Modeling_CitiBike2020.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "mount_file_id": "1LtoYfAzBgL0LoE6N7ab1SUuR3d8hGhKW",
      "authorship_tag": "ABX9TyO5zCN2jLCPTzF5bCNSQ6ed",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Niraj-kumbhar/CitiBike-NYC-2020/blob/main/Modeling_CitiBike2020.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Citi Bike Model Building\n",
        "\n",
        "Data available on : [Citi Bike](https://s3.amazonaws.com/tripdata/index.html)\n",
        "\n",
        "Citi Bike is a privately owned public bicycle sharing system serving the New York City boroughs of the Bronx, Brooklyn, Manhattan, and Queens, as well as Jersey City, New Jersey, and Hoboken, New Jersey.<br>\n",
        "<br>\n",
        "I have created notebook for detailed Analysis on this data, you can find this on [github](https://github.com/Niraj-kumbhar/CitiBike-NYC-2020/blob/main/EDA_CitiBikeHistory2020.ipynb)"
      ],
      "metadata": {
        "id": "wFBXU5YPhwHH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Loading Required Libraries"
      ],
      "metadata": {
        "id": "8cXdu2aHiU8p"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#for handling data\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "pd.set_option('display.max_columns', None)\n",
        "\n",
        "# for vizualization\n",
        "import seaborn as sns\n",
        "sns.set()\n",
        "import matplotlib.ticker as ticker\n",
        "import matplotlib.pyplot as plt\n",
        "from  matplotlib.lines import Line2D\n",
        "%matplotlib inline\n",
        "import plotly.express as px\n",
        "\n",
        "#for handling date\n",
        "import datetime as dt\n",
        "import holidays\n",
        "\n",
        "# for splitting data\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# feature engineering and preprocessing\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "\n",
        "# machine learning and deep learning\n",
        "from xgboost import XGBRegressor\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.metrics import mean_absolute_error, mean_squared_error,r2_score\n",
        "\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.pipeline import Pipeline,make_pipeline\n",
        "# Display Pipeline\n",
        "from sklearn import set_config\n",
        "set_config(display='diagram')\n",
        "\n",
        "import tensorflow as tf\n",
        "import keras\n",
        "from keras.layers import Dense,Input,Embedding,LSTM,Dropout\n",
        "from keras.models import Sequential\n",
        "from keras import Model\n",
        "\n",
        "#for supressing warnings\n",
        "import warnings\n",
        "warnings.filterwarnings(\"always\")"
      ],
      "metadata": {
        "id": "BbUaUQ-RupDS"
      },
      "execution_count": 181,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Loading Data\n",
        "As we have already created final data csv in last EDA notebook, I will import it now"
      ],
      "metadata": {
        "id": "FssgIxR-tZv5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('/content/drive/MyDrive/data/CitiBike2020.csv', index_col=0, parse_dates=['Date'], dtype={'Holiday':'int8'})"
      ],
      "metadata": {
        "id": "STSgSS6mtHnA"
      },
      "execution_count": 182,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kcCTGZBtu05y",
        "outputId": "2d73c636-ecbe-4a96-a7f1-873385b2708d"
      },
      "execution_count": 183,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 366 entries, 0 to 365\n",
            "Data columns (total 4 columns):\n",
            " #   Column   Non-Null Count  Dtype         \n",
            "---  ------   --------------  -----         \n",
            " 0   Date     366 non-null    datetime64[ns]\n",
            " 1   Trips    366 non-null    int64         \n",
            " 2   Weekday  366 non-null    object        \n",
            " 3   Holiday  366 non-null    int8          \n",
            "dtypes: datetime64[ns](1), int64(1), int8(1), object(1)\n",
            "memory usage: 11.8+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "0xYxnw9Fu_-Y",
        "outputId": "d6b9fdbf-6572-41b1-b171-d0bbf480bcbc"
      },
      "execution_count": 184,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Date  Trips    Weekday  Holiday\n",
              "0 2020-01-01  18114  Wednesday        1\n",
              "1 2020-01-02  41556   Thursday        0\n",
              "2 2020-01-03  31873     Friday        0\n",
              "3 2020-01-04  29436   Saturday        0\n",
              "4 2020-01-05  27589     Sunday        0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-56b7cb5a-e752-4153-bbcd-ec3029693589\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Trips</th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020-01-01</td>\n",
              "      <td>18114</td>\n",
              "      <td>Wednesday</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020-01-02</td>\n",
              "      <td>41556</td>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020-01-03</td>\n",
              "      <td>31873</td>\n",
              "      <td>Friday</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020-01-04</td>\n",
              "      <td>29436</td>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020-01-05</td>\n",
              "      <td>27589</td>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-56b7cb5a-e752-4153-bbcd-ec3029693589')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-56b7cb5a-e752-4153-bbcd-ec3029693589 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-56b7cb5a-e752-4153-bbcd-ec3029693589');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 184
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Using Weather data\n",
        "I have downloaded Climate data for 1 station from [NOAA site](https://www.ncdc.noaa.gov/cdo-web/).\n",
        "Lets analyse if it is affecteing no of trips  "
      ],
      "metadata": {
        "id": "oMj246oYujz0"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_weather = pd.read_csv('/content/drive/MyDrive/data/ClimetData.csv', parse_dates=['DATE'])"
      ],
      "metadata": {
        "id": "uwX_iHCmuhsc"
      },
      "execution_count": 185,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_weather.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CDeStgA6vdwW",
        "outputId": "152f8eeb-f6bb-4a24-9e30-5983513d0f70"
      },
      "execution_count": 186,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 366 entries, 0 to 365\n",
            "Data columns (total 44 columns):\n",
            " #   Column           Non-Null Count  Dtype         \n",
            "---  ------           --------------  -----         \n",
            " 0   STATION          366 non-null    object        \n",
            " 1   NAME             366 non-null    object        \n",
            " 2   LATITUDE         366 non-null    float64       \n",
            " 3   LONGITUDE        366 non-null    float64       \n",
            " 4   ELEVATION        366 non-null    float64       \n",
            " 5   DATE             366 non-null    datetime64[ns]\n",
            " 6   AWND             329 non-null    float64       \n",
            " 7   AWND_ATTRIBUTES  329 non-null    object        \n",
            " 8   PGTM             0 non-null      float64       \n",
            " 9   PGTM_ATTRIBUTES  0 non-null      float64       \n",
            " 10  PRCP             366 non-null    float64       \n",
            " 11  PRCP_ATTRIBUTES  366 non-null    object        \n",
            " 12  SNOW             366 non-null    float64       \n",
            " 13  SNOW_ATTRIBUTES  366 non-null    object        \n",
            " 14  SNWD             366 non-null    float64       \n",
            " 15  SNWD_ATTRIBUTES  366 non-null    object        \n",
            " 16  TAVG             0 non-null      float64       \n",
            " 17  TAVG_ATTRIBUTES  0 non-null      float64       \n",
            " 18  TMAX             366 non-null    int64         \n",
            " 19  TMAX_ATTRIBUTES  366 non-null    object        \n",
            " 20  TMIN             366 non-null    int64         \n",
            " 21  TMIN_ATTRIBUTES  366 non-null    object        \n",
            " 22  TSUN             0 non-null      float64       \n",
            " 23  TSUN_ATTRIBUTES  0 non-null      float64       \n",
            " 24  WDF2             329 non-null    float64       \n",
            " 25  WDF2_ATTRIBUTES  329 non-null    object        \n",
            " 26  WDF5             326 non-null    float64       \n",
            " 27  WDF5_ATTRIBUTES  326 non-null    object        \n",
            " 28  WSF2             329 non-null    float64       \n",
            " 29  WSF2_ATTRIBUTES  329 non-null    object        \n",
            " 30  WSF5             326 non-null    float64       \n",
            " 31  WSF5_ATTRIBUTES  326 non-null    object        \n",
            " 32  WT01             153 non-null    float64       \n",
            " 33  WT01_ATTRIBUTES  153 non-null    object        \n",
            " 34  WT02             8 non-null      float64       \n",
            " 35  WT02_ATTRIBUTES  8 non-null      object        \n",
            " 36  WT03             27 non-null     float64       \n",
            " 37  WT03_ATTRIBUTES  27 non-null     object        \n",
            " 38  WT04             0 non-null      float64       \n",
            " 39  WT04_ATTRIBUTES  0 non-null      float64       \n",
            " 40  WT05             0 non-null      float64       \n",
            " 41  WT05_ATTRIBUTES  0 non-null      float64       \n",
            " 42  WT08             41 non-null     float64       \n",
            " 43  WT08_ATTRIBUTES  41 non-null     object        \n",
            "dtypes: datetime64[ns](1), float64(25), int64(2), object(16)\n",
            "memory usage: 125.9+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# checking null in percentage\n",
        "round(df_weather.isnull().sum()/366 *100, 2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AKc_XOgnvpWi",
        "outputId": "c76e3e44-224d-4725-e990-349e65c8d4ed"
      },
      "execution_count": 187,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "STATION              0.00\n",
              "NAME                 0.00\n",
              "LATITUDE             0.00\n",
              "LONGITUDE            0.00\n",
              "ELEVATION            0.00\n",
              "DATE                 0.00\n",
              "AWND                10.11\n",
              "AWND_ATTRIBUTES     10.11\n",
              "PGTM               100.00\n",
              "PGTM_ATTRIBUTES    100.00\n",
              "PRCP                 0.00\n",
              "PRCP_ATTRIBUTES      0.00\n",
              "SNOW                 0.00\n",
              "SNOW_ATTRIBUTES      0.00\n",
              "SNWD                 0.00\n",
              "SNWD_ATTRIBUTES      0.00\n",
              "TAVG               100.00\n",
              "TAVG_ATTRIBUTES    100.00\n",
              "TMAX                 0.00\n",
              "TMAX_ATTRIBUTES      0.00\n",
              "TMIN                 0.00\n",
              "TMIN_ATTRIBUTES      0.00\n",
              "TSUN               100.00\n",
              "TSUN_ATTRIBUTES    100.00\n",
              "WDF2                10.11\n",
              "WDF2_ATTRIBUTES     10.11\n",
              "WDF5                10.93\n",
              "WDF5_ATTRIBUTES     10.93\n",
              "WSF2                10.11\n",
              "WSF2_ATTRIBUTES     10.11\n",
              "WSF5                10.93\n",
              "WSF5_ATTRIBUTES     10.93\n",
              "WT01                58.20\n",
              "WT01_ATTRIBUTES     58.20\n",
              "WT02                97.81\n",
              "WT02_ATTRIBUTES     97.81\n",
              "WT03                92.62\n",
              "WT03_ATTRIBUTES     92.62\n",
              "WT04               100.00\n",
              "WT04_ATTRIBUTES    100.00\n",
              "WT05               100.00\n",
              "WT05_ATTRIBUTES    100.00\n",
              "WT08                88.80\n",
              "WT08_ATTRIBUTES     88.80\n",
              "dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 187
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# as there are a lot of null values present, we will drop columns with null\n",
        "\n",
        "df_climate= df_weather[['DATE','TMAX','TMIN','SNOW']]\n",
        "df_climate.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "mT7ujgGvvxUp",
        "outputId": "406d4e00-8128-4296-8eb5-39116fe7f415"
      },
      "execution_count": 188,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        DATE  TMAX  TMIN  SNOW\n",
              "0 2020-01-01    41    34   0.0\n",
              "1 2020-01-02    49    33   0.0\n",
              "2 2020-01-03    49    44   0.0\n",
              "3 2020-01-04    51    41   0.0\n",
              "4 2020-01-05    42    35   0.0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4b6f66f5-6b53-409d-8782-9092c123aa64\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>DATE</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>SNOW</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020-01-01</td>\n",
              "      <td>41</td>\n",
              "      <td>34</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020-01-02</td>\n",
              "      <td>49</td>\n",
              "      <td>33</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020-01-03</td>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020-01-04</td>\n",
              "      <td>51</td>\n",
              "      <td>41</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020-01-05</td>\n",
              "      <td>42</td>\n",
              "      <td>35</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4b6f66f5-6b53-409d-8782-9092c123aa64')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-4b6f66f5-6b53-409d-8782-9092c123aa64 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-4b6f66f5-6b53-409d-8782-9092c123aa64');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 188
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# plotting normal line chart over Date\n",
        "sns.lineplot(data=df_climate, x='DATE',y='TMAX', color='r')\n",
        "sns.lineplot(data=df_climate, x='DATE',y='TMIN', color='g')\n",
        "plt.rcParams[\"figure.figsize\"] = (40,10)\n",
        "plt.title('Weather throughout year 2020');\n",
        "\n",
        "\n",
        "# Thanks to: https://stackoverflow.com/questions/35906717/marking-specific-dates-when-visualizing-a-time-series\n",
        "ax = plt.gca()\n",
        "label_list = [date for i, date in enumerate(df_climate['DATE']) if df_climate['SNOW'][i]>0.0]\n",
        "\n",
        "for date_point in label_list:\n",
        "    plt.axvline(x=date_point, color='b')\n",
        "\n",
        "\n",
        "legend_elements = [Line2D([0], [0], color='r', lw=4, label='TMAX'),\n",
        "                Line2D([0], [0], color='g', lw=4, label='TMIN'),\n",
        "                Line2D([0], [0], color='b', lw=4, label='Snow')]\n",
        "plt.legend(handles=legend_elements)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "id": "VDnljitMfqg0",
        "outputId": "03bf29a6-f766-4d40-a760-ba46da489914"
      },
      "execution_count": 189,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Lets Join both Datasets"
      ],
      "metadata": {
        "id": "nAc7AfBC0lcD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_modeling = df.merge(df_climate, how='inner', left_on='Date', right_on='DATE')"
      ],
      "metadata": {
        "id": "NZdk32WAx5e0"
      },
      "execution_count": 190,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "fig, axs = plt.subplots(1, 2, figsize=(20,5), sharey=True)\n",
        "\n",
        "sns.scatterplot(x=df_modeling['TMAX'], y=df_modeling['Trips'], ax=axs[0])\n",
        "axs[0].set_xlabel('Tmax, °F', fontsize=18)\n",
        "axs[0].set_ylabel('Trips (% of yearly count)', fontsize=18)\n",
        "#axs[0].yaxis.set_major_formatter(mtick.PercentFormatter(xmax=1))\n",
        "\n",
        "sns.scatterplot(x=df_modeling['TMIN'], y=df_modeling['Trips'], ax=axs[1])\n",
        "axs[1].set_xlabel('Tmin, °F', fontsize=18)\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362
        },
        "id": "jGt4z9KO1DFJ",
        "outputId": "26818075-ecad-47ac-df25-54e38e410d7e"
      },
      "execution_count": 191,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Tmin, °F')"
            ]
          },
          "metadata": {},
          "execution_count": 191
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x360 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,5))\n",
        "axs= sns.scatterplot(x=df_modeling[\"SNOW\"], y=df_modeling[\"Trips\"])\n",
        "axs.set_xlabel('Snowfall, inches', fontsize=18)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 362
        },
        "id": "-Hklv6vw1uZm",
        "outputId": "acf060b3-1201-4e78-ceaf-bef05593b95e"
      },
      "execution_count": 192,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Snowfall, inches')"
            ]
          },
          "metadata": {},
          "execution_count": 192
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Trips increase when temperature increases, until its hotest day \n",
        "* People choose not use bikes on when there is Snowfall"
      ],
      "metadata": {
        "id": "qOPXbmgy3FCG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Feature Engineering\n",
        "We will create some new features using same data\n",
        "* We create new columns to indicate each day (using dummy columns)\n",
        "* will Create new features to indicate day of the year and month\n",
        "* will create column to indicate year\n",
        "* We can drop one DATE column\n",
        "* SNOW can be useful but while predicting for future, you need to have snow prediction also"
      ],
      "metadata": {
        "id": "hIpl000X45Hy"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_modeling.drop(columns=['DATE','SNOW'], inplace=True)\n",
        "df_modeling.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "bgB9SoEm2s83",
        "outputId": "8c2a954e-7e72-41aa-8681-6ef6d946da99"
      },
      "execution_count": 193,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Date  Trips    Weekday  Holiday  TMAX  TMIN\n",
              "0 2020-01-01  18114  Wednesday        1    41    34\n",
              "1 2020-01-02  41556   Thursday        0    49    33\n",
              "2 2020-01-03  31873     Friday        0    49    44\n",
              "3 2020-01-04  29436   Saturday        0    51    41\n",
              "4 2020-01-05  27589     Sunday        0    42    35"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9b431ddd-b5bc-42e8-9a63-747ccb5948f6\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Trips</th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020-01-01</td>\n",
              "      <td>18114</td>\n",
              "      <td>Wednesday</td>\n",
              "      <td>1</td>\n",
              "      <td>41</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020-01-02</td>\n",
              "      <td>41556</td>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020-01-03</td>\n",
              "      <td>31873</td>\n",
              "      <td>Friday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020-01-04</td>\n",
              "      <td>29436</td>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "      <td>51</td>\n",
              "      <td>41</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020-01-05</td>\n",
              "      <td>27589</td>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>42</td>\n",
              "      <td>35</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9b431ddd-b5bc-42e8-9a63-747ccb5948f6')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-9b431ddd-b5bc-42e8-9a63-747ccb5948f6 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-9b431ddd-b5bc-42e8-9a63-747ccb5948f6');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 193
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#changing SNOW feature to binary categorical \n",
        "#df_modeling['SNOW'] = df_modeling['SNOW'].apply(lambda x : 1 if x>0 else 0)"
      ],
      "metadata": {
        "id": "1aV5g5D75FMb"
      },
      "execution_count": 194,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data = df_modeling.copy()\n"
      ],
      "metadata": {
        "id": "m7-rZJEt7BcU"
      },
      "execution_count": 195,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# creating day of the year feature\n",
        "df_modeling['DOY'] = df_modeling['Date'].dt.dayofyear\n",
        "\n",
        "# creating day of the week feature\n",
        "df_modeling['DOW'] = df_modeling['Date'].dt.dayofweek\n",
        "\n",
        "# creating day of the month feature\n",
        "df_modeling['DOM'] = df_modeling['Date'].dt.day\n",
        "\n",
        "# creating year column\n",
        "df_modeling['year'] = df_modeling['Date'].dt.year\n",
        "\n",
        "# creating month number column\n",
        "df_modeling['month_num'] = df_modeling['Date'].dt.month\n",
        "\n",
        "#creating month feature\n",
        "df_modeling['month'] = df_modeling['Date'].dt.month_name()"
      ],
      "metadata": {
        "id": "JQO7Ped9-oty"
      },
      "execution_count": 197,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#data preview\n",
        "df_modeling.head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "SJ81w5FJCTHf",
        "outputId": "6eaded5a-05d4-44c1-fba7-30cc90f38066"
      },
      "execution_count": 198,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Date  Trips    Weekday  Holiday  TMAX  TMIN  DOY  DOW  DOM  year  \\\n",
              "0 2020-01-01  18114  Wednesday        1    41    34    1    2    1  2020   \n",
              "1 2020-01-02  41556   Thursday        0    49    33    2    3    2  2020   \n",
              "2 2020-01-03  31873     Friday        0    49    44    3    4    3  2020   \n",
              "3 2020-01-04  29436   Saturday        0    51    41    4    5    4  2020   \n",
              "4 2020-01-05  27589     Sunday        0    42    35    5    6    5  2020   \n",
              "5 2020-01-06  42738     Monday        0    45    32    6    0    6  2020   \n",
              "6 2020-01-07  46535    Tuesday        0    45    36    7    1    7  2020   \n",
              "7 2020-01-08  42400  Wednesday        0    42    28    8    2    8  2020   \n",
              "8 2020-01-09  39344   Thursday        0    34    23    9    3    9  2020   \n",
              "9 2020-01-10  50800     Friday        0    55    33   10    4   10  2020   \n",
              "\n",
              "   month_num    month  \n",
              "0          1  January  \n",
              "1          1  January  \n",
              "2          1  January  \n",
              "3          1  January  \n",
              "4          1  January  \n",
              "5          1  January  \n",
              "6          1  January  \n",
              "7          1  January  \n",
              "8          1  January  \n",
              "9          1  January  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-18423465-3b0b-4c35-b18b-739ca95b3f00\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Trips</th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020-01-01</td>\n",
              "      <td>18114</td>\n",
              "      <td>Wednesday</td>\n",
              "      <td>1</td>\n",
              "      <td>41</td>\n",
              "      <td>34</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020-01-02</td>\n",
              "      <td>41556</td>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>33</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020-01-03</td>\n",
              "      <td>31873</td>\n",
              "      <td>Friday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020-01-04</td>\n",
              "      <td>29436</td>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "      <td>51</td>\n",
              "      <td>41</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020-01-05</td>\n",
              "      <td>27589</td>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>42</td>\n",
              "      <td>35</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>2020-01-06</td>\n",
              "      <td>42738</td>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>32</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>6</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>2020-01-07</td>\n",
              "      <td>46535</td>\n",
              "      <td>Tuesday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>36</td>\n",
              "      <td>7</td>\n",
              "      <td>1</td>\n",
              "      <td>7</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>2020-01-08</td>\n",
              "      <td>42400</td>\n",
              "      <td>Wednesday</td>\n",
              "      <td>0</td>\n",
              "      <td>42</td>\n",
              "      <td>28</td>\n",
              "      <td>8</td>\n",
              "      <td>2</td>\n",
              "      <td>8</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>2020-01-09</td>\n",
              "      <td>39344</td>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "      <td>34</td>\n",
              "      <td>23</td>\n",
              "      <td>9</td>\n",
              "      <td>3</td>\n",
              "      <td>9</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>2020-01-10</td>\n",
              "      <td>50800</td>\n",
              "      <td>Friday</td>\n",
              "      <td>0</td>\n",
              "      <td>55</td>\n",
              "      <td>33</td>\n",
              "      <td>10</td>\n",
              "      <td>4</td>\n",
              "      <td>10</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-18423465-3b0b-4c35-b18b-739ca95b3f00')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-18423465-3b0b-4c35-b18b-739ca95b3f00 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-18423465-3b0b-4c35-b18b-739ca95b3f00');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 198
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Categorical feature encoding"
      ],
      "metadata": {
        "id": "Vxe6mg7bgFWa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_data = pd.get_dummies(df_modeling, columns=['Weekday','month'], drop_first=True)\n",
        "df_data.head()"
      ],
      "metadata": {
        "id": "HDSE8kXWgBTr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "outputId": "2d2aa42a-6bdf-4ca6-a928-b0a2fae82c02"
      },
      "execution_count": 199,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Date  Trips  Holiday  TMAX  TMIN  DOY  DOW  DOM  year  month_num  \\\n",
              "0 2020-01-01  18114        1    41    34    1    2    1  2020          1   \n",
              "1 2020-01-02  41556        0    49    33    2    3    2  2020          1   \n",
              "2 2020-01-03  31873        0    49    44    3    4    3  2020          1   \n",
              "3 2020-01-04  29436        0    51    41    4    5    4  2020          1   \n",
              "4 2020-01-05  27589        0    42    35    5    6    5  2020          1   \n",
              "\n",
              "   Weekday_Monday  Weekday_Saturday  Weekday_Sunday  Weekday_Thursday  \\\n",
              "0               0                 0               0                 0   \n",
              "1               0                 0               0                 1   \n",
              "2               0                 0               0                 0   \n",
              "3               0                 1               0                 0   \n",
              "4               0                 0               1                 0   \n",
              "\n",
              "   Weekday_Tuesday  Weekday_Wednesday  month_August  month_December  \\\n",
              "0                0                  1             0               0   \n",
              "1                0                  0             0               0   \n",
              "2                0                  0             0               0   \n",
              "3                0                  0             0               0   \n",
              "4                0                  0             0               0   \n",
              "\n",
              "   month_February  month_January  month_July  month_June  month_March  \\\n",
              "0               0              1           0           0            0   \n",
              "1               0              1           0           0            0   \n",
              "2               0              1           0           0            0   \n",
              "3               0              1           0           0            0   \n",
              "4               0              1           0           0            0   \n",
              "\n",
              "   month_May  month_November  month_October  month_September  \n",
              "0          0               0              0                0  \n",
              "1          0               0              0                0  \n",
              "2          0               0              0                0  \n",
              "3          0               0              0                0  \n",
              "4          0               0              0                0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3f9fa129-e02d-4c0b-a98b-4b38f5e5aa8f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Trips</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "      <th>Weekday_Monday</th>\n",
              "      <th>Weekday_Saturday</th>\n",
              "      <th>Weekday_Sunday</th>\n",
              "      <th>Weekday_Thursday</th>\n",
              "      <th>Weekday_Tuesday</th>\n",
              "      <th>Weekday_Wednesday</th>\n",
              "      <th>month_August</th>\n",
              "      <th>month_December</th>\n",
              "      <th>month_February</th>\n",
              "      <th>month_January</th>\n",
              "      <th>month_July</th>\n",
              "      <th>month_June</th>\n",
              "      <th>month_March</th>\n",
              "      <th>month_May</th>\n",
              "      <th>month_November</th>\n",
              "      <th>month_October</th>\n",
              "      <th>month_September</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2020-01-01</td>\n",
              "      <td>18114</td>\n",
              "      <td>1</td>\n",
              "      <td>41</td>\n",
              "      <td>34</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2020-01-02</td>\n",
              "      <td>41556</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>33</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2020-01-03</td>\n",
              "      <td>31873</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2020-01-04</td>\n",
              "      <td>29436</td>\n",
              "      <td>0</td>\n",
              "      <td>51</td>\n",
              "      <td>41</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2020-01-05</td>\n",
              "      <td>27589</td>\n",
              "      <td>0</td>\n",
              "      <td>42</td>\n",
              "      <td>35</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3f9fa129-e02d-4c0b-a98b-4b38f5e5aa8f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-3f9fa129-e02d-4c0b-a98b-4b38f5e5aa8f 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-3f9fa129-e02d-4c0b-a98b-4b38f5e5aa8f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 199
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modelling"
      ],
      "metadata": {
        "id": "6s3BtSMfdUSN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# THANKS TO https://github.com/prateeknigam9/Hinglish_Classifier/blob/main/CaseStudy_Hinglish.ipynb\n",
        "\n",
        "def metric_calc(model, xtrain, ytrain, xtest, ytest):\n",
        "    \"\"\"\n",
        "    To plot the metrics and graph of regression\n",
        "    ==============================================\n",
        "    inputs:\n",
        "      model: Model selected\n",
        "      xtrain: input for train\n",
        "      ytrain: ground truth for xtrain\n",
        "      xtest: input for test\n",
        "      ytest: ground truth for xtest\n",
        "    outputs:\n",
        "      PRINT --> Metrics R2_score, MAE, MSE, RMSE for both Train and validation\n",
        "      RETURN --> Metrics for Validation\n",
        "\n",
        "    \"\"\"\n",
        "    y_pred_train = model.predict(xtrain)\n",
        "    r2_train = r2_score(ytrain,y_pred_train)\n",
        "    mae_train = mean_absolute_error(ytrain,y_pred_train)\n",
        "    mse_train = mean_squared_error(ytrain,y_pred_train)\n",
        "    rmse_train = np.sqrt(mse_train)    \n",
        "    \n",
        "    metrics_train = {}\n",
        "    metrics_train['R2_Score'] = r2_train\n",
        "    metrics_train['MAE'] = mae_train\n",
        "    metrics_train['MSE'] = mse_train\n",
        "    metrics_train['RMSE'] = rmse_train\n",
        "    \n",
        "    y_pred = model.predict(xtest)\n",
        "    r2 = r2_score(ytest,y_pred)\n",
        "    mae = mean_absolute_error(ytest,y_pred)\n",
        "    mse = mean_squared_error(ytest, y_pred)\n",
        "    rmse = np.sqrt(mse)\n",
        "    \n",
        "    metrics = {}\n",
        "    metrics['R2_Score'] = r2\n",
        "    metrics['MAE'] = mae\n",
        "    metrics['MSE'] = mse\n",
        "    metrics['RMSE'] = rmse\n",
        "\n",
        "    print(\"=\"*20,\"Model Metrics - Train\",\"=\"*20)\n",
        "    print(pd.DataFrame([metrics_train]))    \n",
        "    print(\"\\n\")\n",
        "    print(\"=\"*20,\"Model Metrics - Test\",\"=\"*20)\n",
        "    print(pd.DataFrame([metrics]))\n",
        "    print(\"\\n\")\n",
        "    plt.rcParams.update({'figure.figsize': (15, 3)})\n",
        "    fig, ax = plt.subplots()\n",
        "    sns.lineplot(data=ytest, x=ytest.index, y='Trips')\n",
        "    sns.lineplot(x=ytest.index, y=pd.DataFrame(y_pred)[0], color='r')\n",
        "    plt.grid(linestyle='-', linewidth=0.3)\n",
        "    legend_elements = [Line2D([0], [0], color='b', lw=4, label='Actual'),\n",
        "                Line2D([0], [0], color='r', lw=4, label='Predicted')]\n",
        "    plt.legend(handles=legend_elements)\n",
        "    #ax.tick_params(axis='x', rotation=90)\n",
        "    \n",
        "    return metrics\n"
      ],
      "metadata": {
        "id": "WDH8Y9NKz2QW"
      },
      "execution_count": 200,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(df_data.drop(columns=['Trips','Date']), df_data[['Trips']], shuffle=False, test_size=0.2)"
      ],
      "metadata": {
        "id": "-MXSOHyDdT4C"
      },
      "execution_count": 201,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Model Evaluation"
      ],
      "metadata": {
        "id": "MDpi8X2EzWsA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "__Linear Regression__"
      ],
      "metadata": {
        "id": "gLvSoJYKdECN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "lr = LinearRegression()\n",
        "lr.fit(X_train, y_train)\n",
        "lr_metrics = metric_calc(lr, X_train, y_train, X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 397
        },
        "id": "M68zsEK0diEX",
        "outputId": "8fc94e8b-8894-4f26-8842-53af90e7664d"
      },
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "==================== Model Metrics - Train ====================\n",
            "   R2_Score          MAE           MSE          RMSE\n",
            "0  0.734317  9301.066197  1.517908e+08  12320.340932\n",
            "\n",
            "\n",
            "==================== Model Metrics - Test ====================\n",
            "   R2_Score           MAE           MSE          RMSE\n",
            "0 -0.440947  20409.196081  6.472827e+08  25441.751898\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "__Random Forest Regressor__"
      ],
      "metadata": {
        "id": "VtG1FjNIoEMG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rf = RandomForestRegressor(n_estimators=1000, max_depth=8)\n",
        "rf.fit(X_train, y_train)\n",
        "rf_metrics = metric_calc(rf, X_train, y_train, X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 433
        },
        "id": "7YOcOOAtoHxY",
        "outputId": "d73d0f9b-a70c-4933-c859-c61f7a446eef"
      },
      "execution_count": 203,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-203-cbe3ac106029>:2: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples,), for example using ravel().\n",
            "  rf.fit(X_train, y_train)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "==================== Model Metrics - Train ====================\n",
            "   R2_Score          MAE           MSE         RMSE\n",
            "0  0.961255  3424.432375  2.213574e+07  4704.863525\n",
            "\n",
            "\n",
            "==================== Model Metrics - Test ====================\n",
            "   R2_Score           MAE           MSE          RMSE\n",
            "0  0.315311  13879.883341  3.075666e+08  17537.575579\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "__Xgboost Regression__"
      ],
      "metadata": {
        "id": "DYsR2BMihHTf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# hyperparameters set 1\n",
        "xgb1 = XGBRegressor(n_estimators=800, max_depth=10, learning_rate=0.01, n_jobs=-1)\n",
        "xgb1.fit(X_train, y_train)\n",
        "xgb1_metrics = metric_calc(xgb1, X_train, y_train, X_test, y_test)"
      ],
      "metadata": {
        "id": "xCfGzA-WejL7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 415
        },
        "outputId": "99c6f064-7cdc-4618-e33b-6081d37da900"
      },
      "execution_count": 204,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[17:42:31] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "==================== Model Metrics - Train ====================\n",
            "   R2_Score         MAE           MSE       RMSE\n",
            "0  0.999847  190.604643  87407.024633  295.64679\n",
            "\n",
            "\n",
            "==================== Model Metrics - Test ====================\n",
            "   R2_Score           MAE           MSE          RMSE\n",
            "0 -0.120602  18059.543708  5.033818e+08  22436.170761\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x216 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5cAAADECAYAAAALZ9UhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3xb5bnA8d/Rtrwt7+0ktpydkEUGhBECYVN2Wb1ltJRCgbaX0t62ty10QCnQAoUOLruUsGcgKYFMCCGJs2w5y3vbsmVZ1j73D9nO8JJtLTvv9/PJx7HOeqUjyec57/s+jyTLMoIgCIIgCIIgCIIwFopwN0AQBEEQBEEQBEEY/0RwKQiCIAiCIAiCIIyZCC4FQRAEQRAEQRCEMRPBpSAIgiAIgiAIgjBmqnA3YJzRAguAesAT5rYIgiAIgiAIgiCEmhLIAL4CHMcuEMHlyCwANoa7EYIgCIIgCIIgCGF2GrDp2AdEcDky9QBmcxde7+hKuCiVEgAejygBE4nE+Yls4vxEPnGOIps4P5FNnJ/IJs5P5BPnKDQUConExGjoiY2OFZLg0mg0/hG4HMgHZppMpr09jxcBzwMGoBW40WQyHQjHMj95ALxeedTBpeR7z496eyG4xPmJbOL8RD5xjiKbOD+RTZyfyCbOT+QT5yjk+k0TDFVCn7eB04HKEx5/GnjSZDIVAU8Cz4RxmSAIgiAIgiAIgjBKkiyHLrI3Go0VwIUmk2mv0WhMBcoBg8lk8hiNRiW+3sRCQArlMpPJ1OznU8gHjrS3j35YrELhu6Ui7qhEJnF+Ips4P5FPnKPIJs5PZBPnJ7KJ8xP5xDkKDYVCIiEhGqAAqDhuWTga1CMHqDWZTB6Anp91PY+HepkgCIIgCIIgCIIwBiKhzyh4PKOfc3nsPoTIJc5PZPPn/DhdHlQqBYreCRhCSInPUGQT5yeyifMT2cT5CS9ZlrFaO+jutuL1isqAwaRSaUhMTEGpPD5kHGrgaziDy2ogy2g0Ko8ZpprZ87gU4mWCIEwgDpeHnzyzlUVT07jm7MJwN0cQBEEQhAAxm5uRJImkpDSUShWSuIkcFLIs09VlwWxuJjk5w+/twjYs1mQyNQG7gGt7HroW2GkymZpDvSx4z1IQhHD4cn8jHVYn67bXUN/aFe7mCIIgCIIQIE6nnYQEAyqVWgSWQSRJEtHRcbjdzhFtF5Lg0mg0/tloNNYA2cA6o9G4r2fRd4E7jUZjOXBnz++EaZkgCBOALMus215DZpIOjVrBa58eDHeTBEEQBEEIGBlJCmfamJPHaIL3kGaLnQDygSOtrdZRz7kUxV0jmzg/kc2f81Ne3c6fn9/M99rWYZk0k79Z8/nRNXOYlp8Uqmae1MRnKLKJ8xPZxPmJbOL8RIaGhkrS0/PC3YyTxkCvt0IhYTDEQIRlixUEQQi4dduruaBtGyqLmaTdmyhUdfHvTw+KtOSCIAiCIASNxWLhrLOW8thjfxx23Q0bPmP//r1jPuaDD/4vb7zx7zHvJ5BEtlhBECYMc6cD+/YvmWypJOn8C+nY8DmXdHzFH13L2by3ntNmZYa7iYIgCIIgBEhppZmXPjFR32oLyv4zDHquX2lkal7isOuuXbuG6dNnsG7dx9xxxw9Qq9WDrrtx42cUF09l2rQZgWxuRBDBpSAIE8amTfs5p/lLVAWTMVz6DdRpaTT+3z85u7CWNzdoWVCcik4jvvYEQRAEYSJ4YU0ZjebuoO2/vtXGC2vK+N13Fg+77gcfvMv3vncXL774HBs3fs5ZZ62gubmJxx57mJoaX3GKFSvOpaiomE2bNrB9+zbee+8drr76m8iyzJYtG3nggYcA+PDD9/p+P3ToII888nvs9m6cTicXX3wZV131zaA957ESV1mCIEwITqcb/cerUUqQfdt3kBQK4pYsw7JpIwtqvmBLejJrvqzi0tMmhbupgiAIgiBMIAcPHsBi6WDevAW0tbXywQfvctZZK/j1r3/O4sVLefDBhwFob28nISGBZctOp7h4KpdffjXgCyYHk5GRwWOPPYVGo8Fms3HbbTexcOFi8vMLQvLcRkrMuRQEYULYu/o9cqx1sPISNCmpgC/LWer1N4LDzhXuUtZ8WYW50xHmlgqCIAiCEAg3nldMhkEftP1nGPTceF7xsOu9//47nHfeBUiSxPLlZ7J//14aGurZu3f3cb2MCQkJI26D3W7n97//DTfeeDW3334zLS3NHDxYPuL9hIrouRQijqOuFlVcPMqYmHA3RRgnnA0N6D7/gJr4HJZfdv5xy7RZ2SSesxI+XkN6bjZvfn6Imy+cFqaWCoIgCIIQKFPzEnnw1lPD2gaXy8W6dWtQqzWsWfMBAG63e8jeyBMplcrjEg86nUdvhD/zzJMkJRl49tmXUalU3HPPHTidI6s9GUqi51KIGK62Nur/9jSVv/gZLW+9Hu7mCOOE7PFQ+czTuFAgXXwNSkX/rzXDRZeiSkziMsvXbN1TR2VDZxhaKow3sttN1YO/pvPr7eFuiiAIghChNm78nJycPN5660Nef/09Xn/9PR599Ak++eQjZsyYxWuvvdK3bnt7OwDR0dFYrda+x7Oycjh06ABOpxOXy8X69Z/2LbNaO0lNTUOlUnH48EFKSnaF7smNggguhbDzuly0ffg+FT+/H+uO7Uj6aOx19eFuljBOmD/+CLm6gvUZizn1VOOA6yh0OlKu+SZR7U0sth3g358eQNT4FYZjP3IY+5HDmD9ZE+6mCIIgCBHqgw/eZeXKVcc9NmPGLLxeL9/+9m3s2VPCDTdcxU03Xcv7778NwLnnns/atR/zrW99k48+ep8ZM2Yyf/5CbrjhKu6++3vk5+f37eumm27mvffe4qabruHZZ//GnDlzQ/n0RkwSF1gjkg8caW21jrpmnijAe5Qsy3TtLqH51VdwNTehnjGbzRmLiNn6CZPkdmY9/ljI2yTOT2Q78fw4qquofOBXlOlzsJx3DdefO/i8CFmWqX38UaymMp7Kupj/umoRc4tSQtLuk8lE+gy1vvcOre+8BUD+A79Hk54e5haN3UQ6PxOROD+RTZyfyNDQUEl6el64m3HSGOj1VigkDIYYgAKg4rhlIWuZIBzD2dBA7eOPUveXx/AqFJQvv4bfOuewocqJIzoetc2C7PWGu5lCBPO6XNT/42+4NVGsSV7IWfNyhlxfkiRSr7seJTIXWHby2vqDuD3iPSYMzla6H1VyMkgSli82h7s5giAIghDxRHAphJTX3k3z669R8cuf0X2wnIb55/CnxHN5u07L0pkZ/P47p5JblItS9tJS0xju5goRrPXdt3HW1rA2cykFkzLITI4edhtNSipJ519IQdshomoPsX5nbQhaKoxHXqcT++FDxM6bj37adCxbtogbXoIgCIIwDJEtVggJWZbp/GIrza+/hqejnc6iubwmGWlu17CgOIXLTp9EepIvlXRafhbujVBpqiIlNyPMLRciUffBA5jXfIhz1kJKbGncOS/b720Tzzsfy9YtXGj+ihc2ZLJkRjrROnUQWyuMR/ZDB5HdbvTF09Dm5tHw92foLjehL54a7qYJgiAIQsQSwaUQdLZyEy2r/439yGGcqdl8ULQckzeeGQVJ3L58EvnpccetnzE5m2qgsUL0Kgn9eR0OGv75d1QGA+8nnIJB5WH2lGS/t1eo1aRedwOuR//IzIYS3tucwzVnFwaxxcJ4ZCvdDwoFUYWFIClQ6HRYtmwWwaUgCIIgDEEEl0LQOBsaaHljNdadX+OJjmNLwRlsVuQwOT2e/14+meK8xAG30yYbALDUNYWyucI40fTav3G1NKO95S72rjdzxRmTUSikEe0jevoMYhcsZOnXX/PPrZM485Qs0hKDV4RZGH9sZaXoCiah0EUBEDN/IZ1fbSP1uhtQaLVhbp0gCIIgRCYRXAoB5+nspPW9d2j/7FM8ChXb005hY3QxGekJ3LmsgDmFyUjS4MGAQheFR6MDi5lOm5NYvSaErRcijSzLOOvr6DaZ6C4vo/OrbSSecy4ftetRqzo4fXbmqPabcvW1WHfvZkXzNl7/tIA7Lp8V4JYL45XX3o294ghJqy7oeyxuyVIsmzZg3fE1cYuXhLF1giAIghC5RHApBIzX5aR93Tpa3n8X2eGgJKGIDQmzmGLM5u4FOUzNSxwyqDyWMslAvMXKgZoOThHlIk4qsteLs7YGW08w2V1ejsfaCYAqMZGE5WegX3UxW/7+FYumphETNbr5kqqERJIv+wbyq6+wc+fX7JyZEfLSJDZTGc2vvUrmHXeiTjKE9NjC4Gzl5eD1HjcENmpKIerkFCxbNovgUhAEQRAGIYJLYcxkr5fObV9Sv/o1pA4zB/XZbMyfj3FeMfcvyCHLjyyeJ9KnpRJvPkJ5dbsILk8CLrMZ61dfYis30V1ejtfWBYDKYCB65iyijEaiiorRpaciSRIffVGF0+Xl7BEk8hlIwpln07FpI6sav+KNd1MpvGPlqIPVkeo+fJjaPz+G7LBj27+f+GWnheS4wvC6y0qRVCp0k6f0PSYpFMQuXkLb++/iamtDnZQUxhYKgiAIkeSKKy5Co9GgVmvwej3cdNPNrFhx7qj39+GH77Fly0YeeOAhNm36nJKSXdxxxw8GXb+zs5N3332T6667aVTHu+KKi3jooUeZNGnK8CsPQwSXQj9OWzebN+/H1e1AK8loFDJqSUYjyagUMmpkVHhR4UUpe2jfuBFVYw2N2iS+mLQK4+kLuX9uFnHRox/OqjEYSHDvY111ewCfmRCJZFmm9tGHcdbVoU5NI+aUeeiNRqKKjKgNxyfqkSQJryzz6Y4apmTFk5ceO6ZjS0olGbd+B+cffs8lB97nrbfiuOGbYw/y3t54GJfHy5VnDPwl7aiupvaxR1DFxeK2gL3yiAguI4itrBTd5CkoNMd/h8UtXkrbe+/Q+cUWks6/MEytEwRBEHrZykppeukFnA31Qdm/Jj2D1Otv9CuZ2wMP/IFJk6ZQXl7Gd797M/PnLyIhIQEAt9uNSjW6sGvZsuUsW7Z8yHWs1k5eeeWFUQeXgSSCy5OQLMt4OjtxNTfhamrC1dyEs/f/Lc14OjrIGmp7wNXzD8Cm0lMy6SwKzz+Le2ZkoFYpx9xGlcGA2uOksa6VboebKK14q05Uzvo6nHV1pFx7HYlnnzPs+nsOtdJk7ubS0woCcnxtVjZ59/2EQ7/7HdM/f5ld+YnMWTJj1PvbUFLHu5sriNapuGL55H5DwZ0NDdT86WEUWg3ZP/xvGv7vnzgqKsb4LIRA8VitOKqrMFx8ab9lmtRUogqLsGzZTOKqC/we5i8IgiAER+OLz+FqDF5ddGdDPY0vPkfBg3/we5uiomL0ej0PPvhLDIZkqqoqsdlsPPfcK3z00fu8+eZqPB4PMTEx/OhHPyE3Nx+Xy8Wjjz7Ejh3biY9PoLDQ2Le/Y3sxAd5//x1Wr34VALVazUMPPcqf/vQHrFYr3/rWN9HpdDz99LO0tLTw2GMP0djYgMPhYMWKc7nxxm8DUFKyk0ce+T0Ac+acgizLgXrJRHB5MrF8udVXG7CpGdlhP7pAklAlJKJOSaElpYDdSokpMws4ZXo2biRcsgIXEm5ZwumVcPb+9Eo4vJCcbuC/pqSgCOCFVm+PVZzLyqHaDmZMEvPRJirrzh0AxJwy36/1/7O9hvhoDfONqQFrgzYrm/z7f0r5Aw/CC09gTr2PxCkjD14P1nbw4scmorRKuuxuzJ0OkuJ0fctdrS3U/OkhQCb73h+jTk5Bl59P+3/WIbvdSKO8qykEjq3cBLI86F3q2MVLaHrhORwVR9AVTApx6wRBEIRIt2PHdpxOJyqVigMHynniib8RFRVFSclOPv10LU8++Xc0Gg1bt27md7/7NX/967O8884b1NfX8dJLq3G73dxxx61kZPSv9b5jx3ZefPH/eOqpf2AwJGOz2VAqldx7733ccssNPPfcK33rPvDAL/jWt25hzpxTcLlc/OAHtzN16jRmzz6FX/7yp/ziF7/hlFPm85//rOXNN1cH7PmLK5mThHV3CQ3//DvarGzil52GOiUVdWoKmpRUVMnJKNQaNu2u59kPS1l2RgZnrCoO6115VU9yk3i3jfKadhFcTmBdu3aizS9AnThwaZpjNbTZ2H2olYuW5qNSKgLaDn1WFgnf/yEtf3mEukf+QNRPfoIuL9/v7c2dDp58cw9JcVquObuQv7yxh5pma19w6W5vp+aPD+G128n58U/QZPiy3OryCpDdbhx1tehy8wL6nISR6y7bj6TRDBo4xs5fSPMrL9GxZbMILgVBEMIs7YZv0fTyizjr64Kyf01GJqnX3eDXuv/zP/eh0WiJjo7mwQf/wCefrGHatJlERflKWm3evIGDBw9w223fAnwjCTs7LQDs2PE1q1ZdiEqlQqVSce65q9i9e1e/Y2zdupnzzrsAQ08njF4/cBm17u5udu78mvb2o9PLbLYuKioqSEw0oNPpOKXnpv7ZZ5/Dww8/6N8L4oeICC6NRuOFwG8Aqeffr0wm05tGo7EIeB4wAK3AjSaT6UDPNgFfNlHZKyuof+YptNk55Pz3/Sh0un7r7Kto4/k1ZUzLT+TGc41hH+6lNviCyXy9m/IqMe9yonK3m7EfOYzh0m8Mu66128VfXt+NRq1k+ZyhBm6PXv70yey/5BZS3vknlQ/9gbwf/divAMLl9vDkW3uwOz388Jo5JMX6PmPVTVZmTU7GY7VS86eHcVs6yL73x2hzcvu21eb7ekjtFUdEcBkBbGVlRBUWDdqLrNTriZl7Cp3bviD16mvHXW9zzVNP4KiuQRkbizIuDmVsHMrYWFRxvp++3+NQxcWh0OuRFIG9iSMIghBI+uKp5P/mt+FuBnB0zmWvTz5Zg14f1fe7LMMFF1zMLbd8N+htkWUvkiTxj3+80G+u58GDA4U9gbvuD/tfDaPRKAEvAjeYTKY5wA3A80ajUQE8DTxpMpmKgCeBZ47ZNBjLJhxXawu1f34UZXQMWXfdM2BgWdNs5am39pBh0PO9S2cGvEdoNJRxcUgqFTkaJ4frLbjcnnA3SQgCa4nvrlzM3HlDrmd3unl8dQkNbTZ+cOUsEmODV8R+5co5rJ/1DSyyiupHHqb70MEh15dlmRc/KedwnYVbLpxKdkoMep2K5Hgd1U1WPN3d1Dz2CK6mRrK+/wOiJh+f5EedkoJCrxfzLiOAu6MDZ13tsIkb4pYsw9vVhXV3SYhaFhiy203n9u2gkECScNbV0bl9G23vvUPTyy9S//RT1Dz8eyp/8VMO3f19Kn52H7LXG+5mC4IgTAhLl57GmjUf0NTkmyPq8XgoKysFYN68+axZ8yFutxuHw87atWsG3MfixUtZs+YD2tpaAbDZbDgcDqKjo7Hb7bjdbgD0+mhmz57LSy8917dtY2MDra0t5Obm4XA4KCnZCcD69euw9pR8C4RIueXqBeJ7/p8A1APJwClAb4aPfwFPGI3GFHzhdUCXmUymZn8bq1RKjLZjT6EIXY+gx9ZF7eN/Qna6yPvZf6M19B92aO508NjqErRqJfdeM4fY6NCUYRiehNpgIIlu3B6ZysZOjLnDD5scq1CeHwG6du1AnZpKVE7WoL3lLreXJ9/ay+F6C3ddMYuZkw14vYGbeH4ipVLJdVct4o/mbm5q+g+1j/6RnHvuRV9kHHD9tV/VsGl3PZcsy2fhtLS+x3NSY6hvMFP3l0dxVFeRfeddxM6YPsAeJKLyC3BUHkGpHP/vv/H8GbIeKAMgZtq0Ic9F7MwZKOPj6dy6mYQF/s0VjgTOVjPIMsmrVhG/7PS+x2WPB4/VittiwdNpwW2x0LVvLx2bNuFtbUaTnh7GVp9cxvPn52Qgzk/k672UCGB+moCZM+cUbrvte/zkJ/fi8Xhxu12ceeYKiouncvHF3+DgwYNcf/2VxMcnUFw8HbO5td8+TjllPjfc8C3uvvt7SJICjUbNH/7wKElJBlauXMVNN11DbGwcTz/9LL/4xW/485//xI03Xg34As777/8FBkMy//u/D/LII79HkiRmz55LWtrQ3/Mn/k0c6rMQ9uDSZDLJRqPxKuAdo9HYBcQC5wM5QK3JZPL0rOcxGo11PY9LQVjmd3A5HshuNzV/+QvOxkZy7/0R2qz+9QDtTjeP/rsEm93NT2+YhyGuf69mOKkNyehtnaACU1V7SIJLIXQ83d3YSktJPHvFoIGlx+vl6Xf2su9IG7dcOJX5xYFL4jOU3LRYzjpzOs9+Ct+zbKDqkT+Sc/e9RE89vkertNLMK2sPMKcwmUtPP374bK5BR9Hn/6K7u4Gs795O7Ow5gx5Pl59P68dr8LqcKNSjL+EjjI2ttBRFlB5d3tDDkyWlkvhTF9O2bi3uzk5UsWMriRMqrpYWADQpx9cOlpRKVPHxqOLj+x7TpKTSsWkTjro6EVwKgiAM4/XX3+v32M9+9r/9Hlu5chUrV67q97harea++3424L7PP/8izj//or7fL7zwUi68sH9G8/vu+5/jfjcYkvnVrwYeMjx79lxeeOHffb/fe+99A643GmEPLo1Gowq4H7jEZDJtNhqNS4HX8A2PjUgejzzmnhOPJ3i3VGRZpuHZf2IrKyX95lvRGaf2O57H6+XJN/dS3WTlritmkZ0SE9Q2jYYyyYC9toSsadGYqtpD2r5Iey0mos7du5HdbvSz5w74esuyzP99VMb2smauObuQJTMy+j53oTg/qxbl8XVZM89HreBW7WdUP/oImd//AdHTfWVKWjq6eeKNPaQlRXHrhdOQveDB1y7Z42HKl++ittWjvOxaouctHLLNmrx88HjorqyeMElixuNnqKu0lKiiIrwoYJj2xy5eStvHa2jfutWvEjqRwNHsu4eqTEoe9vyo0n1ZCrtratDPnhv0tgnHG4+fn5OJOD+RKxJ7LCeCE9/zQ73OYQ8ugTlApslk2gzQE2B2AXYgy2g0Knt6GJVAJlCNrwcy0MsmjNZ33qJz6xYMl1xG3OKl/ZbLsszLaw+w+1ArN55rZNbkyMzEqjYY8HR0YMyMYXNpCx6vF2WQk0t0dbuwOdx9CVmE4LHu3IkyJrbfHETwvUdfW3+QTbvruXhpPisX5IS8fSqlgm9fMJXfPL+d9TMvZUXpe9T95THU6RnIXi/NZhvf9HgxxGlp+NVb4PUiy16QZWSXC3VnJ+uSFzAzZyaThzmWri+pT8WECS7HG1dbK66mRhLOPMuv9bXZOWhz87Bs3TJugktXSzNIEurERIabSanQRaFKSsJZF5wMjIIgCMLEFP7MLVADZBuNRiOA0WicCqQBB4BdwLU9610L7DSZTM0mk6kp0MuC+gxDqGPTBtref5e4ZaeRdOHFA66zZlsVn+2sZdWpuZwxNzhZNwOhtxxJcTw4nB6qm6xBP+ZLn5Tzy2e3YXe6g36sk5nsdtO1p4ToWbORlMp+yz/YWsnH26o5e142lywbeb3JQMlNi+WiJflsPmyl5ZKbiT11CerkZOq8ehoUccRNyidmUgG6ggJ0hYXojcXop04nevZcUr91M7tTplPT1DXscVRJBpQxsdgrj4TgWQkD6S7zzbfUF0/ze5u4xUtwVBzBUVcbrGYFlKulBVVSkt8ZbjUZmTjHyXMTBOFkIvlu5gpBJ4+iKzjsPZcmk6nBaDTeDrxuNBp73ynfNplMbUaj8bv4Msf+AjADNx6zaTCWjWtd+/bS+MJz6KfPIO36mwacx7attJHV6w+xcGoqly8frj8lvHrLkeRoXQCUV7WTnx4XtOPJsszeI610dbvZuLuec+aHvrfsZNF9oByvzUbM3P7D7dbvqOHNDYdZPD2Na1cUhr0szvmL89hxoJkXN1Tzm1uuZ/OeBl5bf5BvnD+JaUvyh9w269B2qpuGz8AmSRLa/HzsImNs2NjK9qOMiUWT5f8Nt9hFi2le/W8sWzaTcsVVQWxdYLhbW9Ekpwy/Yg9tZhbt5SZkr1eUJBEEIWJoNDra21uIjU1EqVSF/TphopJlma4uCyrVyHJBhD24BDCZTC8DLw/weBmwaJBtAr5sPHNUV1H/1yfQZGaR8d07+t2Z9ni9fG1q5h/vl1KYHc/NF0xFEeEfRlVPcKnr7iQ1IQpTdTsrF+YOs9Xo1bV0YelyoVYqWPtVNWedkhX0YbgnK+vOHUgaDfppM457/It9Dbz0STlzpiTzX+dHxntUpVRw8wXT+PVzX/HY6t1UNFiYb0zhgsXD16TMSY1hR3kzsiwP+8dPl59P24cf4HU4UGiDV2pF6E+WZWxlpUQZjSMKolRxcUTPmInliy0kf+OKiA/AXC0tRE/zv2dWk5mJ7HLham5Gk5Y2/AaCIAghkJiYgtXaQVtbI16vKFUXTCqVhsRE/29KQoQEl8LYuNraqP3zoyiiosi66x6UUUcLtnZYHWwoqeOzXXWYOx1kJUdz5+WzUKv6D0WMNOrEJJAkXK0tFOYYKTnY6tdF+miVVpoBuOqsyby89gBfm5pZOFVcUAWaLMtYd+1AP236cUFUycEW/vlBKcbcBG6/dHpE1FvtlZMaw0VL83l74xGyU6L59gVT/Xof5qTGsKGkjnarc9janLq8AvB6cdRUDzgPNRhcbg9KpSIigvhwcjU14W5rQ7/qwn7LvLI85OsTt2QZXbtLsJXu70v2FIm8LhfudjPq5GS/t9Fk+npxnfV1IrgUBCFiSJJEbGwCsbEJ/Zb1lswQSZfCJ3Ku3oRRq//bX/F2d5N11z2ok5KQZRlTlZmn39nLj57awlsbj5Bp0HPHZTP5328vICYqUmpZDk1SqVAlJOBubaUoJwFrt4u6VlvQjldaaSYlQceKBTmkJen56MuqUY01F4bmqK7C3dZGzJyjQ2JNVWaeensvOakxEXvz4/xT87jm7ELuvnI2Oo1/9+WyU6IB/JovrO1L6hOaeZcut5cf/3Ura7+aUPnMRsVm8hWx1p9Qaqa8up27/7yJrXsbBt02evYcFHo9li2bgtrGsXK3tYEsox7BsFhNRiaAmHcpCIIg+E30XE4A0TNmknL5lXhTM/l0Rw3rd9RS29KFXqvi7HnZnDE3i/QkfbibOSqqJAOuNl9wCXCgup2s5OiAH8frlSmvbmeeMQWFJFvTVlMAACAASURBVHHuwhxeWGPCVNVOcZ6orxlI1p07QJKI7qn7aO508MSbe0iO13HPVbOJ0kbm15JKqRhx1tqc1BgAqps6h83KrEpIQBkfjyNE8y4P1XZg6XLyVVkT5wZxuPl40F1WijI+AfUxRaTLKs08/vpuHC4P201NLJ4xcK1HhVpN7IJFWLZuxtPdfdzIkUjiavXVuBxJz6VSr0eVmDRuEhYJgiAI4Sd6LicA5+Kzef2wzL1PbualT8pRKRX816piHvn+Uq45u3DcBpbgS+rjbm0lNSGK+BgN5dXtQTlOdZOVLrubqflJACydkU6cXs2abVVBOd7JrGvXTqKmFKKKjcPrlfn7e/twe2TuvHwWsfqRTRqPdHqdGkOclprm4TPGSpKELi8/ZBlje4eBH6mz0GlzhuSYkah3vqW++OhQ530VbTy2ugRDvI5Zkw0cqOkYchRD3JKlyE4n1q+/ClWzR8zd4gsuNSMILsE371KUIxEEQRD8JYLLCeCx1SVs3F3P/KIU/ufG+fziW/M5bXYmWnXkDS0cKVWSAbfZN5zLmJOAqbo9KENVey+0p/b0UqpVSs6al83uQ63UNge/BMrJwtXSjKO6iuieIbEfflFJWVU7151TNK5vggwlJzWWGj/L6OjyC3DW1+O124PcKt97PlqnQgb2HWkL+vEilbO+Do/F0jckds/hVh5fvZvURD3//c25nFKUgrXbRUPb4EPydZMmo05Lo2PDZ8jeyEyP72ppBoUCVeLIRmJoMrNwNtRH7PMSBEEQIosILieAH14zlz99fyk3XziNSZlxEyols9pgQHa78VgsFGYnYO500NoR+Avv0kozGQb9cUlXzjolG41KwcfbxJy0QLHu2gVAzJy5HKzt4O2NR1g4NZWlMwcecjgRZKdGU99qw+UePqOdNj8fZBl7VWVQ29TtcHOk3sLpczKJiVKz53BrUI8XyWxlPfMtjVPZdaCFv7yxm8xkX2AZp9dQmB0PMOSoCUmSSDrvfOyHD9O8+t8hafdIuVpbUCcZBqwrOxRtRiay09k3rFYQBEEQhiKCywkgNSFq3CTpGaneciSu1haMPfMuTQEeGuv2eCmvbqc49/g7+jFRapbNymDrvgbarY6AHvNkZd21A01mJu54A8+8s4+kOC03nls8oW6InCgnNRavLFPXMnwyKl1ePkDQ510eqGnH45WZlp/EjElJ7D3ShvckTV7VXVqKKjmZ3W0yT761h5zUGH587dy+79T0JD2xejUHajqG3E/8actJOPsc2td+TPun60LR9BFxtbSgGuGQWKCv7qezNnTzLmWvF+vOHchud8iOKQiCIASGCC6FiKY2+C6G3K2tZKZEE61TcaAmsMFlRUMnDpenb0jssVYuyMEry6zbXhPQY56MPF1ddJebiJ49lxc+NmHudPCdi6ej10VmAp9AGUnGWFV8AqrEpKDPuyytNKNSShRmxTNzkoFOm4vKhs6gHjMSyV4vNlMZXen5/PXtfeRnxPLDq+cSrTt6s06SJAqzE/ya751y9bVEz5lL079exlqyK5hNHzFXS0vf9+lIaDIyAN/w4VDpNpVR9+Sfaf3gvZAdUwgvT3c3Vb97APO6T0SWdkEY50RwKUQ0VVJPz2VbK4qeizxT9dA9CCNVeqQFvbubydpubAfKcbY09y1LTdQzryiFz3bW0u0Qd9HHomt3CXi9HIzNZVtpE5edXsDkrPhwNyvo0hL1aFQKavycu6vNz8ce5J7L0gozU7Li0aiVTC9IQgL2HDr5hsY6aqrx2rr4pFnHlKw47r1qzoA3O4qy42npsGPuHHoEg6RQkHHrd9Hm5lH/t79ir6wIUstHxuty4uloH1Gm2F5KfTSqxMSQZoztfd3MH32As7kpZMcVwsdRXYX90EGaX32Fhr8/HZJ554IgBMfE7jIQxj1lVBQKvR5Xq+/CtygngV0HW+jochIf7X9m0fbPP/NdSHZ14emy4unqwmu14umyYuzuxgg0P7gaAHVKCgW/e7hv23MX5bLd1MzG3fUjLkUhHGXdtQMpNp7nd9sozk1g1aK8cDcpJBQKiayUaL96LsE3NLZr5w48NhtKfeCTHFm7XVQ1Wbn0NF9dzTi9hvyMWPYcbuXiZQUBP14k2/fpl8QCmkIj379qDlrNwPMRC3tLIdW0s3Bq2pD7VGi1ZN15N1W//Q21f36M3J/9HHXS0GVogs3d6kvYNJrgEnz1LkOZMdZRXYUyJhavy0Xzv14m6657QnZsITxcTY0AJJx1Nu3rP8VRU03m9+5Ek54R5pYJgjBSoudSiHi+ciS+ZBKFOb6ergMjmHfpamuj6cXn6Ny6GXvFEbw2G8qYWHSTpxBz6hI2G+ZQPW8l6bd+l/jTTsfV3IzHdnR+3OTMeIqy41n7VRUekTFxVLwuJ11791Cuz0KjVnHrRdNRKCbuPMsTZafEUN1k9Wu4ly7fF+A5gpTUp6wnM/K0vKS+x2ZOMnC4zoK12xWUY0aiz3bV0rijBKs+ke9ct2TQwBIgNy0GrVrJAT9HTagSEsj6wT3ITge1jz+Kp7s7UM0eld5kPKpRDIuFnoyx9XUhyxjrqKpCN2UKhosvoWt3CdZdO0NyXCF8XE1NoFSScvU3yb73x3g6O6l64Fd0RnB5H0EQBiaCSyHiqZIMuNp8d97z0mLRqBUjSurTe5GedfcPKfjtH8j92S/IvvteMm79Du3LLmRj4iwMK1YSt+hUYntKZDgbGo7bx3mL8mi1OPiqTAzRGg1baSmyw8FORQbfPn/qcVl5TwY5qTFYu110dA1fT7I3qY+9IjjzLksrzWg1SvIzYvsemznJcFKVJPna1MxLH5WS72giY/5sNMOUbVIqFEzKjKN8BPO9tVnZZHz3DpwN9dQ//WRYk9O4eob6j7rnMtOXMdbdGvyh016HA2dDPdqcXBLPPgdNRibNr76C13ny1mI9GTibmvqyGeunTiP3579Ck5lJ/V+fpHn1q8ie4bNtC4IQGURwKUS8Y3suVUoFkzPjR9Rzaa+sAElCm5Pbb1lppRlJ8g23hcGTV8yaYiDDoOfjL6tFsoFRqNqwFYekYtLSecwpHN0F7niWkxoD+JfURxkbiyo5OWjzLvdXmjHmJKBSHv36L8iIO6lKkqz5spLp2i5UHhcx06b5tU1RTgI1TVZsdv+DxOjpM0i7/kZs+/bS9MpLYfvucLW0gFKJKmFkNS57aTN9GWNDMe/SUVsDsowuNxdJpSL1m9fjamnGvObDoB9bCB9XcxPq1NS+39VJSWT/+H7izzwL88drqPnTw7g7AptvQRCE4BDBpRDxVIZkvN3dfUNVjTkJVDdZsdn9G8LnqKxAk5GBQtu/t6y0ykx+elxfEg9NSgoolTgb6o9bTyFJnLswl8rGzr5hheHUaLaN6CI3nMyWbhx7S2hIyuPKFcZwNycssnuCy5oRzLt0BCFjbJvFTmObrV/ZHYVCYnpBEnsPt074kiRVjZ0cqrOwJNqXHTfKWOzXdoXZ8cjAwdqRXeDGn7acpPMvpGPDZ5jXfDTS5gaEu7fGpWJ0f/I1GZkAIZl36aiuAui7GaifOo3YBQtp+/B9kdxngpJlGVdT43HBJYBCrSbtuhtJv/k27EcOU/mbX9J98ECYWikIgr9EcClEvN5kGO62o0l9RnKRZ6+qRJvbP3mM3enmSJ3luBIkkkqFJi1twLT7i6enERetYc226lE8i8Bxe7w88Px23thwKKzt8IdXlnnjX58R7e6maMUy1KqRFXCfKKJ1apLitP4n9ckv8M39tfq3vr9Ke+db5vfvwZo5KQmLzUVV48QuSfLZrjrUKgVpllo02TmoYuP82m5yZjwKSRpVKSTDpd8gdsFCWt54jc7t20a8/Vi5WlpGPSQWQBkdjTI+AWcoei6rKlHo9cfND02+6lpQKml+9ZWgH18IPW9XF97ubjQpAyfLilu8hNz7f45CraH64d9jXrdWjCAShAgmssUKEU9l6ClH0tqKNjuHSZlxKBUSpup2Zk0e+oLJ3dGOp70dXW5+v2UHajrweGWK8xKOe1ybkYm9tn9dS7VKydnzsnlrw2Fqmq1kp8SM/kmNQUVDJ112N4dG2IMSDh9/WYX6wD5kSUH2koXhbk5YZafEUO1nOZLepD72ygqip88IWBvKKs3ERKn7elKPNaPA9znbc6iV/HT/Aq7xptvhZuu+BhYVJuFac5D45Wf6va1WoyQvPWZEQ/J7SQoFad++BZfZTMM//oYqMYmoyVNGvJ/RcrU0Ez1rzpj2oc3MwhGCWpeO6iq0OblI0tGEX+rERAwXXULL669hLdlFzOyxPRchsjh7MsWe2HN5LG1ODrk//yUNz/6D5ldfxt1uJuWKq0LVRCFI3BYL7nYzssOJ1+lAdjrwOhx4nU5kh+//stO3TJebT9ySpeFusuAHEVwKEU/dE1z2zrvUqJUUZMT5VdTcUdUzxCqvf89lWaUZpUKiMOv44FKTkUHnzh3IbjeS6viPyJlzs/hgawUfb6vi5gv8m6sVaL3Dcmubu3C5PRHZG9jc3s3a7dWs31HL7e469MXFKKOjw92ssMpJjWHfkTZcbi9q1dCDRnp72h0BDC5lWWZ/pZni3AQUUv9MvXHRGvLTY9lzuI2Llk7MkiRfljbicHo4LdmF7HKhL546ou0LsxP4dEetX+fwRAq1hqw77qLqdw9Q95fHyf3lr1Enjm4O5Eh4nU48FsuYei7Bl9SnY9MGZK931MNrhyN7vThqagYM+hNXrMSyeRPN/3oZ/dRpKDT+l6ISIpurZ7jzUMEl+GquZn7vTur+8hid27eJ4HKc83R3c+QnP0IeLlmXQoGkUNAuy0QVF4e9tJMwPBFcChFPGRuHpFL1ZYwFX0mST7ZV43B50A6R6bG3GPdgyXwmZ8b1K0GgzcgArxdnUxPazMzjlsVEqTltViaf7azlG6dPDkvW07IqX3Dp8cpUN3UxKTNyepkO1nbwybYqvi5vRiFJLM9VE1PeRsycVeFuWtjlpMbg8crUt3aRmxY75LrK6GjUqWkBzRjbZO7G3Ok4bhj4iWZOMvD+1gqs3S5iotQBO3YkkGWZz3bWkp0SQ5K5llYgqqhoRPsozE7gk6+qqWiwUJidMPwGJ1DGxpJ5+x1U/uoXdO0uIWH5GSPex0j13pQbe3CZhexw4G5rRZ2cEoim9eNsaEB2OtEN8H3dm9yn5pGHMK/5EMPFlwalDULouZp6gks/3leSQoGuYBJde/fgdTlRqMVNhvHKUV2F7HRiuOQydJMmo9BokLRaFBotkkaDQuv7KalUuNvaOHL/j2lf+wkpV18b7qYLwxBzLoWIJykUqJKOZowFX1Ifj1fmcJ1lyG0dVZWoU9P6FaPvsruobOykeIAL7b7kFYMMAVu5IAevLLPu69DPvXS5vRys6WDOFN+FYmXD0M8/FLxeme1lTTz44nZ+++LX7K8ws2pRHg/dvoTzE33z92J6SryczEaSMRZAl58f0Iyx+3t6vKfmJw26zszJBmQZ9ldMvJIkFQ2dVDVaOXNuJo7qKtQpqSj1I+tN76uzWzP6Iema7BwUOh2OmtB8f7haeoJLw9gCwt4bbY4gJvXpS+aT2z+4hGOS+3z0gUjuM4G4mppQJSb53RutTksHWcbV1BzklgnB1PsdGLfsdKKnzyCqsAhdbh6a9HTUSUkoo6NRqNVIkoTaYCB24SLaN3yOp6srzC0XhiOCS2FcUBsMuI6psTYlKwEJhp3/ZK+qRDfAkNjyqnZkmQF7cbS95UhOyBjbKyUhivnGVD7bWUe3I7QZW4/UW3C6vSydmUFMlJojDeFLvtLtcLN2ezU/eWYrT729F0uXk2+uKOSPdyzhijN8vbrWnTvQ5uahHmXx9okkNTEKtUrhd3CpzcvH3daK2xKYGwillWYSY7WkJUYNus6kjDiidSr2HJp4JUnW76xFq1Zy6vR0HNXVgwYwQ4nTa8gw6P0akj8YSZLQZGXjHGBedzD0BpeqsfZc9mWMDV5SH0dVpS+pWnrGoOskX3kNKBQiuc8E4mxqRJ3i/80PTVq6b7vGhmHWFCKZo7oKRUwMqgT/RoEknXc+ssNO+2efBrllwliJ4FIYF1RJBlxtRy949ToVOakxmIa4yPNYrbhbWtAOkMyntMqMWqVgUmZ8v2UKnQ5VYtKgPZcAZ8/LptvhDnldwLJKMxJgzE0gPz2WivrQB5eWLier1x/kR09t4V/rDpAQq+WOy2bwu9sWs2J+DjqNb7S922LBfuig6LXsoVQoyEqOpmaESX0cPUO7x8Iry5RVmpmal3hcopQT9ZYk2XOkbUKVJLHZXWzb38iiaWloZTeu5ia02Tmj2ldhdjwHazrG9Ppos7Jx1NSEJOOlq7UFSaVCFd//u24klDExKOPjg1qOxFFdhSYzq99c92Opk5IwXHQJXSW7sO7eFbS2CKHjq3E5cKbYgajTfOu6RHA5rjmqq9Fm5wz5N+lY2uwc9DNm0b5uLV7XMPM0hbASwaUwLqgNBjwdHcjuoz2FRTkJHKrrGLTepb2qEhg8mU9hdvygSTk0GRk4Gwb/w5WfHosE1LfaRvAsxq6sykxOagwxUWryM2Kpa+nC6fKE7Pher8xjq0tYs62KGQVJ/OzGefz0+nnMM6aiUBz/B8J++BDIMvoAZjsd77JTY6husvoVVGhz80CSAjLvsqbJirXbNeR8y14zJxmwdDmpbgxsGZRw2rqvEafbyxlzM3HU1oAsDzgP2x+F2QnYHG7qmkc/NEuTnY3X1oW7ffQ9oP5ytbSgMoy+xuWxtJlZOILUcynLMo6qqgHLRp0occVKNOkZNP/rZXGROc557d14LBY0wyTzOZYyKsp3o0MEl+OW7PXirKsd8fdw0nmr8HRasGzZHKSWCYEQEcGl0WjUGY3GvxqNxgNGo3GP0Wj8W8/jRUajcavRaCzv+Vl4zDYBXyZELpXB4JtjcUxSn6UzM3C7ZV5bf3DAbRyVvuBSd8LFiqXLSU1z15AX2pr0DJz19YMGARq1EkO8jvrW0I39d7k9HKy19M0TzU+PwyvLfg+zDIQNJXVUNHRyy4XTuP3SGUweoOe3V+/wOU1mVqiaF/FyUmLotLmwdA1/QayMikKTlt6XlGoseutb+hNczpjky8S3O8S98sHSm8inICOW/PS4o/P6ckbZc5njG8JVPop6l720WdkAOGuDP+/S1dIcsGHpmoxMnPV1Qelxdbe347F2+jVcWVKpSL3uBlzNzZjXfBTwtgih42r2zZscLlPssdZtr8YWnYirsTFYzRKCzNXoS9410hEkUcZidAWTMH+8BtnrDVLrhLGKiOASeAiwA0Umk2km8POex58GnjSZTEXAk8Azx2wTjGVChOpNPe0+ZmhsXnosKxfmsKGkvu/i+ViOqgpUBgPKmONr+vVmWx0omU8vTUYmssOO29x/v70yDNE0hLDn8lCtBbfHS3Fub3DpyzhaEaJ5l502J298fghjTgKnTht+CJOjrhZVUhLKqMHn+J1sRprUR5ufH5Cey9JKM2mJUSTF6YZdNz5aQ15aLHsnSHB5oKaD2pYuls/x3eRwVFej0OtRjTKdfUq8joQYzZiS+vQGl46a4M+7dLe2BCy7qybraMbYQHP0jDTR5Qzfcwm+5D4x8xfS9uH7fQGKMP701bhM8S+4/M/XNbyy7gCHHdohRxcJkc1R7buxNtKbfJIkkXjeKlxNjVh3fB2MpgkBEPZSJEajMQa4Ecg2mUwygMlkajQajanAKcA5Pav+C3jCaDSmAFKgl5lMJr//OimVEn4OEe/nxKGDgn+0qb477x5zK0rl0dfwG8snsaO8mefXlPHArYuOK0viqKokKi//uPUBTFXt6DRKJmfFoTzhfPSeH12WL3mFu6keXcrAF6FZKdGYqsxICgasGxhopup2JAmm5ieiVEokJ+iIi1ZT2djZ7zkGw5sbDtPt8HDjKiMqP2r8uerr0GZmBbRt4/3zk5fhuyFQ29LF7MLhe5P0BZPo/GIrXkv7qGsiuj1eyqvbWTwj3e9zMWuKgQ+2VGJ3uYnWjawkSaSdow0ldURplSzpef7Ommp0OTl+vYcHJlGUk8CB6nYUCvyeL3QsZXwsqoQEXHU1Qf3seh0OPJ2daFKS+44zlvMTle0L0N0NdehSA1uOxFVTBZJEVF6O369J+rXXcmhPCW3vvU3mrbcFtD3hEmmfn7F647ND7Dncyvcvn0lyfP8bjZ4W36WXLj1t2PP+5f5GXllbTnSUinpzNFM7LeCwjTjr81hMtPMTLs7aalAqicrOQjHC78D4+fNpTUvD/PGHxC9c0O87WJyj0BjqdY6EnsvJQCvwS6PRuN1oNH5mNBqXATlArclk8gD0/KzreTwYy4QIpkpMAknqy3zYS6tW8u3zi2kyd/P2xqM9PJ7ubpyNjejy8/vtq7TSjDE3AeUQc5C0mT0ZY+sHzhgLkGHQ43R7ae2wj/DZjE5ppZn89Fj0Ot89IUmSKEiP40h98MuRHKrtYMOuOlYuzCE7JWbY9WWvF0d9PdosMST2WDFRapJitf6XIynIBxjT0NiK+k7sTg/T8v0PTmdNNuCVZfYdGd8lSTptTr4qbWLpzAy0GiWy14u9pnrU8y17FeUk0NbpoNUy+s++NjsHe5B7LvvKkIwxU2yv3iHujtrAz7u0V1WhSU0d0UgHdVISMTNnYjt4IODtEcbuP9treHdzBRUNnTzw/NfUDpDMzNnUiDImtl+5sBPtO9LGM+/sozAnge9cPB2z2lff2dkghsaOR/bqarTpGSjUI6+nLCkUJJ23CvuRI9hMZUFonTBWYe+5BJTAJGCnyWT6sdFoXAS8B1wZ3mYNzuOR8XrHNufE45k4mRhDQqFCGRePs7W132tXlJPI6bMzWPNFFQuMqeSlx2I7UgGAJifvuPXbLHYa2mwsn5M59DmIjkMRFYW9tm7Q9dISfX8Ma5q6SIodfrjhWDhcHg7VdnDOgpzj2pOXHsvuw63Yut1oNcoh9jB6Xq/M82tMxMVouGhJvl/vXWdzM7LTiTo9Iyjv9fH8+clOjaGqsdOv56DOygVJwnb4CPqZc0Z1vN7hrUXZCX6/bvnpsei1KkoOtDKvyP+5UMeKhHO0YVc9Lo+X02f7Pu/OxkZkhwNNds6Y2jclyzfXuKyincUzRvfZ12RlYSsrxe10IymD89m19xSnVyYl93u+o3n+UlQ0yri4Ib8XR8teVYk2z7/vl2NpsnLo3L4dV1c3Cl1wv4dDKRI+P2Ox+1ALL35iYs6UZC5ZVsCjq0v47Ys7uOeq2RRkxPWt52hsQp2aOuTzPVJv4fHXd5NhiOauy2cC0Kbx7cNeX48mryC4T2YA4/38hJujupqoIuOoX8eYU5egfPNNWj74AF1h8YDriHMUXENNvY+EnssqwI1viComk+lLoAXoBrKMRqMSoOdnJlDd8y/Qy4QIpzYYcLcOPNfnyjOnEKtX838fluL2ePuS+ZyYebB3vuVwiU0kSfIlrxik1iX4ei4BGkKQ1OdgbQcer9w337JXfnocsgxVTcGbd7mhpI7Khk6uPmsKUVr/7kc5a0Uyn8HkpMZQ32rD7Rk+GYFCq0WTmYVjDPMuSyt9GYZj9f4VKAdf2RRfSZLWESVvcXu8rN9RS11L+Itcy7LM57tqmZId39fb3jfPJ3tsPZfZKTFEaZUcGFNSnxxktxtnEJOSuAPccwm+z3Sga116bDZczc39kq/5ozchiCNEdUOF4VU1dvLXd/aRmxrLbRdPIy89lp9efwo6jZKH/rWT0oqjIyJcTU1DzrdsaLPx6GslxEapuffq2eh1avQ6NerkVGSkoH5+hODwWK24zW2jTqoGoFBrSFhxDra9e/q+14XIEfbg0mQytQDr6ZkHaTQai4BUoBzYBVzbs+q1+Ho3m00mU1OglwXzOQqBoTYYcA0SXEbr1Fy/soiqJisfb6vCXlWBMiGhX2230koz0ToV2anDD+30ZYwdvKZbrF5DTJSa+rbgJ/UpqzSjkCQKs49/Pnm9SX2CVO+yN4lPcW4Ci6b6X4es93XrLbwuHJWdEoPHK/sdgOny8rFXVIwqQ6fT5csw7E+W2BPNnGSgw+r0ewiv0+XhiTf38NxHZdz/zBc8vroEU5U5JLUcB1JWaabR3M2Zc47e4HDUVIFCgSZrbO9LhUJiclY85WNI6qPJ7s0YO/agyOH0cKiuf1tcrS1IajXKuLHVuDyWNjMTR11gM8Y6anqTe4w86O+9QO3NAiyEV5vFzmOrS4jWqbjrill9dY9TE/Xcf/08kuN0PLq6hB3lzXhdLtzmtkEzxZo7HTzy6i4kCX549RwSYrR9y7IzE+jUxIhal+PQWD7vx0o44ywkrY62jz8MRLOEABpVcGk0Gs80Go3LA9iO7wI/NRqNe4BXgRtMJlN7z+N3Go3GcuDOnt+P3SbQy4QIpkoy4G5rHTT99DxjKvOKUnhnUwVdRyr63QWXewrJF+cm+pWAR5ORiaejA49t8CAgw6CnPgS9NGVVZgoyYvv1HCbGaomP0VDREJx5l298fgi708N15xSNKHGJo64WVWLSsPNoTka9NzZqBph/NBBdfj6eTgtu88jnPx6s7cDt8Y4yuEwCYI8fWWPtTjePv76bPYdauXZFIZedXsDhegt/eGUnv35+O1/ub8QT4rTx63fVEa1TMb/4aOIZR3U1mvR0FGr/e3EHU5SdQF1LF9bugevsDkeTkQkKBY4xlCNxe7ys31nLT57ZyoMvfE3lCZmj+2pcBjDhmCYzqyeTduDm4/Zmih3NxabKkIwiKiokmXeFoXU7fN8DdqeHu6+YTWKs9rjlibFa7rvuFPLSYnnyrT18sWU/yPKANS677C7+9NourHYX91w1m7Sk4/+W5KXF0KyKxT5EXgQhMvWVgxphGZITKaOjSTh9OZ3bvsTV2jL8BkLI+DXGzWg0fg781GQybTYajfcBVyJB+wAAIABJREFU9wJuo9H4pMlk+u1YG2EymQ4DZwzweBmwaJBtAr5MiGxqgwHZ7cbTaUEVnzDgOtetLOLAM5vwNDagWbDguGXNHXZaLQ7OW+TfhbYm42hSn6jJUwZcJ8OgZ+eB4H6p2Z1uKuo7OW/RwBdeBelxQSlHcqi2gw0l9Zy7MIcsP5L4HMtZV4cmU/RaDiQ9KQqVUjGCciS++UT2ioq+kjz+Ku3p8S7KGfjzMpT4GC25aTHsOdTKBYvzB13PZnfz2OoSDtV1cMuF01g22/e5OXdBLlv2NfDxtmqeeXcfr3+m45wFOZw2K8Pv4dWj1dHlZGd5M2fPy0atOiaDdHU1UVMCU9q4dxTBgZp25haOPHOqQq1Gk5Y+qqBIlmW+NjXzxobDNLbZmJwVR0eXk71HWvtGM4AvuAxUjctevUPdnXW1I34/DsZRVYUyLg5Vwsjfp5Ikoc3OET2XYebxenn6nX3UNndx95WzBh0dFBOl5kfXzOWJt/awfl0JV9K/DInT5eHPr++msc3G3VfOJj89rt9+8tPjKFHH4Ww8gizLAb2BIgSXo7ra93mPH/uIioRzzsX86TrMaz8m9ZrrAtA6IRD87bmcAXzR8/9bgTOBUxG9fkIIqXoukgYbGguQEKPl6unRSMgc8sYet6xsBIXk4Zjgcsh5l9F02lyj7r3wx4Gagedb9spPj6Wh1Ua3wx2wY3q9Mi99Uk5CjIaLl44sWYLs9eKsrxPzLQehVCjISo6mxt/gMjsHlMpRzbssrTRTkNm/x9tfMycZOFhrwWYf+P3daXPy8L92cqTewu2XzGDxjPS+ZRq1kjPmZPHgrYu46/JZGOJ1vPqfA/zoqS2sXn8Qc6djVG3yx6bddXi8MsvnHL3B4bFacbe1jmmez7EmZcahUkpjqnepycoe8bDY8up2fvvi1zz19l4UEtx5+Ux+ev08slNi2F9xfF1eVwBrXPbS9gx1d9YNPmVgpBzVVWMaIqfJzvl/9s48vLGzvNv30b5Y3uRFsuV9kWf3rMlMtkkyWSYhDckAARJICNBCW7rBx9KPljZtgRa+thRoaYEAIVAgGwQy2UkmM5nJMotnt7xb3m15kS1Zu873x7E8iyVrsWTPJLqvy9d4pKNzXkvnHL3P+zzP74evvz9rqr5CiKLIz19s52TXOPfd0sja2sUXHdQqOX+2Zz3rpeIIXuj0zJdZR4LUjn4nn7xjDaurC6Puo7I0hwllLoLfR8iZ+jWYZfnxpUGxO4KysJDcK67E+do+Qq7EvlOzZJ5Eg0sZIFqt1jpAsNlsZ2w2Wx+QmvFaliwpEFkljyXqE6FJJd1gnuoIXDCBPds7SZ5eNS/EE/d4RcUICkVcOxKAoQyK+rT2TiKXCdRboq/yVZsNiEgiCuli3/FBekdmuOeGhqQDk+D4OKLfPz8JzbKQipIc+sYSO2dkKhXqsvKk7UhmvUG6h6ZZVRV9cpYI62olS5KLgxYAp8vHv/z8GAMON5/Zs44tTdH7pmSCQHNDEV+8dxNf/ugW1tYU8txbdj7/Xwf59f6ulMcWi7Aosq9lkKbKfMzGc/536erziaBUyKk259LetwRRH4uFwNgYYW98S5OBMRffeuw4X//ZUcanvTywu4mHPr6NjQ3FCILA6uoC2vud+AMhAMJeD2GXK61iPgBygwG5IRdfmkR9xGAQ3+DAAvG1ZFBXVCD6vAusqrIsD8+/1ccrxwbYfUUlO5sTW1RUKmRsK5URUqj4zdExfvpCm6RM/qyNlg4H993cyNYY9xSQNA+C+dKcwJ/tu7xsEINB/IMDSy6JPZ+CW25D9PuZeuXltO0zy9JINLg8AHwH+CbwFMBcoJm9k2dZNhRG6YskMLF4cOnr7UXQ5zAlaHn0BRuiKCKKImd7J2mqKki4fEaQy1GWlC6auTTNTV6HxjMn6tNqn6S2LBe1MrpdQdVcyVC6SmOnZ/08OSfis21V8jYUkUmnKutxGRNLSQ7Tbj9Otz+h7dXV1Xh7upMSUWnrm0IUE8/UR6OuPBetWsGJi/oux51evvazozicXv7y/etZX5dYAFNblsun37uWr//RdjY1FvP06z2c6o7f05kMp7sncDi97Nx44fl3LrhM36SmwZJHz/AMvrmALlnU5ZKoz2JKpxPTXh5+5ix/+/BbtPU72XNdLV/7o+1cu6HsAq/e1dUFBENh2gekLM68x2Way2IBVGVlactc+gYHIBRCs4SgP6L+G/mMsywfR2yjPPZKB1usxezZWZfUa4Njo+jMJm7bXs2rxwb48g/e5MDJIf7gqmqu32SJ+3q9RbrGA1nF2MsG/8gwYjCY1vuwurwc/foNTL38EmF/Yt+pWTJLoimJB4DPAmPAN+YeawK+lYExZckSFblOh0yrJRincdtn70FbXc17d9Ty2CudHLaNUV6kZ9rtT3qirTKbF52wFOVqUCpkGctcznqD9AzP8J5Fet7y9CoKDOq0BZdPvDon4nOzNaU+lsikM1JWnGUhFRFRn1EXeTXxM4ua6hqm979G0OFAWZxYmeOZ3gmUChn15Qv7lRJFLpOxprqAU13j831NI5OzfPN/jzHrC/HZe5pjZtQXozhfyyfes4r+MRc/ebaVhz5+Rdr6MF89NkCuTsmmxgvfJ19fH3JDbsx+7VRotOTz7Bt2ugenaUohiI8oxvoG+qP2db94uI/HX+1EFEVu2lLBe3ZUk6ONbjreWJGPXCZwtmeSNdWF88GlIsWy2M5BJ88c7CUYChMKSwt04bBISBRpnlZSNdrFV374JmFRyhbrNUry9Cpyc1Tk6SM/kuBYnl6FQadCqVi4nu2zz4l7pJi5PNw6StCnoFAQ8PXZMWzanNJ+Isx6A/zqlU42W4tZF6e8891O1+A03//tGWrLcvnEe1YnJJR3Pv7RUdQWC+/bWYdeq+CxVzq5fmM5d16dWCtGaXUZQUHG7OAg6dNDzpJJ5hf50pi5BCi49Tb6/+VrTL++H+OuXWndd5bkSejb3GazjQN/fdFjz2RkRFmyLIKiMLYdCUA44Mc3OEjhug3cvLWCt86M8rMXbNywWZrEJTsBVJnNuI4dJRwIIFMunNTJZAKlBbqMZS7b+qXsU7xxV5sMaQkuOwec7D8xxK3bKikv0sd/QRT8gwMoCgqQ61J7/bsBS7H03vSNuliTSHA5ZxLu7e1OOLhs7Z2kvjzvAkGbVFhXa+SwbYz+MTcymcA3f3GMUEjk8x/aeIF4TLIoFXIevG0VX330CI+92slHb7EuaZwgZflaOhzsvqIKhfzCQEbq60vvhKbekoeAJOqTSnCpNBYhqNX4o4j6dA46+cXL7aytMfKRmxspytcuui+NSkFtWS5neiaAunn1xFTLYp9/086Z3gksxTnIBAGZTPpRyGR484pQDZ2kXB0koMsFQVoIG5qYpdU+idsbvf9br1Gw2VrMA7tXzT/m67MjqNUx7SgWQxRFHn3BxvRsgD/PKViyqM/o5CzfevwEQ+OznOwa56t/eGXMipF3O2NTHv7j8ePk6lV8Zs96VEm+T2I4TMAxRs7GTQDsvqKKK1ebyM9RJbyoWWXKZVKZi6JvgOxS5uWBr68PQaFAZUrvJ6ZtaERTW8fkC89ReP31CPLsdbuSJKoWqwK+jOQLWQYMIlmG/JPNZovfLJIlS5pQGiU7klj4B6QSK3VlFXKZjI/d1sRDPz7Mbw50Y8zVUJynSep4KnMZhMMERkdRxyjzLCvS0T2UGSuQ1t5JFPL42adqcy7H2h3MeoPoNKllgM4X8bnjquqU9gHgGxrM+lvGwaBTkZ+jSlgxVlVejqBQ4O3pwbBlW9ztp91++sfc7LkucW/SWETEOZ57087JrnHkMoEvfHhj0grC0agrz+OWrZU895adLdbimOIdibL/xBCIcG3zheefGArhHxwg/8b0rmjrNUrKi/Up+10KMhnq8vIFZbGBYJgf7W0lP0fNp+5ck3BWd3V1IU8f6MblCRB0OBBUKuSG5BcAAsEwJ7sn2L7GxP23Ni14ftampb/1Fe5tzkW/dt2C54Oh8HzZt9Ptl353+TjdPcHrJ4f50I2NqFXS5M9n70VtqUCQJe+MNjrlYXo2QJXJQM+wgbCtk3xvAL0menZ3MWz2Sb771ClEUWTPdbU8sa+Llw73LaqU/G4lFA7zH0+cIBgS+fyHN5CrT97aJzgxDqHQBYsKF1uXxKPKZGC/Mpe8bFnsZYOvz47KXIagSK9iuCAIFO6+jcHvfpuZI4fJ3ZY1hVhJEr2b/xdwA/BnwNa5f3cC/5mZYWXJEh2FsWjRzKW3d84vrUoqsaosNXDrFZVz2b/8pMs8I6tr/uHY/UWmQh2OKS+BYGp9V4vRap+kvjw3bvapZi6D1LsEUZ99LQP0jszwwRuTF/GJIIbD+AcHsjYkCVBRYkg4uJQplagsFXi7ExPAabVLAjypZNMupsCgpqIkh0Onh1EpZXzx3k1pCSwjvPeaGkoLdfxob+uSFI+Hxt28dLiPNbWFlFyU5fMPz/X5WNIj5nM+DZZ8OgacKft4qsot+Ab6L+in/e3BbgYdbh7Y3ZTUtbi6ugARKVCK2JCkUtp+tncSnz/ExoboWc/I9R2r71Ihl1GYq6HGnEtzfRHXbijjjqtquG17NaGwSNegFIyL4bCUUa5M7XNp75P284nbV1GxzoreM80/P3ww6TaFAyeG+OYvWsjRKvny/Vu4fXs1zfVF7H2jl5nZbA/Xxbx1dpSBMen8LEuxwiUwNgaAqiT1BbD8HDVuXT4KZ2z/6yyXFpJSbHorSCLoN2xEWWpifO8zSekTZEk/iQaX7wXeY7PZnp1Tin0WuHPu8SxZlg2l0Uh4dpaQxxP1eZ+9B5lOd4H8/h9cVc32NaVcl6CK3fnMB5eLKsbqEYHhiehjShWXJ0DfiCumBcn5RMoTe4ZTy6B6fEGefK2LVVUFiyr0xSM4ISnFZm1I4mMp0TM07iYYSmxSpKmpwdfbk9Ak6kzPJFq1nOo4Zathr4dwIL6Nzs6N5dSYc/nivZsWmJkvFZVSzoO3NTEx7eXxfZ0p7cMx5eGbv2hBLpdx702NC5739c0tOqUYxCxGQ0UePn+I/tHU+q7V5RWEXa55O4Xe4Rn2HrJz1TpT0j1/NeZc1Co5Z3om52xIUiuJbWkfQ62Ux+xRVxhykRsMSSvG1pfnzpURnxMdCnu9aCpS67fsGJhCp1ZgLtLTuFkqtdU6x/jHR44kJBQVFkUee7WDh/eepbEin//70c2UFkjn956ddXj9IX53sDelsV1MyO1+R0x4RVHk2TfslBXp2WRN3ebGPyplGy/2uEwWRUkpMjE8Xwb+TiYwNkbHZz6N+8zplR5KSgSnpwk5nRlZ5AOpEqTwlt14e3uZPXsmI8fIkhiJBpfDwMUzCi0Qe8adJUsGmLcjiVEa6+3tRV1ZdcFqvUop55N3rKG+PPmWf5lajaLQuCJ2JG19U4gkln0y6FQYczX0pth3+dbZEdzeIHdfW7skM+rIZFNtzgaX8agoySEUFhlOsF9XW1NH2OtdVL04QmvvJNaKggvURC9GDAbp+uL/ofMzn8b+Tw8x+vOfMn3odSnLd1EAe/3Gcv7m/i0U5S3e95cqDZZ8btpawStHB+b9aBPF6fLxzV+24A+E+Nw9zfPBwfnM9/mUmqLsYWk0WiSBoLb+1CxJ1OeJ+gRDYX74zFkMeiUfvLEh6X0p5DKsFfmc6Zkg4HCgSCG4DIsixzocrK0tXLRiQmUuwz+UnGKsTqPEUpJD+9x7tdSgv73fSb0lD5kgzFvMfGRDDsZcDf/+qxO8dLgvZkDn84f47pMnefYNOzuby/jLD2y4oJy2vEjP1evM/P5oP2NTS1s4DE5N0fW5v2DmjYNL2s+lwMmuCfrHXNy6rTJpAZ/zCYyOIigUKAqWVl1hmFOMnR1In+/qpcrM0cOEPR4mn9270kNJiUhPdKYylwCG7TuQ5+Yy8cLzGTtGlvgkGlz+FHjOarV+0mq17rZarX8I7AUesVqtN0R+MjfMLFkk5u1IoqxSisEg/v4+NFWp+6VFQ2U2LzqJMhXqEEi/HUlr7yQqhYwac2Jqn9VmAz1DqQWXB04MUVakp7YsdWVROE8pNlsWG5eKudLSREtjNTVzoj5xSmMdTg+jU564ysheey9hlwvduvUISiXO1w8w/MPv0/PlL9L5F39K/79+A8evn8DVcozgMpiU33VtLSUFWh7eexafP7ESc7c3wP/7ZQtOl5+/eP8GLCXRy3V9/X2oysrT3ucDUJirwZirSdnvct6OpL+PvYd66R9z8dFbrCn1DYLUdznlcBKedadkQ9I7PIPT5Y9ZEhtBVV6Of3Ag6WxcgyWPjsFpQuGwpBQrk6VkW+TyBBgan6VhTq1YUVgoVa2MD/PXH9nEhnojP3+pnUeety2oDpiY9vK1nx2hpcPBh25s4CO3WBcIQAG895pa5DKBp15bmh+rp82GGAgw/eabS9rPpcCzb/RSYFBz5Zql9XMHRkclL+kUem3Pp6Re+r4f61yamNPlgPt4CwCzZ0/jG0iPz+xykm6v4WjIlEoKbrgR1/HjSS9+ZUkfiV7VfwQYkBRj/xP4EpALfAr44dzPDzIxwCxZzicyWQpG6bv0Dw1KfVWV1Wk9pspsxj88FLMcUaWUY8zTpD1z2WqfpN6SF1W+PxrVJgOjUx7c3vhljucz4HDTOTjN1evMS8pagqQUK8/LR67PKsXGw2TUoZAL9I0lFlwqS03ItFq8XYtPdA+dkgzF4waXnR0AlN77ESo+/yXqv/1fVP39P1L6wIMYtmwjNDPDxN5nGPzOt+j67J/T981/TmicqaJWSuqx487EymO9/iD//qvjDE/M8pk966hbpDLB12dPu/T9+TRU5NHW70yp7FFuMCDPy2Oys4ffHuzhitWlbGxIvdxwdVUBeUHpnEqlLPZY+xgyQYjrXao2lxH2eAhOJRdUN1jy8flD9I265sU9ZMrkBWE65kprIxUpgiCgtlTg6+tDo1LwJ3ev4/btVexrGeRff9mCyyPdF7uHpvmHRw4zOunhz9+3gZu2VsS87xUY1Ny0tYI3zoykXBUC4Olok/5tPUPYm972ieWkc8CJrW+KW7ZWRA3Gk8E/NpqSQvDFVNSY8MqUzPRdfsFWMoTcbjwd7eRduxNBqWTq9y+u9JCSxtdnl5Tkc9LXsx+NgutvQFAomXzphYweJ0tsErUiScx0KEuWDCPPzUVQKKKK+kTEfNKfuSxD9PsJTk6iNEbvgTIb9QmXNybC9Kyk9nn3qsRXh6vnMpw9wzOsSUJ188CJQeQygR1rl14y6BscRJ3tt0wIuUxGWZE+4cylIJOhqa5ZNHN5xDbGr/d3s7mxmPLixQN8T2cnikIjivyC+f2ryy2oyy3kXX0tAGGfD5/dztRrrzBz6CCByUmUSyxjW4zGinxu3GzhpSP9bLEWY43RbxwIhvj2EyfpHprhj+9au6jKbNDpJDQ9ndFSrEZLPm+cHmF0yhO1LDceqnILA7YudHVr+fCu5Mthz6e8WI9JJom4K1PwuDzW7qDBkhfTTzNCpK/aPziQ1DkRyTS29Tmps/eia1oV5xXRaR+YQi4TLqjsUFsqcL6+HzEcRiaTsee6OsqK9Pxobyv/8JO3uXFzBU/s6yRXp+KzH2nGkoAw1e4rpAD1V6908LkPNqe0AOdpb0eWk0PY5cJ96hSGLVuT3selwN43etFrFAvUmJNFFEUCY6PomhYqESeLMU/DWXUeqtF3tmKs+9RJCIfJvepqxHCY6UMHKbr7/ZfVQq6vvz+ji3wRFLm55O3YgfPQQYreuyclxewsS2NpS09ZsiwzgkyGoqAwas+lz96DoNagXIL6XDTOKcYu3nc5PDFLOE2CDW12KRuQjNpnVemcYmwSK+zBUJiDp4bZUF+Ukpz8+YjhMP6hwWxJbBJUlOTQn2BwCaCpqcU30E/Yv1DBsnPQyfd/e5qaslw+ccfquJNgb1cH2rq6RbeRqdVoGxrIv16y8PB2tic81lTZc10dxfkafrS3FV9gYXlsKBzme785zdneSR68vYlNjYsHUMtRinUuYEqtNHZQnodhdoL7djVg0C3tOhQEgQaDVGWhKExOEGh0ysPAmDtuSSxcGFwmQ2GuhqI8DT2dg4SmptBUprYY2N7vpNpkuMBfUV1RgejzERgbnX9s+xoTX/jwRnyBML94uZ2qUgN/c/+WhAJLAJ1GwR07qjnbO8np7omkxxmancXX30f+ddcj0+txtRxNeh+XAkPjblraHVy/yYJGtbTy8tC0E9HnS8t3tSAIBPKMKJ3xBZwuZ9wnWpAbDGhqaim48SZEvx/n/n0rPayEEYNBaX6wDMElQOHNNyP6/Uzte2VZjpflQmIGl1ar9ex5v/dZrVZ7tJ/lGWaWLOdQGI0xM5eaysol93BcTMSzcbH6fbNRhz8YZsKZHtvXs/ZJ1Mr4ap/nk6NVUpyvoScJz80TnePMzAa4ev3SDY2DkxOIPl9WKTYJKopz5n0AE0FTUwuh0AKz+NEpD//x+AnyclT82Z71cY3fA5OTBCcm0NTVJ3bcykoElQpPR0dC2y8FtUrOx3avYnTKw5P7LszShkWRh59p5Vi7g3tvamTH2vjnrc8+JyKRwUmNuUiPXqOYV0FNhqFxN2+MyVCKIdYXpMdOoUzuxS8oGPIml2VraZPsIZrjBOwgZQfkOYaU+poaLHlMd3UDqQX9gWCYnqEZ6i0XlkJH9hVZUIhQV57H396/hXtvauT/fKg56YW0nRvLKcrT8NirnUkvIHq7OkEU0TZayVnfjPvECcRQ+m2rMs2zb9pRKGTs2mxZ8r4Co1Lwr1qiUmwERUkpet8Mvtl3pu26GArhPnkS/br1UoVJRQVaaxNTv3/5sjmX/EODkgd5BitIzkddbkG3Zi1Tr7yckCJ6lvSy2Cz8k+f9fh/wkRg/WbIsK8pC44LM5bxfWppLYkHqiZLp9HHtSAAG01Qa29o7SUNFXtJ9LdWmXHqSyFweODFEXo6KdbVLM6+Hc2I+6mzmMmHq55RGv/ebU0wn4KenqakFLhT1cXkC/PuvjhMOi/zF+xMzNI/0W2pqEwsuBYUCTXUNno7MZy5BytjfsKmclw73zSuLiqLI/77YzqHTw9x1bS03JjjJ9fXbURQWZrTPRyYINFjyOWob4/dH+wkEEwsSw2GRh/eexamXrj9/mkQ68oNuppQ5tPYml0k91u6gvFi/wCc0FqqyspSERRoq8jE4pQAjleCyd3iGYChMw9z1c2485SAI+Pr6FrymMFfDjZstcT2Do6FUSCW2faMu3jg9nNRrPR3tIAho6+rQN28kPOvG096W9BhWkskZH4dODXPNevOSK1wA/HPBZTp6LgEMFeUIwEB7T1r2d6nh6ewgPOtGv6F5/rH8G28iODGOa07k51Inck1myoYkGgU330rI6cT19lvLdswsEjFnrjab7QCA1WqVAw8Cb9hstn0X/yzXQLNkiaAwGglOTSEGzxmu+4eHEP1+NGkW8wGp7CaeYmzEjmQ4DaI+TpePofFZViXgb3kx1WYDDqd3XrxiMaZcPk50jnPVWvOilhWJErEhiWR6s8SntiyXT7xnFZ2D0/zDj9+OW9KsyM9HUVA4L+oTCIb5zpMncTg9fGbP+vlFjnh4OzsQlEo0SVhAaOsb8PXZCft8Cb9mKbxvZx3GPA0PP3MWfyDEU/u7efloP7duq+Q92xNfRPL19S1Ln8+enVJ/36MvtPHF/z7Ey0f6CQQXzyq8dLiPzoFpbtq9VQqKBvrTMhbZ9ARebS5nehIv43R5ArT1TyVUEhtBVVYuCaklrRibT4lvgpAhP6WgP7LgcLG9lEylQlVqWpC5TAdbV5VQZTLw1GtdcT/X8/F0tKOuqESm0aJfsxZBocDVcizt48skL77dhyjCLdvSExgExkZAEFJSM45GaUQxtiP1YjqvP5iw5/By4z5+DORydKvXzj+W07wRhdHI1GUiWuPrsyMolahK09u2tBi61WtQlZUz+eJz7wiP2cuJuDNKm80WAm4GLs2rLsu7DqWxCESRwOS5iZNvTswnE5lLOKcYGwuDTkWOVsnQxNIzl60p9FtGqDZFRH3il8YePDVMWBTTUhILUuZSnpeXcSW4dxo71pr50n2bEIGvPXokbmZEUyOJ+oRFKevV1jfFg7evorEif9HXnY+nqxN1VXVS1hyaunoIhfD2dCf8mqWgUSn42O4mRiY9fO3Ro/zuYA/XbjDz/uvrEhZVCQf8+IeHMtpvGaG8SM+X7tvE5z7YTHGehp+92Mbnv3eIF97ui9o7OjI5y5OvdbGhzsiVGywoS0vTFhQFHA5UxcW09k0lPGE+3uFAFElKqVZVVkZ4dpaQM7kMaZlRhzkwyVROaqq47f1OSgu0UbNo6oqKBWXj6UAmCHxgZx3j0z5ePpJYtlYMBvF2daJtaJT2odGgW7Uad8uxy2ay6/YGeKVlgK2rSihOMKMdj8DoGEpjUdqsgUrrpet7pj8164mwKPL3Pz7MI8/Z0jKedOM+fhydtQm59tz7L8hk5N+wC0+bLSPne7rx9fehKrcgyJOvHEgVQRAouOlmfH19eGyty3bcLIkL+vwb8PdWqzU1460sWdJIRLH1fDsSr71XWhUzpSdQuhiVuYzQ9DQhd+zMpNmoY8ix9Mxlq30SrVpOZWnyQVpE1Cee36Uoiuw/MUSDJQ9TYfLqltHwDw5klWJTpNqUy9/ev5Vqcy7/89sz/PL37YRiWN9oamoJjI3y9IunePPMCHuuq+XK1Ykr/YYDAXy9PXHFfC5GO9efGSmpXQ5WVReyc2M5vSMzbG0q4aO3NCWl1ukfHIRweNn6fARBYHV1IV+8bzOf/9BGzIU6fvFyO1/43iGee9M+798ZFkV+tLcVuVzgo7dKf5O63IK/f+mZy5DbTdjjIb/chM8fStj7tqXdQX6Oiqok+rwj17u0OedTAAAgAElEQVRvMLlJvej3U+BzYheS99UVRZGOAeeCktj5MVVUEhwfJzSbXt9hkM7HtbWFPHOoJyHLJ6/djuj3o204pwCsb95EwDGGP01Z6kzzytEBfP4Qu69I3wKNf3QEZZr6LQHkWh0elZ5gioqxbfYpRiZmefPsCLPeYPwXLCP+kRH8w0Po1zcveC7v6msRVComX35pBUaWOKIoLlsFycUYrtyO3GBg8oXnlv3Y72YWDS6tVuuH5n79DPB/gJmLxX0yPsIsWS4iooB4vqiPr7cHdUVFxlbFElWMTUvmsneSRkt+SqWqOo2C0gJt3L7LjgEnIxOzactaiqKIb3AwWxK7BHL1Kj73wWZu3GTh+bf6+NdfHo9a3hzpuzy5v4VrN5Rx25XJZet99l7EYDDhfssI8pwcVCbzsvVdRrjnhno+decaPnnHamSy5ARqVqLPJ0JTVQGf//AmvnjvJizFen71Sgef/95Bnn2jl+ffstPWN8U9NzRQYFDPjbGCgGNsyWXHgXEHAOb6SgRIqDQ2EAxxqnuC5oZiZEkE7/NiZ0kqxvr6+xCArrABZ4JiVhGGJ2ZxeQILxHzmx2SxzB8jE7x/Zz2z3iDPHOqNu613zt9SW38uuMyZ65u7HEpj/YEQLx3uY21tIZWl6bNzCKTJ4/J8gvlGVNPjKZW2Hjw9jFwmEAiGebv10rI0cZ+Qeir1GzYseE6u15O7/Spm3jxEaCZ1H9ZME3JOEXLNLNsi3/nIlCrydt6A+8TxRedvWdJLvNnrf8/9ex+wC7iFheI+WbIsK4pCSfwiIuozL+aTgX7LCIkoxpoK9czMBhLqd4zF5IyPkUlPSiWxEarNuXHLYvefGEKtkrO1KT1f8MGJCUSfF1V5NnO5FBRyGffe3MjHbmuivX+Kh378NvaRCycNXeQiAs1aN/fd3Ji07563sxM4l4lMBk19A57ODsQYWdVMoFbK2baqNCXTdl+/HUGlSvtENhkaK/L53Ac38tf3baaq1MBjr3by2CudrKku4JrzFndU5RYQxZQEcs4n4JCCS0NZKZUmQ0LB5ZmeSXyBUFL9liD5Dsv0+nkxr0SJlPGNqAtpT9K+JaLI2xAjuIwsJGQquKwoyWHHWhMvHe5nPI46uKejHWVR8byXLEh905ra2ssiuHz91DDTswFuuyJ97SYht5uw2532a1JZYqLAP52037Q/EOKIbZQr15RiNup4/VRygk2ZxnW8BVVZWUxl3fwbdyEGApe0Lcn8It8ytCdEI3/nDQgKBZMvvbgix383Eu/bWgCIJuSTFfTJslLIlErkeXnzmcvA2Bhhjydlv7REUBZJ/SGLKcaWFUnlpUNLEPVp7Z0EoCkFMZ8I1SYDE9O+mPYWHl+Qt8+Osq2pZMl+ZRH8Q1kxn3RyzfoyvnDvJoKhMF999AhvnZVW0+0jM/zn3g6c2kI26GZTCrg8XR0ojEYU+Yn3aEbQ1tcTdrvxD19aE7BY+Ox21JaKtNsTpUK9JY+/uqeZ//vRzezabOFjt626YGFAXS5l3PwDSwuKgnPBpdJYxOrqAjoHp/H6Fy/1a+lwoFHJk77vCIKAuqx8XswrUXx2OzKdHo/GkLR9S3v/FDlaZcxyfkVBgRTwZii4BHjvNVL1wFP7u2JuI4oinvZ2NOeVxEbIad6Er6eb4NRk0scWRRGn20/noJNjbWOc7ZnAPjLDxLQ3am9vqoTCYZ57s5cacy7WyuTvFbGIeJCq0hxc5laWow95sfeOxt/4PFo6HHh8IXasMXH1OjMd/VJVz6VAaHYWT3tb1JLYCOqycnSr1jD1yu8vEDm8lJj3GrYs3cYmFRR5eRiu3M70wQOEXIn7SmdJnXgzS7nVar2euSAzGjab7ffpGozVav0K8HfAOpvNdspqtV6JlD3VAj3AfTabbXRu27Q/l+XyQWk0zvdc+uxzYj7V1Rk7niCToSw1LZ65nFPqHBqfjdkPFI+z9kn0GgUVKfRbRoh4Y/YMz7C+bqGJ+uHWUXyBENesT18gGMm2ZHsu00ddWR5feWAr333qFN/7zWk6+p0caRtDq1ZQuq6JwJmTiKKYUuZSG2XCmwiR8j5vZ/slbzkjiiK+/j4MW7et9FAuoK4sj7qyhVk3ZXExgkq1ZMXYgGMMmUaDTK9ndVUhz75hp63PGfVeAFL/Z0u7g7W1RpSK5INwVVkZM2+/ndS5KFWaVFJrzptXfk2Ujn4n9eV5MY8lCAJqS0VUO5J0YczTsGuLhefftHPz1oqoJaOB0RFCM9No6xsXPKffsBHHk4/jajlG/s4bLnhOFEWpgmXCw/i0l3GnF4fTO//7+LR3UasbpUJGjlaJXqMkR6tAr1WSo1WyprqQzdbihD+jI7Yxxqa8fOD6+qTvMYvhn+uLTGfPJYCxuoIhYKzTDltrE37dwVPDFBjUWKsKMBn1PL6vk9dPDXH3tcn1pGeC2dOnIBSaL6WORf6Nuxj8zrdwHTt6yd3vQMpcKgqNyHWJqZlngoKbbmH6wH6cr71K4W3vWbFxvFuIF1yqgR8SO7gUgcSv4kWwWq2bgCuB3rn/y4BHgQdsNtsBq9X6ZeDrwIOZeC4df0OW5UNRWISvTwoqvb09IJdnPLBRmc3zhuzRKMrVoFTIlpS5tNknaazIj9v35DrRgsocvVSmstSAgKQYG21Cuf/kEKZCHXXlyYtpxMI/NIg8NzerFJtm8nLUfP7DG/n5S+28dKQfjUrOl+7bjOGMj9G3DhJ0OFAWJ664GZiYIDg5kXS/ZQRlqQlZTg6ejg7yrrkupX0sF8GJCcKzsyvSb5kKgkyGqqwc3xJFfQLjDhRFUhDRYJG8cs/2TsQMLruHpnG6/UmXxEZQlZUTnn2VkNOZUDZcDIXw9feRt/MGGsz57D3Ui9cfTKiKYtrtZ2TSw7UbFl/YUFdU4HxtH2I4nLGs9e3bq9h/fJDHXu3krz6wYUEA5mmXepOjLeSoyspQFpcsCC7DYZFvP3GSY+2OC7bP1Skx5mmwFOtpri/CmKfBmKshL0eFzx/C7ZXaMVyeAG5PcP53lzfAoMPNtNvPvpZBGivy+fCuhrj9k6IosveNXkyFOjY2pqboG4tAxOMyzcGl2iyJms30J55Fd7r9nOqa4JYrKpAJAgUGNWtqCnn95DDvvbo26R7vdOM60YIsJ0dS6l4E/foNKItLmHz5xRUPLkPhMEdsY4xMzHLb9irkMhm+fvuK9Fuej7rcgm71GiZffomCm29Nm1JxlujEe3fdNpstLcHjYlitVjXwXeBDwKtzD28GvBG/TeB7SJnGBzP0XMLI5QKpLuSt9M3qnYKqyIi75RgyGfjtvWgsFhTqpYsZL/b5aMrKcB05jBAOIFMulMCXywVMhTqGJ2aRy5P/nMedXsamvNy8rXLR13v7+xn89rdQWyzU/N1DCyZPOTolJqOO3uGZBfsZGnfT0e/knhvqUaSQpYiFf3AQdVl5Sn93Mrwbrx+5XM7HbmtiXW0hhbkaqs0GvH5psuHv7UZjSnyS5u6R+i31DfUpflYCuvp6vJ0dMV9/qXxGs4NS5kpbtfj1dCmhqajA1XJsSeMNjjtQFRUjlwto5QoaLHmc7Zmc3+fFn8/xDgcyQWBjQ1FKx9VapEW94MggamP8slrf8AhiIICuuoqm0nx+d7CHnuEZ1tQUxn1t56BUQmutyl90rNrKSqb8fkKOUdTmzCiI5+pV3HVtLY++0MYLh/sWCGt5O9uR6/Voy8sQFlwTAoZNmySVT7933mLi6de7Odbu4PbtVTRVFVCUp8GYp0GtXJpQXTgssq9lkMdf7eTvf/w2OzeWs+e6Wgy6hd9jAKe6JrCPuHjw9lUpZbMXI+gYRZGfj1KnSet+NaZSREEg7BhFEBK7Dx1uHZm344qcT9duKOM/nzpFW/9U1HNyue5vYjjM7MkTGNavRxHv85fLKdy1i5H//Tn+vh601TXLMsbz8QdC7D8xxLNv2Bmb8gBS8H7fjTX4h4cxbN6ybPfhWJ+R8dZb6fvX/4f7yFvk7bhqWcbyTmaxa2HlG1EkHgIetdlsPec9VslcFhPAZrM5AJnVai3M0HNZLiOURiNiMEBoehpvbw+aquqMH1NlNoMo4h+JXUVtNuoYdKTWr3GySyrzXRVHzGfsySckw/W+PpwHX4+6TY05N6pi7GvHh5AJAjvWJW5dEQ9RFCUbkqyYT0bZ0lRCbZmUbVaXlyMolXi6Y/d8RcPT0Y6gVKKpTD2bp61vwD88RPASVicE5qsMVqrPJxU0FguhmRmCzuT6ECOIokjA4UBZdC4LuaamEPuoK2YP9tE2B9bKfPTa1Bbn1JYKEARGH38sIUsSb6SNoaKSekseggA2e2Klse39TpRy2byfb8wxVWRW1CfCri0Wtq4q4Ve/7+BU1/gFz3na29HWN8TMnBo2bkQMBnGfOgVIQf5v9ndz9Xoz99xYz/o6I2VF+iUHliBNAq/fVM6/fHo7u7ZY2HdskC/81yFefLsvquXRM4d6yc9Rs2Nt+r4nIvhHR9OetQQQFArCuQXkep0MJ9gz+frJYapKDViKz1XcbGwsQqdWcODEyiqLejo7CLlccUtiI+RdfQ0yjYbJF5dXtMblCfCbA9381Xde55HnbBh0Sj6zZx23XVnJ748O8PKzRyAcRrNCYj7no1+7DlVZGePPP5+0z2wsa7As0YmXucz4MoPVat0ObAG+mOljpYtQSCQcXpoBcih0eRgoX6rIC6QSL1drKyG3G1VFVVrf02j7UpRKK+DegQGU5uiBlKlQx9tnR/H6gigVyU0K3jozQkmBFnOhLubf4unqxHXsKMY778J1vIWxJ59Av3kbMtWFK9BVpQYOnhpm3OklP0eyOgiGwhw4McT6OiMGrSpt71dgYpyw14vSVLZs5/W7/voR5Kgrq/B0dSX1Xsx2dKCpriEsyCHF91A9V1Lrbmsnp3ljzO1W+jPy2O3SJFalWfGxJIqyTAqEZ+196FcnX7YecrkIe70ojEXzf3NEpOd09wTbVpWe2zYkMjI5y6DDzXUbUr92Bb0B8yc/xcjPHqH7K39D4R13UnjL7phlZ57eXgSFAkWJCUEhp6Ikh7a+qYSO3943RbXZgEwQFt1eYTKDTIan145+09aU/q5E+djuJoYcbv7zqVP87QNbKc7XEpyZxj88RO5VV8ccp6qmHllODtNHj+CpXcN//+Y05cU5fPRWK6KYmetHo1LwoRsbuXZ9Gf/7cjuPvtDGK8cG+PCNDayqltbYu4emOdMzyfuvr4v7PqeCf2QU/Zq1Gfn7lKUmCnuH6R6cprRgcf/mQYebnuEZPnhD/QVjkQsytq0u5eDJIe69qRGtOvp5nOl7yvTRYyCXo1mV4Hul1pK74yqcr+3DuOcDKPKiqymni3Gnlxfe7uO144P4AiHW1xnZfUUljRX5CILAhvoiHE4vJw++TgWgLLcs+3042vHyd93M6CM/xnXmLLqmVQnt5+3WUX7wuzN88MYGrt+YXUSPsFh8vmjm0mazpc/YKDbXAauAbqvV2gNYgOeBemC+zsRqtRYBYZvNNgHYM/BclssIpXEuuDx2FAB1VeaUYiOoSk0gCIsqxpqNekRgeMKT1L6nZ/2c7Z1ia1PJouIJjicfR27IpeCmWyh+/z0EJyeZeumFBdtVnSfqE+FU1wTTbv8F9gfpIGJDkLUhWV40NbV4e3sSVggMB/x4e3vR1C5NqEJTXQNy+bL7XSaLr68P9RIytCtBxKPRn2LfZcSG5PzMZbXJgFatiGpJcqxN2j7VfssIhm1XUP3QV9FvaGb8qSew/9ND8xnKi/HZ7ajKLfPBZ4Mln85BZ1x/Qn8gRM/wTEx/y/ORKVWoTKZ5y5NMolEp+NO71wHw7SdO4vOH8M5dG9HEfCIIcjk56zbgPnGc/3zyOGFR5E/uXpuWTGU8yotz+Ow9zfzp3evw+UN84xctfPfJkzimPDz7Ri9atYKdzem/n4d9PkLOqYxZA+VYyij0T9Mbx4oL4NDpYQQBrlhduuC5q9aZ8AfDvN26clqP7hPH0TY0ItctHiSfT/4NuxCDQZyvvZqxcfWPuvj+b0/zhe8d4vdH+9nUWMxDD27jL96/AWtlwfz8RSYIfPz21TSqZvELCjpnL40ex9wrdyDPMTD54vMJbR8WRZ4+0E04LPLT52385LnWlLxU322seFmszWb7us1mK7PZbNU2m60a6Efy0/wGoLVarVfPbfop4LG5349k4LkslxGKueDSfeI4yGRSaVaGkanVKIzGOMFlanYkR21jhEXxgszCxbjPnMbTepbC2+9AptGgszah39DMxN7fEZy58Mu0sjQHQYCeoXOP7z8xSK5exboYwh6pEjFQV2dtSJYVTU0tot+fsA2Er7cXQqG44hDxkKlUaKqq8HZ2LGk/mSTs9RIYG12W+0I6URhykefmpqwYGxgfk/ZjPBcsymQCTZX5nOlZaHvR0j6GpTiHonxtagM+D0VeHmWf/lPMn/4Tgs4p7P/0EI6nniAcOOf7K4oi3jml2AiNFfn4A2HsI4tbBHQPTRMKizSUJ6bErbZUZrwsNkJJgY4/+oM1DDhc/OjZs8y2tyEoFHEVzPXNGwnPzhLu6eITt6+Om21LJ4IgsKmxmH/65BXcdW0tJ7vH+evvv8kR2xg3bCqPmbFbChEbkkwFlxqTCZUYZNg+suh2YVHkjdPDrK0xkjdX2XM+teZcTIU6Xj+5MqWxgbEx/IMDCZfERlCZzOjWrmPq1fTbksx6g3z3qZP87cNvcbTNwY2bLXz9j7bzyTtWYymJLuSnVMhYrfHg1Bv5zq9P0z+28jYgMpWKvJ3X4z5xPCFLrZOd4ww43Nx/axO3b69iX8sg//LzYzhdvmUY7eXLigeXsbDZbGHgI8B/Wa3WdqQM5xcz9VyWywu5To9MqyXs8aAyly0oC80UKpMZ/3DsL5zSQh0CJG3k/HbrKKZCHZbi6FLdoijiePJxFIVG8q7bOf940Z4PEPb5mPjt0xdsr1EpKDPq6Z3LXDrdfk50jrNjrSklb8TF8A0OIjfkIjcsR6FDlgiaWklrzdvdndD2nrlgUFu3dIl9TV0D3p7uS9dXbaAfRHHFTLuXgrq8IvXgMkrmEmB1dSEOp5fRqXMVFTOzftoHnEvOWl6MYfNWqh/6KrnbrmTimd9i/4ev4OmShKSCk5OEXa4L+q/qy6VMZDxLko4BqQ81kcwlSL22wYkJQu7U1buTYW2tkT3X1fHW2VGGj51GU1OLTLl4H+sJiggKMm7KnUq7KmuiKBVy7thRzVc/eSVbrMUU5KrZtTkzfcrzHpcZ6LkEqSwWwD0wuGhPXZt9ivFpH9vXRl/MFQSBq9aZaO93MjK5/J6XrhMtAIv6W8aiYNdNhJxOZo68nbbxjE15+OqjR2hpd3Dn1TV844938KFdDRjzFhdlEkWR4OAAlnWNqJRy/u1Xx5mcWfmgLP/6GxDkciZfXlj1dTHPvWmnMFfNlWtK2XNdHZ+6cw320Rke+slhuofiZ8jfrVwaeerzmMteRn4/CKyLsV3an8tyeaEoNOIf6EezDCWxEVTmMpxttpgS92qlHGOehsEkMpdOt59W+yTv2V4dsyTWdewovp5uSh/4+AUTFnVZGXnXXMfUvlfIv/EmVKXnviyrTQZOdU8giiKHTg0TCotcvS79yon+wQFUl7jn4TsRZVExspwcvN1dcN6CQyy8XZ0oiopQ5C3dEF1bX8/Ui8/jtfeiXWKZbSaIlEOutPx9KqgsFpz7XknJRiPgcCDT6Rb4ya2ulvouz/RMzFdXHO8YRxQlAZN0I8/JwfTxT2LYdgUjj/yYvq/9IwW7bkZTIy2IqCvO3bMLDGqK8zW09U1xy7bYiwHt/U7MRh05CQoPnS/qo7M2LeGvSZzdV1TSNzCB8tlBXA07F922b9TFI6/0cG9hJZaxrpQ8a9NJYa6GP/yDNRk9hj9iQ1ISu0JnKahMUnCpd08y5vRSEiMjf/D0MBqVnI0NsQP6HWvNPPlaFwdPDnPXtRk3TbgA9/EWVCbzBd/niaJbvRZlqYmpl18k94rtSx5LR7+Tbz95glBI5K8+sGG+NzcRgpMThGfd5NXW8JerN/C1nx3l3351nC/dtykjmfFEUeTlY9h2JdOvH6DozrtjWqh1Djqx9U3xwRvq5xflt60qxVSo4ztPnuRrjx7l/lutXJXkvEoURdr7nbi9gUXPwcuZSzZzmSVLPCJ9l+rK6mU7pspkRvT7CU7GbtM1G/VJZS6P2EYRRdi2KvpqrhgOM/7rJ1CZzORu37HgeeOd70VQKHA8eWGFd5XJgNPtZ3LGx/4Tg9SV51JWlF4TY1EU8Q8Nosqwx2iWhQiCgKa6Vgou4yCKIp7ODrQp+ltejLZO8u7zdlyapbG+vj5kOh2KwvSWgC8H6nILot8/7weYDMFxB0rjwmDRVKijwKC+oDS2pcNBgUFNVRzPw6WgX7eeqof+ibxrr2PyxecZ+uH/gCCgrrgwM9Zoyae93xkz2xQWRTr6nTQkmLWEcwsLvr7lKY0F6Zr8oFWNnDB7+2WMxsh6RUoMdRoFDbuuITjuwJ9itvpyIjA6ikynR65P7/dQBEVBISgUFAam56t2LsYfCHHENspma/Gi/a0FBjVrqgs5eGqIcJLKoksh7PUwa2tFv2FDSq8XZDLyb9yFt6trvmIgVQ6dHuZf/vcoWrWCL9+/JanAEs5de2pLBZWlBv7krrUMjbv57lMnV7xvseCmWxD9/kX7U597045OreCai3x1K0sN/M39W6gvz+WHz5zl5y+1JaQmOzwxy1OvdfGF7x3i6z87yrefOMkR29hS/5RLkmxwmeWyJdJ3ubyZS2mFyj8UW3LfbJS8LhP9Qnr77ChlRXrKi6Ovnk0fOoh/cBDjXXcjyBd+GSry8im4+VZcRw7Plz4CVJsltcnfHx1gaHyWa9anP7sYnJwk7PGgzmYuVwRNTQ3+wQHCXu+i2wUnJghNTaFJQ0ksIPnUFRXj6bw0RX18/X2oLRUrmglKlYh1im8g+aAo4HCgKFoYXAqCwKqqAlp7JwmLIv5AiFPd4zQ3FGX8PZJrtZR+5AEsn/sCysJCNNU1yDQXZpQaKvJxeQIxLSSGHG5mfUEaLIln3eV5+chzDMvWdxkh2CPdg4d0pXznSUng53xEUeSHz5xh3Onlj9+7luKtmwFwtRxb1nGuBIGx0Yz1W4IUWClLSikMzMQMLls6HHh8IXasiW+zctU6M+PTPmy9C/uVM4X79CkIhdBviK3EHY+8HVch02qZfHZvSq8PiyJPvdbF9397hrqyPL780S2YCpPvBz5XQSLd09bWGLn/1ibO9Ezy42dbk7YDSSfqigp0a9czsfd3BMbHFzw/MjHLUdsY18foPzboVPzVPc3s2mLhpcP9/OsvjzMzu9DuaXrWz0uH+/iHnxzmr//nDX53qIfSQh0fv30V1SYDP9p7FoczOQHIy4FscJnlskVtqUCm1S5r6du54HJxUR9/MMyEc/EJP8DkjI+2PkklNhrhQIDxp59CXVVNzqYtMfdTeMtu5Lm5jD32y/kbdkVJDjJB4IW37aiUspjHWAqRIDubuVwZNDW1IIp4e3sW3c4732/ZkL5j19Xj6Whf0QlCNMRweD64vBxRmcskH9skFWMlj8uxqJlLkEpjXZ4AfSMuzvRM4g+E095vuRi6plVU/+PXsXx+ocxBJCPZ3h/d3zPyeKL9liAF1CqLZdmDS09HO6qych7Ys4kBh5uH95694Bp57i07x9odvP/6ehos+Sjy89HU1r47gsvRUVQZDC4B1CYTJeEZekeiB5cHTw1TYFBjjeMnDZKKslat4MDJ+MIv6cJ9vAWZTo92CcJrMo2Wglt24zp2hOk3DyX1Wn8gxH//5jS/PdjD1evNfPaDzQmXol+Mr78PZXHxBYtJV683896razh4apin9iemF5ApSu/9CKIoMvKThxd8jz3/lh25XFi0/1ghl/HhXY08eNsq2vud/MNPDmMfmcEfCPHW2RG+9dhxPvud1/n5S+2EQmE+cH093/zjq/jsPc1ctc7Mp+5cQ1gU+Z+nz6x4JjfdZIPLLJcteddcR83XvrFgFTyTKAy5yHJy4tqRAAwmUBp7xDaKCDEDP+f+fQTHxym6+32LZhhkGg3GO+/C29GOu2XOnkUpp6xITzAksrWpJCM9Dv4BSak023O5MmhrIqI+i5fGero6EFSq+axYWo5d30DI6SQ4JyJzqRAYG0X0+S47G5IIMrUaZUlJ0mWSIdcMot+Psih6D8+qKqmk7XTPBEfbxtCq5fMemMuFIJcjUy4UXzMVSr2U7X3RRX3a+53k6pQxe+hioa6oxD/QjxgKxd84DYjhMN7ODrQNDaytMfK+6+p4u3WUZ9+UMjitvZM8/monW5pKuGnLuWsxp3kTvp5uApPLlyFbbsRgkMC4I6OZS5BEfQzeaexDC8usp91+TnVNcOWaUmQJZOxVSjnbVpVwpG0Ujy/z4mViOIz75An069ZFrVJKhsLdt6Opq2f00UcIjCd2j3a6fPzzz49xuHWU919fx8d2Ny1JANDX14fasvA+fMdV1Vyz3szvDvawryUxtfNMoCwupvh99zB75jTO1/bNP+50+zlwcpgda81R1YQv5ur1Zr547yaCoTBf/ekR/vI7B/jeb05jH3Vx89YKHvr4Nv7uwW3cekUlBYZz+ysp0HH/rU10DDj5zYGVDbTTTTa4zHLZIshkMRuxM0k8xVjTnGDGcAKiPm+1jmIp1kfthQz7fEz87mm01iZ0q+MLLeRdfS0qk5mxxx+bV/GsNkv9VJkoiQXwDQ0gNxhQGJI3fM+ydOQGA8ri4rjBpbezE011TUxj+1TQ1ktZ0EutNPZcn8/lGVyC1HeZrGJsMIZSbIQCgxqzUceprgla2h2sqzWmXTk6VQRBoMGSFzNz2TEwRYMlP+kSXrWlAjEQwD+yuDVFunqsTRIAACAASURBVPAPDBD2eOb9LW+9opKtTSU88WonB04M8b2nT1NaoONju5su+Fv0zVIJpPsdnL0MjDtAFFFmSCk2gqrUhEwMI5+ZWqBM+ubZEcKimFBJbISr1pnxB8IctmXe89Lb3UVoZiYlldiLEeRyzJ/4IxBFhn/wP4hxegL7Rl384yOHGXC4+JO717H7iqollcyHfT4CoyNRK8sEQeAjt1hZW1vIT59voy3GotJykHfdTrRNqxj71S/mg/CXj/QTCoW5ZVvi1S+1Zbl85YGtNDcUsamhmM99sJlvfHoH77++HkuMlieQfFav3WBm76FeTnfH1vK43Lg0vlmyZLmMUJnNi/Zc5upU5GiVDMXoH4owMe2lo9/J1hjelpMvvUBoejpu1jKCIJdT9L4PEBgZxrn/NQCu21DGTVsqkhLCSAb/4KBUxpdlxdDU1C5qRxIO+PHae9GkWdVVVV6OTKvFc4mJ+vj67SCToSq/fM9LVbmFwOgoYV/isv3zNiQxymJBsiQ53T2B0+2neRlLYhOhwZLP6JSHqYv846ZcPsamvEmVxEaITGz9y1Qa6+loA0DbIC28CILAg7etorxYz8N7z+Lzh/iTu9ctqCJRmctQlpTiOv4ODi4jNiQZUoqNoJqzIyn0LxT1OXRqmMrSnJj6BtGoK8ultFDH68tQGus+3gIyGfq16TE0UBYXU3LvR/C0tzHx7DMxtzve4eCrjx4hFBb50r2b2ZQGW5xzdlDRAzSFXMYfv3cteq2C3x9dOTErQSbD9MCDAIz8+GE8vgCvHO1nY2PxfBVaouTlqPnUnWv5+HtWs7q6EJksseD8Q7saMRfp+f7vzuB0L+zbvBzJBpdZsiSJymQmNDNDyBXbENhk1DHkWDxzeXhOJSxaSWzI5WLyub3oNzQn1Xuh39CMttHK+NO/Juz1UFeex4d2NWREtEMURcmGpDzbb7mSaKprCU6ME3RGX/319fRCKLSkHp5oCDIZmto6vJdg5lJlMkUtv7xcUFssMKfEnCiR4DKaoE+EiCWJXCawvvbSUtJtrJDEei7OXnbM/T8ZMZ8IKnMZyOXL1nfpaW9Hnp+P4rwAX62S86d71lNlMvDx21dRHqVKRRAEcpo34mk9S8jzzhP3gPNsSDKcuVSapOC1MDB9Qd/loMNNz/BMUllLkD6bq9eZaOubiqn+my5cx1vQNjSmVU3XcOUODFu3Mf70r6NWuJztneQ/njiBqUDH39y/lSpTetSjI9fcYhUkGpWCzY3FHO8Yxx9YntL1aCiLiin+wD3Mnj1Dyy+exu0NsvuK5at8USvlfOrONXh8QX7w29PLqk6cKbLBZZYsSRLJ1C3Wd1lm1MXNXL59doTKkpyoKmwTzz9L2Oul6K49SY1NEASK3ncPoZlpJp57NqnXJktwakpSis1mLlcUzXzfZfTspadLyixq0hxcAmjr6vH1919SE+JYfT6XE+ryORuNJER9AuMOZHo9cm3svkRrRQGCANbKfHSa1EQ6MkVlaQ4qhWxBiVx7vxOVQkZlafItEDKlEpXJPK9amWk8HW1o6xsXLOaV5Gv5ygNb2bKIqJq+eSNiMIj71MlMD3NFCIyOIqhUyPMyU0UTQZ5jQKbTYZHPXpC5PHR6GEGQyhCTZfsaEwKSGFCmCMzZ0ejXp2ZBEgtBECi5734UeXkM/eC/FyiL//b1bvJz1Hzx3k0X9AMuFV9fHzKtdtHFLoAtTSX4AiFOdi1UbF1O8q7diXbVagyvP0ezUaCuPLPn6cVYinP48K4GTvdM8uwbvct67EyQDS6zZEmSecXY4dhZBVOhnpnZAC5PIOrzDqeHzsFptkbxtgxOTTH18osYtl2ZkuKltrYWw9ZtTL7wHMGpzAlE+AcjYj7ZzOVKoq6qApksZt+lt7MDZXExitz098Vq6hsktdol+qmli5DbTXBifFkVpDOBsqQEQaVKqu8y4IjucXk+Oo2CB3Y38b6d6S2RTgcKuYzaslza+y8OLqeoMeem3B+qtlQsS+YyMD5OcGJiviQ2WbR19chycpg5ejTNI7s0CIyOoCwuybj1jSAIqEpNlIZd9MxlLsOiyBunh1lbY0xIoOViCnM1rK4u4PWTwxnLKrmPtwCQswQLkljI9XpMH/9DAqOjjP3qf+cf7x6aptU+xU1bKlCrliYgdDH+BO2grJX55GiVvN2a+Z7WxRAEgaEddyCKcNPw63F7VDPBtRvK2LaqhKde656v2LhcyQaXWbIkidJYhKBUxrUjARiKIepzuDV2Sez4M08jhkIY77wr5TEa734fYiiE4zdPpbyPeGRtSC4NZCoVaktF1OBSFEU8nZ1oatOftQRpIQNBwNNxaZTGnvNVu7wzl4JMhspchj+JzGXQMRZTzOd8dm4sX/ZV+URprMinb9Q1r8zp84ewj7hoqEh9vGpLBcHJyUXbGNJB5BrQNjSm9HpBLidn/QZcJ47PC7K9kwiMjWW83zKCsrQUg2cKp8vPlMtHm32K8Wkf29emfnzJ89KLzZ4Z8RnXieMoS0tRmZIr200UnbWJgltvw/naPmaOHgHguTftaNVyrmtOb/WRKIr4+vtQJaBOLpfJ2Gxd+dJYURR55uwMR6q3I+/twLnv1WUfgyAIfPSWJox5av776VMxkxOXA9ngMkuWJBFkMpSlpkUVY81zfTVDMexI3m4docpkoKTgwpJY/9goztf2ScqvS5BsVxWXkH/9jUwf2I9vIDNS3/7BAeQ5hoxkxLIkh6amBm9P94LV1uDEOCHnFNq6zGSqZBqtFNh2XhqiPvN9Ppd55hLmMm4DiWXcRFGUbB5i2JBcLjRY8hFF6ByQVu27hqYJiyL15cn3W0aInAuZzl56OtqQaTSoy1O3+9E3byI8O8tsW1saR7byiOEwgbFRlCXLc36qSk0oXE4U4SD2kRkOnh5Go5KzsSH1429sLEarlnPgROzv/VQJe714Ws+SkwaV2MUouvMu1JVVjDzyI4Z7hzhsG2Vnc3nabcqCDofUMpNge8LWS6A09nTPBH2jLurecwu6NWsZe/yXBMbGln0cOo2CT925limXnx9d5JF7OZENLrNkSQH1/2/vzuPrKuvEj3/O3W/2m33f05OmG3SjUJBNNkEUkQEGxW0UdJBx1Jnxpw6C+vLnTx1XGJYZt1EEB4cfMqNYhYJQgVIopS1tniZp9qVJm6VZ7pJ775k/7k2atkma5q5tvu/XK6/knuXeJ/nec3K/5zzP9zlFxdjcDAcWs2nWO5cDw25ae0fZOEuX2CNPP4VmNpPz7usjbmPOdddjcjgYeOLXMTlBeXt6ZH7LJOGoqiY4McFk//Fdi9wtsRtvOf3atbV4DrYkpBvRibydnZjTM7BkLj4ZSRb2klICR4/iHz16ym0DR49iTE6ecnxTsqsuzsCkaRwIdwlr6hpGA2pLFn8Bazq5jPG4S3dTE47qmojmJ0xdsRLNYmX0LJuSxD88hOH3Y43TncupirGuyVGaukZ4Q/WzTs/Dbl18bOxWMxvqC9ixvx+PL7I7y0YwSNDjwT96lMkjRzj62qsYfj+pa2KbXGoWC0WfuBPD56PzkUcwAe9cH/0LcZ72NmDhF/mSoWvsM692kJVmY9OKIgpu/wiaptH3sx/H7P+af2SE8b178HZ24B89etxntKqiDN5/SQ1vNh1m687EzQMaiejPqi7EEmAtLGL09R0EJ32zVqU0mTQKs1NmvXO5o7EfW9DH2swA43v34B8aZHJwEP/gIKOvvoLrqmuwZEU+ubk5LY2cd7+Hgf98nJEXnifr0ssifs4pU5Vi0zecF7XnFIt3rKhPy3HdqjwtLWjhbrOx4qytY+T5rfi6u6LaHdUIBpnsP4Snox1fby+Gz4cR8GMEAqFug/5A6LH/2DJPayuOysqotSGRprqU+bq6sCxvmHfbqfHPpxpzmeycdgtlBWk0hYv6NHeNUJKXGlHxIUtmFub0jNMqjnS6AhPj+Lq7SF/33oiex2S3k7pqFSPbXiLz4sti1kUy3ibjVCl2irUglMRW2Txs3dmF2xs47Sqxs9m8qpAX3+phx/5+LlhZNO+2k0ND9P/y50wO9GP4JglO+jB8PoI+HwRO7v5pSkmdnjs4lmyFRaTfcBPGrx/lxvyqqBTxCUxM4G46wETjftyN+/F2daJZLAu+iz/VNfbVtw/hmwxgi+AiwGK0942yv32Imy6twWoxQU4OeTffyqGf/5SRF7aSddk7o/I6hmHgaW5m+PnnGH1jx3HvA81iCVWazszC4nJxTmYWHpOHt55qpSawnooNq2M+XjmaJLkUYhHsRcVgGAz+/neY09IgEMQwghAIhK50BYNsPtzHSKuHAU0RGB/DPziIf2iI8kMDfDbgY+Q7MD1kW9MwZ2SQsmIl2Ve/K2rtzHrnlUzs30f/449iLy+P2nQUgZERghMTMg1JkrAVFaPZHXhaD5Jx/ubp5e6WZhyVVRHdTTmVqfeUu7lp0cml4ffj7enG29GOt6MdT0cH3s4OjKl5HjUNzWIJfZktYDajWczTj0PfzdhLSsi8+JIo/WaJNfXBzNvdRUo4uQyMjeHt6cbX3R36Hv4KjIYKl0zdsTmT1ZVm8uddPfgmA7T0jHBeQ+S/k720LKZ3Lj0tLWAYUUkOCv/6Nlrvu5fu+79P+RfvwZxycjXxM81UchnJUI/TYQsnlxVWD695A7jS7egVkV+wrS3JpMDlZNue3nmTS09HOz0/+j6BCTepK1eiWa2YbHY0mxXNasNks6FZrWg2G6bwd3txKZolPh/JX3PUYE4tpW7fi3g7Lz3tYQRBrxd3c9N0MulpawXDQLNYcNTWkfOeG0hbcw4m+8IT1/X1+fx5Vw97Dh5hnR6f98mUZ7a3h8aerjn2eSbjwncw+sbrDPzmP0lZtRpbBBdGgl4vo9tfZfj55/B2dmByOsm69HLSzjmXwNgY/uFh/MND4a9hvF2d+PfsYbnXw3LA98hLDNv+Htc50a0kHEuSXAqxCPbKKjCZGPzv3865TRUQQGP42b2YnClYsrMJunLZPZ5OtV5Gw5oaLC4X1uxsLFmumPxj0UwmCv/mDjq+fi89D95PxT/fhyUKpeC94TslMg1JctBMJhyVlcdNRxL0+fB2duC68uqYvrYlJxdzVhbu5mayLr18wftNqEaOvvJyKKHs7pq+iqvZHTjKy8ncfBH28gocFRWh5DlOH7yShSUzE3N6OiMv/pnxt3bh7ekmMHKsgqDJ4cBWXELqmnOxF5fgqKo6K+50LSvN4tnXu3h5bx9ub4C6KBQfspeVMbz1WYxAICYXWtzNTRCe9zVS1txcSu+6i/Zvf4u+f3+Y4rv+Ds10Zo9g8vUfArMZS3Z85lY1OZyYM7PIC4QuumxaUYApCnd9NE1j86oinnzxIMNjXrJmqTw7tnsXvQ8/iDkllfIvfDHpiov5JgM8u7Ob+o3XUr/zcXr//WHKv3QPJtvJPbCMYBD/4BEmBwaYHBjA138IT0sz7oMtofO12Yyjqprsa99NSv1yHDU1i55fuH5G19h4JpcDw252NPZz1cZyUhzH/sdomkbB7R+h/Stf4tBPf0zp5//ptI9D36FDjLywlZG/vBS+GF9K/gc/TMam8xeUeAfcbpr3t/Pof+/m5swSIr88Ej9L67+1EFFiy8+n5vv3Y0xOhj6smDQ0kxlMptAJyGTitcYBHn76be776EbK8kNztP3Py2086z3Id265gIwMR1zaak5NpeiTd9H5f79O7yMPUvrZf4j4A5avRyrFJhtHVTXDz/6R4OQkJqsVb3sbBAJRu1s9F03TcNbU4m5ZeMXY8X1v0/PD76HZ7DgqK3FdcRWO8grs5RWhaTjO8A/T0ZLSsJKxXTvRbDZSV6zCVlKCvaQEW3EJFlf2GdVNaqHqSkPJ5B+2dxz3OBL2sjIMvx9fXx/2GPS2cDcdwF5ecVp3auaToteTf+tt9P/yPzjy1JPkvu/9UXneRJkc6MeamxvX49pWUECmZ4RMl42LVkfvIujq2hyefPEg+9uHOP+ErrZDz/2Jgcd/hb2snJK7PxOV4S3R9pc9vYy5J7n8opUUrvkY3d//Lv2P/ZLUVWuYHOgPJ5Lh70cOH9+F12zGXlqG64qrSKlfjrO2DpMjOp9jEtU19o+vdWLSNK6YZeypNTs71D32Zz9h+PnncF1+xSmfzwgGGd+7m+GtW5nYuxvMZtLXriPz0stx1p08B+58zE4n+tp6vtBQk3TzEp+KJJdCLNKpuivNnI5kKrnc0dhPbUkm2XFKLKc4yiso+OCH6fvJv3H4yd+Qd9PNET2fr6c7NGG7VIpNGo6qqtAH6K5OHFXVx4r5ROFuyqk4a+sYe+N1/MNDmHOy593W3dJMzwM/xFpYRNk/fAFzamrM23emKvr4HRiGcVYmkXPJTLOT73LSP+QmK81GTmbk58qpqpXers6oJ5ehsb4HybwkemPaAbIuuQxvZweDv/8f7KVlpG88c8e3T/b3Y82LTzGfKbbCQnw7d/K9ey6M6vNWFKST6rSwr21wOrk0gkEGHv8Vw1ufJfWccyn6+J1Ru9AQTcGgwZbXOqkqymBZWRaa5iLr8isYfu5PHH3pRSA09tOal4e9vJy0deux5uVhy8sPzZXsyo7pEItjXWMHWafHvrLw6ISPl3b3cP6KwjnHnmZsvojR119n4PFfcfi/nji2YvqcrIV+DD82gkEMrxdzZhY517+XzHdcHPFFhjMtsQRJLoWImYLsFDSgL1zUp/fIOJ39Y9x6eewH7c8m44LNuA+2MLTlGRzV1aSv27Do5/L19mAvLllSH3qT3VRRH3frQRxV1XhaWrDm5cdlqhhHTeg97W5uxp6zcc7tvF2ddP/ge1gysyj9+89JYrkAS/EYW1aaRf+Qm7rSrKj8/raiIjCbQ+Muz9sUhRYe42lvw5icxFkb/R4C+bd+AF9PD30/+zHWggIcFZVRf41YMwyDyYF+nHXx/b9nLSgkMDZKYGwsVBchSkwmjYaKbPa1DWEYBobXS+8jDzK++y1cV1xF7k03J23Pi50HBugfdvOpS2qmj6u8m24mZcUKLBlZWPPyEnpOnuoa+3q4um8sGYbBH3d04vMHueq8ubsua5pG0cc+zvDzzxH0eoFwVVcDMAyma7waxvQ6Z3UtaWvXLbmhHDMt3d9ciBizW83kZDroCU9HsqOxH43Q1blEybv5VrwdbfT95MfYikqwL2IqEcMw8HZ3k75h8cmpiD6LKxtzZhae1oMYhoH7YPN0IZhYc5SXo9lsuFuayTpv9uTS199P1/e+g8luo/Sznz8rpgsRsVFXmsm2Pb3URqFLLIQqMdqKimMy16W7OdQdPBaVPjWLhaJP3hUaM//ADyn/8r1n3LzCgdFRgh5P3KYhmTJV3MrXfwhnFJNLgIZKFzsa++lt7WHyFw/j7e4i/7bbo1qRPdoMw+CZ7R3kZzlZu+xY4qZZLDGfX3OhzCYTa5flsX1fbLvG9hwe5/HnmtjbOsh6PY+S3PkTanN6OjnXR1YJeqmR5FKIGCrKSZ2+c7mjsZ+60syolP5eLJPVStGdd9Hxta/Q+68/ovzL92ByOE/rOQJHRwhOjMt4yySjaRqOqio8rQfxHz5MYGQk5uMtp1/bYsFRWYWnefZxl/7hIbq/+22MQIDSz/0j1tz4TKYuzkxranNZWZ3NumXRe5/Yy8qY2L8vdLfJ5yPodhP0uMPfPQTc7mPLPB5sBYWkNDRgTpn/g6e7uQlrfkHMLpZYMjIo/tu76fx/36D3wfsp/dw/xuyOiGEY4TlTfRiTkxh+P8HJydDP4cdTPwf9k1gys7AVF2PNmXs85eRAfKchmTJVMXbyUB/OKA8NWFGVTYH3CCM/+CbWwCQld3+G1JWro/oa0Xagc5jW3qN88MplmEzJ2xtiw/J8XnwrNl1jx9yTPL2tla07u7HbzNxyWS2XrVvYdCni9EhyKUQMFeWkoDqG6BoYo3tgnNuuWJboJmHNzqbojk/R9S/fou+nP6bozr89ra5nU8V87JJcJh1HVTXju95kfO/u0OM4JZcQunMzuOUZgj7fcZUHA2NjdH332/hHRyn7/D/K+0acUkaqjc/+VXTvpthLyxh95WWaPvHRcBe2BTCZcNbUkrJiJakrV2MvLz8uiTIMA09TE6mrYztFgKOikoIPf5S+Rx6i/7FHKfjghyJ+TiMQwNfbg7ejA094CiBvZwdBt/u0n0uz2bAVFGIrLsZWFPqyFxdjzctnsv8QEL9pSKZY8/JB0/Ad6ov6c6e0Kz7QvQWfzUn1F74U03mEo+WZ7R2kOa1sXjX//JyJFouusYFgkBfe7OGplw4y4fVz8TklvPeiKjJSFlfZVpxawpNLXddzgF8ANYAPaALuUEoN6Lq+CXgYcAJtwAeUUv3h/aK+TohoK8xJwecPsmV7R6hLbBwGqS9ESv1ycm+8icO/+U+G/vgHsq+6ZsH7Tk1DYltEl1oRW1PjLoe3Podmty94EuuovHZNLQQCeFoPkqLXAxD0uOn+wXeZ7O+n5DOfm26fEPGWsel8AkdH0CxWTA4nJqcDk9MZ/tmJyRF6bHY40Ww2vB0djO/dzfjePRx56kmOPPUk5vQMUlauJHXlKlIaVhAcHycwNhqX8YQZGzfh7exk6JnfYS8rI+s0CghNTUvk7ejA2xmaR9bX1Ynh9wOh5NBeWkr6xk3Yiosx2R2heRit1tA8jBbL9OPpL7MZ/9BQeK7VHry9vbibmxjd/uqxFzabQ9VENQ1LnHsraBYL1tw8Jg8dWtD2Qa8X/9AggbGx0Nd46HsoxjMfj+Ht7saTWcCv8y/hm8XJf+er+/A4u1uO8J4Lq+JWhXWxprvG7o9O19i32wZ5/Nkmug+Ps7zCxS2X100XWBSxk/DkktAI2G8ppV4A0HX928A3dV3/OPBL4MNKqW26rn8Z+CbwUV3XTdFeF9ffWCwZxTmhLlWvvH0IvTyLzFnmxUoU11XX4DnYwuH/egJHRSUp9cvn3NYIBvF1dzOhGhl56c+YUlIxZ0RnPJSIHkdlJRAquOTU62Na2e9EU11wJ5qbSdHrCU766L7/h3ja2yj+1KfnfX8JEWuWzCzybrplwds76+pw1tWRe8ON+I8eZeLtvYzv3cPEnj2MvvJyKGFyucLbxqdHSu4NN+Lr7qL/sUexFZeQskyfdbug14u7pRn3gUbcSh2bl5BQNVB7eTlZl70Te3k59vIKbAWFizpXWLNzTup6H/R48PX14evtxtfbi7enG2t2NiZr/CteWgsK8c2RXAY9btxNTUyoRtwHGvG0tUEwePKGZjPmlFTMaWmY09Kw5ueTtnoNgzXncfiZZtoPjVJVlNzjYLds78BmMXHZ2jOj18iG+si7xh4anOBXf2piV/Nh8rIc3PW+VZxbl7skC6QlQsKTS6XUIPDCjEWvAp8E1gEepdS28PKHCN1p/GiM1i2Y2ayx2PdnMvd1F9GPT0leKLkMGgYbGwowm5Mp/hrFf/Nx2r56H72PPEjVvfdhdYWmkTCCQbzhZHJi/34mVCOB8VBhImteHrnXXYfFEv+KeHL8zM+cnoatqAhfby8ptbVxfb+ZM9OxFRXhbm5CM4L0/dtDuBv3U/zxT5C5bm3c2iHmJ8fQ6TO7MrFfuBnXhZsxgkE8bW2M7d3D+J7dWHNzcRQXRe1D67zxMZspueNO2r72VXofvJ+qe+7FmptL0OtlormJicZGJhr3425tDSWTJhOOykpyrrwKZ20tjvIKLDk5Mf2AbU51Yq2pIrWmKmavsVD2okKGmxQmUyjpdTcdYKKxkXE1I5k0m3FWV5PzrmuxFxeHksjUtOlk0uR0Hvf3moqP/agHnmmmsWMoaoWnYmHwqIdX3u7jknNLyEpgvYfTsaLaRZrTyhsH+tnYcHrdqf2BIL/e2sKW7R1YLSb+6tIartxYjjUBn1fOdvOdqxKeXM4UvrP4SeBpoBxon1qnlDqs67pJ1/XsWKwLJ7lCRFVGqo00p5VxzyQbElgldi5mp5PSuz5N69fuo/uB+8nYdH4ooWxUBMZGAbDm5pJ27lpS6utJ1eux5uYmuNViPs6qany9vTGZGuGUr11bx9jON+j5yY8Z27mTgts+QOYFm+PeDiFiRTOZcFZX46yuJu/698T99c0pKZTe/Xe0fe2rdPzLtzGnpuFum5lMVpFz1dWk1NfjrK3D7Dy9gm1nE1thIYbXS+u9XwlVCp5OJmvIufY6UvV6nLW1i5qTMjPNTml+Gm+3DnLdBZXRb3yU/On1LoKGwdXzTLeRbMwmE+v0068aaxgGP/ndfv6yp493nFPMjRdXk5VEvcWWkqRKLoEfAWPA/cANCW7LnAIBg2BwgQUB5nkOkbyiGZ/q4gxMmkaqw5qUcbcUFlP4kY/R+9C/4m5pwZKdQ+rq1Tj15aTo+kmVPZPhd0iGNiQr5/IGjr7xOvaq2rj/nRw1tYy89CLD27aR854byLz0nRKrJCVxSW7zxceSX0jRHXfS+/CDmFLTyL7qGpzL9FCidEL176UcZ0d1HabUVEx2O9nvupYUfTmO6prjkkmDxf2NAgGDhgoXW3d24/b4k3Iso9vr5/mdXazX88nJcJxR74X1ej5/3tXDW81Hjps6ZT5PvXSQv+zp433vqOY9F1URCBhn1O98ppmvLlrSJJe6rn8HqAPerZQK6rreAVTMWJ8LBJVSg7FYF+vfTyxdd71v1YKLEyZK+vqNWL6YiyUjXaaJOMOlb7qA1DXnYk5JiftrO/V60DSyr7iS7Ouuj/vrC7FUpK5cTe2PHkx0M5KavayM2h88ELPnb6h08ccdnTR3j9BQmR2z11msP+/qwe0NnFF3LafUV4Srxjb2Lyi53La7l6f/0saFq4q4/sLK2DdQzCspOiHruv4NQuMh36uU8oYXvwE4dV2/MPz4TuCJGK4TIiYsZtMZ0d/fWV0tieVZQNO0hCSWALa8fOq+/0PysFH7vgAADTpJREFUb7lVCicIIc5qy8qyMJs09rUNJbopJ/EHgvzp9U7qy7OSvuDQbEJVY3N5s/kwvsnAvNvuaxvk539opKHSxe1X6/K/Jwkk/M6lrusrgP8DHABe1nUdoFUpdYOu6x8EHtZ13UF42hCA8J3NqK4TQggROUvG1AeZJL9dL4QQEXDYLFQXZ7CvbZDQbHqxcXTcR2PHEI0dwxzsHsFk0nDaLaTYLTjs5umfnTO+ugfGGBr18qGr62PWrljbUF/Ai2/1srd1cM67l10DYzzw//dQlJPCp967Cos5+S/kLwUJTy6VUm8Ds15mUEq9DKyK1zohhBBCCCEWoqEym6e3tTLumSTVEZ3pViY8k6iOYfa3D9HYMUTXQKhSu8NmpqYkE5Om4fb66RufYMLrx+Pz4/aefHevNC+NVdXJ1113oU7VNXZo1Mv3n3gLm9XMZ25aQ4oj4SmNCJNICCGEEEIIcZqWV7j47bZWGtuHWKcvriK8PxBkf/tQKJlsH6L90CiGATaLibrSTM5rKKC+wkVlYTpm0+x35oKGgccbwO314/b5cXv95LtSzuguolNdY7fv7z+paqzH5+cHv3mLcY+fL/z1WrIzHAlsqTiRJJdCCCGEEEKcpuriDOw2M/vaFp9c/scWxbbdvZhNGjUlmVy/uYr68iyqizMXXK/BpGmkOCxn3d279fX5J3WNDQSDPPTbt+nqH+fu96+mojA9wa0UJzq73oVCCCGEEELEgcVsQi/LYl/74or6HB528/KePi4+p5hbLq/DnoRTmiRSfbnruK6xhmHw6J+a2N1yhNuv1lldk5PoJopZyMhXIYQQQgghFqGhwsWhwQkGj3pOe98tr3WiaXD95ipJLGdhMYe6xu5qPsykP8AftnfwwpvdXLOpnEvOKUl088QcJLkUQgghhBBiEabmuDzdKUmOTvh4aXcP568sxJVuj0XTzgrr6/Px+AL8xxbFEy+0sHF5PjdeHLvqvCJyklwKIYQQQgixCCV5qWSkWNnXPnha+z33eheT/iDXnFceo5adHerLXaQ6LPxlTx91pZl87NrlmM7gQkVLgSSXQgghhBBCLIKmaSyvzGZ/2xCGsbD5fT0+P1t3dnHusjyKclJj3MIzm8Vs4tK1JVQUpvPpG1djtUj34WQnyaUQQgghhBCL1FDhYmTcR8/h8QVt/+KuHsY9fq7ZJHctF+KGi6q550PrSXNGZy5REVuSXAohhBBCCLFIyytdwMLGXfoDQbbs6KS+PIua4sxYN+2soGnaGT1n51IjyaUQQgghhBCLlJvpJN/lZF/bqcddvvJ2H0OjXt61qSIOLRMi/iS5FEIIIYQQIgINldmozmH8geCc2wQNg2de7aA8P40VVdlxbJ0Q8SPJpRBCCCGEEBFoqHDh8QVo6x2dc5tdTYfpG5zgmk0V0s1TnLUkuRRCCCGEECIC9RUuNJiza6xhGPzulXbyshysr8+Lb+OEiCNJLoUQQgghhIhAmtNKeWE6+9pnL+qjOoZp7T3K1edVYDbJx29x9pJ3txBCCCGEEBFqqHDR0j2Cx+c/ad3vt7eTkWJl88rCBLRMiPiR5FIIIYQQQogINVRmEwgaHOgcOW55x6FR9h4c5IoNZdis5gS1Toj4kORSCCGEEEKICNWVZmIxm9jffvy4y9+/2o7DZubSc0sS1DIh4keSSyGEEEIIISJks5qpLclgX9uxcZf9w252NPZzybklpDisCWydEPEhyaUQQgghhBBR0FCZTWf/GEfHfQBs2d6B2aRxxfqyBLdMiPiQ5FIIIYQQQogoaKjMBqCxY4iRcR8v7e7lgpVFuNLtCW6ZEPFhSXQDhBBCCCGEOBtUFqbjtFvY1zZIZ/8YgUCQq88rT3SzhIibJZlc6rq+DPg5kAMcAW5XSjUltlVCCCGEEOJMZjJp1JdnsefgIB5fgHV6HoXZKYlulhBxs1S7xT4EPKCUWgY8ADyc4PYIIYQQQoizQENlNkOjXtxeP9dsqkh0c4SIqyV351LX9XxgLXBFeNFjwP26rucppQYW8hxms4amLe71TaZF7ijiQuKT3CQ+yU9ilNwkPslN4pPcFhqfldWhcZcNlS5qSzNj2SRxAjmG4mO+v/NSvHNZBnQrpQIA4e894eVCCCGEEEIsWlFOCjdeXM1tVy5LdFOEiLsld+cyGgIBg2DQiPg5RPKS+CQ3iU/ykxglN4lPcpP4JLeFxOfa8ysXvK2IPvm7x5Yxz593Kd657ARKdF03A4S/F4eXCyGEEEIIIYRYhCWXXCql+oFdwK3hRbcCby50vKUQQgghhBBCiJMt1W6xdwI/13X9HmAIuD3B7RFCCCGEEEKIM9qSTC6VUo3AeYluhxBCCCGEEEKcLZZkchkBM0RW5nhq3/kGworEkfgkN4lP8pMYJTeJT3KT+CQ3iU/ykxjFx4xcyHziOkkuT08RgMuVmuh2CCGEEEIIIUQiFQEtMxdohqT2p8MObAB6gUCC2yKEEEIIIYQQ8WYmlFjuALwzV0hyKYQQQgghhBAiYktuKhIhhBBCCCGEENEnyaUQQgghhBBCiIhJcimEEEIIIYQQImKSXAohhBBCCCGEiJgkl0IIIYQQQgghIibJpRBCCCGEEEKIiElyKYQQQgghhBAiYpJcCiGEEEIIIYSImCXRDTib6LqeA/wCqAF8QBNwh1JqQNf1jwB/D5iBg8CHlFKD4f02AQ8DTqAN+IBSqj/+v8HZT9f1p4AqIAiMAZ9WSu3SdX0Z8HMgBzgC3K6UagrvM+c6EX3zxOg7wI1AJbBKKbV3xj4SoziZLT5AJ3Oc+8L7yDkuTuY5fmZdHt5Hjp84mS8O4fVfAe5lxjlOjp/4mucYagM84S+Af1JKbQnvIzGKk3ni4wC+B7yTUIxeUUp9IryPnOPiSO5cRpcBfEsppSulVgEtwDd1XV8OfB24XCm1AtgOfANA13UT8Evgb5VSy4AXgW8mpPVLw4eUUmuUUucC3wF+El7+EPBAOAYPEPonwQLWieibK0ZPAe8A2mfZR2IUP7PFZ9ZzH8g5LgHmOn7mWg5y/MTTnHHQdX0tsIkZ5zg5fhJivmPl/Uqpc8JfU4mlxCi+5orPtwgllcvC/4f+ecY+co6LI0kuo0gpNaiUemHGoleBCmAlsGvqKj7we+C28M/rAI9Salv48UPAX8WhuUuSUmpkxsNMIKjrej6wFngsvPwxYK2u63nzrYtXm5ea2WIUXr5NKdV54vYSo/iaLT7znPtAznFxNc/xM+tyOX7ia5442Al96P3kCbvI8RNnc8VoHhKjOJrjc1wacDvwz0opI7zdIZBzXCJIt9gYCV/J+iTwNPAWsEHX9SpC3SX+GkjTdT0bKGfGVUql1GFd1026rmdPdZsV0aXr+r8DVwIacDVQBnQrpQIASqmArus94eXaPOsGZnt+EblZYjSf+eInMYqB+eJzwrkP5BwXd3PFZ47lcvzE2Rxx+CrwS6VUm67rMzeX4ycB5jnHParrugZsA76olBpGYhR3s8SnhlB316/oun4poe6yXw4n/HKOizO5cxk7PyL05r5fKXUAuBv4NaEr+lMnG3+C2rakKaX+RilVDnwR+Hai2yNOJjFKbqeIz/S5L+4NE8Dc8ZHjKjmcGAdd188H1gP/mtiWiSlzHCsXKaXWABsIJTVyjkuQWeJjBqqBN5VS64F/Ap7UdT0jgc1csiS5jIFw4ZE64Gal1FSXpMeVUhuVUucBzxK6inIU6OBY9zF0Xc8l3M0sAU1fUpRSvwAuBbqAEl3XzQDh78WEipR0zrNOxNhUjMLFsuYiMUqQE+Mz27kPOcclzFzHzwnL5fhJkBn/gy4DlgOt4aIxpcAWXdevRI6fhJp5rEwNy1BKeQldCNgc3kxilCAzjqFuQjdsHgsv3w4cBpYh57i4k+QyynRd/wah/vfvDZ+AppYXhr87gPsIDUIGeANw6rp+YfjxncAT8Wvx0qHrepqu62UzHr+b0F3kfmAXcGt41a2Ern4NhKu9zboufi1fOuaJ0Zz/pCVG8TNffOY69yHnuLiZJz6+ueImx0/8zBOfbyilipVSlUqpSkIXPK9SSv0ROX7iap4YeXRdzwwv04BbCB03IDGKm1N8jnseuCK8fBmQDzTLOS7+NMMwEt2Gs4au6yuAvcABwB1e3KqUukHX9WcIXdmyAY8D90xd2dd1/QJClascHCthfSjOzT/r6bpeAPwWSAUChE5In1dK7dR1vZ5QmWoXMESoTLUK7zfnOhFdp4jRD4H3AYWErkgeUaHqyxKjOJkrPoCXOc594f3kHBcH88Sne7blSqmd4f3k+ImD+c5vJ2zXBlynjk1FIsdPnMxzDA0D/0Wo+6UZ2AfcrZTqDe8nMYqDU3xGqCZUOTYHmAS+pJR6JryfnOPiSJJLIYQQQgghhBARk26xQgghhBBCCCEiJsmlEEIIIYQQQoiISXIphBBCCCGEECJiklwKIYQQQgghhIiYJJdCCCGEEEIIISImyaUQQgghhBBCiIhJcimEEEIIIYQQImL/C9+H8jWcMXN9AAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# hyperparamters set 2\n",
        "xgb2 = XGBRegressor(n_estimators=1000, max_depth=7, learning_rate=0.001, n_jobs=-1)\n",
        "xgb2.fit(X_train, y_train)\n",
        "xgb2_metrics = metric_calc(xgb2, X_train, y_train, X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 415
        },
        "id": "puvCi8Oh7mx7",
        "outputId": "511b8d5b-cdd8-451a-9764-f6a468c9cfaf"
      },
      "execution_count": 205,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[17:42:36] WARNING: /workspace/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
            "==================== Model Metrics - Train ====================\n",
            "   R2_Score           MAE           MSE          RMSE\n",
            "0   0.02252  20808.901875  5.584568e+08  23631.690582\n",
            "\n",
            "\n",
            "==================== Model Metrics - Test ====================\n",
            "   R2_Score           MAE           MSE          RMSE\n",
            "0 -1.290425  28110.187441  1.028874e+09  32076.064493\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Metric Evaluation"
      ],
      "metadata": {
        "id": "X3TL8hNczg7V"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "ML_comparison = pd.DataFrame([lr_metrics,rf_metrics,xgb1_metrics,xgb2_metrics],\n",
        "                             index=['LinearRegression','RandomForest Regressor','XGBoostRegressor1','XGBoostRegressor2'])\n",
        "ML_comparison.sort_values(by='RMSE')"
      ],
      "metadata": {
        "id": "Vdp9Tisu8Bnq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "outputId": "2a73b2e2-6384-4d29-d864-5b4d20d64237"
      },
      "execution_count": 206,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                        R2_Score           MAE           MSE          RMSE\n",
              "RandomForest Regressor  0.315311  13879.883341  3.075666e+08  17537.575579\n",
              "XGBoostRegressor1      -0.120602  18059.543708  5.033818e+08  22436.170761\n",
              "LinearRegression       -0.440947  20409.196081  6.472827e+08  25441.751898\n",
              "XGBoostRegressor2      -1.290425  28110.187441  1.028874e+09  32076.064493"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-bf0c7895-e969-435e-b018-fbb1d9792903\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>R2_Score</th>\n",
              "      <th>MAE</th>\n",
              "      <th>MSE</th>\n",
              "      <th>RMSE</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>RandomForest Regressor</th>\n",
              "      <td>0.315311</td>\n",
              "      <td>13879.883341</td>\n",
              "      <td>3.075666e+08</td>\n",
              "      <td>17537.575579</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoostRegressor1</th>\n",
              "      <td>-0.120602</td>\n",
              "      <td>18059.543708</td>\n",
              "      <td>5.033818e+08</td>\n",
              "      <td>22436.170761</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LinearRegression</th>\n",
              "      <td>-0.440947</td>\n",
              "      <td>20409.196081</td>\n",
              "      <td>6.472827e+08</td>\n",
              "      <td>25441.751898</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBoostRegressor2</th>\n",
              "      <td>-1.290425</td>\n",
              "      <td>28110.187441</td>\n",
              "      <td>1.028874e+09</td>\n",
              "      <td>32076.064493</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-bf0c7895-e969-435e-b018-fbb1d9792903')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-bf0c7895-e969-435e-b018-fbb1d9792903 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-bf0c7895-e969-435e-b018-fbb1d9792903');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 206
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* RandomForestRegressor is performing well, and able to capture the trends\n",
        "* We can start our final model training with __RandomForest Regressor__"
      ],
      "metadata": {
        "id": "scdvB34-Gp7Q"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model Finalization"
      ],
      "metadata": {
        "id": "Zv89g8U9IZG2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now we will train our model on whole year 2020 data and will test it on Q1 of 2021.\n",
        "* We can get Holiday information for Q1 of 2021 by using same Holiday API.\n",
        "* We can extract all date features from date column"
      ],
      "metadata": {
        "id": "RVZl-GbOJoKN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df_modeling.tail()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "nXoqDyuDGKST",
        "outputId": "d255419d-392f-4998-c031-f7bf6929eb7b"
      },
      "execution_count": 207,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          Date  Trips    Weekday  Holiday  TMAX  TMIN  DOY  DOW  DOM  year  \\\n",
              "361 2020-12-27  23764     Sunday        0    37    24  362    6   27  2020   \n",
              "362 2020-12-28  38784     Monday        0    50    35  363    0   28  2020   \n",
              "363 2020-12-29  33202    Tuesday        0    43    30  364    1   29  2020   \n",
              "364 2020-12-30  35705  Wednesday        0    43    28  365    2   30  2020   \n",
              "365 2020-12-31  26334   Thursday        0    48    36  366    3   31  2020   \n",
              "\n",
              "     month_num     month  \n",
              "361         12  December  \n",
              "362         12  December  \n",
              "363         12  December  \n",
              "364         12  December  \n",
              "365         12  December  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f33901df-4885-4f29-9dad-b92e374dc54e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Trips</th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "      <th>month</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>361</th>\n",
              "      <td>2020-12-27</td>\n",
              "      <td>23764</td>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>37</td>\n",
              "      <td>24</td>\n",
              "      <td>362</td>\n",
              "      <td>6</td>\n",
              "      <td>27</td>\n",
              "      <td>2020</td>\n",
              "      <td>12</td>\n",
              "      <td>December</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>362</th>\n",
              "      <td>2020-12-28</td>\n",
              "      <td>38784</td>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>50</td>\n",
              "      <td>35</td>\n",
              "      <td>363</td>\n",
              "      <td>0</td>\n",
              "      <td>28</td>\n",
              "      <td>2020</td>\n",
              "      <td>12</td>\n",
              "      <td>December</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>363</th>\n",
              "      <td>2020-12-29</td>\n",
              "      <td>33202</td>\n",
              "      <td>Tuesday</td>\n",
              "      <td>0</td>\n",
              "      <td>43</td>\n",
              "      <td>30</td>\n",
              "      <td>364</td>\n",
              "      <td>1</td>\n",
              "      <td>29</td>\n",
              "      <td>2020</td>\n",
              "      <td>12</td>\n",
              "      <td>December</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>364</th>\n",
              "      <td>2020-12-30</td>\n",
              "      <td>35705</td>\n",
              "      <td>Wednesday</td>\n",
              "      <td>0</td>\n",
              "      <td>43</td>\n",
              "      <td>28</td>\n",
              "      <td>365</td>\n",
              "      <td>2</td>\n",
              "      <td>30</td>\n",
              "      <td>2020</td>\n",
              "      <td>12</td>\n",
              "      <td>December</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>365</th>\n",
              "      <td>2020-12-31</td>\n",
              "      <td>26334</td>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "      <td>48</td>\n",
              "      <td>36</td>\n",
              "      <td>366</td>\n",
              "      <td>3</td>\n",
              "      <td>31</td>\n",
              "      <td>2020</td>\n",
              "      <td>12</td>\n",
              "      <td>December</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f33901df-4885-4f29-9dad-b92e374dc54e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f33901df-4885-4f29-9dad-b92e374dc54e 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-f33901df-4885-4f29-9dad-b92e374dc54e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 207
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_test = pd.DataFrame(columns=['Date', 'Weekday', 'Holiday', 'TMAX', 'TMIN', 'DOY', 'DOW', 'DOM', 'month', 'year', 'month_num'])"
      ],
      "metadata": {
        "id": "xUpF1l3GJijw"
      },
      "execution_count": 208,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_test['Date'] = pd.date_range(start='1/1/2021',end='31/3/2021')\n",
        "df_test['Weekday'] = df_test.Date.dt.day_name()\n",
        "df_test['Holiday'] = [1 if d in holidays.US() else 0 for d in df_test[\"Date\"]]\n",
        "df_test['DOY'] = df_test.Date.dt.dayofyear\n",
        "df_test['DOW'] = df_test.Date.dt.dayofweek\n",
        "df_test['DOM'] = df_test.Date.dt.day\n",
        "df_test['month'] = df_test.Date.dt.month_name()\n",
        "df_test['year'] = df_test.Date.dt.year\n",
        "df_test['month_num'] = df_test.Date.dt.month"
      ],
      "metadata": {
        "id": "Jd7i2mj4dYJa"
      },
      "execution_count": 209,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_test.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "iZKQiEtgdkIU",
        "outputId": "c5bb2d67-4589-4b3e-f53f-b92484be3f13"
      },
      "execution_count": 210,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        Date   Weekday  Holiday TMAX TMIN  DOY  DOW  DOM    month  year  \\\n",
              "0 2021-01-01    Friday        1  NaN  NaN    1    4    1  January  2021   \n",
              "1 2021-01-02  Saturday        0  NaN  NaN    2    5    2  January  2021   \n",
              "2 2021-01-03    Sunday        0  NaN  NaN    3    6    3  January  2021   \n",
              "3 2021-01-04    Monday        0  NaN  NaN    4    0    4  January  2021   \n",
              "4 2021-01-05   Tuesday        0  NaN  NaN    5    1    5  January  2021   \n",
              "\n",
              "   month_num  \n",
              "0          1  \n",
              "1          1  \n",
              "2          1  \n",
              "3          1  \n",
              "4          1  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-767cc40d-abc9-49af-94f0-3dccd5d71277\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Date</th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>month</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2021-01-01</td>\n",
              "      <td>Friday</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2021-01-02</td>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2021-01-03</td>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>3</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2021-01-04</td>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2021-01-05</td>\n",
              "      <td>Tuesday</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-767cc40d-abc9-49af-94f0-3dccd5d71277')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-767cc40d-abc9-49af-94f0-3dccd5d71277 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-767cc40d-abc9-49af-94f0-3dccd5d71277');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 210
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "jan_min = int(df_modeling[df_modeling['month_num']==1]['TMIN'].mean())\n",
        "feb_min = int(df_modeling[df_modeling['month_num']==2]['TMIN'].mean())\n",
        "mar_min = int(df_modeling[df_modeling['month_num']==3]['TMIN'].mean())\n",
        "# avgerage by previous year month\n",
        "jan_max = int(df_modeling[df_modeling['month_num']==1]['TMAX'].mean())\n",
        "feb_max = int(df_modeling[df_modeling['month_num']==2]['TMAX'].mean())\n",
        "mar_max = int(df_modeling[df_modeling['month_num']==3]['TMAX'].mean())"
      ],
      "metadata": {
        "id": "0pBOFa3Du-WR"
      },
      "execution_count": 211,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_test['TMIN'] = df_test['month_num'].apply(lambda x: jan_min if x==1 else(feb_min if x==2 else mar_min))\n",
        "df_test['TMAX'] = df_test['month_num'].apply(lambda x: jan_max if x==1 else(feb_max if x==2 else mar_max))"
      ],
      "metadata": {
        "id": "PImXPxcMtva6"
      },
      "execution_count": 212,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test = df_test.drop(columns=['Date'])\n",
        "features = test.columns.to_list()\n",
        "test.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "OAbaj3oUutKc",
        "outputId": "e5d41124-e46f-4cfd-fe90-2ee8f97602a9"
      },
      "execution_count": 222,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    Weekday  Holiday  TMAX  TMIN  DOY  DOW  DOM    month  year  month_num\n",
              "0    Friday        1    45    33    1    4    1  January  2021          1\n",
              "1  Saturday        0    45    33    2    5    2  January  2021          1\n",
              "2    Sunday        0    45    33    3    6    3  January  2021          1\n",
              "3    Monday        0    45    33    4    0    4  January  2021          1\n",
              "4   Tuesday        0    45    33    5    1    5  January  2021          1"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-014f77b8-6e7d-4f09-9a0b-3e9b64be81bd\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>month</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Friday</td>\n",
              "      <td>1</td>\n",
              "      <td>45</td>\n",
              "      <td>33</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>33</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>33</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>3</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Monday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>33</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Tuesday</td>\n",
              "      <td>0</td>\n",
              "      <td>45</td>\n",
              "      <td>33</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>January</td>\n",
              "      <td>2021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-014f77b8-6e7d-4f09-9a0b-3e9b64be81bd')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-014f77b8-6e7d-4f09-9a0b-3e9b64be81bd 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-014f77b8-6e7d-4f09-9a0b-3e9b64be81bd');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 222
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X = df_modeling[features]\n",
        "y = df_modeling['Trips']\n",
        "X.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "ZNPTtMC13_rf",
        "outputId": "aac15080-c438-40bf-edfc-062f994d1125"
      },
      "execution_count": 223,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     Weekday  Holiday  TMAX  TMIN  DOY  DOW  DOM    month  year  month_num\n",
              "0  Wednesday        1    41    34    1    2    1  January  2020          1\n",
              "1   Thursday        0    49    33    2    3    2  January  2020          1\n",
              "2     Friday        0    49    44    3    4    3  January  2020          1\n",
              "3   Saturday        0    51    41    4    5    4  January  2020          1\n",
              "4     Sunday        0    42    35    5    6    5  January  2020          1"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f3daa27f-969c-46c9-92dc-5f877725a904\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Weekday</th>\n",
              "      <th>Holiday</th>\n",
              "      <th>TMAX</th>\n",
              "      <th>TMIN</th>\n",
              "      <th>DOY</th>\n",
              "      <th>DOW</th>\n",
              "      <th>DOM</th>\n",
              "      <th>month</th>\n",
              "      <th>year</th>\n",
              "      <th>month_num</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Wednesday</td>\n",
              "      <td>1</td>\n",
              "      <td>41</td>\n",
              "      <td>34</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>January</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Thursday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>33</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>January</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Friday</td>\n",
              "      <td>0</td>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>January</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Saturday</td>\n",
              "      <td>0</td>\n",
              "      <td>51</td>\n",
              "      <td>41</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>January</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Sunday</td>\n",
              "      <td>0</td>\n",
              "      <td>42</td>\n",
              "      <td>35</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>January</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f3daa27f-969c-46c9-92dc-5f877725a904')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f3daa27f-969c-46c9-92dc-5f877725a904 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-f3daa27f-969c-46c9-92dc-5f877725a904');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 223
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Making pipeline"
      ],
      "metadata": {
        "id": "J9kWsIQa0dBH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# for getting dummy\n",
        "ohe = ColumnTransformer([('ohe_week_month',OneHotEncoder(),[0,7])],remainder='passthrough')\n",
        "\n",
        "#Creating model\n",
        "model = RandomForestRegressor(n_estimators=1000, max_depth=8)\n",
        "# pipeline\n",
        "pipe = Pipeline([\n",
        "    ('columnTransformer',ohe),\n",
        "    ('model',model)\n",
        "])"
      ],
      "metadata": {
        "id": "z-Q2MIP3v6vI"
      },
      "execution_count": 224,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pipe.fit(X,y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 192
        },
        "id": "kU1F8Jt64_Vs",
        "outputId": "487b9c05-4d9f-4342-e6c6-2f251d5a4e02"
      },
      "execution_count": 225,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('columnTransformer',\n",
              "                 ColumnTransformer(remainder='passthrough',\n",
              "                                   transformers=[('ohe_week_month',\n",
              "                                                  OneHotEncoder(), [0, 7])])),\n",
              "                ('model',\n",
              "                 RandomForestRegressor(max_depth=8, n_estimators=1000))])"
            ],
            "text/html": [
              "<style>#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec {color: black;background-color: white;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec pre{padding: 0;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-toggleable {background-color: white;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-estimator:hover {background-color: #d4ebff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-item {z-index: 1;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 2em;bottom: 0;left: 50%;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel-item {display: flex;flex-direction: column;position: relative;background-color: white;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-parallel-item:only-child::after {width: 0;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;position: relative;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-label label {font-family: monospace;font-weight: bold;background-color: white;display: inline-block;line-height: 1.2em;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-label-container {position: relative;z-index: 2;text-align: center;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-d0595fc5-0d2d-4753-81a6-9d79b51e18ec\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;columnTransformer&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   transformers=[(&#x27;ohe_week_month&#x27;,\n",
              "                                                  OneHotEncoder(), [0, 7])])),\n",
              "                (&#x27;model&#x27;,\n",
              "                 RandomForestRegressor(max_depth=8, n_estimators=1000))])</pre><b>Please rerun this cell to show the HTML repr or trust the notebook.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"ad08e9ca-a30e-43d6-9cb2-bdda78c4ef96\" type=\"checkbox\" ><label for=\"ad08e9ca-a30e-43d6-9cb2-bdda78c4ef96\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;columnTransformer&#x27;,\n",
              "                 ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                                   transformers=[(&#x27;ohe_week_month&#x27;,\n",
              "                                                  OneHotEncoder(), [0, 7])])),\n",
              "                (&#x27;model&#x27;,\n",
              "                 RandomForestRegressor(max_depth=8, n_estimators=1000))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"406daeab-3d31-4918-9286-28b55c932363\" type=\"checkbox\" ><label for=\"406daeab-3d31-4918-9286-28b55c932363\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">columnTransformer: ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(remainder=&#x27;passthrough&#x27;,\n",
              "                  transformers=[(&#x27;ohe_week_month&#x27;, OneHotEncoder(), [0, 7])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"3917928a-d4d1-49dd-a495-67732aa4835f\" type=\"checkbox\" ><label for=\"3917928a-d4d1-49dd-a495-67732aa4835f\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ohe_week_month</label><div class=\"sk-toggleable__content\"><pre>[0, 7]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"f6a10b17-2a96-486b-b28e-e0b616e333b2\" type=\"checkbox\" ><label for=\"f6a10b17-2a96-486b-b28e-e0b616e333b2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder()</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"56dc3724-ac25-48c4-85c1-a7a9a53976ea\" type=\"checkbox\" ><label for=\"56dc3724-ac25-48c4-85c1-a7a9a53976ea\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">remainder</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Holiday&#x27;, &#x27;TMAX&#x27;, &#x27;TMIN&#x27;, &#x27;DOY&#x27;, &#x27;DOW&#x27;, &#x27;DOM&#x27;, &#x27;year&#x27;, &#x27;month_num&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"3cc356d5-01b9-45b9-8142-b167f1a1c016\" type=\"checkbox\" ><label for=\"3cc356d5-01b9-45b9-8142-b167f1a1c016\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">passthrough</label><div class=\"sk-toggleable__content\"><pre>passthrough</pre></div></div></div></div></div></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"3057486b-4856-40ce-be1b-928b53db193e\" type=\"checkbox\" ><label for=\"3057486b-4856-40ce-be1b-928b53db193e\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">RandomForestRegressor</label><div class=\"sk-toggleable__content\"><pre>RandomForestRegressor(max_depth=8, n_estimators=1000)</pre></div></div></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 225
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# prediction\n",
        "y_pred = pipe.predict(test)\n",
        "df_test['Trips'] = y_pred"
      ],
      "metadata": {
        "id": "XNOSGHwz5faB"
      },
      "execution_count": 228,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.rcParams.update({'figure.figsize': (15, 5)})\n",
        "fig, ax = plt.subplots()\n",
        "sns.lineplot(data=df_modeling.tail(120), x='Date', y='Trips')\n",
        "sns.lineplot(data=df_test, x='Date', y='Trips', color='r')\n",
        "plt.grid(linestyle='-', linewidth=0.3)\n",
        "legend_elements = [Line2D([0], [0], color='b', lw=4, label='Actual'),\n",
        "                Line2D([0], [0], color='r', lw=4, label='Predicted')]\n",
        "plt.legend(handles=legend_elements)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "id": "9H43JRkf6TPC",
        "outputId": "cf96f80f-7f34-42ff-8598-ad102cc3496a"
      },
      "execution_count": 234,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f9076f64100>"
            ]
          },
          "metadata": {},
          "execution_count": 234
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Reference:\n",
        "* YouTube channel: [CodeEmporium](https://www.youtube.com/watch?v=Wsfz3i1AXzY)\n",
        "* Python Holidays Library : [Holidays](https://pypi.org/project/holidays/)\n",
        "* Weather Data: [NOAA](https://www.ncdc.noaa.gov/cdo-web/)\n",
        "* Data Vizualization : \n",
        "  * https://stackoverflow.com/questions/35906717/marking-specific-dates-when-visualizing-a-time-series\n",
        "  * https://matplotlib.org/stable/gallery/text_labels_and_annotations/custom_legends.html\n",
        "* Handling large data in Pandas : [Article_link](https://pythonspeed.com/articles/pandas-load-less-data/)\n",
        "* Visualization cheatsheet: [link](https://www.data-to-viz.com/)\n",
        "* Analysis by Others: [NYC-Taxi](https://medium.com/analytics-vidhya/exploratory-data-analysis-of-nyc-taxi-trip-duration-dataset-using-python-257fdef2749e) [NYC-Data science academy](https://nycdatascience.com/blog/student-works/embrace-the-new-normal-citi-bike-jersey-city-and-hoboken-data-analysis/)\n",
        "* `metric_calc` function inspired from [github.com/prateeknigam9/](https://github.com/prateeknigam9/Hinglish_Classifier/blob/main/CaseStudy_Hinglish.ipynb)"
      ],
      "metadata": {
        "id": "qjzD_bsq9Btg"
      }
    }
  ]
}