n",
              " 'aging',\n",
              " 'aerial',\n",
              " 'zen',\n",
              " 'statesmen',\n",
              " 'star',\n",
              " 'quality',\n",
              " 'plains',\n",
              " 'moon',\n",
              " 'meditation',\n",
              " 'martin',\n",
              " 'mammals',\n",
              " 'lost',\n",
              " 'inventions',\n",
              " 'fishing',\n",
              " 'eastern',\n",
              " 'butterflies',\n",
              " 'best',\n",
              " 'answers',\n",
              " 'acting',\n",
              " 'x',\n",
              " 'whales',\n",
              " 'treasure',\n",
              " 'texts',\n",
              " 'snakes',\n",
              " 'safety',\n",
              " 'research',\n",
              " 'quilts',\n",
              " ...]"
            ]
          },
          "metadata": {},
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TrainvectorizedText = vectorizer(train_data[\"subjects\"].values)"
      ],
      "metadata": {
        "id": "J4TnmG9SWKOu"
      },
      "execution_count": 92,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "vectorizer.adapt(test_data[\"subjects\"].values)"
      ],
      "metadata": {
        "id": "VTVC10PzWWFG"
      },
      "execution_count": 93,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "TestvectorizedText = vectorizer(test_data[\"subjects\"].values)"
      ],
      "metadata": {
        "id": "9VpdE3HZWcs9"
      },
      "execution_count": 94,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import scipy"
      ],
      "metadata": {
        "id": "fFpfpwCCWfWi"
      },
      "execution_count": 95,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_data.columns"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3EQLNx1RWix2",
        "outputId": "647e36c5-7521-4d1b-b446-52277bdc3336"
      },
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'usage_class', 'checkout_type', 'checkout_year', 'checkout_month',\n",
              "       'checkouts', 'title', 'subjects', 'publication_year', 'material_type'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 97
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Y_train = pd.get_dummies(train_data.material_type)"
      ],
      "metadata": {
        "id": "GYAeW-TuWkkW"
      },
      "execution_count": 98,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y_test = pd.get_dummies(test_data.material_type)"
      ],
      "metadata": {
        "id": "mLfieZJTWs6e"
      },
      "execution_count": 99,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "hfMJTq__WxoO"
      },
      "execution_count": 100,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#MCSHBAP - map, cache, suffle, batch, prefetch\n",
        "dataset = tf.data.Dataset.from_tensor_slices((TrainvectorizedText, Y_train))\n",
        "dataset = dataset.cache()\n",
        "dataset = dataset.shuffle(160000)\n",
        "dataset = dataset.batch(16)\n",
        "dataset = dataset.prefetch(8) #helps bottleneck"
      ],
      "metadata": {
        "id": "vb9M1pdrW0Om"
      },
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset.as_numpy_iterator().next()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vN098CHbW1ju",
        "outputId": "f9549e93-34e3-4fb2-fdb6-21271ac77657"
      },
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[  26,    2,  331, ...,    0,    0,    0],\n",
              "        [ 202, 2153,  178, ...,    0,    0,    0],\n",
              "        [5063,   82,    3, ...,    0,    0,    0],\n",
              "        ...,\n",
              "        [3164, 1383,    2, ...,    0,    0,    0],\n",
              "        [   0,    0,    0, ...,    0,    0,    0],\n",
              "        [  19,   15, 3038, ...,    0,    0,    0]]),\n",
              " array([[1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 0, 0, 0, 0, 1, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0]], dtype=uint8))"
            ]
          },
          "metadata": {},
          "execution_count": 102
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "batch_x, batch_y = dataset.as_numpy_iterator().next()"
      ],
      "metadata": {
        "id": "ejteifNcW4Hu"
      },
      "execution_count": 103,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "batch_x\n",
        "batch_x.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ws9k-3ZcW7D2",
        "outputId": "86e8f630-8d6a-4caa-d7e9-2c122fc20415"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(16, 1800)"
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "batch_y.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sgP63py3W8Wu",
        "outputId": "ba63c500-fe4a-4650-d8a4-4b44a3fc843b"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(16, 8)"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(train_data)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aBeN3c9pW-pe",
        "outputId": "ce0401bc-9270-4b4f-cdc2-40007bbf0802"
      },
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "31653"
            ]
          },
          "metadata": {},
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train = dataset.take(int(len(dataset)*.7))\n",
        "val = dataset.skip(int(len(dataset)*.7)).take(int(len(dataset)*.2))\n",
        "test = dataset.skip(int(len(dataset)*.9)).take(int(len(dataset)*.1))"
      ],
      "metadata": {
        "id": "lqAZAephXB82"
      },
      "execution_count": 108,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "len(train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "19_YLKe_XFmW",
        "outputId": "3780973d-8fb9-4091-c8d0-7fe5681fe173"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1385"
            ]
          },
          "metadata": {},
          "execution_count": 109
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_generator = train.as_numpy_iterator()"
      ],
      "metadata": {
        "id": "3TURLabvXHem"
      },
      "execution_count": 110,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_generator.next()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1iWMUYORXJL3",
        "outputId": "97f6d2eb-b887-4c47-d4fb-bee0ebeeea77"
      },
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[ 7323,  4821,    18, ...,     0,     0,     0],\n",
              "        [    8,     7,    10, ...,     0,     0,     0],\n",
              "        [   99,     0,     0, ...,     0,     0,     0],\n",
              "        ...,\n",
              "        [ 1820,  1600,   264, ...,     0,     0,     0],\n",
              "        [10517,   473,    45, ...,     0,     0,     0],\n",
              "        [11367,  4158,     0, ...,     0,     0,     0]]),\n",
              " array([[1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 0, 0, 1, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 1, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 0, 0, 0, 0, 0, 1],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 0, 0, 1, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [1, 0, 0, 0, 0, 0, 0, 0],\n",
              "        [0, 0, 0, 0, 0, 0, 1, 0]], dtype=uint8))"
            ]
          },
          "metadata": {},
          "execution_count": 111
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import LSTM, Dropout, Bidirectional, Embedding, Dense"
      ],
      "metadata": {
        "id": "PnuIkxXVXLA2"
      },
      "execution_count": 112,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = Sequential()\n",
        "# Creating the embedding layer\n",
        "model.add(Embedding(MAX_WORDS+1, 32))\n",
        "# Bidirectional LSTM Layer\n",
        "model.add(Bidirectional(LSTM(32, activation='tanh')))\n",
        "# Feature extractor Fully connected layers\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dense(128, activation='relu'))\n",
        "# Final layer\n",
        "model.add(Dense(8, activation='sigmoid'))"
      ],
      "metadata": {
        "id": "7ajjbkGCXOv2"
      },
      "execution_count": 113,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(loss='BinaryCrossentropy', optimizer='Adam', metrics=['accuracy'])"
      ],
      "metadata": {
        "id": "ddYHviG8XSJ-"
      },
      "execution_count": 114,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X9sHafjXXVS-",
        "outputId": "cfdef527-cf8e-4d71-f7a2-e15001a5449b"
      },
      "execution_count": 115,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding (Embedding)       (None, None, 32)          6400032   \n",
            "                                                                 \n",
            " bidirectional (Bidirection  (None, 64)                16640     \n",
            " al)                                                             \n",
            "                                                                 \n",
            " dense (Dense)               (None, 128)               8320      \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 256)               33024     \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               32896     \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 8)                 1032      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 6491944 (24.76 MB)\n",
            "Trainable params: 6491944 (24.76 MB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(train, epochs=34, validation_data = val)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k4l_NRf8XWte",
        "outputId": "9a6029ed-a96b-47a9-b578-f8e993232545"
      },
      "execution_count": 156,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/34\n",
            "1385/1385 [==============================] - 132s 96ms/step - loss: 0.0356 - accuracy: 0.9497 - val_loss: 0.0325 - val_accuracy: 0.9535\n",
            "Epoch 2/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0342 - accuracy: 0.9519 - val_loss: 0.0319 - val_accuracy: 0.9571\n",
            "Epoch 3/34\n",
            "1385/1385 [==============================] - 131s 95ms/step - loss: 0.0341 - accuracy: 0.9520 - val_loss: 0.0331 - val_accuracy: 0.9511\n",
            "Epoch 4/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0331 - accuracy: 0.9509 - val_loss: 0.0305 - val_accuracy: 0.9527\n",
            "Epoch 5/34\n",
            "1385/1385 [==============================] - 133s 96ms/step - loss: 0.0322 - accuracy: 0.9536 - val_loss: 0.0301 - val_accuracy: 0.9546\n",
            "Epoch 6/34\n",
            "1385/1385 [==============================] - 133s 96ms/step - loss: 0.0323 - accuracy: 0.9538 - val_loss: 0.0301 - val_accuracy: 0.9559\n",
            "Epoch 7/34\n",
            "1385/1385 [==============================] - 136s 98ms/step - loss: 0.0308 - accuracy: 0.9552 - val_loss: 0.0301 - val_accuracy: 0.9571\n",
            "Epoch 8/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0323 - accuracy: 0.9540 - val_loss: 0.0294 - val_accuracy: 0.9595\n",
            "Epoch 9/34\n",
            "1385/1385 [==============================] - 131s 94ms/step - loss: 0.0298 - accuracy: 0.9559 - val_loss: 0.0281 - val_accuracy: 0.9589\n",
            "Epoch 10/34\n",
            "1385/1385 [==============================] - 131s 95ms/step - loss: 0.0298 - accuracy: 0.9566 - val_loss: 0.0306 - val_accuracy: 0.9551\n",
            "Epoch 11/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0300 - accuracy: 0.9574 - val_loss: 0.0307 - val_accuracy: 0.9565\n",
            "Epoch 12/34\n",
            "1385/1385 [==============================] - 131s 95ms/step - loss: 0.0295 - accuracy: 0.9576 - val_loss: 0.0287 - val_accuracy: 0.9563\n",
            "Epoch 13/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0289 - accuracy: 0.9591 - val_loss: 0.0267 - val_accuracy: 0.9622\n",
            "Epoch 14/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0285 - accuracy: 0.9585 - val_loss: 0.0248 - val_accuracy: 0.9653\n",
            "Epoch 15/34\n",
            "1385/1385 [==============================] - 131s 95ms/step - loss: 0.0281 - accuracy: 0.9588 - val_loss: 0.0269 - val_accuracy: 0.9611\n",
            "Epoch 16/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0281 - accuracy: 0.9593 - val_loss: 0.0253 - val_accuracy: 0.9646\n",
            "Epoch 17/34\n",
            "1385/1385 [==============================] - 132s 96ms/step - loss: 0.0263 - accuracy: 0.9615 - val_loss: 0.0261 - val_accuracy: 0.9616\n",
            "Epoch 18/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0263 - accuracy: 0.9623 - val_loss: 0.0267 - val_accuracy: 0.9619\n",
            "Epoch 19/34\n",
            "1385/1385 [==============================] - 128s 93ms/step - loss: 0.0265 - accuracy: 0.9608 - val_loss: 0.0247 - val_accuracy: 0.9611\n",
            "Epoch 20/34\n",
            "1385/1385 [==============================] - 128s 92ms/step - loss: 0.0272 - accuracy: 0.9603 - val_loss: 0.0244 - val_accuracy: 0.9668\n",
            "Epoch 21/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0251 - accuracy: 0.9634 - val_loss: 0.0242 - val_accuracy: 0.9623\n",
            "Epoch 22/34\n",
            "1385/1385 [==============================] - 128s 93ms/step - loss: 0.0256 - accuracy: 0.9624 - val_loss: 0.0254 - val_accuracy: 0.9625\n",
            "Epoch 23/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0264 - accuracy: 0.9618 - val_loss: 0.0237 - val_accuracy: 0.9650\n",
            "Epoch 24/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0253 - accuracy: 0.9640 - val_loss: 0.0234 - val_accuracy: 0.9658\n",
            "Epoch 25/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0252 - accuracy: 0.9631 - val_loss: 0.0261 - val_accuracy: 0.9647\n",
            "Epoch 26/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0258 - accuracy: 0.9631 - val_loss: 0.0241 - val_accuracy: 0.9634\n",
            "Epoch 27/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0243 - accuracy: 0.9633 - val_loss: 0.0244 - val_accuracy: 0.9634\n",
            "Epoch 28/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0249 - accuracy: 0.9642 - val_loss: 0.0232 - val_accuracy: 0.9655\n",
            "Epoch 29/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0245 - accuracy: 0.9634 - val_loss: 0.0242 - val_accuracy: 0.9641\n",
            "Epoch 30/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0244 - accuracy: 0.9642 - val_loss: 0.0247 - val_accuracy: 0.9642\n",
            "Epoch 31/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0237 - accuracy: 0.9657 - val_loss: 0.0240 - val_accuracy: 0.9652\n",
            "Epoch 32/34\n",
            "1385/1385 [==============================] - 134s 97ms/step - loss: 0.0241 - accuracy: 0.9646 - val_loss: 0.0239 - val_accuracy: 0.9641\n",
            "Epoch 33/34\n",
            "1385/1385 [==============================] - 129s 93ms/step - loss: 0.0244 - accuracy: 0.9638 - val_loss: 0.0219 - val_accuracy: 0.9676\n",
            "Epoch 34/34\n",
            "1385/1385 [==============================] - 130s 94ms/step - loss: 0.0238 - accuracy: 0.9640 - val_loss: 0.0265 - val_accuracy: 0.9595\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from matplotlib import pyplot as plt\n",
        "plt.figure(figsize=(15,12))\n",
        "pd.DataFrame(history.history).plot()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "id": "SPaaqqZmXbT-",
        "outputId": "209e8fc5-6ad0-450e-e75e-a0a09430b085"
      },
      "execution_count": 157,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x1200 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGdCAYAAADAAnMpAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABHeklEQVR4nO3deZgU5b328W9V9To768yAI4sCEgVEEMQ9whHxyAE1eQlyBDVoTMCoxERIEExMxKgoxiUcTdRjBDUa8RgxRCViEsQNnKgRCSCbAYZFnX2mu6ue94/uaWjZZoChGLg/XM9V1bV0/bqmmLr7qeoeyxhjEBEREfGJ7XcBIiIicnRTGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHwV8LuAxvA8j40bN5Kbm4tlWX6XIyIiIo1gjKGyspIOHTpg23vu/2gRYWTjxo2UlJT4XYaIiIjshw0bNnDMMcfscX6Tw8hf//pX7rrrLpYuXcqmTZuYN28eI0eO3Os6ixYtYtKkSfzzn/+kpKSEqVOncsUVVzR6m7m5uUDyxeTl5TW1ZBEREfFBRUUFJSUl6fP4njQ5jFRXV9OnTx+uuuoqLrnkkn0uv2bNGv7zP/+Ta6+9ljlz5rBw4ULGjx9PcXExQ4cObdQ2Gy7N5OXlKYyIiIi0MPu6xaLJYWTYsGEMGzas0cvPnj2bLl26MHPmTAB69uzJ3//+d+69995GhxERERE5cjX7p2mWLFnCkCFDMqYNHTqUJUuW7HGd+vp6KioqMpqIiIgcmZo9jGzevJnCwsKMaYWFhVRUVFBbW7vbdWbMmEF+fn666eZVERGRI9dh+T0jU6ZMoby8PN02bNjgd0kiIiLSTJr9o71FRUWUlZVlTCsrKyMvL49oNLrbdcLhMOFwuLlLExERkcNAs/eMDBo0iIULF2ZMe/XVVxk0aFBzb1pERERagCaHkaqqKkpLSyktLQWSH90tLS1l/fr1QPISy9ixY9PLX3vttXz66af86Ec/4pNPPuGhhx7i97//PTfeeOPBeQUiIiLSojU5jLz33nv07duXvn37AjBp0iT69u3LtGnTANi0aVM6mAB06dKF+fPn8+qrr9KnTx9mzpzJb37zG32sV0RERACwjDHG7yL2paKigvz8fMrLy/WlZyIiIi1EY8/fh+WnaUREROTooTAiIiIivlIYEREREV81+/eMiIgczowxuNu2EVu3DhyHYIcOBNq2xXIcv0sT2SMvFsPdto3E9u0ktm0jsW0b7vbtuJWVhLt2JdKrF+Hjjmsxx7HCiMhhzCQSuF98QeLzL3C/+Bz388+T459/TuKLz3FT41Y4TLC4iEBhEcGiQgJFxemhk5O9f9v2PNzy8p1+4W3H3b4Nr7YOp6AAp1UrnFYFBFq1So4XFGAFDt9fKW5lJbG164itXbtL86qrMxcOBAi2b0+guJhgcXFy3xYXEywqJtghOc3Oy8PyXKivAMsCJwxOCJym7wNjDF51NV55OW5lJW55BW5FOV5FReZ4RSWmvg4Ti2MSiR0tHsck4hBPPY7VY+IxTDwOiQTGdbECDlYwiBUKJofhMHYoiBUKY4XDWKEQVjiCHQ5jhSNY4ShWJIIVDu00LYQdjiTXjTSst9N4OIwdDn9lHxgwZi9DkkPPxatKvk6vshy3shyvshKvqhKvqgqvuhq3uhqvuibZamrxauowiQRWKIQdCaWG4R31RSLJeqNR7EgEK5qFHY1iRbOSf0XWS2DcBHgueKmhm0hO91xwd5ruJcB1MbFaTH09pr42uZ9jsfTQi+3Y517DzyLhgmewAjZWwMEOOsmfRSiAFQxgBYPYwUBq/6d+HqEQViiMFwO3qp5EZbK5FbUkKmpIlFfj1dTv87iyImGi3ToR6d6ZaPdORLsfS6BtAVb65+Lt1Ax0Oh2yWjf5+D0YDt/fHCJHOGMMia1bia1ZS2zNGmJr1hDf+O902HA//xy3vPyAt2Pn5BAoKiRYVJwcFhYRLC7CadMGr6o69Y5qG4lt25OhY/s23G3bSXz+OSQSTdtWbi5OQR6BvByc3ChOThgnK4iTFcC2veQvfjf5C9+kf/G7GHenX/aelzopJMetQAArGEr/giYYwgqGsUKR1AkyghWKQjiaOoFm48U9Yv8uI/bZJmIbNhJb/2/cL77cc+GWRbBDBzAe8bIySCSIb9xIfONGdv8XtMAKGIJZCYJRD2wD6c8lWoCNsez0OFbDtNRjLIwHbp3Bq3Nx69yd1pcjj0XyB5xItYPENgTCHoGISyDq4YRd7IChvjxI3edBvLp6aj78FzUf/iu9ihN2ibaJE2kdI9o6TqRNnEDYS84cv1BhRI4MxvOS7/AqK3ErK5Pv5iorcSsq8CqrcCsr8Coqk8PKquQ7Nif5TgEnkDzxBBxwHCwnsGO642AFA8npgSBO61YE2rVLvntt1w6ndWv/uiN3fnfX8G7PeJCog3gtXuXnxD79lNiatdSvW09sw7+JfbaZ2MateLWxfT+/BU52BCcvSiA3CycvCycvm0B+Dk5eDk5+LiaeIL7tSxLbyol/Xk5iewXx7RV4NfV4VVXEVlURW7V6v16eHQ0SyA0lW04IKwBuVS1uVT1uTRy31sOtM4CVfCdbWUl8v7bU/JyISzg3QSjdXEK5CYI5CezAJjBe6kdnE69xSNQ4xBta9Y7Hbr2DSVjEKoLEGvVHxQ3g7nMpyzbYIQ8n5OEEdxoPGexgctwOGCzbpPKNwbKT61k2WNaOxwRDWJFsrGgOViiKcT1MPIEXi2PiCUwskRy6XvJdfDyBcZMhybgWnmthPAvj7vR4p6HxLLzUvIzprsV+fWGEBXbA4AQ87BDYQbBDFnbIxg7bOGEbOxzAjgSwI0GcSAg7GsRynORriiV2DOMuJuYmh3EPL+4lhwmDlzCYhAFjJUNiRrOTPSYNzU5Ow7KwGoap3oxkCyV7mkLhnXqYGkJyFlYkihWJgu0ke1PqapM9KvX1ePV1yR6V+lhyXjyGicVTP5/k67BDNk52gEBOgECWTSDLwsmyCUQNgajBDrhYJgFuLBXs4+DGwbIwxiJWblG7xVC7xaOuzKNum4tb71C10aFqYyS964P5DtHiEG3WlhE5Zj9+dgeBwshXGGPwKiuT1+C2biOxbSuJrVuTXdVbt6WvzdmRCKGuXQl17UK4a1dCXboSKjkGKxj0+yVkMK6bOvDrU12LXxmvq0/9h0iN19dj6uvw6mOYuuR/lozxutT82prktPo6vLo6TF0dbnUNXlUN+/eb6ADZFoGcMIH8yI4TZ26AQHaAQI5DIMfBCZF87bEYXn0cUx9P/hKOJfBiqWHDL62GX2IJA17q9Vipd79W8rHV8FbWImNokdwF8aoA9ZUOiZq9/DezDMHs5AkxnJc8KQbCHk7YIxBJDp2Qh7WnW83rUg2Sb7jbp1qKF7eI1zokauz0STVRmxrW2ThBQyDi4kSS2wtE3OR2G8bDHlYjMp7xwI3buPU2bszBdXNIeFHceAQ3HkievI0Njo1lOWDb4CSHluOA7WDZDdMDyXHHActOXX6IpX5ZxzCJ1Mk0dYmCVFe4SbjJE67rYVkeoTwvHTZCOXFCOTGcwF7elZrku0MrECRY2JZgVhvIagPZbSGrbWq8DWS1xXNySVRDvCJO4svaZC+PSWB5HpgEqTM4eKmTRfpx8hIAJoFlGZysEE5WEDsaxIkEsAPsWMbbaXkvnuotikMgDJE8CKdaejw3NZ6fHA+E9v2D+6qGbbix5NDbz3fx1k7/KSxrpyG7Pt55GScETnCnebK/LCCcagWpaV59PfWffELtBx9S++EH1H3wIbG1a4mXu8TLa2mdf6xv9R7VYeTz3z1J/epVyW7qnYKGqd/3tTiA2n/8I3NCIECopIRQl86Eu3RODjt3ItSl847r9nv4z2Y8D1Nbi1dXh1dbmxyvrcWrTZ74verqZKtJjVdV4VV+iVfxJV5lBV511Y75dfV4tTG8uAuuP32/6Xd4QYMdAidsYUccnIid/KUbDWJHQ1iWB/FUIIrXJ086rpfqZEi9SzMWpC5pGpN855Wos5Ot1sGts8GDREUdiYq6fdbWiOo52P81nLAhVGARahUg1CZMqF0W4fa5BNvnYUezIRiFYFby2PC8nU5CXz0pJdInufTj1Duh5Ds4JzW0wXawLZtwqjVMy1jOtnc83nmevfMyTubyTmink2A+RPKwwnkEInkEwnkQyj58TyYNvVa77NvUMJSVfG37qN8GQql2RLFTx0Ewsu9lpcWxw2GiffoQ7dMnPc2tqKDuo4+o/eBDwj16+FbbUf0NrGsvHUntP1fsdp4dCRDICSbfVUchEEnghOMEgnUEnGq8mEd9RYBYZYBYRYD6igDG3fMnpZ1I8h0agElYeK6T7NZMpLo3D+JlxD2yDLZjsFIteY5JjdsGK9X9m1yG1DImY5hcdqf1goHkTW1BBzsQw7HrsJ167INxxSQQhXAOhFItnJM80YWykydEOwhOAGMcErWGRFWcREWcRFWMREUsGU7Ka0lU1JL4shqvNpbqUg1ih0Opm/F2vuEtghUJY0eSN+7Z0SysaDT5Lp1kr1lyhB1DY5IPTcPlmdQMyyZ4TAmh47oR6tqVQKtWB2GHiIi0LI09fx/VPSN5hf8mm8pkN3V0Rxd1IOImu0ubwBhI1NjUVwaJVewIKLHKQOrdu0NtXePO0JaTui7sGOxAMgjYwdR4MDkvPR4OYWdlYWdnY+fkYOfmY+cVYOe1ws7OxQoGsBvu2g4ESHaJ2jveSe/ymOS79EAk+U49GN3RAtEd7+CDkeRjezcBzPPArU/eM5HYyzBeC3ZgD4Ejh8YmGgsIppqIiLQ8R3UYaX3OCbA5vo/rrw3j+TvNz02erIGG650WELQsgjtf/wSwLNyqamLr1hPb8BmWZbBDDlbQTvUmgBW0sIMWdsDDsj0sLwaJ2I6TtvEg2mo3rSDZrX+4sW2wU8FFRERkH47qyzQiIiLSfPSH8kRERKRFUBgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+Gq/wsiDDz5I586diUQiDBw4kHfeeWevy8+aNYsePXoQjUYpKSnhxhtvpK6ubr8KFhERkSNLk8PIM888w6RJk5g+fTrLli2jT58+DB06lC1btux2+blz5zJ58mSmT5/O8uXL+e1vf8szzzzDj3/84wMuXkRERFq+JoeRe+65h6uvvporr7ySr33ta8yePZusrCweffTR3S7/5ptvcsYZZ3DZZZfRuXNnzj//fEaPHr3P3hQRERE5OjQpjMRiMZYuXcqQIUN2PIFtM2TIEJYsWbLbdU4//XSWLl2aDh+ffvopL7/8MhdeeOEet1NfX09FRUVGExERkSNToCkLb9u2Ddd1KSwszJheWFjIJ598stt1LrvsMrZt28aZZ56JMYZEIsG1116718s0M2bM4Kc//WlTShMREZEWqtk/TbNo0SJuv/12HnroIZYtW8bzzz/P/Pnzue222/a4zpQpUygvL0+3DRs2NHeZIiIi4pMm9Yy0bdsWx3EoKyvLmF5WVkZRUdFu17nlllu4/PLLGT9+PAC9evWiurqaa665hp/85CfY9q55KBwOEw6Hm1KaiIiItFBN6hkJhUL069ePhQsXpqd5nsfChQsZNGjQbtepqanZJXA4jgOAMaap9YqIiMgRpkk9IwCTJk1i3Lhx9O/fnwEDBjBr1iyqq6u58sorARg7diwdO3ZkxowZAAwfPpx77rmHvn37MnDgQFatWsUtt9zC8OHD06FEREREjl5NDiOjRo1i69atTJs2jc2bN3PyySezYMGC9E2t69evz+gJmTp1KpZlMXXqVP7973/Trl07hg8fzi9+8YuD9ypERESkxbJMC7hWUlFRQX5+PuXl5eTl5fldjoiIiDRCY8/f+ts0IiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcBvwsQEZHDk+u6xONxv8uQw1gwGMRxnAN+HoURERHJYIxh8+bNfPnll36XIi1AQUEBRUVFWJa138+hMCIiIhkagkj79u3Jyso6oJOMHLmMMdTU1LBlyxYAiouL9/u5FEZERCTNdd10EGnTpo3f5chhLhqNArBlyxbat2+/35ds9usG1gcffJDOnTsTiUQYOHAg77zzzl6X//LLL5kwYQLFxcWEw2G6d+/Oyy+/vF8Fi4hI82m4RyQrK8vnSqSlaDhWDuT+oib3jDzzzDNMmjSJ2bNnM3DgQGbNmsXQoUNZsWIF7du332X5WCzGf/zHf9C+fXuee+45OnbsyLp16ygoKNjvokVEpHnp0ow01sE4VpocRu655x6uvvpqrrzySgBmz57N/PnzefTRR5k8efIuyz/66KN8/vnnvPnmmwSDQQA6d+58YFWLiIjIEaNJl2lisRhLly5lyJAhO57AthkyZAhLlizZ7TovvvgigwYNYsKECRQWFnLSSSdx++2347ruHrdTX19PRUVFRhMREdmbc889lxtuuMHvMmQ/NCmMbNu2Ddd1KSwszJheWFjI5s2bd7vOp59+ynPPPYfrurz88svccsstzJw5k5///Od73M6MGTPIz89Pt5KSkqaUKSIiIi1Is38Dq+d5tG/fnocffph+/foxatQofvKTnzB79uw9rjNlyhTKy8vTbcOGDc1dpoiIiPikSWGkbdu2OI5DWVlZxvSysjKKiop2u05xcTHdu3fP+LhPz5492bx5M7FYbLfrhMNh8vLyMpqIiEhjffHFF4wdO5ZWrVqRlZXFsGHDWLlyZXr+unXrGD58OK1atSI7O5sTTzwx/SnPL774gjFjxtCuXTui0SjdunXjscce8+ulHBWadANrKBSiX79+LFy4kJEjRwLJno+FCxcyceLE3a5zxhlnMHfuXDzPw7aT2edf//oXxcXFhEKhA6teRESanTGG2vie7/NrLtGgs9+f1LjiiitYuXIlL774Inl5edx8881ceOGFfPzxxwSDQSZMmEAsFuOvf/0r2dnZfPzxx+Tk5ABwyy238PHHH/OnP/2Jtm3bsmrVKmpraw/mS5OvaPKnaSZNmsS4cePo378/AwYMYNasWVRXV6c/XTN27Fg6duzIjBkzAPjud7/LAw88wPXXX891113HypUruf322/n+979/cF+JiIg0i9q4y9em/fmQb/fjnw0lK9T07+ZsCCGLFy/m9NNPB2DOnDmUlJTwwgsv8M1vfpP169dz6aWX0qtXLwC6du2aXn/9+vX07duX/v37A/oE6KHQ5J/yqFGj2Lp1K9OmTWPz5s2cfPLJLFiwIH1T6/r169M9IAAlJSX8+c9/5sYbb6R379507NiR66+/nptvvvngvQoREZGU5cuXEwgEGDhwYHpamzZt6NGjB8uXLwfg+9//Pt/97nd55ZVXGDJkCJdeeim9e/cGkm+iL730UpYtW8b555/PyJEj06FGmsd+fR38xIkT93hZZtGiRbtMGzRoEG+99db+bEpERHwWDTp8/LOhvmy3uYwfP56hQ4cyf/58XnnlFWbMmMHMmTO57rrrGDZsGOvWrePll1/m1VdfZfDgwUyYMIG777672eo52jX7p2lERKRlsyyLrFDgkLf9vV+kZ8+eJBIJ3n777fS07du3s2LFCr72ta+lp5WUlHDttdfy/PPP84Mf/IBHHnkkPa9du3aMGzeOJ598klmzZvHwww/v/w6UfdIfyhMRkSNKt27dGDFiBFdffTX/8z//Q25uLpMnT6Zjx46MGDECgBtuuIFhw4bRvXt3vvjiC15//XV69uwJwLRp0+jXrx8nnngi9fX1vPTSS+l50jzUMyIiIkecxx57jH79+nHRRRcxaNAgjDG8/PLL6T9L4rouEyZMoGfPnlxwwQV0796dhx56CEh+cnTKlCn07t2bs88+G8dxePrpp/18OUc8yxhj/C5iXyoqKsjPz6e8vFzfOSIi0ozq6upYs2YNXbp0IRKJ+F2OtAB7O2Yae/5Wz4iIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERFpBvF43O8SWgyFEREROSIsWLCAM888k4KCAtq0acNFF13E6tWr0/M/++wzRo8eTevWrcnOzqZ///68/fbb6fl//OMfOfXUU4lEIrRt25aLL744Pc+yLF544YWM7RUUFPD4448DsHbtWizL4plnnuGcc84hEokwZ84ctm/fzujRo+nYsSNZWVn06tWLp556KuN5PM/jzjvv5PjjjyccDnPsscfyi1/8AoDzzjuPiRMnZiy/detWQqEQCxcuPBi77bAQ8LsAERE5zBkD8ZpDv91gFlhWoxevrq5m0qRJ9O7dm6qqKqZNm8bFF19MaWkpNTU1nHPOOXTs2JEXX3yRoqIili1bhud5AMyfP5+LL76Yn/zkJzzxxBPEYjFefvnlJpc8efJkZs6cSd++fYlEItTV1dGvXz9uvvlm8vLymD9/PpdffjnHHXccAwYMAGDKlCk88sgj3HvvvZx55pls2rSJTz75BIDx48czceJEZs6cSTgcBuDJJ5+kY8eOnHfeeU2u73BlGWOM30XsS2P/BLGIiByY3f45+Fg13N7h0Bfz440Qyt7v1bdt20a7du348MMPefPNN7nppptYu3YtrVu33mXZ008/na5du/Lkk0/u9rksy2LevHmMHDkyPa2goIBZs2ZxxRVXsHbtWrp06cKsWbO4/vrr91rXRRddxAknnMDdd99NZWUl7dq144EHHmD8+PG7LFtXV0eHDh2YPXs2/+///T8A+vTpwyWXXML06dObsDeaz26PmZTGnr91mUZERI4IK1euZPTo0XTt2pW8vDw6d+4MwPr16yktLaVv3767DSIApaWlDB48+IBr6N+/f8Zj13W57bbb6NWrF61btyYnJ4c///nPrF+/HoDly5dTX1+/x21HIhEuv/xyHn30UQCWLVvGRx99xBVXXHHAtR5OdJlGRET2LpiV7KXwY7tNMHz4cDp16sQjjzxChw4d8DyPk046iVgsRjQa3eu6+5pvWRZfvZCwuxtUs7Mze3Luuusu7rvvPmbNmkWvXr3Izs7mhhtuIBaLNWq7kLxUc/LJJ/PZZ5/x2GOPcd5559GpU6d9rteSqGdERET2zrKSl0sOdWvC/SLbt29nxYoVTJ06lcGDB9OzZ0+++OKL9PzevXtTWlrK559/vtv1e/fuvdcbQtu1a8emTZvSj1euXElNzb7vo1m8eDEjRozgv//7v+nTpw9du3blX//6V3p+t27diEaje912r1696N+/P4888ghz587lqquu2ud2WxqFERERafFatWpFmzZtePjhh1m1ahV/+ctfmDRpUnr+6NGjKSoqYuTIkSxevJhPP/2UP/zhDyxZsgSA6dOn89RTTzF9+nSWL1/Ohx9+yC9/+cv0+ueddx4PPPAA77//Pu+99x7XXnstwWBwn3V169aNV199lTfffJPly5fzne98h7KysvT8SCTCzTffzI9+9COeeOIJVq9ezVtvvcVvf/vbjOcZP348d9xxB8aYjE/5HCkURkREpMWzbZunn36apUuXctJJJ3HjjTdy1113peeHQiFeeeUV2rdvz4UXXkivXr244447cBwHgHPPPZdnn32WF198kZNPPpnzzjuPd955J73+zJkzKSkp4ayzzuKyyy7jpptuIitr35eRpk6dyimnnMLQoUM599xz04FoZ7fccgs/+MEPmDZtGj179mTUqFFs2bIlY5nRo0cTCAQYPXr0LjeJHgn0aRoREUnb2ycjxD9r167luOOO49133+WUU07xu5wMB+PTNLqBVURE5DAVj8fZvn07U6dO5bTTTjvsgsjBoss0IiIih6nFixdTXFzMu+++y+zZs/0up9moZ0REROQwde655+7ykeIjkXpGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgREREBOnfuzKxZsxq1rGVZvPDCC81az9FEYURERER8pTAiIiIivlIYERGRFu/hhx+mQ4cOeJ6XMX3EiBFcddVVrF69mhEjRlBYWEhOTg6nnnoqr7322kHb/ocffsh5551HNBqlTZs2XHPNNVRVVaXnL1q0iAEDBpCdnU1BQQFnnHEG69atA+Af//gHX//618nNzSUvL49+/frx3nvvHbTaWgKFERER2StjDDXxmkPemvI16N/85jfZvn07r7/+enra559/zoIFCxgzZgxVVVVceOGFLFy4kPfff58LLriA4cOHs379+gPeP9XV1QwdOpRWrVrx7rvv8uyzz/Laa68xceJEABKJBCNHjuScc87hgw8+YMmSJVxzzTVYlgXAmDFjOOaYY3j33XdZunQpkydPJhgMHnBdLYn+No2IiOxVbaKWgXMHHvLtvn3Z22QFsxq1bKtWrRg2bBhz585l8ODBADz33HO0bduWr3/969i2TZ8+fdLL33bbbcybN48XX3wxHRr219y5c6mrq+OJJ54gOzsbgAceeIDhw4fzy1/+kmAwSHl5ORdddBHHHXccAD179kyvv379en74wx9ywgknANCtW7cDqqclUs+IiIgcEcaMGcMf/vAH6uvrAZgzZw7f+ta3sG2bqqoqbrrpJnr27ElBQQE5OTksX778oPSMLF++nD59+qSDCMAZZ5yB53msWLGC1q1bc8UVVzB06FCGDx/Offfdx6ZNm9LLTpo0ifHjxzNkyBDuuOMOVq9efcA1tTTqGRERkb2KBqK8fdnbvmy3KYYPH44xhvnz53Pqqafyt7/9jXvvvReAm266iVdffZW7776b448/nmg0yje+8Q1isVhzlL6Lxx57jO9///ssWLCAZ555hqlTp/Lqq69y2mmnceutt3LZZZcxf/58/vSnPzF9+nSefvppLr744kNS2+FAYURERPbKsqxGXy7xUyQS4ZJLLmHOnDmsWrWKHj16cMoppwCwePFirrjiivQJvqqqirVr1x6U7fbs2ZPHH3+c6urqdO/I4sWLsW2bHj16pJfr27cvffv2ZcqUKQwaNIi5c+dy2mmnAdC9e3e6d+/OjTfeyOjRo3nssceOqjCiyzQiInLEGDNmDPPnz+fRRx9lzJgx6endunXj+eefp7S0lH/84x9cdtllu3zy5kC2GYlEGDduHB999BGvv/461113HZdffjmFhYWsWbOGKVOmsGTJEtatW8crr7zCypUr6dmzJ7W1tUycOJFFixaxbt06Fi9ezLvvvptxT8nRQD0jIiJyxDjvvPNo3bo1K1as4LLLLktPv+eee7jqqqs4/fTTadu2LTfffDMVFRUHZZtZWVn8+c9/5vrrr+fUU08lKyuLSy+9lHvuuSc9/5NPPuF///d/2b59O8XFxUyYMIHvfOc7JBIJtm/fztixYykrK6Nt27Zccskl/PSnPz0otbUUlmnKZ6d8UlFRQX5+PuXl5eTl5fldjojIEauuro41a9bQpUsXIpGI3+VIC7C3Y6ax529dphERERFfKYyIiIjsZM6cOeTk5Oy2nXjiiX6Xd0TSPSMiIiI7+a//+i8GDtz9l7wdbd+MeqgojIiIiOwkNzeX3Nxcv8s4qugyjYiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIiQOfOnZk1a5bfZRyVFEZERETEVwojIiIiLZzrunie53cZ+01hRERE9soYg1dTc8hbU/6o/MMPP0yHDh12OSGPGDGCq666itWrVzNixAgKCwvJycnh1FNP5bXXXtvvfXLPPffQq1cvsrOzKSkp4Xvf+x5VVVUZyyxevJhzzz2XrKwsWrVqxdChQ/niiy8A8DyPO++8k+OPP55wOMyxxx7LL37xCwAWLVqEZVl8+eWX6ecqLS3FsizWrl0LwOOPP05BQQEvvvgiX/va1wiHw6xfv553332X//iP/6Bt27bk5+dzzjnnsGzZsoy6vvzyS77zne9QWFhIJBLhpJNO4qWXXqK6upq8vDyee+65jOVfeOEFsrOzqays3O/9tS/79XXwDz74IHfddRebN2+mT58+3H///QwYMGCf6z399NOMHj2aESNG8MILL+zPpkVE5BAztbWsOKXfId9uj2VLsbKyGrXsN7/5Ta677jpef/11Bg8eDMDnn3/OggULePnll6mqquLCCy/kF7/4BeFwmCeeeILhw4ezYsUKjj322CbXZts2v/rVr+jSpQuffvop3/ve9/jRj37EQw89BCTDw+DBg7nqqqu47777CAQCvP7667iuC8CUKVN45JFHuPfeeznzzDPZtGkTn3zySZNqqKmp4Ze//CW/+c1vaNOmDe3bt+fTTz9l3Lhx3H///RhjmDlzJhdeeCErV64kNzcXz/MYNmwYlZWVPPnkkxx33HF8/PHHOI5DdnY23/rWt3jsscf4xje+kd5Ow+Pm/Ir8JoeRZ555hkmTJjF79mwGDhzIrFmzGDp0KCtWrKB9+/Z7XG/t2rXcdNNNnHXWWQdUsIiIyFe1atWKYcOGMXfu3HQYee6552jbti1f//rXsW2bPn36pJe/7bbbmDdvHi+++CITJ05s8vZuuOGG9Hjnzp35+c9/zrXXXpsOI3feeSf9+/dPPwbSf/G3srKS++67jwceeIBx48YBcNxxx3HmmWc2qYZ4PM5DDz2U8brOO++8jGUefvhhCgoKeOONN7jooot47bXXeOedd1i+fDndu3cHoGvXrunlx48fz+mnn86mTZsoLi5my5YtvPzyywfUi9QYTQ4j99xzD1dffTVXXnklALNnz2b+/Pk8+uijTJ48ebfruK7LmDFj+OlPf8rf/va3jK4nERE5vFnRKD2WLfVlu00xZswYrr76ah566CHC4TBz5szhW9/6FrZtU1VVxa233sr8+fPZtGkTiUSC2tpa1q9fv1+1vfbaa8yYMYNPPvmEiooKEokEdXV11NTUkJWVRWlpKd/85jd3u+7y5cupr69Ph6b9FQqF6N27d8a0srIypk6dyqJFi9iyZQuu61JTU5N+naWlpRxzzDHpIPJVAwYM4MQTT+R///d/mTx5Mk8++SSdOnXi7LPPPqBa96VJ94zEYjGWLl3KkCFDdjyBbTNkyBCWLFmyx/V+9rOf0b59e7797W83ajv19fVUVFRkNBER8YdlWdhZWYe8WZbVpDqHDx+OMYb58+ezYcMG/va3vzFmzBgAbrrpJubNm8ftt9/O3/72N0pLS+nVqxexWKzJ+2Pt2rVcdNFF9O7dmz/84Q8sXbqUBx98ECD9fNG9BKm9zYPkeRXIuGcmHo/v9nm+uo/GjRtHaWkp9913H2+++SalpaW0adOmUXU1GD9+PI8//jiQvERz5ZVXNvln0VRNCiPbtm3DdV0KCwszphcWFrJ58+bdrvP3v/+d3/72tzzyyCON3s6MGTPIz89Pt5KSkqaUKSIiR6FIJMIll1zCnDlzeOqpp+jRowennHIKkLyZ9IorruDiiy+mV69eFBUVpW8GbaqlS5fieR4zZ87ktNNOo3v37mzcuDFjmd69e7Nw4cLdrt+tWzei0ege57dr1w6ATZs2paeVlpY2qrbFixfz/e9/nwsvvJATTzyRcDjMtm3bMur67LPP+Ne//rXH5/jv//5v1q1bx69+9Ss+/vjj9KWk5tSsn6aprKzk8ssv55FHHqFt27aNXm/KlCmUl5en24YNG5qxShEROVKMGTMmfetAQ68IJAPA888/T2lpKf/4xz+47LLL9vujsMcffzzxeJz777+fTz/9lN/97nfMnj07Y5kpU6bw7rvv8r3vfY8PPviATz75hF//+tds27aNSCTCzTffzI9+9COeeOIJVq9ezVtvvcVvf/vb9POXlJRw6623snLlSubPn8/MmTMbVVu3bt343e9+x/Lly3n77bcZM2ZMRm/IOeecw9lnn82ll17Kq6++ypo1a/jTn/7EggUL0su0atWKSy65hB/+8Iecf/75HHPMMfu1n5rENEF9fb1xHMfMmzcvY/rYsWPNf/3Xf+2y/Pvvv28A4zhOulmWZSzLMo7jmFWrVjVqu+Xl5QYw5eXlTSlXRESaqLa21nz88cemtrbW71L2i+u6pri42ABm9erV6elr1qwxX//61000GjUlJSXmgQceMOecc465/vrr08t06tTJ3HvvvY3azj333GOKi4tNNBo1Q4cONU888YQBzBdffJFeZtGiReb000834XDYFBQUmKFDh6bnu65rfv7zn5tOnTqZYDBojj32WHP77ben1/373/9uevXqZSKRiDnrrLPMs88+awCzZs0aY4wxjz32mMnPz9+lrmXLlpn+/fubSCRiunXrZp599tldXtf27dvNlVdeadq0aWMikYg56aSTzEsvvZTxPAsXLjSA+f3vf7/PfbG3Y6ax52/LmCZ8kBsYOHAgAwYM4P777weSn5U+9thjmThx4i43sNbV1bFq1aqMaVOnTk3fSdy9e3dCodA+t1lRUUF+fj7l5eXk5eU1pVwREWmCuro61qxZQ5cuXYhEIn6XIz753e9+x4033sjGjRv3eZ7e2zHT2PN3kz9NM2nSJMaNG0f//v0ZMGAAs2bNorq6Ov3pmrFjx9KxY0dmzJiR/jKVnRUUFADsMl1ERET8VVNTw6ZNm7jjjjv4zne+06gOg4OhyfeMjBo1irvvvptp06Zx8sknU1payoIFC9I3ta5fvz7jphsREZGWZM6cOeTk5Oy2NXxXyJHqzjvv5IQTTqCoqIgpU6Ycsu02+TKNH3SZRkTk0NBlmuSHL8rKynY7LxgM0qlTp0Nc0eHNl8s0IiIiR7Lc3Nxm/epz2ZX+UJ6IiOyiBXSay2HiYBwrCiMiIpIWDAaB5I2MIo3RcKw0HDv7Q5dpREQkzXEcCgoK2LJlCwBZ+/G17HJ0MMZQU1PDli1bKCgowHGc/X4uhREREclQVFQEkA4kIntTUFCQPmb2l8KIiIhksCyL4uJi2rdvv9s/0CbSIBgMHlCPSAOFERER2S3HcQ7KiUZkX3QDq4iIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIiv9iuMPPjgg3Tu3JlIJMLAgQN555139rjsI488wllnnUWrVq1o1aoVQ4YM2evyIiIicnRpchh55plnmDRpEtOnT2fZsmX06dOHoUOHsmXLlt0uv2jRIkaPHs3rr7/OkiVLKCkp4fzzz+ff//73ARcvIiIiLZ9ljDFNWWHgwIGceuqpPPDAAwB4nkdJSQnXXXcdkydP3uf6ruvSqlUrHnjgAcaOHduobVZUVJCfn095eTl5eXlNKVdERER80tjzd5N6RmKxGEuXLmXIkCE7nsC2GTJkCEuWLGnUc9TU1BCPx2nduvUel6mvr6eioiKjiYiIyJGpSWFk27ZtuK5LYWFhxvTCwkI2b97cqOe4+eab6dChQ0ag+aoZM2aQn5+fbiUlJU0pU0RERFqQQ/ppmjvuuIOnn36aefPmEYlE9rjclClTKC8vT7cNGzYcwipFRETkUAo0ZeG2bdviOA5lZWUZ08vKyigqKtrrunfffTd33HEHr732Gr17997rsuFwmHA43JTSREREpIVqUs9IKBSiX79+LFy4MD3N8zwWLlzIoEGD9rjenXfeyW233caCBQvo37///lcrIiIiR5wm9YwATJo0iXHjxtG/f38GDBjArFmzqK6u5sorrwRg7NixdOzYkRkzZgDwy1/+kmnTpjF37lw6d+6cvrckJyeHnJycg/hSREREpCVqchgZNWoUW7duZdq0aWzevJmTTz6ZBQsWpG9qXb9+Pba9o8Pl17/+NbFYjG984xsZzzN9+nRuvfXWA6teREREWrwmf8+IH/Q9IyIiIi1Ps3zPiIiIiMjBpjAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivlIYEREREV8pjIiIiIivFEZERETEVwojIiIi4iuFEREREfGVwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfKYyIiIiIrxRGRERExFcKIyIiIuIrhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPhKYURERER8pTAiIiIivgr4XYCf/viPjWyvqicacogEky2aMbST4w3zAzYBR/lNRETkYDqqw8iDS/7Equ1bwdgYbDA2NAwzplmpaQ4B2yYcCBJwACuOsRJYVgLSLQ6Wu9N45nSHCEGyCVg5hKxswk4OYSuHaCCXqJNDOOAQcmxCAZtgahgK2ISc5OP0NMfKeJwcZk6zLQtjDCb1eo0BMKkhmNQ0k5rQsFw4YBMNJQNZNOgQTg2DjoVlWU3ez65nqIu71MZdamNuxnjAscmPBsiLBsmPBgkHnAP5kYqISAt0VIeRWN6fiEZX7Ne6if3cZgKo38t8Ux/BuFkYN5psXhTjhXcKShbJcGQlH2Nh0oEpNW/n5XbRiDBhdr+MZUPQtgmkglDAtggGbIK2hW3ZJFxIuBZx1yKesIi7EEskpzXUb1KhbsfjAHgBMEGMFyTkhMkLZ5EfySI/GiY/FVLyIoHkMBokLxLEtnfUmPDi1LnV1LmV1LpV1LpV1HnV1CWqqHOrqPOSw3qvNvnyjMHD2ymEmVRoMzSENZP6hzE4VpCwEyXsZBNxokTsLMJOFlmBbKKBbKKBLKJONlnB5LSsYDZZgSgBxyZg2zi2RcC2kkOnYdzeMS09tLFtcGwL20q25Dj7FQJFRFqKozqMfL1LH1Z9mU3CJHA9F894JEwCz/NwjUvCS6SnuZ5Hwkukprs4lkPQDhK0QwSdEEErNbRDhOwQATtI0A7vNB7CsQLUxKupjFdSFa+gKl5BdaKSmkQlsdSJ0nLqsJw6n/fM3rmptovdHE3BVGuKulTbbBzwApiqIFQEMSYIXjAZZuz61L6qxbJjTdxC8zPGAi+E8cLJMJkaGi8Mbnj303cZD6WXt6wATiqg2DbJcXvnwGLh2GSEmIYgY9tech9ZdVhOPdj1YNeBFcMmgGXCWISwTQiLMJYJYZkgFiEabitr6E2DZFjKCjlkhwJkhTOH2eEA2SGHrNQwErQJBz1CAXDs5P8r17h4xsXDw039n2qYtmOei2s8POMStiPJ0OdkEwlECTtZ2Nipnr1kePQaAqQBLzUt7nq4niHuGhKeR8IzJFyD63k7prkmOd0zJFyPoGOTGwmkX0tuJDnMSbVI0G7WYGiMSdcccz0SbrLWuOulHhtsi3SPaTjg7NRzun89l3uS8BLEvTgRJ6IwfASLu3HKasrYVL2JXm17EQlEfKnDMmbnXzOHp4qKCvLz8ykvLycvL8/vcppF3ItTUV9BRayC8vryjGFNvAbXuMlfVCYZmjzj4eHheamh2bXtuECTtLsfdWOW8VK/IN3UL+2dh+lmPGzLYFkeWB6W5WFwwfKA5HjDyWZHqEsQc2PUu/XUuXXUJ+qJefsfLBwTxSELhywCZKeGyeaQjUMYy7KxUv+wLGwskr9nbSyL9DzLSo/hEiduaol7NSRMLQnqUsNaXFOHSy0udXhWHR51YB38/1LGczIDSjq4hMALJ7dp12HZ9emG0/A4fgDbDaa2EcSYEHihZBi03HSzrFQ83flxerx5fr0k6wqDG9lDmAuS7Hlr6CFs6FFkp/HMeSajN9HaqYewYZi89BlykiEgEnQIBRzCjoNnJfBMHNfE8YjjmhgeCTySjz2TwFjJcUMcQwKDB14IL7WPPS+I6wYxbijVU5jc38kQHtrpsUNmD2hqaFkEbYeQ4xByAoQch2DAIWQHCAZssGvwrBpcqwbPqsZY1elxz6rGtapxqU4tU41n1aZ2toVNJBlaTQTLi4DJ3N/GDeO5ITw3jOeGk2/WAhahAAQdCAYsgo6VHHcsAgEI2hBIPXYccGxwrCAWQWyT7C21CGCZIJhAqgWTr98EMV4g2YwFdsMxmMCyXQwJLCs5NFYCSIDl4pEcN7gYK5Hq/3STPwtjMJaHMS7gYSyDZ1Lz8NJDxwoQsAM4tkPQDiTflDrJaUE7QMjZ8TjcMG45Oy6LY+F5qV5ZY6XD81fneQYcK0TQihK0o8khUSwCGCxck+zR9TyDm7rc7hmDZ8jodXWpo8rdSpW7lcrEViriW/kyXkZ5bCtfxMooj21PnweeGPo0fYtOPKj/Vxt7/j6qe0YOJ0E7SJtoG9pE2/hdiq8841Hv1lOfSAUUt566RHLYMB5zY0SDUfJD+eSF8sgL55EdzCZg+384G2OoTdRSHa9OtkQ1NfEaquPVVMWr0uMNrSZRQ1Wsaqdpyfm1iRqqEzXE3ORFPct2wa7Boma/awtYQcJOFhEni7CdRciOkDBxYl4dca+emFdHzK0jbnZcSLTs+AGFmd0x5iv3Z+3jfi3LjqV6c+pTwWenugJVB7W2xqpn75dbm8oCnFQ7GGKpdlBYBo9aaAgneyky9VMESJ32d8MAB/eQOqoYY+8Igu5uelRNACtQgR38Ejv4RaN62o0XwMQLWLFlO32LDsGL2A3/f3uL7MS2bKKBKNFA1O9S9otlWcl7R4JZtKPdAT9f3ItTE6/JDDipELNzsAnYAbKCWeQEc8gOZpMdzM4Yzw5mE3JCjdqmZzzqEnXUJmozWsO0uBcnaAdT7wSDBJ0gAStA0AmmLl3uNC813vDYtg7sMkfMjVEVr8rYFw2tIexVxauoS9Rl3BtkMMnewp3HU/cKNYzvPL/hManLP57xiHvJyyYx1yXheqlLP8lpATtIwM68TBuyw4QaLt06yekhJ0zYST52LBuXGK5JhsG4qSfu1RJLBcN6N9nq3F33v2e8zN5Sz0td1koOjcnsMQVDxMkmK5BLlpNLViCPrEAuUSc3NS2PaCCHqJObbllOHgE7iEsdCVNH3NQQN3XEvVrippZ6r4aYW0u9lxyvS9RQ5yab67mYVG9TQ/MMeF7muOclx13PwvVSvRRWQ+9RfEfPEvFkzxPJ3qfkfouz47Z7sLBxrAA2AWwrgE0Q2wrgEEhe5iTV04KTvDyJk2rJ3iXLcrCMjWUlg3Cy37QhXtlYOGBIX0pMmESqtzc59IyLSyLVm+LuGJJI1WnAauhn23mcjB5EC9K9q4YYnlWPRx2elXpjYnng1GI5tY2+Bu6YLEK0IeC1IWBa47itsL3WkGgFiQI8NwfXhT7tTt7f/5oHTGFE5DAWtIPkh/PJD+cfsm3alp0OVIebkBOitdOa1pHWfpciPjPGkPASJEyCkB3CsY/sT+K5nktNIvkGJP1GJFFNdSw1TIXyereedtF2FGUX0SG7A8U5xWQHs/0uf58URkREpMWxLCvZG9fkW+RbJsd2yA3lkhvK9buUZqFv8BIRERFf7VcYefDBB+ncuTORSISBAwfyzjvv7HX5Z599lhNOOIFIJEKvXr14+eWX96tYEREROfI0OYw888wzTJo0ienTp7Ns2TL69OnD0KFD2bJly26Xf/PNNxk9ejTf/va3ef/99xk5ciQjR47ko48+OuDiRUREpOVr8veMDBw4kFNPPZUHHngAAM/zKCkp4brrrmPy5Mm7LD9q1Ciqq6t56aWX0tNOO+00Tj75ZGbPnt2obR4N3zMiIiJypGns+btJPSOxWIylS5cyZMiQHU9g2wwZMoQlS5bsdp0lS5ZkLA8wdOjQPS4PUF9fT0VFRUYTERGRI1OTwsi2bdtwXZfCwsKM6YWFhWzevHm362zevLlJywPMmDGD/Pz8dCspKWlKmSIiItKCHJafppkyZQrl5eXptmHDBr9LEhERkWbSpO8Zadu2LY7jUFZWljG9rKyMoqLdf4dsUVFRk5YHCIfDhMPhppQmIiIiLVSTekZCoRD9+vVj4cKF6Wme57Fw4UIGDRq023UGDRqUsTzAq6++usflRURE5OjS5G9gnTRpEuPGjaN///4MGDCAWbNmUV1dzZVXXgnA2LFj6dixIzNmzADg+uuv55xzzmHmzJn853/+J08//TTvvfceDz/88MF9JSIiItIiNTmMjBo1iq1btzJt2jQ2b97MySefzIIFC9I3qa5fvx7b3tHhcvrppzN37lymTp3Kj3/8Y7p168YLL7zASSeddPBehYiIiLRYTf6eET/oe0ZERERanmb5nhERERGRg61F/NXehs4bffmZiIhIy9Fw3t7XRZgWEUYqKysB9OVnIiIiLVBlZSX5+fl7nN8i7hnxPI+NGzeSm5uLZVkH7XkrKiooKSlhw4YNuhclRfskk/ZHJu2PXWmfZNL+yHS07w9jDJWVlXTo0CHjwy1f1SJ6Rmzb5phjjmm258/LyzsqD5K90T7JpP2RSftjV9onmbQ/Mh3N+2NvPSINdAOriIiI+EphRERERHx1VIeRcDjM9OnT9XdwdqJ9kkn7I5P2x660TzJpf2TS/micFnEDq4iIiBy5juqeEREREfGfwoiIiIj4SmFEREREfKUwIiIiIr5SGBERERFfHdVh5MEHH6Rz585EIhEGDhzIO++843dJvrj11luxLCujnXDCCX6XdUj99a9/Zfjw4XTo0AHLsnjhhRcy5htjmDZtGsXFxUSjUYYMGcLKlSv9KfYQ2Nf+uOKKK3Y5Zi644AJ/ij0EZsyYwamnnkpubi7t27dn5MiRrFixImOZuro6JkyYQJs2bcjJyeHSSy+lrKzMp4qbV2P2x7nnnrvLMXLttdf6VHHz+vWvf03v3r3T37I6aNAg/vSnP6XnH03Hxv46asPIM888w6RJk5g+fTrLli2jT58+DB06lC1btvhdmi9OPPFENm3alG5///vf/S7pkKqurqZPnz48+OCDu51/55138qtf/YrZs2fz9ttvk52dzdChQ6mrqzvElR4a+9ofABdccEHGMfPUU08dwgoPrTfeeIMJEybw1ltv8eqrrxKPxzn//POprq5OL3PjjTfyxz/+kWeffZY33niDjRs3cskll/hYdfNpzP4AuPrqqzOOkTvvvNOnipvXMcccwx133MHSpUt57733OO+88xgxYgT//Oc/gaPr2Nhv5ig1YMAAM2HChPRj13VNhw4dzIwZM3ysyh/Tp083ffr08buMwwZg5s2bl37seZ4pKioyd911V3ral19+acLhsHnqqad8qPDQ+ur+MMaYcePGmREjRvhSz+Fgy5YtBjBvvPGGMSZ5PASDQfPss8+ml1m+fLkBzJIlS/wq85D56v4wxphzzjnHXH/99f4V5bNWrVqZ3/zmN0f9sdFYR2XPSCwWY+nSpQwZMiQ9zbZthgwZwpIlS3yszD8rV66kQ4cOdO3alTFjxrB+/Xq/SzpsrFmzhs2bN2ccL/n5+QwcOPCoPV4AFi1aRPv27enRowff/e532b59u98lHTLl5eUAtG7dGoClS5cSj8czjpETTjiBY4899qg4Rr66PxrMmTOHtm3bctJJJzFlyhRqamr8KO+Qcl2Xp59+murqagYNGnTUHxuN1SL+au/Btm3bNlzXpbCwMGN6YWEhn3zyiU9V+WfgwIE8/vjj9OjRg02bNvHTn/6Us846i48++ojc3Fy/y/Pd5s2bAXZ7vDTMO9pccMEFXHLJJXTp0oXVq1fz4x//mGHDhrFkyRIcx/G7vGbleR433HADZ5xxBieddBKQPEZCoRAFBQUZyx4Nx8ju9gfAZZddRqdOnejQoQMffPABN998MytWrOD555/3sdrm8+GHHzJo0CDq6urIyclh3rx5fO1rX6O0tPSoPTaa4qgMI5Jp2LBh6fHevXszcOBAOnXqxO9//3u+/e1v+1iZHK6+9a1vpcd79epF7969Oe6441i0aBGDBw/2sbLmN2HCBD766KOj7r6qPdnT/rjmmmvS47169aK4uJjBgwezevVqjjvuuENdZrPr0aMHpaWllJeX89xzzzFu3DjeeOMNv8tqMY7KyzRt27bFcZxd7mYuKyujqKjIp6oOHwUFBXTv3p1Vq1b5XcphoeGY0PGyZ127dqVt27ZH/DEzceJEXnrpJV5//XWOOeaY9PSioiJisRhffvllxvJH+jGyp/2xOwMHDgQ4Yo+RUCjE8ccfT79+/ZgxYwZ9+vThvvvuO2qPjaY6KsNIKBSiX79+LFy4MD3N8zwWLlzIoEGDfKzs8FBVVcXq1aspLi72u5TDQpcuXSgqKso4XioqKnj77bd1vKR89tlnbN++/Yg9ZowxTJw4kXnz5vGXv/yFLl26ZMzv168fwWAw4xhZsWIF69evPyKPkX3tj90pLS0FOGKPka/yPI/6+vqj7tjYb37fQeuXp59+2oTDYfP444+bjz/+2FxzzTWmoKDAbN682e/SDrkf/OAHZtGiRWbNmjVm8eLFZsiQIaZt27Zmy5Ytfpd2yFRWVpr333/fvP/++wYw99xzj3n//ffNunXrjDHG3HHHHaagoMD83//9n/nggw/MiBEjTJcuXUxtba3PlTePve2PyspKc9NNN5klS5aYNWvWmNdee82ccsopplu3bqaurs7v0pvFd7/7XZOfn28WLVpkNm3alG41NTXpZa699lpz7LHHmr/85S/mvffeM4MGDTKDBg3yserms6/9sWrVKvOzn/3MvPfee2bNmjXm//7v/0zXrl3N2Wef7XPlzWPy5MnmjTfeMGvWrDEffPCBmTx5srEsy7zyyivGmKPr2NhfR20YMcaY+++/3xx77LEmFAqZAQMGmLfeesvvknwxatQoU1xcbEKhkOnYsaMZNWqUWbVqld9lHVKvv/66AXZp48aNM8YkP957yy23mMLCQhMOh83gwYPNihUr/C26Ge1tf9TU1Jjzzz/ftGvXzgSDQdOpUydz9dVXH9FBfnf7AjCPPfZYepna2lrzve99z7Rq1cpkZWWZiy++2GzatMm/opvRvvbH+vXrzdlnn21at25twuGwOf74480Pf/hDU15e7m/hzeSqq64ynTp1MqFQyLRr184MHjw4HUSMObqOjf1lGWPMoeuHEREREcl0VN4zIiIiIocPhRERERHxlcKIiIiI+EphRERERHylMCIiIiK+UhgRERERXymMiIiIiK8URkRERMRXCiMiIiLiK4URERER8ZXCiIiIiPjq/wP2Ia1+k/IgsAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.metrics import Precision, Recall, CategoricalAccuracy"
      ],
      "metadata": {
        "id": "GvXCfuW2bQ_w"
      },
      "execution_count": 158,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.evaluate(TrainvectorizedText)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V2G82HUpbVAU",
        "outputId": "13988fc3-1c6e-4379-8210-5b3166e500b0"
      },
      "execution_count": 159,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "990/990 [==============================] - 2s 2ms/step - loss: 0.0000e+00 - accuracy: 0.0000e+00\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.0, 0.0]"
            ]
          },
          "metadata": {},
          "execution_count": 159
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred = (model.predict(TestvectorizedText) > 0.5).astype(int)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pgMXKLlbbWpB",
        "outputId": "5c5cb57e-6335-4494-909b-59451225d91e"
      },
      "execution_count": 160,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "660/660 [==============================] - 25s 39ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred"
      ],
      "metadata": {
        "id": "uV0sK4OUm7In"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred = pd.DataFrame(Y_pred)"
      ],
      "metadata": {
        "id": "Bl8Pgw0dm8w7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred.shape"
      ],
      "metadata": {
        "id": "Tqwd6EbxnZQk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "##\n",
        "output = {\n",
        "    0: \"BOOK\",\n",
        "    1: \"CR\",\n",
        "    2: \"MIXED\" ,\n",
        "    3: \"MUSIC\" ,\n",
        "    4: \"SOUNDCASS\" ,\n",
        "    5: \"SOUNDDISC\" ,\n",
        "    6: \"VIDEOCASS\" ,\n",
        "    7: \"VIDEODISC\"\n",
        "}"
      ],
      "metadata": {
        "id": "0OONeXMSndAd"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Y_pred = pd.DataFrame(Y_pred)"
      ],
      "metadata": {
        "id": "GZMNvsBNn5xZ"
      },
      "execution_count": 165,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import OneHotEncoder\n",
        "enc = OneHotEncoder(handle_unknown='ignore')"
      ],
      "metadata": {
        "id": "0FIdK0DuoDYQ"
      },
      "execution_count": 166,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred = Y_pred.idxmax(axis=1)"
      ],
      "metadata": {
        "id": "e8YMDx6VoE63"
      },
      "execution_count": 167,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_test = Y_test.idxmax(axis=1)"
      ],
      "metadata": {
        "id": "_InpmdyroGWv"
      },
      "execution_count": 168,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred = pd.DataFrame(new_Y_pred)"
      ],
      "metadata": {
        "id": "Ah-84Df_oIhS"
      },
      "execution_count": 169,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_test = pd.DataFrame(new_Y_test)"
      ],
      "metadata": {
        "id": "GfyppqHsoJ_K"
      },
      "execution_count": 170,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kjiMkNYkoLdR",
        "outputId": "ad24b005-ac8b-4d11-958e-cf210392b302"
      },
      "execution_count": 171,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    15409\n",
              "5     2354\n",
              "6     1778\n",
              "7      823\n",
              "4      550\n",
              "3       87\n",
              "2       63\n",
              "1       38\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 171
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred = pd.DataFrame(new_Y_pred)"
      ],
      "metadata": {
        "id": "SR_KDz22oOZn"
      },
      "execution_count": 172,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred = new_Y_pred.rename(columns ={0: \"PREDICT\"} )"
      ],
      "metadata": {
        "id": "1FZ5yleZoSxi"
      },
      "execution_count": 173,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fxFpvykOoVYY",
        "outputId": "a1a756a0-463b-4159-e479-ae26f292a95f"
      },
      "execution_count": 174,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "PREDICT\n",
              "0          15409\n",
              "5           2354\n",
              "6           1778\n",
              "7            823\n",
              "4            550\n",
              "3             87\n",
              "2             63\n",
              "1             38\n",
              "dtype: int64"
            ]
          },
          "execution_count": 174,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_test = pd.DataFrame(new_Y_test)"
      ],
      "metadata": {
        "id": "6n7oRnlzoXLz"
      },
      "execution_count": 175,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_test = new_Y_test.rename(columns ={0: \"PREDICT\"} )"
      ],
      "metadata": {
        "id": "G6IiTBuUod2j"
      },
      "execution_count": 176,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print('Baseline: Accuracy: ', round(accuracy_score(new_Y_test, new_Y_pred)*100, 2))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OYlRhJNBogF5",
        "outputId": "b24e82fa-6f15-4801-d715-7fd04ce23a18"
      },
      "execution_count": 177,
      "outputs": [
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Baseline: Accuracy:  73.02\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ahaVeXQioiHt",
        "outputId": "1795493f-a089-41b1-8889-de69d1f64c4b"
      },
      "execution_count": 178,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "PREDICT\n",
              "0          15409\n",
              "5           2354\n",
              "6           1778\n",
              "7            823\n",
              "4            550\n",
              "3             87\n",
              "2             63\n",
              "1             38\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 178
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred[\"PREDICT\"] = new_Y_pred[\"PREDICT\"].map(output)"
      ],
      "metadata": {
        "id": "j8YM7l1zok6o"
      },
      "execution_count": 179,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_test[\"PREDICT\"] = new_Y_test[\"PREDICT\"].map(output)"
      ],
      "metadata": {
        "id": "1I5lqmCQo6Zq"
      },
      "execution_count": 180,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred[\"PREDICT\"].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9-XCD98Fo9K2",
        "outputId": "54f9a092-07f8-412f-dfd5-081a2c428afc"
      },
      "execution_count": 181,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "BOOK         15409\n",
              "SOUNDDISC     2354\n",
              "VIDEOCASS     1778\n",
              "VIDEODISC      823\n",
              "SOUNDCASS      550\n",
              "MUSIC           87\n",
              "MIXED           63\n",
              "CR              38\n",
              "Name: PREDICT, dtype: int64"
            ]
          },
          "execution_count": 181,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred[\"PREDICT\"].unique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XErBc4YjpFjZ",
        "outputId": "3e771413-ed4f-4fae-90c5-2c1498a79562"
      },
      "execution_count": 182,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "array(['SOUNDDISC', 'BOOK', 'SOUNDCASS', 'VIDEOCASS', 'VIDEODISC',\n",
              "       'MUSIC', 'MIXED', 'CR'], dtype=object)"
            ]
          },
          "execution_count": 182,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "keys = ['BOOK', 'SOUNDDISC', 'SOUNDCASS', 'VIDEOCASS', 'VIDEODISC',\n",
        "       'MUSIC', 'CR', 'MIXED']\n",
        "\n",
        "explode = [0, 0.3, 0.1, 0.1, 0.1, 0.4, 0.4, 0.1]"
      ],
      "metadata": {
        "id": "OpAJyKDPpMAB"
      },
      "execution_count": 183,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "all = new_Y_pred[\"PREDICT\"].value_counts()"
      ],
      "metadata": {
        "id": "R2v3WuPvpdER"
      },
      "execution_count": 184,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize = [15,8])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        },
        "id": "ivAgiB6553g6",
        "outputId": "79052b61-eeb3-4e78-ae29-ab874f80ac98"
      },
      "execution_count": 186,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Figure size 1500x800 with 0 Axes>"
            ]
          },
          "metadata": {},
          "execution_count": 186
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x800 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plotting data on chart\n",
        "plt.pie(all, labels=keys, explode=explode, autopct='%.1f%%')\n",
        "\n",
        "# displaying chart\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "geXUUsA_pmJB",
        "outputId": "73c3699a-ad1e-4e60-f766-7a7cabf9d5a9"
      },
      "execution_count": 187,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "new_Y_pred.to_csv(\"Predict.csv\")"
      ],
      "metadata": {
        "id": "vSpb_LtFpobq"
      },
      "execution_count": 188,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "oHR-UQDy9EMv"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}