Bulldozer_price_regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "48a53f6b",
   "metadata": {},
   "source": [
    "# Predicting the sales price of bulldozers using Machine Learning.\n",
    "[Source : ZTM Python for Data Science Course]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5e3be36",
   "metadata": {},
   "source": [
    "## 1. Problem Definition\n",
    "> How well can we predict the future sale price of a bulldozer given its characteristics and previous record of sales.(Regression Problem).\n",
    "\n",
    "## 2. Data\n",
    "The data is used from kaggle's blue book for bulldozers, link: https://www.kaggle.com/c/bluebook-for-bulldozers/data\n",
    "\n",
    "There are three main datasets:\n",
    "\n",
    "* Train.csv is the training set, which contains data through the end of 2011.\n",
    "* Valid.csv is the validation set, which contains data from January 1, 2012 - April 30, 2012 You make predictions on this set  throughout the majority of the competition. Your score on this set is used to create the public leaderboard.\n",
    "* Test.csv is the test set, which won't be released until the last week of the competition. It contains data from May 1, 2012 - November 2012. Your score on the test set determines your final rank for the competition.\n",
    "\n",
    "## 3. Evaluation\n",
    "The evaluation metric for this competition is the RMSLE (root mean squared log error) between the actual and predicted auction prices. For more on the evaluation check: https://www.kaggle.com/c/bluebook-for-bulldozers/overview/evaluation\n",
    "\n",
    "**Note:** Our goal for this project will be to build a machine learning model which minimises the RMSLE.\n",
    "\n",
    "## 4. Features\n",
    "Kaggle provides the data dictionary describing all the features of a dataset. You can view this data dictionary here-\n",
    "    https://www.kaggle.com/competitions/bluebook-for-bulldozers/data\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e42dbdce",
   "metadata": {},
   "source": [
    "## 5. Exploring Our Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0e1a9f8",
   "metadata": {},
   "source": [
    "### i. Import required libraries "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "b5354478",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c346b8e6",
   "metadata": {},
   "source": [
    "### ii. Import Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "71721a18",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing training and validation datasets.\n",
    "df = pd.read_csv('Data/bluebook-for-bulldozers/TrainAndValid.csv',\n",
    "                 low_memory=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "697ebf14",
   "metadata": {},
   "source": [
    "### iii. Find the information about data like how many columns, rows, data types etc. and visualize it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8781735e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 412698 entries, 0 to 412697\n",
      "Data columns (total 53 columns):\n",
      " #   Column                    Non-Null Count   Dtype  \n",
      "---  ------                    --------------   -----  \n",
      " 0   SalesID                   412698 non-null  int64  \n",
      " 1   SalePrice                 412698 non-null  float64\n",
      " 2   MachineID                 412698 non-null  int64  \n",
      " 3   ModelID                   412698 non-null  int64  \n",
      " 4   datasource                412698 non-null  int64  \n",
      " 5   auctioneerID              392562 non-null  float64\n",
      " 6   YearMade                  412698 non-null  int64  \n",
      " 7   MachineHoursCurrentMeter  147504 non-null  float64\n",
      " 8   UsageBand                 73670 non-null   object \n",
      " 9   saledate                  412698 non-null  object \n",
      " 10  fiModelDesc               412698 non-null  object \n",
      " 11  fiBaseModel               412698 non-null  object \n",
      " 12  fiSecondaryDesc           271971 non-null  object \n",
      " 13  fiModelSeries             58667 non-null   object \n",
      " 14  fiModelDescriptor         74816 non-null   object \n",
      " 15  ProductSize               196093 non-null  object \n",
      " 16  fiProductClassDesc        412698 non-null  object \n",
      " 17  state                     412698 non-null  object \n",
      " 18  ProductGroup              412698 non-null  object \n",
      " 19  ProductGroupDesc          412698 non-null  object \n",
      " 20  Drive_System              107087 non-null  object \n",
      " 21  Enclosure                 412364 non-null  object \n",
      " 22  Forks                     197715 non-null  object \n",
      " 23  Pad_Type                  81096 non-null   object \n",
      " 24  Ride_Control              152728 non-null  object \n",
      " 25  Stick                     81096 non-null   object \n",
      " 26  Transmission              188007 non-null  object \n",
      " 27  Turbocharged              81096 non-null   object \n",
      " 28  Blade_Extension           25983 non-null   object \n",
      " 29  Blade_Width               25983 non-null   object \n",
      " 30  Enclosure_Type            25983 non-null   object \n",
      " 31  Engine_Horsepower         25983 non-null   object \n",
      " 32  Hydraulics                330133 non-null  object \n",
      " 33  Pushblock                 25983 non-null   object \n",
      " 34  Ripper                    106945 non-null  object \n",
      " 35  Scarifier                 25994 non-null   object \n",
      " 36  Tip_Control               25983 non-null   object \n",
      " 37  Tire_Size                 97638 non-null   object \n",
      " 38  Coupler                   220679 non-null  object \n",
      " 39  Coupler_System            44974 non-null   object \n",
      " 40  Grouser_Tracks            44875 non-null   object \n",
      " 41  Hydraulics_Flow           44875 non-null   object \n",
      " 42  Track_Type                102193 non-null  object \n",
      " 43  Undercarriage_Pad_Width   102916 non-null  object \n",
      " 44  Stick_Length              102261 non-null  object \n",
      " 45  Thumb                     102332 non-null  object \n",
      " 46  Pattern_Changer           102261 non-null  object \n",
      " 47  Grouser_Type              102193 non-null  object \n",
      " 48  Backhoe_Mounting          80712 non-null   object \n",
      " 49  Blade_Type                81875 non-null   object \n",
      " 50  Travel_Controls           81877 non-null   object \n",
      " 51  Differential_Type         71564 non-null   object \n",
      " 52  Steering_Controls         71522 non-null   object \n",
      "dtypes: float64(3), int64(5), object(45)\n",
      "memory usage: 166.9+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "ec6fd039",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['SalesID', 'SalePrice', 'MachineID', 'ModelID', 'datasource',\n",
       "       'auctioneerID', 'YearMade', 'MachineHoursCurrentMeter', 'UsageBand',\n",
       "       'saledate', 'fiModelDesc', 'fiBaseModel', 'fiSecondaryDesc',\n",
       "       'fiModelSeries', 'fiModelDescriptor', 'ProductSize',\n",
       "       'fiProductClassDesc', 'state', 'ProductGroup', 'ProductGroupDesc',\n",
       "       'Drive_System', 'Enclosure', 'Forks', 'Pad_Type', 'Ride_Control',\n",
       "       'Stick', 'Transmission', 'Turbocharged', 'Blade_Extension',\n",
       "       'Blade_Width', 'Enclosure_Type', 'Engine_Horsepower', 'Hydraulics',\n",
       "       'Pushblock', 'Ripper', 'Scarifier', 'Tip_Control', 'Tire_Size',\n",
       "       'Coupler', 'Coupler_System', 'Grouser_Tracks', 'Hydraulics_Flow',\n",
       "       'Track_Type', 'Undercarriage_Pad_Width', 'Stick_Length', 'Thumb',\n",
       "       'Pattern_Changer', 'Grouser_Type', 'Backhoe_Mounting', 'Blade_Type',\n",
       "       'Travel_Controls', 'Differential_Type', 'Steering_Controls'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3926ae12",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10,6))\n",
    "ax.scatter(df['saledate'][:1000], df['SalePrice'][:1000],\n",
    "           c=df['SalePrice'][:1000], cmap='Blues');\n",
    "plt.xlabel('Date of Sale');\n",
    "plt.ylabel('Sale price of bulldozers in USD');\n",
    "plt.title('Sale date vs. Sale price');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0934c13f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.SalePrice.plot.hist();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fe28b151",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      11/16/2006 0:00\n",
       "1       3/26/2004 0:00\n",
       "2       2/26/2004 0:00\n",
       "3       5/19/2011 0:00\n",
       "4       7/23/2009 0:00\n",
       "            ...       \n",
       "995     7/16/2009 0:00\n",
       "996     6/14/2007 0:00\n",
       "997     9/22/2005 0:00\n",
       "998     7/28/2005 0:00\n",
       "999     6/16/2011 0:00\n",
       "Name: saledate, Length: 1000, dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.saledate[:1000]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8f9434c",
   "metadata": {},
   "source": [
    "### Parsing dates\n",
    "\n",
    "When we work with time series data, we want to enrich the time and date component as much as possible.\n",
    "\n",
    "We can do that by telling pandas which of our columns has dates in it using the `parse_dates` parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ca2632a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import dataset again with parse date\n",
    "df = pd.read_csv('Data/bluebook-for-bulldozers/TrainAndValid.csv',\n",
    "                  low_memory=False,\n",
    "                  parse_dates=['saledate'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d96091e8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        2006-11-16\n",
       "1        2004-03-26\n",
       "2        2004-02-26\n",
       "3        2011-05-19\n",
       "4        2009-07-23\n",
       "            ...    \n",
       "412693   2012-03-07\n",
       "412694   2012-01-28\n",
       "412695   2012-01-28\n",
       "412696   2012-03-07\n",
       "412697   2012-01-28\n",
       "Name: saledate, Length: 412698, dtype: datetime64[ns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.saledate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e1dcb022",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10,6))\n",
    "ax.scatter(df['saledate'][:1000], df['SalePrice'][:1000],\n",
    "           c=df['SalePrice'][:1000],\n",
    "           cmap='Blues');\n",
    "plt.xlabel('Date of Sale');\n",
    "plt.ylabel('Sale price of bulldozers in USD');\n",
    "plt.title('Sale date vs. Sale price');"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f4cd320",
   "metadata": {},
   "source": [
    "### iv. Find about missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4e07ee7b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SalesID                          0\n",
       "SalePrice                        0\n",
       "MachineID                        0\n",
       "ModelID                          0\n",
       "datasource                       0\n",
       "auctioneerID                 20136\n",
       "YearMade                         0\n",
       "MachineHoursCurrentMeter    265194\n",
       "UsageBand                   339028\n",
       "saledate                         0\n",
       "fiModelDesc                      0\n",
       "fiBaseModel                      0\n",
       "fiSecondaryDesc             140727\n",
       "fiModelSeries               354031\n",
       "fiModelDescriptor           337882\n",
       "ProductSize                 216605\n",
       "fiProductClassDesc               0\n",
       "state                            0\n",
       "ProductGroup                     0\n",
       "ProductGroupDesc                 0\n",
       "Drive_System                305611\n",
       "Enclosure                      334\n",
       "Forks                       214983\n",
       "Pad_Type                    331602\n",
       "Ride_Control                259970\n",
       "Stick                       331602\n",
       "Transmission                224691\n",
       "Turbocharged                331602\n",
       "Blade_Extension             386715\n",
       "Blade_Width                 386715\n",
       "Enclosure_Type              386715\n",
       "Engine_Horsepower           386715\n",
       "Hydraulics                   82565\n",
       "Pushblock                   386715\n",
       "Ripper                      305753\n",
       "Scarifier                   386704\n",
       "Tip_Control                 386715\n",
       "Tire_Size                   315060\n",
       "Coupler                     192019\n",
       "Coupler_System              367724\n",
       "Grouser_Tracks              367823\n",
       "Hydraulics_Flow             367823\n",
       "Track_Type                  310505\n",
       "Undercarriage_Pad_Width     309782\n",
       "Stick_Length                310437\n",
       "Thumb                       310366\n",
       "Pattern_Changer             310437\n",
       "Grouser_Type                310505\n",
       "Backhoe_Mounting            331986\n",
       "Blade_Type                  330823\n",
       "Travel_Controls             330821\n",
       "Differential_Type           341134\n",
       "Steering_Controls           341176\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3b782e91",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>SalePrice</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>saledate</th>\n",
       "      <th>...</th>\n",
       "      <th>Undercarriage_Pad_Width</th>\n",
       "      <th>Stick_Length</th>\n",
       "      <th>Thumb</th>\n",
       "      <th>Pattern_Changer</th>\n",
       "      <th>Grouser_Type</th>\n",
       "      <th>Backhoe_Mounting</th>\n",
       "      <th>Blade_Type</th>\n",
       "      <th>Travel_Controls</th>\n",
       "      <th>Differential_Type</th>\n",
       "      <th>Steering_Controls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1139246</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>999089</td>\n",
       "      <td>3157</td>\n",
       "      <td>121</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2004</td>\n",
       "      <td>68.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>2006-11-16</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>Conventional</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1139248</td>\n",
       "      <td>57000.0</td>\n",
       "      <td>117657</td>\n",
       "      <td>77</td>\n",
       "      <td>121</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1996</td>\n",
       "      <td>4640.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>2004-03-26</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>Conventional</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1139249</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>434808</td>\n",
       "      <td>7009</td>\n",
       "      <td>121</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2001</td>\n",
       "      <td>2838.0</td>\n",
       "      <td>High</td>\n",
       "      <td>2004-02-26</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1139251</td>\n",
       "      <td>38500.0</td>\n",
       "      <td>1026470</td>\n",
       "      <td>332</td>\n",
       "      <td>121</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2001</td>\n",
       "      <td>3486.0</td>\n",
       "      <td>High</td>\n",
       "      <td>2011-05-19</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1139253</td>\n",
       "      <td>11000.0</td>\n",
       "      <td>1057373</td>\n",
       "      <td>17311</td>\n",
       "      <td>121</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2007</td>\n",
       "      <td>722.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>2009-07-23</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 53 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  SalePrice  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1139246    66000.0     999089     3157         121           3.0      2004   \n",
       "1  1139248    57000.0     117657       77         121           3.0      1996   \n",
       "2  1139249    10000.0     434808     7009         121           3.0      2001   \n",
       "3  1139251    38500.0    1026470      332         121           3.0      2001   \n",
       "4  1139253    11000.0    1057373    17311         121           3.0      2007   \n",
       "\n",
       "   MachineHoursCurrentMeter UsageBand   saledate  ... Undercarriage_Pad_Width  \\\n",
       "0                      68.0       Low 2006-11-16  ...                     NaN   \n",
       "1                    4640.0       Low 2004-03-26  ...                     NaN   \n",
       "2                    2838.0      High 2004-02-26  ...                     NaN   \n",
       "3                    3486.0      High 2011-05-19  ...                     NaN   \n",
       "4                     722.0    Medium 2009-07-23  ...                     NaN   \n",
       "\n",
       "  Stick_Length Thumb Pattern_Changer Grouser_Type Backhoe_Mounting Blade_Type  \\\n",
       "0          NaN   NaN             NaN          NaN              NaN        NaN   \n",
       "1          NaN   NaN             NaN          NaN              NaN        NaN   \n",
       "2          NaN   NaN             NaN          NaN              NaN        NaN   \n",
       "3          NaN   NaN             NaN          NaN              NaN        NaN   \n",
       "4          NaN   NaN             NaN          NaN              NaN        NaN   \n",
       "\n",
       "  Travel_Controls Differential_Type Steering_Controls  \n",
       "0             NaN          Standard      Conventional  \n",
       "1             NaN          Standard      Conventional  \n",
       "2             NaN               NaN               NaN  \n",
       "3             NaN               NaN               NaN  \n",
       "4             NaN               NaN               NaN  \n",
       "\n",
       "[5 rows x 53 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()  ## will not show all the columns when there are so many columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "2c52c4fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SalesID</th>\n",
       "      <td>1139246</td>\n",
       "      <td>1139248</td>\n",
       "      <td>1139249</td>\n",
       "      <td>1139251</td>\n",
       "      <td>1139253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>66000.0</td>\n",
       "      <td>57000.0</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>38500.0</td>\n",
       "      <td>11000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineID</th>\n",
       "      <td>999089</td>\n",
       "      <td>117657</td>\n",
       "      <td>434808</td>\n",
       "      <td>1026470</td>\n",
       "      <td>1057373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ModelID</th>\n",
       "      <td>3157</td>\n",
       "      <td>77</td>\n",
       "      <td>7009</td>\n",
       "      <td>332</td>\n",
       "      <td>17311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datasource</th>\n",
       "      <td>121</td>\n",
       "      <td>121</td>\n",
       "      <td>121</td>\n",
       "      <td>121</td>\n",
       "      <td>121</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>auctioneerID</th>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YearMade</th>\n",
       "      <td>2004</td>\n",
       "      <td>1996</td>\n",
       "      <td>2001</td>\n",
       "      <td>2001</td>\n",
       "      <td>2007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <td>68.0</td>\n",
       "      <td>4640.0</td>\n",
       "      <td>2838.0</td>\n",
       "      <td>3486.0</td>\n",
       "      <td>722.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UsageBand</th>\n",
       "      <td>Low</td>\n",
       "      <td>Low</td>\n",
       "      <td>High</td>\n",
       "      <td>High</td>\n",
       "      <td>Medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saledate</th>\n",
       "      <td>2006-11-16 00:00:00</td>\n",
       "      <td>2004-03-26 00:00:00</td>\n",
       "      <td>2004-02-26 00:00:00</td>\n",
       "      <td>2011-05-19 00:00:00</td>\n",
       "      <td>2009-07-23 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDesc</th>\n",
       "      <td>521D</td>\n",
       "      <td>950FII</td>\n",
       "      <td>226</td>\n",
       "      <td>PC120-6E</td>\n",
       "      <td>S175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiBaseModel</th>\n",
       "      <td>521</td>\n",
       "      <td>950</td>\n",
       "      <td>226</td>\n",
       "      <td>PC120</td>\n",
       "      <td>S175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiSecondaryDesc</th>\n",
       "      <td>D</td>\n",
       "      <td>F</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelSeries</th>\n",
       "      <td>NaN</td>\n",
       "      <td>II</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-6E</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDescriptor</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductSize</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Medium</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Small</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiProductClassDesc</th>\n",
       "      <td>Wheel Loader - 110.0 to 120.0 Horsepower</td>\n",
       "      <td>Wheel Loader - 150.0 to 175.0 Horsepower</td>\n",
       "      <td>Skid Steer Loader - 1351.0 to 1601.0 Lb Operat...</td>\n",
       "      <td>Hydraulic Excavator, Track - 12.0 to 14.0 Metr...</td>\n",
       "      <td>Skid Steer Loader - 1601.0 to 1751.0 Lb Operat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>North Carolina</td>\n",
       "      <td>New York</td>\n",
       "      <td>Texas</td>\n",
       "      <td>New York</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroup</th>\n",
       "      <td>WL</td>\n",
       "      <td>WL</td>\n",
       "      <td>SSL</td>\n",
       "      <td>TEX</td>\n",
       "      <td>SSL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroupDesc</th>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Skid Steer Loaders</td>\n",
       "      <td>Track Excavators</td>\n",
       "      <td>Skid Steer Loaders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drive_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure</th>\n",
       "      <td>EROPS w AC</td>\n",
       "      <td>EROPS w AC</td>\n",
       "      <td>OROPS</td>\n",
       "      <td>EROPS w AC</td>\n",
       "      <td>EROPS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Forks</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pad_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ride_Control</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Transmission</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turbocharged</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Extension</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Engine_Horsepower</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics</th>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>Auxiliary</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>Auxiliary</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pushblock</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ripper</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Scarifier</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tip_Control</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tire_Size</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>23.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Tracks</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics_Flow</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Track_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Undercarriage_Pad_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick_Length</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thumb</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pattern_Changer</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Backhoe_Mounting</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Travel_Controls</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Differential_Type</th>\n",
       "      <td>Standard</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Steering_Controls</th>\n",
       "      <td>Conventional</td>\n",
       "      <td>Conventional</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                 0  \\\n",
       "SalesID                                                    1139246   \n",
       "SalePrice                                                  66000.0   \n",
       "MachineID                                                   999089   \n",
       "ModelID                                                       3157   \n",
       "datasource                                                     121   \n",
       "auctioneerID                                                   3.0   \n",
       "YearMade                                                      2004   \n",
       "MachineHoursCurrentMeter                                      68.0   \n",
       "UsageBand                                                      Low   \n",
       "saledate                                       2006-11-16 00:00:00   \n",
       "fiModelDesc                                                   521D   \n",
       "fiBaseModel                                                    521   \n",
       "fiSecondaryDesc                                                  D   \n",
       "fiModelSeries                                                  NaN   \n",
       "fiModelDescriptor                                              NaN   \n",
       "ProductSize                                                    NaN   \n",
       "fiProductClassDesc        Wheel Loader - 110.0 to 120.0 Horsepower   \n",
       "state                                                      Alabama   \n",
       "ProductGroup                                                    WL   \n",
       "ProductGroupDesc                                      Wheel Loader   \n",
       "Drive_System                                                   NaN   \n",
       "Enclosure                                               EROPS w AC   \n",
       "Forks                                          None or Unspecified   \n",
       "Pad_Type                                                       NaN   \n",
       "Ride_Control                                   None or Unspecified   \n",
       "Stick                                                          NaN   \n",
       "Transmission                                                   NaN   \n",
       "Turbocharged                                                   NaN   \n",
       "Blade_Extension                                                NaN   \n",
       "Blade_Width                                                    NaN   \n",
       "Enclosure_Type                                                 NaN   \n",
       "Engine_Horsepower                                              NaN   \n",
       "Hydraulics                                                 2 Valve   \n",
       "Pushblock                                                      NaN   \n",
       "Ripper                                                         NaN   \n",
       "Scarifier                                                      NaN   \n",
       "Tip_Control                                                    NaN   \n",
       "Tire_Size                                      None or Unspecified   \n",
       "Coupler                                        None or Unspecified   \n",
       "Coupler_System                                                 NaN   \n",
       "Grouser_Tracks                                                 NaN   \n",
       "Hydraulics_Flow                                                NaN   \n",
       "Track_Type                                                     NaN   \n",
       "Undercarriage_Pad_Width                                        NaN   \n",
       "Stick_Length                                                   NaN   \n",
       "Thumb                                                          NaN   \n",
       "Pattern_Changer                                                NaN   \n",
       "Grouser_Type                                                   NaN   \n",
       "Backhoe_Mounting                                               NaN   \n",
       "Blade_Type                                                     NaN   \n",
       "Travel_Controls                                                NaN   \n",
       "Differential_Type                                         Standard   \n",
       "Steering_Controls                                     Conventional   \n",
       "\n",
       "                                                                 1  \\\n",
       "SalesID                                                    1139248   \n",
       "SalePrice                                                  57000.0   \n",
       "MachineID                                                   117657   \n",
       "ModelID                                                         77   \n",
       "datasource                                                     121   \n",
       "auctioneerID                                                   3.0   \n",
       "YearMade                                                      1996   \n",
       "MachineHoursCurrentMeter                                    4640.0   \n",
       "UsageBand                                                      Low   \n",
       "saledate                                       2004-03-26 00:00:00   \n",
       "fiModelDesc                                                 950FII   \n",
       "fiBaseModel                                                    950   \n",
       "fiSecondaryDesc                                                  F   \n",
       "fiModelSeries                                                   II   \n",
       "fiModelDescriptor                                              NaN   \n",
       "ProductSize                                                 Medium   \n",
       "fiProductClassDesc        Wheel Loader - 150.0 to 175.0 Horsepower   \n",
       "state                                               North Carolina   \n",
       "ProductGroup                                                    WL   \n",
       "ProductGroupDesc                                      Wheel Loader   \n",
       "Drive_System                                                   NaN   \n",
       "Enclosure                                               EROPS w AC   \n",
       "Forks                                          None or Unspecified   \n",
       "Pad_Type                                                       NaN   \n",
       "Ride_Control                                   None or Unspecified   \n",
       "Stick                                                          NaN   \n",
       "Transmission                                                   NaN   \n",
       "Turbocharged                                                   NaN   \n",
       "Blade_Extension                                                NaN   \n",
       "Blade_Width                                                    NaN   \n",
       "Enclosure_Type                                                 NaN   \n",
       "Engine_Horsepower                                              NaN   \n",
       "Hydraulics                                                 2 Valve   \n",
       "Pushblock                                                      NaN   \n",
       "Ripper                                                         NaN   \n",
       "Scarifier                                                      NaN   \n",
       "Tip_Control                                                    NaN   \n",
       "Tire_Size                                                     23.5   \n",
       "Coupler                                        None or Unspecified   \n",
       "Coupler_System                                                 NaN   \n",
       "Grouser_Tracks                                                 NaN   \n",
       "Hydraulics_Flow                                                NaN   \n",
       "Track_Type                                                     NaN   \n",
       "Undercarriage_Pad_Width                                        NaN   \n",
       "Stick_Length                                                   NaN   \n",
       "Thumb                                                          NaN   \n",
       "Pattern_Changer                                                NaN   \n",
       "Grouser_Type                                                   NaN   \n",
       "Backhoe_Mounting                                               NaN   \n",
       "Blade_Type                                                     NaN   \n",
       "Travel_Controls                                                NaN   \n",
       "Differential_Type                                         Standard   \n",
       "Steering_Controls                                     Conventional   \n",
       "\n",
       "                                                                          2  \\\n",
       "SalesID                                                             1139249   \n",
       "SalePrice                                                           10000.0   \n",
       "MachineID                                                            434808   \n",
       "ModelID                                                                7009   \n",
       "datasource                                                              121   \n",
       "auctioneerID                                                            3.0   \n",
       "YearMade                                                               2001   \n",
       "MachineHoursCurrentMeter                                             2838.0   \n",
       "UsageBand                                                              High   \n",
       "saledate                                                2004-02-26 00:00:00   \n",
       "fiModelDesc                                                             226   \n",
       "fiBaseModel                                                             226   \n",
       "fiSecondaryDesc                                                         NaN   \n",
       "fiModelSeries                                                           NaN   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                             NaN   \n",
       "fiProductClassDesc        Skid Steer Loader - 1351.0 to 1601.0 Lb Operat...   \n",
       "state                                                              New York   \n",
       "ProductGroup                                                            SSL   \n",
       "ProductGroupDesc                                         Skid Steer Loaders   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                             OROPS   \n",
       "Forks                                                   None or Unspecified   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                            NaN   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                        Auxiliary   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                                  NaN   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                 None or Unspecified   \n",
       "Coupler_System                                          None or Unspecified   \n",
       "Grouser_Tracks                                          None or Unspecified   \n",
       "Hydraulics_Flow                                                    Standard   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                                        NaN   \n",
       "Blade_Type                                                              NaN   \n",
       "Travel_Controls                                                         NaN   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "\n",
       "                                                                          3  \\\n",
       "SalesID                                                             1139251   \n",
       "SalePrice                                                           38500.0   \n",
       "MachineID                                                           1026470   \n",
       "ModelID                                                                 332   \n",
       "datasource                                                              121   \n",
       "auctioneerID                                                            3.0   \n",
       "YearMade                                                               2001   \n",
       "MachineHoursCurrentMeter                                             3486.0   \n",
       "UsageBand                                                              High   \n",
       "saledate                                                2011-05-19 00:00:00   \n",
       "fiModelDesc                                                        PC120-6E   \n",
       "fiBaseModel                                                           PC120   \n",
       "fiSecondaryDesc                                                         NaN   \n",
       "fiModelSeries                                                           -6E   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                           Small   \n",
       "fiProductClassDesc        Hydraulic Excavator, Track - 12.0 to 14.0 Metr...   \n",
       "state                                                                 Texas   \n",
       "ProductGroup                                                            TEX   \n",
       "ProductGroupDesc                                           Track Excavators   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                        EROPS w AC   \n",
       "Forks                                                                   NaN   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                            NaN   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                          2 Valve   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                                  NaN   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                 None or Unspecified   \n",
       "Coupler_System                                                          NaN   \n",
       "Grouser_Tracks                                                          NaN   \n",
       "Hydraulics_Flow                                                         NaN   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                                        NaN   \n",
       "Blade_Type                                                              NaN   \n",
       "Travel_Controls                                                         NaN   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "\n",
       "                                                                          4  \n",
       "SalesID                                                             1139253  \n",
       "SalePrice                                                           11000.0  \n",
       "MachineID                                                           1057373  \n",
       "ModelID                                                               17311  \n",
       "datasource                                                              121  \n",
       "auctioneerID                                                            3.0  \n",
       "YearMade                                                               2007  \n",
       "MachineHoursCurrentMeter                                              722.0  \n",
       "UsageBand                                                            Medium  \n",
       "saledate                                                2009-07-23 00:00:00  \n",
       "fiModelDesc                                                            S175  \n",
       "fiBaseModel                                                            S175  \n",
       "fiSecondaryDesc                                                         NaN  \n",
       "fiModelSeries                                                           NaN  \n",
       "fiModelDescriptor                                                       NaN  \n",
       "ProductSize                                                             NaN  \n",
       "fiProductClassDesc        Skid Steer Loader - 1601.0 to 1751.0 Lb Operat...  \n",
       "state                                                              New York  \n",
       "ProductGroup                                                            SSL  \n",
       "ProductGroupDesc                                         Skid Steer Loaders  \n",
       "Drive_System                                                            NaN  \n",
       "Enclosure                                                             EROPS  \n",
       "Forks                                                   None or Unspecified  \n",
       "Pad_Type                                                                NaN  \n",
       "Ride_Control                                                            NaN  \n",
       "Stick                                                                   NaN  \n",
       "Transmission                                                            NaN  \n",
       "Turbocharged                                                            NaN  \n",
       "Blade_Extension                                                         NaN  \n",
       "Blade_Width                                                             NaN  \n",
       "Enclosure_Type                                                          NaN  \n",
       "Engine_Horsepower                                                       NaN  \n",
       "Hydraulics                                                        Auxiliary  \n",
       "Pushblock                                                               NaN  \n",
       "Ripper                                                                  NaN  \n",
       "Scarifier                                                               NaN  \n",
       "Tip_Control                                                             NaN  \n",
       "Tire_Size                                                               NaN  \n",
       "Coupler                                                 None or Unspecified  \n",
       "Coupler_System                                          None or Unspecified  \n",
       "Grouser_Tracks                                          None or Unspecified  \n",
       "Hydraulics_Flow                                                    Standard  \n",
       "Track_Type                                                              NaN  \n",
       "Undercarriage_Pad_Width                                                 NaN  \n",
       "Stick_Length                                                            NaN  \n",
       "Thumb                                                                   NaN  \n",
       "Pattern_Changer                                                         NaN  \n",
       "Grouser_Type                                                            NaN  \n",
       "Backhoe_Mounting                                                        NaN  \n",
       "Blade_Type                                                              NaN  \n",
       "Travel_Controls                                                         NaN  \n",
       "Differential_Type                                                       NaN  \n",
       "Steering_Controls                                                       NaN  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## To see all the columns we should use\n",
    "df.head().T"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0361eeb5",
   "metadata": {},
   "source": [
    "## Sort DataFrame by saledate\n",
    "When working with time series data, it's a good idea to sort it by date."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "31653386",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "205615   1989-01-17\n",
       "274835   1989-01-31\n",
       "141296   1989-01-31\n",
       "212552   1989-01-31\n",
       "62755    1989-01-31\n",
       "54653    1989-01-31\n",
       "81383    1989-01-31\n",
       "204924   1989-01-31\n",
       "135376   1989-01-31\n",
       "113390   1989-01-31\n",
       "113394   1989-01-31\n",
       "116419   1989-01-31\n",
       "32138    1989-01-31\n",
       "127610   1989-01-31\n",
       "76171    1989-01-31\n",
       "127000   1989-01-31\n",
       "128130   1989-01-31\n",
       "127626   1989-01-31\n",
       "55455    1989-01-31\n",
       "55454    1989-01-31\n",
       "Name: saledate, dtype: datetime64[ns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sort_values(by=['saledate'], inplace=True, ascending=True)\n",
    "df.saledate.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "722a71bd",
   "metadata": {},
   "source": [
    "## Make a copy of original DataFrame\n",
    "We make copy of the original dataframe so when we manipulate the copy, we still have our original dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "50115ae8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_tmp = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4993cdae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>205615</th>\n",
       "      <th>274835</th>\n",
       "      <th>141296</th>\n",
       "      <th>212552</th>\n",
       "      <th>62755</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SalesID</th>\n",
       "      <td>1646770</td>\n",
       "      <td>1821514</td>\n",
       "      <td>1505138</td>\n",
       "      <td>1671174</td>\n",
       "      <td>1329056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>9500.0</td>\n",
       "      <td>14000.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>22000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineID</th>\n",
       "      <td>1126363</td>\n",
       "      <td>1194089</td>\n",
       "      <td>1473654</td>\n",
       "      <td>1327630</td>\n",
       "      <td>1336053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ModelID</th>\n",
       "      <td>8434</td>\n",
       "      <td>10150</td>\n",
       "      <td>4139</td>\n",
       "      <td>8591</td>\n",
       "      <td>4089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datasource</th>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>auctioneerID</th>\n",
       "      <td>18.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YearMade</th>\n",
       "      <td>1974</td>\n",
       "      <td>1980</td>\n",
       "      <td>1978</td>\n",
       "      <td>1980</td>\n",
       "      <td>1984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UsageBand</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saledate</th>\n",
       "      <td>1989-01-17 00:00:00</td>\n",
       "      <td>1989-01-31 00:00:00</td>\n",
       "      <td>1989-01-31 00:00:00</td>\n",
       "      <td>1989-01-31 00:00:00</td>\n",
       "      <td>1989-01-31 00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDesc</th>\n",
       "      <td>TD20</td>\n",
       "      <td>A66</td>\n",
       "      <td>D7G</td>\n",
       "      <td>A62</td>\n",
       "      <td>D3B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiBaseModel</th>\n",
       "      <td>TD20</td>\n",
       "      <td>A66</td>\n",
       "      <td>D7</td>\n",
       "      <td>A62</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiSecondaryDesc</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>G</td>\n",
       "      <td>NaN</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelSeries</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDescriptor</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductSize</th>\n",
       "      <td>Medium</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Large</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiProductClassDesc</th>\n",
       "      <td>Track Type Tractor, Dozer - 105.0 to 130.0 Hor...</td>\n",
       "      <td>Wheel Loader - 120.0 to 135.0 Horsepower</td>\n",
       "      <td>Track Type Tractor, Dozer - 190.0 to 260.0 Hor...</td>\n",
       "      <td>Wheel Loader - Unidentified</td>\n",
       "      <td>Track Type Tractor, Dozer - 20.0 to 75.0 Horse...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <td>Texas</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroup</th>\n",
       "      <td>TTT</td>\n",
       "      <td>WL</td>\n",
       "      <td>TTT</td>\n",
       "      <td>WL</td>\n",
       "      <td>TTT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroupDesc</th>\n",
       "      <td>Track Type Tractors</td>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Track Type Tractors</td>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Track Type Tractors</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drive_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure</th>\n",
       "      <td>OROPS</td>\n",
       "      <td>OROPS</td>\n",
       "      <td>OROPS</td>\n",
       "      <td>EROPS</td>\n",
       "      <td>OROPS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Forks</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pad_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ride_Control</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Transmission</th>\n",
       "      <td>Direct Drive</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turbocharged</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Extension</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Engine_Horsepower</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics</th>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pushblock</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ripper</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Scarifier</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tip_Control</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tire_Size</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Tracks</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics_Flow</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Track_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Undercarriage_Pad_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick_Length</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thumb</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pattern_Changer</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Backhoe_Mounting</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Type</th>\n",
       "      <td>Straight</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Straight</td>\n",
       "      <td>NaN</td>\n",
       "      <td>PAT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Travel_Controls</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Differential_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Steering_Controls</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Conventional</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Conventional</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                     205615  \\\n",
       "SalesID                                                             1646770   \n",
       "SalePrice                                                            9500.0   \n",
       "MachineID                                                           1126363   \n",
       "ModelID                                                                8434   \n",
       "datasource                                                              132   \n",
       "auctioneerID                                                           18.0   \n",
       "YearMade                                                               1974   \n",
       "MachineHoursCurrentMeter                                                NaN   \n",
       "UsageBand                                                               NaN   \n",
       "saledate                                                1989-01-17 00:00:00   \n",
       "fiModelDesc                                                            TD20   \n",
       "fiBaseModel                                                            TD20   \n",
       "fiSecondaryDesc                                                         NaN   \n",
       "fiModelSeries                                                           NaN   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                          Medium   \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 105.0 to 130.0 Hor...   \n",
       "state                                                                 Texas   \n",
       "ProductGroup                                                            TTT   \n",
       "ProductGroupDesc                                        Track Type Tractors   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                             OROPS   \n",
       "Forks                                                                   NaN   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                   Direct Drive   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                          2 Valve   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                  None or Unspecified   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                                 NaN   \n",
       "Coupler_System                                                          NaN   \n",
       "Grouser_Tracks                                                          NaN   \n",
       "Hydraulics_Flow                                                         NaN   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                        None or Unspecified   \n",
       "Blade_Type                                                         Straight   \n",
       "Travel_Controls                                         None or Unspecified   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "\n",
       "                                                            274835  \\\n",
       "SalesID                                                    1821514   \n",
       "SalePrice                                                  14000.0   \n",
       "MachineID                                                  1194089   \n",
       "ModelID                                                      10150   \n",
       "datasource                                                     132   \n",
       "auctioneerID                                                  99.0   \n",
       "YearMade                                                      1980   \n",
       "MachineHoursCurrentMeter                                       NaN   \n",
       "UsageBand                                                      NaN   \n",
       "saledate                                       1989-01-31 00:00:00   \n",
       "fiModelDesc                                                    A66   \n",
       "fiBaseModel                                                    A66   \n",
       "fiSecondaryDesc                                                NaN   \n",
       "fiModelSeries                                                  NaN   \n",
       "fiModelDescriptor                                              NaN   \n",
       "ProductSize                                                    NaN   \n",
       "fiProductClassDesc        Wheel Loader - 120.0 to 135.0 Horsepower   \n",
       "state                                                      Florida   \n",
       "ProductGroup                                                    WL   \n",
       "ProductGroupDesc                                      Wheel Loader   \n",
       "Drive_System                                                   NaN   \n",
       "Enclosure                                                    OROPS   \n",
       "Forks                                          None or Unspecified   \n",
       "Pad_Type                                                       NaN   \n",
       "Ride_Control                                   None or Unspecified   \n",
       "Stick                                                          NaN   \n",
       "Transmission                                                   NaN   \n",
       "Turbocharged                                                   NaN   \n",
       "Blade_Extension                                                NaN   \n",
       "Blade_Width                                                    NaN   \n",
       "Enclosure_Type                                                 NaN   \n",
       "Engine_Horsepower                                              NaN   \n",
       "Hydraulics                                                 2 Valve   \n",
       "Pushblock                                                      NaN   \n",
       "Ripper                                                         NaN   \n",
       "Scarifier                                                      NaN   \n",
       "Tip_Control                                                    NaN   \n",
       "Tire_Size                                      None or Unspecified   \n",
       "Coupler                                        None or Unspecified   \n",
       "Coupler_System                                                 NaN   \n",
       "Grouser_Tracks                                                 NaN   \n",
       "Hydraulics_Flow                                                NaN   \n",
       "Track_Type                                                     NaN   \n",
       "Undercarriage_Pad_Width                                        NaN   \n",
       "Stick_Length                                                   NaN   \n",
       "Thumb                                                          NaN   \n",
       "Pattern_Changer                                                NaN   \n",
       "Grouser_Type                                                   NaN   \n",
       "Backhoe_Mounting                                               NaN   \n",
       "Blade_Type                                                     NaN   \n",
       "Travel_Controls                                                NaN   \n",
       "Differential_Type                                         Standard   \n",
       "Steering_Controls                                     Conventional   \n",
       "\n",
       "                                                                     141296  \\\n",
       "SalesID                                                             1505138   \n",
       "SalePrice                                                           50000.0   \n",
       "MachineID                                                           1473654   \n",
       "ModelID                                                                4139   \n",
       "datasource                                                              132   \n",
       "auctioneerID                                                           99.0   \n",
       "YearMade                                                               1978   \n",
       "MachineHoursCurrentMeter                                                NaN   \n",
       "UsageBand                                                               NaN   \n",
       "saledate                                                1989-01-31 00:00:00   \n",
       "fiModelDesc                                                             D7G   \n",
       "fiBaseModel                                                              D7   \n",
       "fiSecondaryDesc                                                           G   \n",
       "fiModelSeries                                                           NaN   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                           Large   \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 190.0 to 260.0 Hor...   \n",
       "state                                                               Florida   \n",
       "ProductGroup                                                            TTT   \n",
       "ProductGroupDesc                                        Track Type Tractors   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                             OROPS   \n",
       "Forks                                                                   NaN   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                       Standard   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                          2 Valve   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                  None or Unspecified   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                                 NaN   \n",
       "Coupler_System                                                          NaN   \n",
       "Grouser_Tracks                                                          NaN   \n",
       "Hydraulics_Flow                                                         NaN   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                        None or Unspecified   \n",
       "Blade_Type                                                         Straight   \n",
       "Travel_Controls                                         None or Unspecified   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "\n",
       "                                               212552  \\\n",
       "SalesID                                       1671174   \n",
       "SalePrice                                     16000.0   \n",
       "MachineID                                     1327630   \n",
       "ModelID                                          8591   \n",
       "datasource                                        132   \n",
       "auctioneerID                                     99.0   \n",
       "YearMade                                         1980   \n",
       "MachineHoursCurrentMeter                          NaN   \n",
       "UsageBand                                         NaN   \n",
       "saledate                          1989-01-31 00:00:00   \n",
       "fiModelDesc                                       A62   \n",
       "fiBaseModel                                       A62   \n",
       "fiSecondaryDesc                                   NaN   \n",
       "fiModelSeries                                     NaN   \n",
       "fiModelDescriptor                                 NaN   \n",
       "ProductSize                                       NaN   \n",
       "fiProductClassDesc        Wheel Loader - Unidentified   \n",
       "state                                         Florida   \n",
       "ProductGroup                                       WL   \n",
       "ProductGroupDesc                         Wheel Loader   \n",
       "Drive_System                                      NaN   \n",
       "Enclosure                                       EROPS   \n",
       "Forks                             None or Unspecified   \n",
       "Pad_Type                                          NaN   \n",
       "Ride_Control                      None or Unspecified   \n",
       "Stick                                             NaN   \n",
       "Transmission                                      NaN   \n",
       "Turbocharged                                      NaN   \n",
       "Blade_Extension                                   NaN   \n",
       "Blade_Width                                       NaN   \n",
       "Enclosure_Type                                    NaN   \n",
       "Engine_Horsepower                                 NaN   \n",
       "Hydraulics                                    2 Valve   \n",
       "Pushblock                                         NaN   \n",
       "Ripper                                            NaN   \n",
       "Scarifier                                         NaN   \n",
       "Tip_Control                                       NaN   \n",
       "Tire_Size                         None or Unspecified   \n",
       "Coupler                           None or Unspecified   \n",
       "Coupler_System                                    NaN   \n",
       "Grouser_Tracks                                    NaN   \n",
       "Hydraulics_Flow                                   NaN   \n",
       "Track_Type                                        NaN   \n",
       "Undercarriage_Pad_Width                           NaN   \n",
       "Stick_Length                                      NaN   \n",
       "Thumb                                             NaN   \n",
       "Pattern_Changer                                   NaN   \n",
       "Grouser_Type                                      NaN   \n",
       "Backhoe_Mounting                                  NaN   \n",
       "Blade_Type                                        NaN   \n",
       "Travel_Controls                                   NaN   \n",
       "Differential_Type                            Standard   \n",
       "Steering_Controls                        Conventional   \n",
       "\n",
       "                                                                     62755   \n",
       "SalesID                                                             1329056  \n",
       "SalePrice                                                           22000.0  \n",
       "MachineID                                                           1336053  \n",
       "ModelID                                                                4089  \n",
       "datasource                                                              132  \n",
       "auctioneerID                                                           99.0  \n",
       "YearMade                                                               1984  \n",
       "MachineHoursCurrentMeter                                                NaN  \n",
       "UsageBand                                                               NaN  \n",
       "saledate                                                1989-01-31 00:00:00  \n",
       "fiModelDesc                                                             D3B  \n",
       "fiBaseModel                                                              D3  \n",
       "fiSecondaryDesc                                                           B  \n",
       "fiModelSeries                                                           NaN  \n",
       "fiModelDescriptor                                                       NaN  \n",
       "ProductSize                                                             NaN  \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 20.0 to 75.0 Horse...  \n",
       "state                                                               Florida  \n",
       "ProductGroup                                                            TTT  \n",
       "ProductGroupDesc                                        Track Type Tractors  \n",
       "Drive_System                                                            NaN  \n",
       "Enclosure                                                             OROPS  \n",
       "Forks                                                                   NaN  \n",
       "Pad_Type                                                                NaN  \n",
       "Ride_Control                                                            NaN  \n",
       "Stick                                                                   NaN  \n",
       "Transmission                                                       Standard  \n",
       "Turbocharged                                                            NaN  \n",
       "Blade_Extension                                                         NaN  \n",
       "Blade_Width                                                             NaN  \n",
       "Enclosure_Type                                                          NaN  \n",
       "Engine_Horsepower                                                       NaN  \n",
       "Hydraulics                                                          2 Valve  \n",
       "Pushblock                                                               NaN  \n",
       "Ripper                                                  None or Unspecified  \n",
       "Scarifier                                                               NaN  \n",
       "Tip_Control                                                             NaN  \n",
       "Tire_Size                                                               NaN  \n",
       "Coupler                                                                 NaN  \n",
       "Coupler_System                                                          NaN  \n",
       "Grouser_Tracks                                                          NaN  \n",
       "Hydraulics_Flow                                                         NaN  \n",
       "Track_Type                                                              NaN  \n",
       "Undercarriage_Pad_Width                                                 NaN  \n",
       "Stick_Length                                                            NaN  \n",
       "Thumb                                                                   NaN  \n",
       "Pattern_Changer                                                         NaN  \n",
       "Grouser_Type                                                            NaN  \n",
       "Backhoe_Mounting                                        None or Unspecified  \n",
       "Blade_Type                                                              PAT  \n",
       "Travel_Controls                                                       Lever  \n",
       "Differential_Type                                                       NaN  \n",
       "Steering_Controls                                                       NaN  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.head().T"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c4eb65d",
   "metadata": {},
   "source": [
    "## 6. Feture Engineering\n",
    "Adding or removing column/s to enrich our dataset and get maximum result out of it is called **\"Feture Engineering\".**\n",
    "> In this project, we can add `saleYear`, `saleMonth` and `saleDay` column to know the yearly, monthly and daily sales.\n",
    "> Also add dayofweek and dayofyear columns and remove the `saledate` column because we have used datetime parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3ecb0f08",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_tmp['saleYear'] = df_tmp.saledate.dt.year\n",
    "df_tmp['saleMonth'] = df_tmp.saledate.dt.month\n",
    "df_tmp['saleDay'] = df_tmp.saledate.dt.day\n",
    "df_tmp['saleDayofWeek'] = df_tmp.saledate.dt.dayofweek\n",
    "df_tmp['saleDayofYear'] = df_tmp.saledate.dt.dayofyear"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "12c0ec46",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_tmp.drop('saledate', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "51c5829a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Florida           67320\n",
       "Texas             53110\n",
       "California        29761\n",
       "Washington        16222\n",
       "Georgia           14633\n",
       "Maryland          13322\n",
       "Mississippi       13240\n",
       "Ohio              12369\n",
       "Illinois          11540\n",
       "Colorado          11529\n",
       "New Jersey        11156\n",
       "North Carolina    10636\n",
       "Tennessee         10298\n",
       "Alabama           10292\n",
       "Pennsylvania      10234\n",
       "South Carolina     9951\n",
       "Arizona            9364\n",
       "New York           8639\n",
       "Connecticut        8276\n",
       "Minnesota          7885\n",
       "Missouri           7178\n",
       "Nevada             6932\n",
       "Louisiana          6627\n",
       "Kentucky           5351\n",
       "Maine              5096\n",
       "Indiana            4124\n",
       "Arkansas           3933\n",
       "New Mexico         3631\n",
       "Utah               3046\n",
       "Unspecified        2801\n",
       "Wisconsin          2745\n",
       "New Hampshire      2738\n",
       "Virginia           2353\n",
       "Idaho              2025\n",
       "Oregon             1911\n",
       "Michigan           1831\n",
       "Wyoming            1672\n",
       "Montana            1336\n",
       "Iowa               1336\n",
       "Oklahoma           1326\n",
       "Nebraska            866\n",
       "West Virginia       840\n",
       "Kansas              667\n",
       "Delaware            510\n",
       "North Dakota        480\n",
       "Alaska              430\n",
       "Massachusetts       347\n",
       "Vermont             300\n",
       "South Dakota        244\n",
       "Hawaii              118\n",
       "Rhode Island         83\n",
       "Puerto Rico          42\n",
       "Washington DC         2\n",
       "Name: state, dtype: int64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check the values of different columns.\n",
    "df_tmp.state.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "ec958fa1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None or Unspecified    71447\n",
       "Differential Steer      5257\n",
       "Finger Tip              2693\n",
       "2 Pedal                 1144\n",
       "Lever                    902\n",
       "Pedal                    423\n",
       "1 Speed                   11\n",
       "Name: Travel_Controls, dtype: int64"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.Travel_Controls.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "de55e0f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Conventional           70774\n",
       "Command Control          594\n",
       "Four Wheel Standard      139\n",
       "Wheel                     14\n",
       "No                         1\n",
       "Name: Steering_Controls, dtype: int64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.Steering_Controls.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "796dfb8a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None or Unspecified    85405\n",
       "Yes                     8185\n",
       "Multi Shank             8071\n",
       "Single Shank            5284\n",
       "Name: Ripper, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.Ripper.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e7dac90",
   "metadata": {},
   "source": [
    "## 7. Modelling\n",
    "We have done enough Exploratory Data Analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2b9a92f",
   "metadata": {},
   "source": [
    "## i. Turning Data into numbers.\n",
    "One way we can turn all of our data into numbers is by converting them into pandas categories.\n",
    "If the datatype is string than we can use `pandas.api.types` to convert string into pandas datatypes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b484276",
   "metadata": {},
   "source": [
    "#### **Note:** If we use `df.info()` function than we can not know whether there is string or not because it will return `object` datatype for all the datatypes accept numbers. So we can use pandas api types to know which columns have string values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "b4b0e2b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 412698 entries, 205615 to 409203\n",
      "Data columns (total 57 columns):\n",
      " #   Column                    Non-Null Count   Dtype  \n",
      "---  ------                    --------------   -----  \n",
      " 0   SalesID                   412698 non-null  int64  \n",
      " 1   SalePrice                 412698 non-null  float64\n",
      " 2   MachineID                 412698 non-null  int64  \n",
      " 3   ModelID                   412698 non-null  int64  \n",
      " 4   datasource                412698 non-null  int64  \n",
      " 5   auctioneerID              392562 non-null  float64\n",
      " 6   YearMade                  412698 non-null  int64  \n",
      " 7   MachineHoursCurrentMeter  147504 non-null  float64\n",
      " 8   UsageBand                 73670 non-null   object \n",
      " 9   fiModelDesc               412698 non-null  object \n",
      " 10  fiBaseModel               412698 non-null  object \n",
      " 11  fiSecondaryDesc           271971 non-null  object \n",
      " 12  fiModelSeries             58667 non-null   object \n",
      " 13  fiModelDescriptor         74816 non-null   object \n",
      " 14  ProductSize               196093 non-null  object \n",
      " 15  fiProductClassDesc        412698 non-null  object \n",
      " 16  state                     412698 non-null  object \n",
      " 17  ProductGroup              412698 non-null  object \n",
      " 18  ProductGroupDesc          412698 non-null  object \n",
      " 19  Drive_System              107087 non-null  object \n",
      " 20  Enclosure                 412364 non-null  object \n",
      " 21  Forks                     197715 non-null  object \n",
      " 22  Pad_Type                  81096 non-null   object \n",
      " 23  Ride_Control              152728 non-null  object \n",
      " 24  Stick                     81096 non-null   object \n",
      " 25  Transmission              188007 non-null  object \n",
      " 26  Turbocharged              81096 non-null   object \n",
      " 27  Blade_Extension           25983 non-null   object \n",
      " 28  Blade_Width               25983 non-null   object \n",
      " 29  Enclosure_Type            25983 non-null   object \n",
      " 30  Engine_Horsepower         25983 non-null   object \n",
      " 31  Hydraulics                330133 non-null  object \n",
      " 32  Pushblock                 25983 non-null   object \n",
      " 33  Ripper                    106945 non-null  object \n",
      " 34  Scarifier                 25994 non-null   object \n",
      " 35  Tip_Control               25983 non-null   object \n",
      " 36  Tire_Size                 97638 non-null   object \n",
      " 37  Coupler                   220679 non-null  object \n",
      " 38  Coupler_System            44974 non-null   object \n",
      " 39  Grouser_Tracks            44875 non-null   object \n",
      " 40  Hydraulics_Flow           44875 non-null   object \n",
      " 41  Track_Type                102193 non-null  object \n",
      " 42  Undercarriage_Pad_Width   102916 non-null  object \n",
      " 43  Stick_Length              102261 non-null  object \n",
      " 44  Thumb                     102332 non-null  object \n",
      " 45  Pattern_Changer           102261 non-null  object \n",
      " 46  Grouser_Type              102193 non-null  object \n",
      " 47  Backhoe_Mounting          80712 non-null   object \n",
      " 48  Blade_Type                81875 non-null   object \n",
      " 49  Travel_Controls           81877 non-null   object \n",
      " 50  Differential_Type         71564 non-null   object \n",
      " 51  Steering_Controls         71522 non-null   object \n",
      " 52  saleYear                  412698 non-null  int64  \n",
      " 53  saleMonth                 412698 non-null  int64  \n",
      " 54  saleDay                   412698 non-null  int64  \n",
      " 55  saleDayofWeek             412698 non-null  int64  \n",
      " 56  saleDayofYear             412698 non-null  int64  \n",
      "dtypes: float64(3), int64(10), object(44)\n",
      "memory usage: 182.6+ MB\n"
     ]
    }
   ],
   "source": [
    "df_tmp.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "5e2a050f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.api.types.is_string_dtype(df_tmp['UsageBand'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "98b942fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the columns which contain strings.\n",
    "for label,content in df_tmp.items():\n",
    "    if pd.api.types.is_string_dtype(content):\n",
    "        df_tmp[label] = content.astype('category').cat.as_ordered()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "64347403",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 412698 entries, 205615 to 409203\n",
      "Data columns (total 57 columns):\n",
      " #   Column                    Non-Null Count   Dtype   \n",
      "---  ------                    --------------   -----   \n",
      " 0   SalesID                   412698 non-null  int64   \n",
      " 1   SalePrice                 412698 non-null  float64 \n",
      " 2   MachineID                 412698 non-null  int64   \n",
      " 3   ModelID                   412698 non-null  int64   \n",
      " 4   datasource                412698 non-null  int64   \n",
      " 5   auctioneerID              392562 non-null  float64 \n",
      " 6   YearMade                  412698 non-null  int64   \n",
      " 7   MachineHoursCurrentMeter  147504 non-null  float64 \n",
      " 8   UsageBand                 73670 non-null   category\n",
      " 9   fiModelDesc               412698 non-null  category\n",
      " 10  fiBaseModel               412698 non-null  category\n",
      " 11  fiSecondaryDesc           271971 non-null  category\n",
      " 12  fiModelSeries             58667 non-null   category\n",
      " 13  fiModelDescriptor         74816 non-null   category\n",
      " 14  ProductSize               196093 non-null  category\n",
      " 15  fiProductClassDesc        412698 non-null  category\n",
      " 16  state                     412698 non-null  category\n",
      " 17  ProductGroup              412698 non-null  category\n",
      " 18  ProductGroupDesc          412698 non-null  category\n",
      " 19  Drive_System              107087 non-null  category\n",
      " 20  Enclosure                 412364 non-null  category\n",
      " 21  Forks                     197715 non-null  category\n",
      " 22  Pad_Type                  81096 non-null   category\n",
      " 23  Ride_Control              152728 non-null  category\n",
      " 24  Stick                     81096 non-null   category\n",
      " 25  Transmission              188007 non-null  category\n",
      " 26  Turbocharged              81096 non-null   category\n",
      " 27  Blade_Extension           25983 non-null   category\n",
      " 28  Blade_Width               25983 non-null   category\n",
      " 29  Enclosure_Type            25983 non-null   category\n",
      " 30  Engine_Horsepower         25983 non-null   category\n",
      " 31  Hydraulics                330133 non-null  category\n",
      " 32  Pushblock                 25983 non-null   category\n",
      " 33  Ripper                    106945 non-null  category\n",
      " 34  Scarifier                 25994 non-null   category\n",
      " 35  Tip_Control               25983 non-null   category\n",
      " 36  Tire_Size                 97638 non-null   category\n",
      " 37  Coupler                   220679 non-null  category\n",
      " 38  Coupler_System            44974 non-null   category\n",
      " 39  Grouser_Tracks            44875 non-null   category\n",
      " 40  Hydraulics_Flow           44875 non-null   category\n",
      " 41  Track_Type                102193 non-null  category\n",
      " 42  Undercarriage_Pad_Width   102916 non-null  category\n",
      " 43  Stick_Length              102261 non-null  category\n",
      " 44  Thumb                     102332 non-null  category\n",
      " 45  Pattern_Changer           102261 non-null  category\n",
      " 46  Grouser_Type              102193 non-null  category\n",
      " 47  Backhoe_Mounting          80712 non-null   category\n",
      " 48  Blade_Type                81875 non-null   category\n",
      " 49  Travel_Controls           81877 non-null   category\n",
      " 50  Differential_Type         71564 non-null   category\n",
      " 51  Steering_Controls         71522 non-null   category\n",
      " 52  saleYear                  412698 non-null  int64   \n",
      " 53  saleMonth                 412698 non-null  int64   \n",
      " 54  saleDay                   412698 non-null  int64   \n",
      " 55  saleDayofWeek             412698 non-null  int64   \n",
      " 56  saleDayofYear             412698 non-null  int64   \n",
      "dtypes: category(44), float64(3), int64(10)\n",
      "memory usage: 63.2 MB\n"
     ]
    }
   ],
   "source": [
    "df_tmp.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d48205e",
   "metadata": {},
   "source": [
    "We can see from the above that all the objects are converted into category datatypes and ordered in alphabetical order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "eeffb9e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Alabama', 'Alaska', 'Arizona', 'Arkansas', 'California', 'Colorado',\n",
       "       'Connecticut', 'Delaware', 'Florida', 'Georgia', 'Hawaii', 'Idaho',\n",
       "       'Illinois', 'Indiana', 'Iowa', 'Kansas', 'Kentucky', 'Louisiana',\n",
       "       'Maine', 'Maryland', 'Massachusetts', 'Michigan', 'Minnesota',\n",
       "       'Mississippi', 'Missouri', 'Montana', 'Nebraska', 'Nevada',\n",
       "       'New Hampshire', 'New Jersey', 'New Mexico', 'New York',\n",
       "       'North Carolina', 'North Dakota', 'Ohio', 'Oklahoma', 'Oregon',\n",
       "       'Pennsylvania', 'Puerto Rico', 'Rhode Island', 'South Carolina',\n",
       "       'South Dakota', 'Tennessee', 'Texas', 'Unspecified', 'Utah', 'Vermont',\n",
       "       'Virginia', 'Washington', 'Washington DC', 'West Virginia', 'Wisconsin',\n",
       "       'Wyoming'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.state.cat.categories"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6cbabd09",
   "metadata": {},
   "source": [
    "#### As by accessing categories of states column, we can see that it is looking like string but under the hood it is the numbers which we can access by `cat.codes`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "eb2ad577",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "205615    43\n",
       "274835     8\n",
       "141296     8\n",
       "212552     8\n",
       "62755      8\n",
       "          ..\n",
       "410879     4\n",
       "412476     4\n",
       "411927     4\n",
       "407124     4\n",
       "409203     4\n",
       "Length: 412698, dtype: int8"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.state.cat.codes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d27af934",
   "metadata": {},
   "source": [
    "## ii. Filling missing values.\n",
    "one way to check is that look for the percentage of how much data is missing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "2185322b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SalesID                      0.000000\n",
      "SalePrice                    0.000000\n",
      "MachineID                    0.000000\n",
      "ModelID                      0.000000\n",
      "datasource                   0.000000\n",
      "auctioneerID                 4.879113\n",
      "YearMade                     0.000000\n",
      "MachineHoursCurrentMeter    64.258610\n",
      "UsageBand                   82.149174\n",
      "fiModelDesc                  0.000000\n",
      "fiBaseModel                  0.000000\n",
      "fiSecondaryDesc             34.099269\n",
      "fiModelSeries               85.784520\n",
      "fiModelDescriptor           81.871490\n",
      "ProductSize                 52.485110\n",
      "fiProductClassDesc           0.000000\n",
      "state                        0.000000\n",
      "ProductGroup                 0.000000\n",
      "ProductGroupDesc             0.000000\n",
      "Drive_System                74.051970\n",
      "Enclosure                    0.080931\n",
      "Forks                       52.092087\n",
      "Pad_Type                    80.349796\n",
      "Ride_Control                62.992794\n",
      "Stick                       80.349796\n",
      "Transmission                54.444412\n",
      "Turbocharged                80.349796\n",
      "Blade_Extension             93.704113\n",
      "Blade_Width                 93.704113\n",
      "Enclosure_Type              93.704113\n",
      "Engine_Horsepower           93.704113\n",
      "Hydraulics                  20.006155\n",
      "Pushblock                   93.704113\n",
      "Ripper                      74.086378\n",
      "Scarifier                   93.701448\n",
      "Tip_Control                 93.704113\n",
      "Tire_Size                   76.341538\n",
      "Coupler                     46.527727\n",
      "Coupler_System              89.102443\n",
      "Grouser_Tracks              89.126431\n",
      "Hydraulics_Flow             89.126431\n",
      "Track_Type                  75.237825\n",
      "Undercarriage_Pad_Width     75.062637\n",
      "Stick_Length                75.221348\n",
      "Thumb                       75.204144\n",
      "Pattern_Changer             75.221348\n",
      "Grouser_Type                75.237825\n",
      "Backhoe_Mounting            80.442842\n",
      "Blade_Type                  80.161038\n",
      "Travel_Controls             80.160553\n",
      "Differential_Type           82.659475\n",
      "Steering_Controls           82.669652\n",
      "saleYear                     0.000000\n",
      "saleMonth                    0.000000\n",
      "saleDay                      0.000000\n",
      "saleDayofWeek                0.000000\n",
      "saleDayofYear                0.000000\n",
      "dtype: float64%\n"
     ]
    }
   ],
   "source": [
    "print(f\"{(df_tmp.isna().sum()/len(df_tmp))*100}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "870e9d07",
   "metadata": {},
   "source": [
    "## Save Preprocessed data.\n",
    "This is a good practice to export the dataframe to csv file and than whenever you need it next day, you don't have to run all the code again and you can import that preprocessed dataframe and start your coding."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "cf86266c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Export current tmp dataframe.\n",
    "df_tmp.to_csv(\"Data/bluebook-for-bulldozers/train_tmp.csv\",\n",
    "              index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "4ac997d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SalesID</th>\n",
       "      <td>1646770</td>\n",
       "      <td>1821514</td>\n",
       "      <td>1505138</td>\n",
       "      <td>1671174</td>\n",
       "      <td>1329056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>9500.0</td>\n",
       "      <td>14000.0</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>22000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineID</th>\n",
       "      <td>1126363</td>\n",
       "      <td>1194089</td>\n",
       "      <td>1473654</td>\n",
       "      <td>1327630</td>\n",
       "      <td>1336053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ModelID</th>\n",
       "      <td>8434</td>\n",
       "      <td>10150</td>\n",
       "      <td>4139</td>\n",
       "      <td>8591</td>\n",
       "      <td>4089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datasource</th>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>auctioneerID</th>\n",
       "      <td>18.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YearMade</th>\n",
       "      <td>1974</td>\n",
       "      <td>1980</td>\n",
       "      <td>1978</td>\n",
       "      <td>1980</td>\n",
       "      <td>1984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UsageBand</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDesc</th>\n",
       "      <td>TD20</td>\n",
       "      <td>A66</td>\n",
       "      <td>D7G</td>\n",
       "      <td>A62</td>\n",
       "      <td>D3B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiBaseModel</th>\n",
       "      <td>TD20</td>\n",
       "      <td>A66</td>\n",
       "      <td>D7</td>\n",
       "      <td>A62</td>\n",
       "      <td>D3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiSecondaryDesc</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>G</td>\n",
       "      <td>NaN</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelSeries</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiModelDescriptor</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductSize</th>\n",
       "      <td>Medium</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Large</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fiProductClassDesc</th>\n",
       "      <td>Track Type Tractor, Dozer - 105.0 to 130.0 Hor...</td>\n",
       "      <td>Wheel Loader - 120.0 to 135.0 Horsepower</td>\n",
       "      <td>Track Type Tractor, Dozer - 190.0 to 260.0 Hor...</td>\n",
       "      <td>Wheel Loader - Unidentified</td>\n",
       "      <td>Track Type Tractor, Dozer - 20.0 to 75.0 Horse...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <td>Texas</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "      <td>Florida</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroup</th>\n",
       "      <td>TTT</td>\n",
       "      <td>WL</td>\n",
       "      <td>TTT</td>\n",
       "      <td>WL</td>\n",
       "      <td>TTT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ProductGroupDesc</th>\n",
       "      <td>Track Type Tractors</td>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Track Type Tractors</td>\n",
       "      <td>Wheel Loader</td>\n",
       "      <td>Track Type Tractors</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Drive_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure</th>\n",
       "      <td>OROPS</td>\n",
       "      <td>OROPS</td>\n",
       "      <td>OROPS</td>\n",
       "      <td>EROPS</td>\n",
       "      <td>OROPS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Forks</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pad_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ride_Control</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Transmission</th>\n",
       "      <td>Direct Drive</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Turbocharged</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Extension</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Enclosure_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Engine_Horsepower</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics</th>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "      <td>2 Valve</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pushblock</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ripper</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Scarifier</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tip_Control</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tire_Size</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler</th>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Coupler_System</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Tracks</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Hydraulics_Flow</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Track_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Undercarriage_Pad_Width</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stick_Length</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thumb</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pattern_Changer</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Grouser_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Backhoe_Mounting</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blade_Type</th>\n",
       "      <td>Straight</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Straight</td>\n",
       "      <td>NaN</td>\n",
       "      <td>PAT</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Travel_Controls</th>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Lever</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Differential_Type</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Steering_Controls</th>\n",
       "      <td>NaN</td>\n",
       "      <td>Conventional</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Conventional</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saleYear</th>\n",
       "      <td>1989</td>\n",
       "      <td>1989</td>\n",
       "      <td>1989</td>\n",
       "      <td>1989</td>\n",
       "      <td>1989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saleMonth</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saleDay</th>\n",
       "      <td>17</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saleDayofWeek</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>saleDayofYear</th>\n",
       "      <td>17</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                          0  \\\n",
       "SalesID                                                             1646770   \n",
       "SalePrice                                                            9500.0   \n",
       "MachineID                                                           1126363   \n",
       "ModelID                                                                8434   \n",
       "datasource                                                              132   \n",
       "auctioneerID                                                           18.0   \n",
       "YearMade                                                               1974   \n",
       "MachineHoursCurrentMeter                                                NaN   \n",
       "UsageBand                                                               NaN   \n",
       "fiModelDesc                                                            TD20   \n",
       "fiBaseModel                                                            TD20   \n",
       "fiSecondaryDesc                                                         NaN   \n",
       "fiModelSeries                                                           NaN   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                          Medium   \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 105.0 to 130.0 Hor...   \n",
       "state                                                                 Texas   \n",
       "ProductGroup                                                            TTT   \n",
       "ProductGroupDesc                                        Track Type Tractors   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                             OROPS   \n",
       "Forks                                                                   NaN   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                   Direct Drive   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                          2 Valve   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                  None or Unspecified   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                                 NaN   \n",
       "Coupler_System                                                          NaN   \n",
       "Grouser_Tracks                                                          NaN   \n",
       "Hydraulics_Flow                                                         NaN   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                        None or Unspecified   \n",
       "Blade_Type                                                         Straight   \n",
       "Travel_Controls                                         None or Unspecified   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "saleYear                                                               1989   \n",
       "saleMonth                                                                 1   \n",
       "saleDay                                                                  17   \n",
       "saleDayofWeek                                                             1   \n",
       "saleDayofYear                                                            17   \n",
       "\n",
       "                                                                 1  \\\n",
       "SalesID                                                    1821514   \n",
       "SalePrice                                                  14000.0   \n",
       "MachineID                                                  1194089   \n",
       "ModelID                                                      10150   \n",
       "datasource                                                     132   \n",
       "auctioneerID                                                  99.0   \n",
       "YearMade                                                      1980   \n",
       "MachineHoursCurrentMeter                                       NaN   \n",
       "UsageBand                                                      NaN   \n",
       "fiModelDesc                                                    A66   \n",
       "fiBaseModel                                                    A66   \n",
       "fiSecondaryDesc                                                NaN   \n",
       "fiModelSeries                                                  NaN   \n",
       "fiModelDescriptor                                              NaN   \n",
       "ProductSize                                                    NaN   \n",
       "fiProductClassDesc        Wheel Loader - 120.0 to 135.0 Horsepower   \n",
       "state                                                      Florida   \n",
       "ProductGroup                                                    WL   \n",
       "ProductGroupDesc                                      Wheel Loader   \n",
       "Drive_System                                                   NaN   \n",
       "Enclosure                                                    OROPS   \n",
       "Forks                                          None or Unspecified   \n",
       "Pad_Type                                                       NaN   \n",
       "Ride_Control                                   None or Unspecified   \n",
       "Stick                                                          NaN   \n",
       "Transmission                                                   NaN   \n",
       "Turbocharged                                                   NaN   \n",
       "Blade_Extension                                                NaN   \n",
       "Blade_Width                                                    NaN   \n",
       "Enclosure_Type                                                 NaN   \n",
       "Engine_Horsepower                                              NaN   \n",
       "Hydraulics                                                 2 Valve   \n",
       "Pushblock                                                      NaN   \n",
       "Ripper                                                         NaN   \n",
       "Scarifier                                                      NaN   \n",
       "Tip_Control                                                    NaN   \n",
       "Tire_Size                                      None or Unspecified   \n",
       "Coupler                                        None or Unspecified   \n",
       "Coupler_System                                                 NaN   \n",
       "Grouser_Tracks                                                 NaN   \n",
       "Hydraulics_Flow                                                NaN   \n",
       "Track_Type                                                     NaN   \n",
       "Undercarriage_Pad_Width                                        NaN   \n",
       "Stick_Length                                                   NaN   \n",
       "Thumb                                                          NaN   \n",
       "Pattern_Changer                                                NaN   \n",
       "Grouser_Type                                                   NaN   \n",
       "Backhoe_Mounting                                               NaN   \n",
       "Blade_Type                                                     NaN   \n",
       "Travel_Controls                                                NaN   \n",
       "Differential_Type                                         Standard   \n",
       "Steering_Controls                                     Conventional   \n",
       "saleYear                                                      1989   \n",
       "saleMonth                                                        1   \n",
       "saleDay                                                         31   \n",
       "saleDayofWeek                                                    1   \n",
       "saleDayofYear                                                   31   \n",
       "\n",
       "                                                                          2  \\\n",
       "SalesID                                                             1505138   \n",
       "SalePrice                                                           50000.0   \n",
       "MachineID                                                           1473654   \n",
       "ModelID                                                                4139   \n",
       "datasource                                                              132   \n",
       "auctioneerID                                                           99.0   \n",
       "YearMade                                                               1978   \n",
       "MachineHoursCurrentMeter                                                NaN   \n",
       "UsageBand                                                               NaN   \n",
       "fiModelDesc                                                             D7G   \n",
       "fiBaseModel                                                              D7   \n",
       "fiSecondaryDesc                                                           G   \n",
       "fiModelSeries                                                           NaN   \n",
       "fiModelDescriptor                                                       NaN   \n",
       "ProductSize                                                           Large   \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 190.0 to 260.0 Hor...   \n",
       "state                                                               Florida   \n",
       "ProductGroup                                                            TTT   \n",
       "ProductGroupDesc                                        Track Type Tractors   \n",
       "Drive_System                                                            NaN   \n",
       "Enclosure                                                             OROPS   \n",
       "Forks                                                                   NaN   \n",
       "Pad_Type                                                                NaN   \n",
       "Ride_Control                                                            NaN   \n",
       "Stick                                                                   NaN   \n",
       "Transmission                                                       Standard   \n",
       "Turbocharged                                                            NaN   \n",
       "Blade_Extension                                                         NaN   \n",
       "Blade_Width                                                             NaN   \n",
       "Enclosure_Type                                                          NaN   \n",
       "Engine_Horsepower                                                       NaN   \n",
       "Hydraulics                                                          2 Valve   \n",
       "Pushblock                                                               NaN   \n",
       "Ripper                                                  None or Unspecified   \n",
       "Scarifier                                                               NaN   \n",
       "Tip_Control                                                             NaN   \n",
       "Tire_Size                                                               NaN   \n",
       "Coupler                                                                 NaN   \n",
       "Coupler_System                                                          NaN   \n",
       "Grouser_Tracks                                                          NaN   \n",
       "Hydraulics_Flow                                                         NaN   \n",
       "Track_Type                                                              NaN   \n",
       "Undercarriage_Pad_Width                                                 NaN   \n",
       "Stick_Length                                                            NaN   \n",
       "Thumb                                                                   NaN   \n",
       "Pattern_Changer                                                         NaN   \n",
       "Grouser_Type                                                            NaN   \n",
       "Backhoe_Mounting                                        None or Unspecified   \n",
       "Blade_Type                                                         Straight   \n",
       "Travel_Controls                                         None or Unspecified   \n",
       "Differential_Type                                                       NaN   \n",
       "Steering_Controls                                                       NaN   \n",
       "saleYear                                                               1989   \n",
       "saleMonth                                                                 1   \n",
       "saleDay                                                                  31   \n",
       "saleDayofWeek                                                             1   \n",
       "saleDayofYear                                                            31   \n",
       "\n",
       "                                                    3  \\\n",
       "SalesID                                       1671174   \n",
       "SalePrice                                     16000.0   \n",
       "MachineID                                     1327630   \n",
       "ModelID                                          8591   \n",
       "datasource                                        132   \n",
       "auctioneerID                                     99.0   \n",
       "YearMade                                         1980   \n",
       "MachineHoursCurrentMeter                          NaN   \n",
       "UsageBand                                         NaN   \n",
       "fiModelDesc                                       A62   \n",
       "fiBaseModel                                       A62   \n",
       "fiSecondaryDesc                                   NaN   \n",
       "fiModelSeries                                     NaN   \n",
       "fiModelDescriptor                                 NaN   \n",
       "ProductSize                                       NaN   \n",
       "fiProductClassDesc        Wheel Loader - Unidentified   \n",
       "state                                         Florida   \n",
       "ProductGroup                                       WL   \n",
       "ProductGroupDesc                         Wheel Loader   \n",
       "Drive_System                                      NaN   \n",
       "Enclosure                                       EROPS   \n",
       "Forks                             None or Unspecified   \n",
       "Pad_Type                                          NaN   \n",
       "Ride_Control                      None or Unspecified   \n",
       "Stick                                             NaN   \n",
       "Transmission                                      NaN   \n",
       "Turbocharged                                      NaN   \n",
       "Blade_Extension                                   NaN   \n",
       "Blade_Width                                       NaN   \n",
       "Enclosure_Type                                    NaN   \n",
       "Engine_Horsepower                                 NaN   \n",
       "Hydraulics                                    2 Valve   \n",
       "Pushblock                                         NaN   \n",
       "Ripper                                            NaN   \n",
       "Scarifier                                         NaN   \n",
       "Tip_Control                                       NaN   \n",
       "Tire_Size                         None or Unspecified   \n",
       "Coupler                           None or Unspecified   \n",
       "Coupler_System                                    NaN   \n",
       "Grouser_Tracks                                    NaN   \n",
       "Hydraulics_Flow                                   NaN   \n",
       "Track_Type                                        NaN   \n",
       "Undercarriage_Pad_Width                           NaN   \n",
       "Stick_Length                                      NaN   \n",
       "Thumb                                             NaN   \n",
       "Pattern_Changer                                   NaN   \n",
       "Grouser_Type                                      NaN   \n",
       "Backhoe_Mounting                                  NaN   \n",
       "Blade_Type                                        NaN   \n",
       "Travel_Controls                                   NaN   \n",
       "Differential_Type                            Standard   \n",
       "Steering_Controls                        Conventional   \n",
       "saleYear                                         1989   \n",
       "saleMonth                                           1   \n",
       "saleDay                                            31   \n",
       "saleDayofWeek                                       1   \n",
       "saleDayofYear                                      31   \n",
       "\n",
       "                                                                          4  \n",
       "SalesID                                                             1329056  \n",
       "SalePrice                                                           22000.0  \n",
       "MachineID                                                           1336053  \n",
       "ModelID                                                                4089  \n",
       "datasource                                                              132  \n",
       "auctioneerID                                                           99.0  \n",
       "YearMade                                                               1984  \n",
       "MachineHoursCurrentMeter                                                NaN  \n",
       "UsageBand                                                               NaN  \n",
       "fiModelDesc                                                             D3B  \n",
       "fiBaseModel                                                              D3  \n",
       "fiSecondaryDesc                                                           B  \n",
       "fiModelSeries                                                           NaN  \n",
       "fiModelDescriptor                                                       NaN  \n",
       "ProductSize                                                             NaN  \n",
       "fiProductClassDesc        Track Type Tractor, Dozer - 20.0 to 75.0 Horse...  \n",
       "state                                                               Florida  \n",
       "ProductGroup                                                            TTT  \n",
       "ProductGroupDesc                                        Track Type Tractors  \n",
       "Drive_System                                                            NaN  \n",
       "Enclosure                                                             OROPS  \n",
       "Forks                                                                   NaN  \n",
       "Pad_Type                                                                NaN  \n",
       "Ride_Control                                                            NaN  \n",
       "Stick                                                                   NaN  \n",
       "Transmission                                                       Standard  \n",
       "Turbocharged                                                            NaN  \n",
       "Blade_Extension                                                         NaN  \n",
       "Blade_Width                                                             NaN  \n",
       "Enclosure_Type                                                          NaN  \n",
       "Engine_Horsepower                                                       NaN  \n",
       "Hydraulics                                                          2 Valve  \n",
       "Pushblock                                                               NaN  \n",
       "Ripper                                                  None or Unspecified  \n",
       "Scarifier                                                               NaN  \n",
       "Tip_Control                                                             NaN  \n",
       "Tire_Size                                                               NaN  \n",
       "Coupler                                                                 NaN  \n",
       "Coupler_System                                                          NaN  \n",
       "Grouser_Tracks                                                          NaN  \n",
       "Hydraulics_Flow                                                         NaN  \n",
       "Track_Type                                                              NaN  \n",
       "Undercarriage_Pad_Width                                                 NaN  \n",
       "Stick_Length                                                            NaN  \n",
       "Thumb                                                                   NaN  \n",
       "Pattern_Changer                                                         NaN  \n",
       "Grouser_Type                                                            NaN  \n",
       "Backhoe_Mounting                                        None or Unspecified  \n",
       "Blade_Type                                                              PAT  \n",
       "Travel_Controls                                                       Lever  \n",
       "Differential_Type                                                       NaN  \n",
       "Steering_Controls                                                       NaN  \n",
       "saleYear                                                               1989  \n",
       "saleMonth                                                                 1  \n",
       "saleDay                                                                  31  \n",
       "saleDayofWeek                                                             1  \n",
       "saleDayofYear                                                            31  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Import preprocessed data.\n",
    "df_tmp = pd.read_csv(\"Data/bluebook-for-bulldozers/train_tmp.csv\",\n",
    "                     low_memory=False)\n",
    "df_tmp.head().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f8d7cb76",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SalesID                          0\n",
       "SalePrice                        0\n",
       "MachineID                        0\n",
       "ModelID                          0\n",
       "datasource                       0\n",
       "auctioneerID                 20136\n",
       "YearMade                         0\n",
       "MachineHoursCurrentMeter    265194\n",
       "UsageBand                   339028\n",
       "fiModelDesc                      0\n",
       "fiBaseModel                      0\n",
       "fiSecondaryDesc             140727\n",
       "fiModelSeries               354031\n",
       "fiModelDescriptor           337882\n",
       "ProductSize                 216605\n",
       "fiProductClassDesc               0\n",
       "state                            0\n",
       "ProductGroup                     0\n",
       "ProductGroupDesc                 0\n",
       "Drive_System                305611\n",
       "Enclosure                      334\n",
       "Forks                       214983\n",
       "Pad_Type                    331602\n",
       "Ride_Control                259970\n",
       "Stick                       331602\n",
       "Transmission                224691\n",
       "Turbocharged                331602\n",
       "Blade_Extension             386715\n",
       "Blade_Width                 386715\n",
       "Enclosure_Type              386715\n",
       "Engine_Horsepower           386715\n",
       "Hydraulics                   82565\n",
       "Pushblock                   386715\n",
       "Ripper                      305753\n",
       "Scarifier                   386704\n",
       "Tip_Control                 386715\n",
       "Tire_Size                   315060\n",
       "Coupler                     192019\n",
       "Coupler_System              367724\n",
       "Grouser_Tracks              367823\n",
       "Hydraulics_Flow             367823\n",
       "Track_Type                  310505\n",
       "Undercarriage_Pad_Width     309782\n",
       "Stick_Length                310437\n",
       "Thumb                       310366\n",
       "Pattern_Changer             310437\n",
       "Grouser_Type                310505\n",
       "Backhoe_Mounting            331986\n",
       "Blade_Type                  330823\n",
       "Travel_Controls             330821\n",
       "Differential_Type           341134\n",
       "Steering_Controls           341176\n",
       "saleYear                         0\n",
       "saleMonth                        0\n",
       "saleDay                          0\n",
       "saleDayofWeek                    0\n",
       "saleDayofYear                    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "079dbbac",
   "metadata": {},
   "source": [
    "## a. Filling missing  numerical values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "73cf8767",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SalesID\n",
      "SalePrice\n",
      "MachineID\n",
      "ModelID\n",
      "datasource\n",
      "auctioneerID\n",
      "YearMade\n",
      "MachineHoursCurrentMeter\n",
      "saleYear\n",
      "saleMonth\n",
      "saleDay\n",
      "saleDayofWeek\n",
      "saleDayofYear\n"
     ]
    }
   ],
   "source": [
    "for label,content in df_tmp.items():\n",
    "    if pd.api.types.is_numeric_dtype(content):\n",
    "        print(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "19fb264b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auctioneerID\n",
      "MachineHoursCurrentMeter\n"
     ]
    }
   ],
   "source": [
    "# Check for which numeric columns have null values.\n",
    "for label,content in df_tmp.items():\n",
    "    if pd.api.types.is_numeric_dtype(content):\n",
    "        if pd.isnull(content).sum():\n",
    "            print(label)\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "53dd068a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fill numeric rows with the median value.\n",
    "for label,content in df_tmp.items():\n",
    "    if pd.api.types.is_numeric_dtype(content):\n",
    "        if pd.isnull(content).sum():\n",
    "            # Add a binary column which tells us if data was missing or not.\n",
    "            df_tmp[label+'_is_missing'] = pd.isnull(content)\n",
    "            # Fill missing numeric values with median.\n",
    "            df_tmp[label] = content.fillna(content.median())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "264ae1e3",
   "metadata": {},
   "source": [
    "#### Demonstrate how median is more robust than mean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4691373d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100.0, 999100.8991008991, 100.0, 100.0)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Demonstarte how median is more robust than mean.\n",
    "hundreds = np.full((1000,),100)\n",
    "hundreds_billions = np.append(hundreds, 1000000000)\n",
    "np.mean(hundreds),np.mean(hundreds_billions),np.median(hundreds),np.median(hundreds_billions)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "afb8967e",
   "metadata": {},
   "source": [
    "### By seeing this example, we can say that just by adding one number i.e. 1000000000, mean is getting very much bigger but there is not difference in median value so it is always good to fill missing numeric values with median value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "dc2c3231",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    392562\n",
       "True      20136\n",
       "Name: auctioneerID_is_missing, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's check how many examples were missing\n",
    "df_tmp.auctioneerID_is_missing.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "faa04f76",
   "metadata": {},
   "source": [
    "## b. Turn categorical valiables into numbers and fill missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bf660c17",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "UsageBand\n",
      "fiModelDesc\n",
      "fiBaseModel\n",
      "fiSecondaryDesc\n",
      "fiModelSeries\n",
      "fiModelDescriptor\n",
      "ProductSize\n",
      "fiProductClassDesc\n",
      "state\n",
      "ProductGroup\n",
      "ProductGroupDesc\n",
      "Drive_System\n",
      "Enclosure\n",
      "Forks\n",
      "Pad_Type\n",
      "Ride_Control\n",
      "Stick\n",
      "Transmission\n",
      "Turbocharged\n",
      "Blade_Extension\n",
      "Blade_Width\n",
      "Enclosure_Type\n",
      "Engine_Horsepower\n",
      "Hydraulics\n",
      "Pushblock\n",
      "Ripper\n",
      "Scarifier\n",
      "Tip_Control\n",
      "Tire_Size\n",
      "Coupler\n",
      "Coupler_System\n",
      "Grouser_Tracks\n",
      "Hydraulics_Flow\n",
      "Track_Type\n",
      "Undercarriage_Pad_Width\n",
      "Stick_Length\n",
      "Thumb\n",
      "Pattern_Changer\n",
      "Grouser_Type\n",
      "Backhoe_Mounting\n",
      "Blade_Type\n",
      "Travel_Controls\n",
      "Differential_Type\n",
      "Steering_Controls\n"
     ]
    }
   ],
   "source": [
    "# Check for labels which aren't numeric.\n",
    "for label,content in df_tmp.items():\n",
    "    if pd.api.types.is_numeric_dtype(content) == False:\n",
    "        print(label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0478a544",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([43,  8,  8, ...,  4,  4,  4], dtype=int8)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Categorical(df_tmp['state']).codes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "eee05430",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Turn categorical variable into numbers and fill missing\n",
    "for label,content in df_tmp.items():\n",
    "    # Check for the non-numeric columns\n",
    "    if not pd.api.types.is_numeric_dtype(content):\n",
    "        # Add binary column to indicate whether sample had missing value or not\n",
    "        df_tmp[label+\"_is_missing\"] = pd.isnull(content)\n",
    "        # Fill missing values with categorical codes and add 1 because for missing values code will be negative i.e. -1\n",
    "        df_tmp[label] = pd.Categorical(content).codes + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "60e489b8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1, -1, -1, ..., -1, -1, -1], dtype=int8)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Negative code value in missing data\n",
    "pd.Categorical(df_tmp[\"UsageBand\"]).codes - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "072f24c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 0, 0, 0], dtype=int8)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Categorical(df_tmp[\"UsageBand\"]).codes # From this we can say that 0 means missing value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "32379298",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SalesID                         0\n",
       "SalePrice                       0\n",
       "MachineID                       0\n",
       "ModelID                         0\n",
       "datasource                      0\n",
       "                               ..\n",
       "Backhoe_Mounting_is_missing     0\n",
       "Blade_Type_is_missing           0\n",
       "Travel_Controls_is_missing      0\n",
       "Differential_Type_is_missing    0\n",
       "Steering_Controls_is_missing    0\n",
       "Length: 103, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "71767941",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['SalesID', 'SalePrice', 'MachineID', 'ModelID', 'datasource',\n",
       "       'auctioneerID', 'YearMade', 'MachineHoursCurrentMeter', 'UsageBand',\n",
       "       'fiModelDesc',\n",
       "       ...\n",
       "       'Undercarriage_Pad_Width_is_missing', 'Stick_Length_is_missing',\n",
       "       'Thumb_is_missing', 'Pattern_Changer_is_missing',\n",
       "       'Grouser_Type_is_missing', 'Backhoe_Mounting_is_missing',\n",
       "       'Blade_Type_is_missing', 'Travel_Controls_is_missing',\n",
       "       'Differential_Type_is_missing', 'Steering_Controls_is_missing'],\n",
       "      dtype='object', length=103)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "482e5c3b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 412698 entries, 0 to 412697\n",
      "Columns: 103 entries, SalesID to Steering_Controls_is_missing\n",
      "dtypes: bool(46), float64(3), int16(4), int64(10), int8(40)\n",
      "memory usage: 77.9 MB\n"
     ]
    }
   ],
   "source": [
    "df_tmp.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c840fdff",
   "metadata": {},
   "source": [
    "#### Now that all of our data is numeric as well as our dataframe has no missing values, we should be able to build a machine learning model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "7051847b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>SalePrice</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>fiModelDesc</th>\n",
       "      <th>...</th>\n",
       "      <th>Undercarriage_Pad_Width_is_missing</th>\n",
       "      <th>Stick_Length_is_missing</th>\n",
       "      <th>Thumb_is_missing</th>\n",
       "      <th>Pattern_Changer_is_missing</th>\n",
       "      <th>Grouser_Type_is_missing</th>\n",
       "      <th>Backhoe_Mounting_is_missing</th>\n",
       "      <th>Blade_Type_is_missing</th>\n",
       "      <th>Travel_Controls_is_missing</th>\n",
       "      <th>Differential_Type_is_missing</th>\n",
       "      <th>Steering_Controls_is_missing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1646770</td>\n",
       "      <td>9500.0</td>\n",
       "      <td>1126363</td>\n",
       "      <td>8434</td>\n",
       "      <td>132</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1974</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>4593</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1821514</td>\n",
       "      <td>14000.0</td>\n",
       "      <td>1194089</td>\n",
       "      <td>10150</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1980</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1820</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1505138</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>1473654</td>\n",
       "      <td>4139</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1978</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2348</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1671174</td>\n",
       "      <td>16000.0</td>\n",
       "      <td>1327630</td>\n",
       "      <td>8591</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1980</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1819</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1329056</td>\n",
       "      <td>22000.0</td>\n",
       "      <td>1336053</td>\n",
       "      <td>4089</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1984</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2119</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 103 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  SalePrice  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1646770     9500.0    1126363     8434         132          18.0      1974   \n",
       "1  1821514    14000.0    1194089    10150         132          99.0      1980   \n",
       "2  1505138    50000.0    1473654     4139         132          99.0      1978   \n",
       "3  1671174    16000.0    1327630     8591         132          99.0      1980   \n",
       "4  1329056    22000.0    1336053     4089         132          99.0      1984   \n",
       "\n",
       "   MachineHoursCurrentMeter  UsageBand  fiModelDesc  ...  \\\n",
       "0                       0.0          0         4593  ...   \n",
       "1                       0.0          0         1820  ...   \n",
       "2                       0.0          0         2348  ...   \n",
       "3                       0.0          0         1819  ...   \n",
       "4                       0.0          0         2119  ...   \n",
       "\n",
       "   Undercarriage_Pad_Width_is_missing  Stick_Length_is_missing  \\\n",
       "0                                True                     True   \n",
       "1                                True                     True   \n",
       "2                                True                     True   \n",
       "3                                True                     True   \n",
       "4                                True                     True   \n",
       "\n",
       "   Thumb_is_missing  Pattern_Changer_is_missing  Grouser_Type_is_missing  \\\n",
       "0              True                        True                     True   \n",
       "1              True                        True                     True   \n",
       "2              True                        True                     True   \n",
       "3              True                        True                     True   \n",
       "4              True                        True                     True   \n",
       "\n",
       "   Backhoe_Mounting_is_missing  Blade_Type_is_missing  \\\n",
       "0                        False                  False   \n",
       "1                         True                   True   \n",
       "2                        False                  False   \n",
       "3                         True                   True   \n",
       "4                        False                  False   \n",
       "\n",
       "   Travel_Controls_is_missing  Differential_Type_is_missing  \\\n",
       "0                       False                          True   \n",
       "1                        True                         False   \n",
       "2                       False                          True   \n",
       "3                        True                         False   \n",
       "4                       False                          True   \n",
       "\n",
       "   Steering_Controls_is_missing  \n",
       "0                          True  \n",
       "1                         False  \n",
       "2                          True  \n",
       "3                         False  \n",
       "4                          True  \n",
       "\n",
       "[5 rows x 103 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb8bbfe1",
   "metadata": {},
   "source": [
    "## Modelling Step-1: Splitting Data into train/valid sets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05e24be1",
   "metadata": {},
   "source": [
    "According to the Kaggle data page, the validation set and test set are split according to dates.\n",
    "\n",
    "This makes sense since we're working on a time series problem.\n",
    "\n",
    "E.g. using past events to try and predict future events.\n",
    "\n",
    "Knowing this, randomly splitting our data into train and test sets using something like train_test_split() wouldn't work.\n",
    "\n",
    "Instead, we split our data into training, validation and test sets using the date each sample occured.\n",
    "\n",
    "In our case:\n",
    "\n",
    "* Training = all samples up until 2011\n",
    "* Valid = all samples form January 1, 2012 - April 30, 2012\n",
    "* Test = all samples from May 1, 2012 - November 2012"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9ec819df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2009    43849\n",
       "2008    39767\n",
       "2011    35197\n",
       "2010    33390\n",
       "2007    32208\n",
       "2006    21685\n",
       "2005    20463\n",
       "2004    19879\n",
       "2001    17594\n",
       "2000    17415\n",
       "2002    17246\n",
       "2003    15254\n",
       "1998    13046\n",
       "1999    12793\n",
       "2012    11573\n",
       "1997     9785\n",
       "1996     8829\n",
       "1995     8530\n",
       "1994     7929\n",
       "1993     6303\n",
       "1992     5519\n",
       "1991     5109\n",
       "1989     4806\n",
       "1990     4529\n",
       "Name: saleYear, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_tmp.saleYear.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "37163159",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11573, 401125)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Split data into training and validation sets\n",
    "df_val = df_tmp[df_tmp.saleYear == 2012]\n",
    "df_train = df_tmp[df_tmp.saleYear != 2012]\n",
    "\n",
    "len(df_val), len(df_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "badbb64b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((401125, 102), (401125,), (11573, 102), (11573,))"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Split data into X and y\n",
    "X_train,y_train = df_train.drop('SalePrice', axis=1), df_train['SalePrice']\n",
    "X_valid,y_valid = df_val.drop('SalePrice', axis=1), df_val['SalePrice']\n",
    "X_train.shape, y_train.shape, X_valid.shape, y_valid.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35f38458",
   "metadata": {},
   "source": [
    "## Modelling Step-2: Building an Evaluation function as per requirements"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53982f25",
   "metadata": {},
   "source": [
    "RMSLE = generally you don't care as much if you're off by $10 as much as you'd care if you were off by 10%, you care more about ratios rather than differences. MAE (mean absolute error) is more about exact differences.\n",
    "\n",
    "It's important to understand the evaluation metric you're going for.\n",
    "\n",
    "Since Scikit-Learn doesn't have a function built-in for RMSLE, we'll create our own.\n",
    "\n",
    "We can do this by taking the square root of Scikit-Learn's mean_squared_log_error (MSLE). MSLE is the same as taking the log of mean squared error (MSE).\n",
    "\n",
    "We'll also calculate the MAE and R^2 for fun."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ea694812",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_squared_log_error, mean_absolute_error\n",
    "\n",
    "def rmsle(y_test, y_preds):\n",
    "    return np.sqrt(mean_squared_log_error(y_test, y_preds))\n",
    "\n",
    "# Create function to evaluate our model.\n",
    "def show_scores(model):\n",
    "    train_preds = model.predict(X_train)\n",
    "    val_preds = model.predict(X_valid)\n",
    "    scores = {\"Training MAE\": mean_absolute_error(y_train, train_preds),\n",
    "              \"Valid MAE\": mean_absolute_error(y_valid, val_preds),\n",
    "              \"Training RMSLE\": rmsle(y_train, train_preds),\n",
    "              \"Valid RMSLE\": rmsle(y_valid, val_preds),\n",
    "              \"Training R^2\": model.score(X_train, y_train),\n",
    "              \"Valid R^2\": model.score(X_valid, y_valid)}\n",
    "    return scores"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfed4205",
   "metadata": {},
   "source": [
    "## Modelling Step-3: Testing on a subset to tune Hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8126c20e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change max samples in RandomForestRegressor\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "model = RandomForestRegressor(n_jobs=-1,\n",
    "                              max_samples=10000)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf3e48f7",
   "metadata": {},
   "source": [
    "Setting max_samples to 10000 means every n_estimator (default 100) in our RandomForestRegressor will only see 10000 random samples from our DataFrame instead of the entire 400,000.\n",
    "\n",
    "In other words, we'll be looking at 40x less samples which means we'll get faster computation speeds but we should expect our results to worsen (simple the model has less samples to learn patterns from)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "028b6cda",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: total: 35.8 s\n",
      "Wall time: 7.57 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_samples=10000, n_jobs=-1)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "# Cutting down the max number of samples each tree can see improves training time\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "db85c902",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Training MAE': 5559.501292464943,\n",
       " 'Valid MAE': 7146.917208156917,\n",
       " 'Training RMSLE': 0.2575612841223517,\n",
       " 'Valid RMSLE': 0.29203596131569337,\n",
       " 'Training R^2': 0.8607413993755985,\n",
       " 'Valid R^2': 0.8336784306238554}"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "show_scores(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cda1c5c2",
   "metadata": {},
   "source": [
    "## Modelling Step-4: Hyperparameter tuning with RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "e7d7cc24",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 20 candidates, totalling 100 fits\n",
      "CPU times: total: 9min 32s\n",
      "Wall time: 9min 33s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=RandomForestRegressor(), n_iter=20,\n",
       "                   param_distributions={'max_depth': [None, 3, 5, 10],\n",
       "                                        'max_features': [0.5, 1, 'sqrt',\n",
       "                                                         'auto'],\n",
       "                                        'max_samples': [10000],\n",
       "                                        'min_samples_leaf': array([ 1,  3,  5,  7,  9, 11, 13, 15, 17, 19]),\n",
       "                                        'min_samples_split': array([ 2,  4,  6,  8, 10, 12, 14, 16, 18]),\n",
       "                                        'n_estimators': array([10, 20, 30, 40, 50, 60, 70, 80, 90])},\n",
       "                   verbose=True)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "\n",
    "# Different RandomForestClassifier hyperparameters\n",
    "rf_grid = {\"n_estimators\": np.arange(10, 100, 10),\n",
    "           \"max_depth\": [None, 3, 5, 10],\n",
    "           \"min_samples_split\": np.arange(2, 20, 2),\n",
    "           \"min_samples_leaf\": np.arange(1, 20, 2),\n",
    "           \"max_features\": [0.5, 1, \"sqrt\", \"auto\"],\n",
    "           \"max_samples\": [10000]}\n",
    "\n",
    "rs_model = RandomizedSearchCV(RandomForestRegressor(),\n",
    "                              param_distributions=rf_grid,\n",
    "                              n_iter=20,\n",
    "                              cv=5,\n",
    "                              verbose=True)\n",
    "\n",
    "rs_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "bad840e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_estimators': 50,\n",
       " 'min_samples_split': 14,\n",
       " 'min_samples_leaf': 7,\n",
       " 'max_samples': 10000,\n",
       " 'max_features': 0.5,\n",
       " 'max_depth': None}"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find the best parameters from the RandomizedSearch \n",
    "rs_model.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "9cbbb80f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Training MAE': 6204.604672990465,\n",
       " 'Valid MAE': 7588.973167445535,\n",
       " 'Training RMSLE': 0.28163565502500026,\n",
       " 'Valid RMSLE': 0.30704440081236817,\n",
       " 'Training R^2': 0.8290257833004695,\n",
       " 'Valid R^2': 0.8128158966902216}"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate the RandomizedSearch model\n",
    "show_scores(rs_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "250044a3",
   "metadata": {},
   "source": [
    "### Train a model with the best parameters\n",
    "In a model I prepared earlier, I tried 100 different combinations of hyperparameters (setting n_iter to 100 in RandomizedSearchCV) and found the best results came from the ones you see below.\n",
    "\n",
    "Note: This kind of search on my computer (n_iter = 100) took ~2-hours. So it's kind of a set and come back later experiment.\n",
    "\n",
    "We'll instantiate a new model with these discovered hyperparameters and reset the max_samples back to its original value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "c554024c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: total: 6min 33s\n",
      "Wall time: 56 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_features=0.5, min_samples_split=14, n_estimators=90,\n",
       "                      n_jobs=-1)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "# Most ideal hyperparameters\n",
    "ideal_model = RandomForestRegressor(n_estimators=90,\n",
    "                                    min_samples_leaf=1,\n",
    "                                    min_samples_split=14,\n",
    "                                    max_features=0.5,\n",
    "                                    n_jobs=-1,\n",
    "                                    max_samples=None)\n",
    "ideal_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "fd8a429a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Training MAE': 2930.9911522274256,\n",
       " 'Valid MAE': 5927.361868807335,\n",
       " 'Training RMSLE': 0.14352809526801028,\n",
       " 'Valid RMSLE': 0.24560802458274109,\n",
       " 'Training R^2': 0.9595699366884156,\n",
       " 'Valid R^2': 0.8825154127870516}"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "show_scores(ideal_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "8fc956b1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: total: 3min 3s\n",
      "Wall time: 25.9 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_features=0.5, min_samples_leaf=3, n_estimators=40,\n",
       "                      n_jobs=-1)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "# Faster model\n",
    "fast_model = RandomForestRegressor(n_estimators=40,\n",
    "                                   min_samples_leaf=3,\n",
    "                                   max_features=0.5,\n",
    "                                   n_jobs=-1)\n",
    "fast_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "fd82a669",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Training MAE': 2538.3700030826235,\n",
       " 'Valid MAE': 5901.842139046333,\n",
       " 'Training RMSLE': 0.12935673426211658,\n",
       " 'Valid RMSLE': 0.24371831811918682,\n",
       " 'Training R^2': 0.9673419229814186,\n",
       " 'Valid R^2': 0.8821707184043303}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "show_scores(fast_model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f22dd48",
   "metadata": {},
   "source": [
    "## Modelling Step-5: Make Predictions on test data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e362a8c",
   "metadata": {},
   "source": [
    "Now we've got a trained model, it's time to make predictions on the test data.\n",
    "\n",
    "Remember what we've done.\n",
    "\n",
    "Our model is trained on data prior to 2011. However, the test data is from May 1 2012 to November 2012.\n",
    "\n",
    "So what we're doing is trying to use the patterns our model has learned in the training data to predict the sale price of a Bulldozer with characteristics it's never seen before but are assumed to be similar to that of those in the training data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "8089d469",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>saledate</th>\n",
       "      <th>fiModelDesc</th>\n",
       "      <th>...</th>\n",
       "      <th>Undercarriage_Pad_Width</th>\n",
       "      <th>Stick_Length</th>\n",
       "      <th>Thumb</th>\n",
       "      <th>Pattern_Changer</th>\n",
       "      <th>Grouser_Type</th>\n",
       "      <th>Backhoe_Mounting</th>\n",
       "      <th>Blade_Type</th>\n",
       "      <th>Travel_Controls</th>\n",
       "      <th>Differential_Type</th>\n",
       "      <th>Steering_Controls</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1227829</td>\n",
       "      <td>1006309</td>\n",
       "      <td>3168</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1999</td>\n",
       "      <td>3688.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>2012-05-03</td>\n",
       "      <td>580G</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1227844</td>\n",
       "      <td>1022817</td>\n",
       "      <td>7271</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1000</td>\n",
       "      <td>28555.0</td>\n",
       "      <td>High</td>\n",
       "      <td>2012-05-10</td>\n",
       "      <td>936</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Standard</td>\n",
       "      <td>Conventional</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1227847</td>\n",
       "      <td>1031560</td>\n",
       "      <td>22805</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2004</td>\n",
       "      <td>6038.0</td>\n",
       "      <td>Medium</td>\n",
       "      <td>2012-05-10</td>\n",
       "      <td>EC210BLC</td>\n",
       "      <td>...</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>9' 6\"</td>\n",
       "      <td>Manual</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>Double</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1227848</td>\n",
       "      <td>56204</td>\n",
       "      <td>1269</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2006</td>\n",
       "      <td>8940.0</td>\n",
       "      <td>High</td>\n",
       "      <td>2012-05-10</td>\n",
       "      <td>330CL</td>\n",
       "      <td>...</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>Manual</td>\n",
       "      <td>Yes</td>\n",
       "      <td>Triple</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1227863</td>\n",
       "      <td>1053887</td>\n",
       "      <td>22312</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2005</td>\n",
       "      <td>2286.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>2012-05-10</td>\n",
       "      <td>650K</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>PAT</td>\n",
       "      <td>None or Unspecified</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1227829    1006309     3168         121             3      1999   \n",
       "1  1227844    1022817     7271         121             3      1000   \n",
       "2  1227847    1031560    22805         121             3      2004   \n",
       "3  1227848      56204     1269         121             3      2006   \n",
       "4  1227863    1053887    22312         121             3      2005   \n",
       "\n",
       "   MachineHoursCurrentMeter UsageBand   saledate fiModelDesc  ...  \\\n",
       "0                    3688.0       Low 2012-05-03        580G  ...   \n",
       "1                   28555.0      High 2012-05-10         936  ...   \n",
       "2                    6038.0    Medium 2012-05-10    EC210BLC  ...   \n",
       "3                    8940.0      High 2012-05-10       330CL  ...   \n",
       "4                    2286.0       Low 2012-05-10        650K  ...   \n",
       "\n",
       "  Undercarriage_Pad_Width         Stick_Length   Thumb      Pattern_Changer  \\\n",
       "0                     NaN                  NaN     NaN                  NaN   \n",
       "1                     NaN                  NaN     NaN                  NaN   \n",
       "2     None or Unspecified                9' 6\"  Manual  None or Unspecified   \n",
       "3     None or Unspecified  None or Unspecified  Manual                  Yes   \n",
       "4                     NaN                  NaN     NaN                  NaN   \n",
       "\n",
       "  Grouser_Type     Backhoe_Mounting Blade_Type      Travel_Controls  \\\n",
       "0          NaN                  NaN        NaN                  NaN   \n",
       "1          NaN                  NaN        NaN                  NaN   \n",
       "2       Double                  NaN        NaN                  NaN   \n",
       "3       Triple                  NaN        NaN                  NaN   \n",
       "4          NaN  None or Unspecified        PAT  None or Unspecified   \n",
       "\n",
       "  Differential_Type Steering_Controls  \n",
       "0               NaN               NaN  \n",
       "1          Standard      Conventional  \n",
       "2               NaN               NaN  \n",
       "3               NaN               NaN  \n",
       "4               NaN               NaN  \n",
       "\n",
       "[5 rows x 52 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv(\"Data/bluebook-for-bulldozers/Test.csv\",\n",
    "                      parse_dates=[\"saledate\"])\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ef00606",
   "metadata": {},
   "source": [
    "### i. Preprocessing the data \n",
    "#### As our test dataset and train dataset are different we have to manipulate test data and make it look same as train data i.e. same number of columns and no missing values and all numeric values"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a3377be",
   "metadata": {},
   "source": [
    "Our model has been trained on data formatted in the same way as the training data.\n",
    "\n",
    "This means in order to make predictions on the test data, we need to take the same steps we used to preprocess the training data to preprocess the test data.\n",
    "\n",
    "Remember: Whatever you do to the training data, you have to do to the test data.\n",
    "\n",
    "Let's create a function for doing so (by copying the preprocessing steps we used above).\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "b7c3c7bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_data(df):\n",
    "    # Add datetime parameters for saledate\n",
    "    df[\"saleYear\"] = df.saledate.dt.year\n",
    "    df[\"saleMonth\"] = df.saledate.dt.month\n",
    "    df[\"saleDay\"] = df.saledate.dt.day\n",
    "    df[\"saleDayofWeek\"] = df.saledate.dt.dayofweek\n",
    "    df[\"saleDayofYear\"] = df.saledate.dt.dayofyear\n",
    "\n",
    "    # Drop original saledate\n",
    "    df.drop(\"saledate\", axis=1, inplace=True)\n",
    "    \n",
    "    # Fill numeric rows with the median\n",
    "    for label, content in df.items():\n",
    "        if pd.api.types.is_numeric_dtype(content):\n",
    "            if pd.isnull(content).sum():\n",
    "                df[label+\"_is_missing\"] = pd.isnull(content)\n",
    "                df[label] = content.fillna(content.median())\n",
    "                \n",
    "        # Turn categorical variables into numbers\n",
    "        if not pd.api.types.is_numeric_dtype(content):\n",
    "            df[label+\"_is_missing\"] = pd.isnull(content)\n",
    "            # We add the +1 because pandas encodes missing categories as -1\n",
    "            df[label] = pd.Categorical(content).codes+1        \n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f22ab725",
   "metadata": {},
   "source": [
    "Question: Where would this function break?\n",
    "\n",
    "Hint: What if the test data had different missing values to the training data?\n",
    "\n",
    "Now we've got a function for preprocessing data, let's preprocess the test dataset into the same format as our training dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "319c3ef9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>fiModelDesc</th>\n",
       "      <th>fiBaseModel</th>\n",
       "      <th>...</th>\n",
       "      <th>Undercarriage_Pad_Width_is_missing</th>\n",
       "      <th>Stick_Length_is_missing</th>\n",
       "      <th>Thumb_is_missing</th>\n",
       "      <th>Pattern_Changer_is_missing</th>\n",
       "      <th>Grouser_Type_is_missing</th>\n",
       "      <th>Backhoe_Mounting_is_missing</th>\n",
       "      <th>Blade_Type_is_missing</th>\n",
       "      <th>Travel_Controls_is_missing</th>\n",
       "      <th>Differential_Type_is_missing</th>\n",
       "      <th>Steering_Controls_is_missing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1227829</td>\n",
       "      <td>1006309</td>\n",
       "      <td>3168</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1999</td>\n",
       "      <td>3688.0</td>\n",
       "      <td>2</td>\n",
       "      <td>499</td>\n",
       "      <td>180</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1227844</td>\n",
       "      <td>1022817</td>\n",
       "      <td>7271</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1000</td>\n",
       "      <td>28555.0</td>\n",
       "      <td>1</td>\n",
       "      <td>831</td>\n",
       "      <td>292</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1227847</td>\n",
       "      <td>1031560</td>\n",
       "      <td>22805</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2004</td>\n",
       "      <td>6038.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1177</td>\n",
       "      <td>404</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1227848</td>\n",
       "      <td>56204</td>\n",
       "      <td>1269</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2006</td>\n",
       "      <td>8940.0</td>\n",
       "      <td>1</td>\n",
       "      <td>287</td>\n",
       "      <td>113</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1227863</td>\n",
       "      <td>1053887</td>\n",
       "      <td>22312</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2005</td>\n",
       "      <td>2286.0</td>\n",
       "      <td>2</td>\n",
       "      <td>566</td>\n",
       "      <td>196</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 101 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1227829    1006309     3168         121             3      1999   \n",
       "1  1227844    1022817     7271         121             3      1000   \n",
       "2  1227847    1031560    22805         121             3      2004   \n",
       "3  1227848      56204     1269         121             3      2006   \n",
       "4  1227863    1053887    22312         121             3      2005   \n",
       "\n",
       "   MachineHoursCurrentMeter  UsageBand  fiModelDesc  fiBaseModel  ...  \\\n",
       "0                    3688.0          2          499          180  ...   \n",
       "1                   28555.0          1          831          292  ...   \n",
       "2                    6038.0          3         1177          404  ...   \n",
       "3                    8940.0          1          287          113  ...   \n",
       "4                    2286.0          2          566          196  ...   \n",
       "\n",
       "   Undercarriage_Pad_Width_is_missing  Stick_Length_is_missing  \\\n",
       "0                                True                     True   \n",
       "1                                True                     True   \n",
       "2                               False                    False   \n",
       "3                               False                    False   \n",
       "4                                True                     True   \n",
       "\n",
       "   Thumb_is_missing  Pattern_Changer_is_missing  Grouser_Type_is_missing  \\\n",
       "0              True                        True                     True   \n",
       "1              True                        True                     True   \n",
       "2             False                       False                    False   \n",
       "3             False                       False                    False   \n",
       "4              True                        True                     True   \n",
       "\n",
       "   Backhoe_Mounting_is_missing  Blade_Type_is_missing  \\\n",
       "0                         True                   True   \n",
       "1                         True                   True   \n",
       "2                         True                   True   \n",
       "3                         True                   True   \n",
       "4                        False                  False   \n",
       "\n",
       "   Travel_Controls_is_missing  Differential_Type_is_missing  \\\n",
       "0                        True                          True   \n",
       "1                        True                         False   \n",
       "2                        True                          True   \n",
       "3                        True                          True   \n",
       "4                       False                          True   \n",
       "\n",
       "   Steering_Controls_is_missing  \n",
       "0                          True  \n",
       "1                         False  \n",
       "2                          True  \n",
       "3                          True  \n",
       "4                          True  \n",
       "\n",
       "[5 rows x 101 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = preprocess_data(df_test)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "06c30988",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>fiModelDesc</th>\n",
       "      <th>fiBaseModel</th>\n",
       "      <th>...</th>\n",
       "      <th>Undercarriage_Pad_Width_is_missing</th>\n",
       "      <th>Stick_Length_is_missing</th>\n",
       "      <th>Thumb_is_missing</th>\n",
       "      <th>Pattern_Changer_is_missing</th>\n",
       "      <th>Grouser_Type_is_missing</th>\n",
       "      <th>Backhoe_Mounting_is_missing</th>\n",
       "      <th>Blade_Type_is_missing</th>\n",
       "      <th>Travel_Controls_is_missing</th>\n",
       "      <th>Differential_Type_is_missing</th>\n",
       "      <th>Steering_Controls_is_missing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1646770</td>\n",
       "      <td>1126363</td>\n",
       "      <td>8434</td>\n",
       "      <td>132</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1974</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>4593</td>\n",
       "      <td>1744</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1821514</td>\n",
       "      <td>1194089</td>\n",
       "      <td>10150</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1980</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1820</td>\n",
       "      <td>559</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1505138</td>\n",
       "      <td>1473654</td>\n",
       "      <td>4139</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1978</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2348</td>\n",
       "      <td>713</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1671174</td>\n",
       "      <td>1327630</td>\n",
       "      <td>8591</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1980</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1819</td>\n",
       "      <td>558</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1329056</td>\n",
       "      <td>1336053</td>\n",
       "      <td>4089</td>\n",
       "      <td>132</td>\n",
       "      <td>99.0</td>\n",
       "      <td>1984</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2119</td>\n",
       "      <td>683</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 102 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1646770    1126363     8434         132          18.0      1974   \n",
       "1  1821514    1194089    10150         132          99.0      1980   \n",
       "2  1505138    1473654     4139         132          99.0      1978   \n",
       "3  1671174    1327630     8591         132          99.0      1980   \n",
       "4  1329056    1336053     4089         132          99.0      1984   \n",
       "\n",
       "   MachineHoursCurrentMeter  UsageBand  fiModelDesc  fiBaseModel  ...  \\\n",
       "0                       0.0          0         4593         1744  ...   \n",
       "1                       0.0          0         1820          559  ...   \n",
       "2                       0.0          0         2348          713  ...   \n",
       "3                       0.0          0         1819          558  ...   \n",
       "4                       0.0          0         2119          683  ...   \n",
       "\n",
       "   Undercarriage_Pad_Width_is_missing  Stick_Length_is_missing  \\\n",
       "0                                True                     True   \n",
       "1                                True                     True   \n",
       "2                                True                     True   \n",
       "3                                True                     True   \n",
       "4                                True                     True   \n",
       "\n",
       "   Thumb_is_missing  Pattern_Changer_is_missing  Grouser_Type_is_missing  \\\n",
       "0              True                        True                     True   \n",
       "1              True                        True                     True   \n",
       "2              True                        True                     True   \n",
       "3              True                        True                     True   \n",
       "4              True                        True                     True   \n",
       "\n",
       "   Backhoe_Mounting_is_missing  Blade_Type_is_missing  \\\n",
       "0                        False                  False   \n",
       "1                         True                   True   \n",
       "2                        False                  False   \n",
       "3                         True                   True   \n",
       "4                        False                  False   \n",
       "\n",
       "   Travel_Controls_is_missing  Differential_Type_is_missing  \\\n",
       "0                       False                          True   \n",
       "1                        True                         False   \n",
       "2                       False                          True   \n",
       "3                        True                         False   \n",
       "4                       False                          True   \n",
       "\n",
       "   Steering_Controls_is_missing  \n",
       "0                          True  \n",
       "1                         False  \n",
       "2                          True  \n",
       "3                         False  \n",
       "4                          True  \n",
       "\n",
       "[5 rows x 102 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "5ca2867c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ankit\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\base.py:493: FutureWarning: The feature names should match those that were passed during fit. Starting version 1.2, an error will be raised.\n",
      "Feature names seen at fit time, yet now missing:\n",
      "- auctioneerID_is_missing\n",
      "\n",
      "  warnings.warn(message, FutureWarning)\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "X has 101 features, but RandomForestRegressor is expecting 102 features as input.",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Input \u001b[1;32mIn [55]\u001b[0m, in \u001b[0;36m<cell line: 2>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Make predictions on the test dataset using the best model\u001b[39;00m\n\u001b[1;32m----> 2\u001b[0m test_preds \u001b[38;5;241m=\u001b[39m \u001b[43mideal_model\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpredict\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdf_test\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32m~\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\ensemble\\_forest.py:971\u001b[0m, in \u001b[0;36mForestRegressor.predict\u001b[1;34m(self, X)\u001b[0m\n\u001b[0;32m    969\u001b[0m check_is_fitted(\u001b[38;5;28mself\u001b[39m)\n\u001b[0;32m    970\u001b[0m \u001b[38;5;66;03m# Check data\u001b[39;00m\n\u001b[1;32m--> 971\u001b[0m X \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_validate_X_predict\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    973\u001b[0m \u001b[38;5;66;03m# Assign chunk of trees to jobs\u001b[39;00m\n\u001b[0;32m    974\u001b[0m n_jobs, _, _ \u001b[38;5;241m=\u001b[39m _partition_estimators(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mn_estimators, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mn_jobs)\n",
      "File \u001b[1;32m~\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\ensemble\\_forest.py:579\u001b[0m, in \u001b[0;36mBaseForest._validate_X_predict\u001b[1;34m(self, X)\u001b[0m\n\u001b[0;32m    576\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m    577\u001b[0m \u001b[38;5;124;03mValidate X whenever one tries to predict, apply, predict_proba.\"\"\"\u001b[39;00m\n\u001b[0;32m    578\u001b[0m check_is_fitted(\u001b[38;5;28mself\u001b[39m)\n\u001b[1;32m--> 579\u001b[0m X \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_validate_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdtype\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mDTYPE\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maccept_sparse\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mcsr\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mreset\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[0;32m    580\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m issparse(X) \u001b[38;5;129;01mand\u001b[39;00m (X\u001b[38;5;241m.\u001b[39mindices\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m!=\u001b[39m np\u001b[38;5;241m.\u001b[39mintc \u001b[38;5;129;01mor\u001b[39;00m X\u001b[38;5;241m.\u001b[39mindptr\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m!=\u001b[39m np\u001b[38;5;241m.\u001b[39mintc):\n\u001b[0;32m    581\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mNo support for np.int64 index based sparse matrices\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "File \u001b[1;32m~\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\base.py:585\u001b[0m, in \u001b[0;36mBaseEstimator._validate_data\u001b[1;34m(self, X, y, reset, validate_separately, **check_params)\u001b[0m\n\u001b[0;32m    582\u001b[0m     out \u001b[38;5;241m=\u001b[39m X, y\n\u001b[0;32m    584\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m no_val_X \u001b[38;5;129;01mand\u001b[39;00m check_params\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mensure_2d\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mTrue\u001b[39;00m):\n\u001b[1;32m--> 585\u001b[0m     \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_check_n_features\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mreset\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mreset\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    587\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m out\n",
      "File \u001b[1;32m~\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\base.py:400\u001b[0m, in \u001b[0;36mBaseEstimator._check_n_features\u001b[1;34m(self, X, reset)\u001b[0m\n\u001b[0;32m    397\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m\n\u001b[0;32m    399\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m n_features \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mn_features_in_:\n\u001b[1;32m--> 400\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[0;32m    401\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mX has \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mn_features\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m features, but \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    402\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mis expecting \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mn_features_in_\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m features as input.\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m    403\u001b[0m     )\n",
      "\u001b[1;31mValueError\u001b[0m: X has 101 features, but RandomForestRegressor is expecting 102 features as input."
     ]
    }
   ],
   "source": [
    "# Make predictions on the test dataset using the best model\n",
    "test_preds = ideal_model.predict(df_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c437c3f",
   "metadata": {},
   "source": [
    "We've found an error and it's because our test dataset (after preprocessing) has 101 columns where as, our training dataset (X_train) has 102 columns (after preprocessing).\n",
    "\n",
    "Let's find the difference."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "a86c3db4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'auctioneerID_is_missing'}"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can find how the columns differ using sets\n",
    "set(X_train.columns) - set(df_test.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b83e1ec4",
   "metadata": {},
   "source": [
    "In this case, it's because the test dataset wasn't missing any auctioneerID fields.\n",
    "\n",
    "To fix it, we'll add a column to the test dataset called auctioneerID_is_missing and fill it with False, since none of the auctioneerID fields are missing in the test dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "d3c4f1aa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>MachineID</th>\n",
       "      <th>ModelID</th>\n",
       "      <th>datasource</th>\n",
       "      <th>auctioneerID</th>\n",
       "      <th>YearMade</th>\n",
       "      <th>MachineHoursCurrentMeter</th>\n",
       "      <th>UsageBand</th>\n",
       "      <th>fiModelDesc</th>\n",
       "      <th>fiBaseModel</th>\n",
       "      <th>...</th>\n",
       "      <th>Stick_Length_is_missing</th>\n",
       "      <th>Thumb_is_missing</th>\n",
       "      <th>Pattern_Changer_is_missing</th>\n",
       "      <th>Grouser_Type_is_missing</th>\n",
       "      <th>Backhoe_Mounting_is_missing</th>\n",
       "      <th>Blade_Type_is_missing</th>\n",
       "      <th>Travel_Controls_is_missing</th>\n",
       "      <th>Differential_Type_is_missing</th>\n",
       "      <th>Steering_Controls_is_missing</th>\n",
       "      <th>auctioneerID_is_missing</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1227829</td>\n",
       "      <td>1006309</td>\n",
       "      <td>3168</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1999</td>\n",
       "      <td>3688.0</td>\n",
       "      <td>2</td>\n",
       "      <td>499</td>\n",
       "      <td>180</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1227844</td>\n",
       "      <td>1022817</td>\n",
       "      <td>7271</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>1000</td>\n",
       "      <td>28555.0</td>\n",
       "      <td>1</td>\n",
       "      <td>831</td>\n",
       "      <td>292</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1227847</td>\n",
       "      <td>1031560</td>\n",
       "      <td>22805</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2004</td>\n",
       "      <td>6038.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1177</td>\n",
       "      <td>404</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1227848</td>\n",
       "      <td>56204</td>\n",
       "      <td>1269</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2006</td>\n",
       "      <td>8940.0</td>\n",
       "      <td>1</td>\n",
       "      <td>287</td>\n",
       "      <td>113</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1227863</td>\n",
       "      <td>1053887</td>\n",
       "      <td>22312</td>\n",
       "      <td>121</td>\n",
       "      <td>3</td>\n",
       "      <td>2005</td>\n",
       "      <td>2286.0</td>\n",
       "      <td>2</td>\n",
       "      <td>566</td>\n",
       "      <td>196</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 102 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalesID  MachineID  ModelID  datasource  auctioneerID  YearMade  \\\n",
       "0  1227829    1006309     3168         121             3      1999   \n",
       "1  1227844    1022817     7271         121             3      1000   \n",
       "2  1227847    1031560    22805         121             3      2004   \n",
       "3  1227848      56204     1269         121             3      2006   \n",
       "4  1227863    1053887    22312         121             3      2005   \n",
       "\n",
       "   MachineHoursCurrentMeter  UsageBand  fiModelDesc  fiBaseModel  ...  \\\n",
       "0                    3688.0          2          499          180  ...   \n",
       "1                   28555.0          1          831          292  ...   \n",
       "2                    6038.0          3         1177          404  ...   \n",
       "3                    8940.0          1          287          113  ...   \n",
       "4                    2286.0          2          566          196  ...   \n",
       "\n",
       "   Stick_Length_is_missing  Thumb_is_missing  Pattern_Changer_is_missing  \\\n",
       "0                     True              True                        True   \n",
       "1                     True              True                        True   \n",
       "2                    False             False                       False   \n",
       "3                    False             False                       False   \n",
       "4                     True              True                        True   \n",
       "\n",
       "   Grouser_Type_is_missing  Backhoe_Mounting_is_missing  \\\n",
       "0                     True                         True   \n",
       "1                     True                         True   \n",
       "2                    False                         True   \n",
       "3                    False                         True   \n",
       "4                     True                        False   \n",
       "\n",
       "   Blade_Type_is_missing  Travel_Controls_is_missing  \\\n",
       "0                   True                        True   \n",
       "1                   True                        True   \n",
       "2                   True                        True   \n",
       "3                   True                        True   \n",
       "4                  False                       False   \n",
       "\n",
       "   Differential_Type_is_missing  Steering_Controls_is_missing  \\\n",
       "0                          True                          True   \n",
       "1                         False                         False   \n",
       "2                          True                          True   \n",
       "3                          True                          True   \n",
       "4                          True                          True   \n",
       "\n",
       "   auctioneerID_is_missing  \n",
       "0                    False  \n",
       "1                    False  \n",
       "2                    False  \n",
       "3                    False  \n",
       "4                    False  \n",
       "\n",
       "[5 rows x 102 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Match test dataset columns to training dataset\n",
    "df_test[\"auctioneerID_is_missing\"] = False\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6199883e",
   "metadata": {},
   "source": [
    "Now the test dataset matches the training dataset, we should be able to make predictions on it using our trained model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "083850cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ankit\\Desktop\\ML_DS_Projects\\Regression_project\\env\\lib\\site-packages\\sklearn\\base.py:493: FutureWarning: The feature names should match those that were passed during fit. Starting version 1.2, an error will be raised.\n",
      "Feature names must be in the same order as they were in fit.\n",
      "\n",
      "  warnings.warn(message, FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# Make predictions on the test dataset using the best model\n",
    "test_preds = ideal_model.predict(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "fc5412d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>SalesID</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1227829</td>\n",
       "      <td>18923.523509</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1227844</td>\n",
       "      <td>18984.474057</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1227847</td>\n",
       "      <td>47890.526859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1227848</td>\n",
       "      <td>62904.298239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1227863</td>\n",
       "      <td>48892.670853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12452</th>\n",
       "      <td>6643171</td>\n",
       "      <td>45993.611618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12453</th>\n",
       "      <td>6643173</td>\n",
       "      <td>14117.748375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12454</th>\n",
       "      <td>6643184</td>\n",
       "      <td>14562.053315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12455</th>\n",
       "      <td>6643186</td>\n",
       "      <td>18942.861394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12456</th>\n",
       "      <td>6643196</td>\n",
       "      <td>28965.723511</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12457 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       SalesID     SalePrice\n",
       "0      1227829  18923.523509\n",
       "1      1227844  18984.474057\n",
       "2      1227847  47890.526859\n",
       "3      1227848  62904.298239\n",
       "4      1227863  48892.670853\n",
       "...        ...           ...\n",
       "12452  6643171  45993.611618\n",
       "12453  6643173  14117.748375\n",
       "12454  6643184  14562.053315\n",
       "12455  6643186  18942.861394\n",
       "12456  6643196  28965.723511\n",
       "\n",
       "[12457 rows x 2 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create DataFrame compatible with Kaggle submission requirements\n",
    "df_preds = pd.DataFrame()\n",
    "df_preds[\"SalesID\"] = df_test[\"SalesID\"]\n",
    "df_preds[\"SalePrice\"] = test_preds\n",
    "df_preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "97758147",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Export to csv...\n",
    "df_preds.to_csv(\"Data/bluebook-for-bulldozers/predictions.csv\",\n",
    "                index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1a753db",
   "metadata": {},
   "source": [
    "## Modelling Step-6: Feature Importance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9a930fe",
   "metadata": {},
   "source": [
    "Since we've built a model which is able to make predictions. The people you share these predictions with (or yourself) might be curious of what parts of the data led to these predictions.\n",
    "\n",
    "This is where feature importance comes in. Feature importance seeks to figure out which different attributes of the data were most important when it comes to predicting the target variable.\n",
    "\n",
    "In our case, after our model learned the patterns in the data, which bulldozer sale attributes were most important for predicting its overall sale price?\n",
    "\n",
    "Beware: the default feature importances for random forests can lead to non-ideal results.\n",
    "\n",
    "To find which features were most important of a machine learning model, a good idea is to search something like \"[MODEL NAME] feature importance\".\n",
    "\n",
    "Doing this for our RandomForestRegressor leads us to find the feature_importances_ attribute.\n",
    "\n",
    "Let's check it out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "ae0dbb15",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3.45498198e-02, 1.82508792e-02, 4.39575134e-02, 1.66705052e-03,\n",
       "       3.35604241e-03, 2.07342512e-01, 3.27540982e-03, 1.08757252e-03,\n",
       "       3.90729680e-02, 4.29506279e-02, 6.07689200e-02, 4.85225973e-03,\n",
       "       1.83412074e-02, 1.63430850e-01, 4.03621315e-02, 5.95135377e-03,\n",
       "       2.26793149e-03, 3.07131387e-03, 2.95584109e-03, 6.76990931e-02,\n",
       "       5.77223353e-04, 3.85311981e-04, 1.47172620e-03, 1.68749229e-04,\n",
       "       1.04296444e-03, 2.17673808e-05, 1.79213105e-03, 5.55761147e-03,\n",
       "       4.15843979e-03, 1.07496163e-04, 2.79363274e-03, 4.46934892e-03,\n",
       "       3.79232518e-03, 1.05733856e-03, 2.52988039e-03, 1.22619959e-02,\n",
       "       8.10768497e-04, 1.26986094e-02, 1.55792433e-03, 2.39774669e-03,\n",
       "       7.77353605e-04, 9.30824348e-04, 1.69532194e-03, 6.57065002e-04,\n",
       "       4.78846407e-04, 3.51472727e-04, 2.67507153e-04, 2.46624062e-03,\n",
       "       1.14823862e-03, 2.03114101e-04, 4.88816285e-04, 7.29575855e-02,\n",
       "       3.80996997e-03, 5.71658981e-03, 2.89853400e-03, 9.89401780e-03,\n",
       "       2.54331715e-04, 1.53388225e-03, 3.11782655e-04, 0.00000000e+00,\n",
       "       0.00000000e+00, 2.52502131e-03, 1.30093790e-03, 9.55826693e-03,\n",
       "       1.79890311e-02, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
       "       0.00000000e+00, 4.47672592e-05, 7.94280140e-06, 3.48037400e-04,\n",
       "       7.00777671e-05, 1.07881212e-04, 2.66875048e-05, 2.62915906e-04,\n",
       "       1.71391601e-04, 1.13598406e-03, 2.65224913e-03, 1.90764949e-03,\n",
       "       1.06057340e-03, 4.26075829e-03, 1.10064568e-03, 1.40274275e-03,\n",
       "       6.15817615e-04, 1.83042965e-03, 3.88415772e-03, 3.30717460e-04,\n",
       "       1.15203035e-02, 1.55896629e-03, 1.55985630e-03, 6.99790368e-05,\n",
       "       5.88211442e-05, 4.72677499e-05, 5.48583496e-05, 4.08154050e-05,\n",
       "       6.47007179e-05, 2.59606149e-04, 1.10986210e-04, 1.61387534e-04,\n",
       "       1.00829702e-04, 9.11546569e-05])"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find feature importance of our best model\n",
    "ideal_model.feature_importances_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "4ad0df1b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting package metadata (current_repodata.json): ...working... done\n",
      "Solving environment: ...working... done\n",
      "\n",
      "## Package Plan ##\n",
      "\n",
      "  environment location: C:\\Users\\ankit\\Desktop\\ML_DS_Projects\\Regression_project\\env\n",
      "\n",
      "  added / updated specs:\n",
      "    - seaborn\n",
      "\n",
      "\n",
      "The following NEW packages will be INSTALLED:\n",
      "\n",
      "  seaborn            pkgs/main/noarch::seaborn-0.11.2-pyhd3eb1b0_0\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "==> WARNING: A newer version of conda exists. <==\n",
      "  current version: 4.11.0\n",
      "  latest version: 4.13.0\n",
      "\n",
      "Please update conda by running\n",
      "\n",
      "    $ conda update -n base -c defaults conda\n",
      "\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The following packages will be UPDATED:\n",
      "\n",
      "  certifi                          2021.10.8-py39haa95532_2 --> 2022.5.18.1-py39haa95532_0\n",
      "  openssl                                 1.1.1n-h2bbff1b_0 --> 1.1.1o-h2bbff1b_0\n",
      "\n",
      "\n",
      "Preparing transaction: ...working... done\n",
      "Verifying transaction: ...working... done\n",
      "Executing transaction: ...working... done\n"
     ]
    }
   ],
   "source": [
    "#Install Seaborn package in current environment (if you don't have it)\n",
    "import sys\n",
    "!conda install --yes --prefix {sys.prefix} seaborn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "45f35696",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "\n",
    "# Helper function for plotting feature importance\n",
    "def plot_features(columns, importances, n=20):\n",
    "    df = (pd.DataFrame({\"features\": columns,\n",
    "                        \"feature_importance\": importances})\n",
    "          .sort_values(\"feature_importance\", ascending=False)\n",
    "          .reset_index(drop=True))\n",
    "    \n",
    "    sns.barplot(x=\"feature_importance\",\n",
    "                y=\"features\",\n",
    "                data=df[:n],\n",
    "                orient=\"h\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "72b35e46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgMAAAEHCAYAAAAzokXcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABGtElEQVR4nO3de7xc0/3/8ddbXBJCgoSqIu63iJBEq9S9WqpFS1NFSy9+blX6pdUvVfSCL98qokiVqHsVFaqkjYhLXRK5B6WIb9W9IhF38fn9sdbIzmRmzpyTOWfOyXk/H495nJm191577ckwa9be+70UEZiZmVn3tVSzG2BmZmbN5c6AmZlZN+fOgJmZWTfnzoCZmVk3586AmZlZN7d0sxtg1lr9+vWLAQMGNLsZZmZdyiOPPPJqRPSvtMydAetyBgwYwMSJE5vdDDOzLkXSs9WWuTNgXc4Hr7zGKxdd1exmmJl1qP5HHNRudfuagSWUkvsk7VEo+6qkOxaz3pB0ZeH10pJekXRbK+u5W9LQxWmLmZk1hjsDS6hI0ZKHA7+S1FPSCsAvgKPaUp+kHvnpm8BASb3y688C/17c9pqZWfO4M7AEi4gZwK3Aj4CfAlcBJ0maIGmypL0BJA2QdK+kSfnx6Vy+k6Rxkq4Bpheq/gvwhfz8AODa0gJJ20j6e67/75I2zuW9JF0naZqk64FehW12l/RA3vcNknq325tiZmaLcGdgyXca8HVgD6AncFdEDAN2Bs7OIwYvA5+NiK2B4cD5he23AU6KiM0KZdcBX5PUExgEPFRY9jiwQ0RsBZwC/DKXHwG8FRGDSCMUQwAk9QNOBnbL+58I/KD8ICQdJmmipIn/mTe37e+GmZktwhcQLuEi4s38S3we8FXgi5KOz4t7AmsDzwMjJA0G5gMbFap4OCKeKatzmqQBpFGB28t22Qe4QtKGQADL5PIdyJ2MvP20XP4pYDPgfkkAywIPVDiOkcBIgMHrrOfZtczMGsidge7hw/wQ8JWI+EdxoaRTgZeALUmjRe8UFr9Zpc7RwDnATsCqhfKfAeMiYt/cYbi7sKzSl7iAv0bEAfUdipmZNZpPE3QvdwLfU/4JLmmrXN4HeCEiPgQOBnpU2b7oMuD0iJheVt6HBRcUHlIovwc4MO93IOn0AsCDwHaSNsjLlpdUHJkwM7N25pGB7uVnwK+BablDMAvYC/gNcKOk/YFxVB8N+EhEPAecV2HR/5BOE/wAuKtQfhFweT49MAV4ONfziqRDgGslLZfXPRl4otq+l+6/Srveb2tm1t0o3YFm1nUMHTo0nEBoZtY6kh6JiIr5Lh4ZsC7n/Vde4MWLft7sZphZB/jYESc3uwndgq8ZMDMz6+bcGeiCJM2XNEXSjBzSs/xi1NWmWGBJfSUdWXi9lKTzc5um52CjdfOy2yX1bWsbzcysfbkz0DW9HRGDI2Ig8B4pdvgjhejg9tQXOLLwejjwcWBQRGwB7Au8DhARe0bE6x3QJjMzawN3Brq+e4ENyqOD83wEl+df6ZMl7QwtxgLPKzzfT9Ko/Hx1STdLmpofnwbOBNbPIxRnA2uw4PZEIuK5iJidt58lqZ+kw/P6UyQ9I2lcXu44YjOzJvIFhF2YpKVJMcOlmQi3AQZGxDOS/gsgIraQtAkwJt+//1EssKRBwKQ6dnU+MD4HCfUAegMn5n0Nzm35BHCfpM8AY4GrImJysZKIuBi4WNIypNsOf1UWR/ympB+R4ohPLzvWw4DDANZcpU8r3iUzM2uJRwa6pl6SppBy/P8P+F0uL0YHbw9cCRARjwPPkmKGdyBNWERETAOm0bJdSDkBRMT8iJhTvkLOHdgY+DEp7XCspF2r1HceaY6EW1k4jngK8E1gnQr1j4yIoRExdNXeK9TRZDMzq5dHBrqmt0u/yEtyqGAxLEg1tq8WLlEs79naRkXEu6QZDf8i6SVgH9IoQbGdh5C+7I8utNNxxGZmTeSRgSVXMf53I9KERP+geiwwwEuSNpW0FOkCwJKxpNMLSOohaSXgDWDF0gqStpb08fx8qVzvs8UGSRoCHA8cVLq2AMcRm5k1nUcGlly/IZ2fnw58ABwSEe9KqhgLnJ0I3Ab8C5hBujYA4PvASEnfJs1qeEREPCDpfkkzSKMBY4HfFiKFHwZGlLXpaGAVYFweyZgYEd9pbRzxMv3XcBCJmVkDOY7YuhzHEZuZtZ7jiG2J8s7L/+TxC/dudjNsMWxy1C3NboKZFfiaAWuRpFGS9qux/LCcWVB6vZKkp0oJhGZm1rm5M2CN8FvgE5J2y69PBy4r3ObYah2UomhmZrgz0G1JWkHSn3Oi4AxJwyWdkucUmCFppPJVfmXbDZE0XtIjku6UtEakC0+OAH6d5znYFThb0kGSHs6Jg5eUvuAlXSRpoqSZkk4r1D0rt+E+YP8OeivMzLo9dwa6r88Dz0fElnmOgzuAERExLL/uBexV3CAnB14A7BcRQ4DLgF/ARwFGd5LuKjgGWJ80X8F2ORNhPvmWRuCkfBHLIGDHnIRY8k5EbB8R17XHQZuZ2aLcGei+pgO7STpL0mdyquDOkh7KtyPuAmxets3GwEDgrzkt8GTgE4XlFwL/johxpNGBIcCEvO6uwHp5va9KmgRMzvvYrFDH9VSQr0uYKGni7HnvtfmgzcxsUb6boJuKiCdyCNCewBmSxgBHAUMj4l+STmXRFEIBMyNi2yrVfpgfpXWviIgfL1RBuqjweGBYRMzOkyEV91NMUSy2dyQwEmDg2n19P6yZWQN5ZKCbymmBb0XEVcA5wNZ50at51sBKdw/8A+gvadtcxzKSykcPSsYC+0laLa+7iqR1gJVIX/hzJK1OmmjJzMyayCMD3dcWpIv8PgTeJ10AuA/p9MEsYEL5BhHxXr7F8HxJfUifn18DMyus+6ikk0mzJS6V93FURDwoaXLe5mng/sYfmpmZtYYTCK3LcQKhmVnr1Uog9GkCMzOzbs6nCazLeePVJ7n7t19odjOsip2+++dmN8HMWskjA0swSfNz4E/pcWIb65klqV+j22dmZp2DRwaWbG/nwJ9OS9LSEfFBs9thZtadeWSgG8q/9E+TNEnSdEmb5PLeki7PZdMkfaXCtj/IccUzJB2byxaJNi7sp19+PlTS3fn5qTnueAzwe0n9Jd2Yo5AnSNqug94KMzPDIwNLul45/a/kjIgoJfy9GhFbSzqSFAL0HeAnwJyI2AJA0srFynJI0aHAJ0mhQg9JGk9KFnw+Ir6Q1+tTR9uGANtHxNuSrgHOjYj7JK1NijXetG2HbGZmreXOwJKt1mmCm/LfR4Av5+e7AV8rrRARs8u22R64OSLeBJB0E/AZ0rwG50g6C7gtIu6to22jI+Ltwn43K8yLtJKkFSPijVKBpMOAwwBWX6U8GNHMzBaHTxN0X+/mv/NZ0CkUUCt4YpFZDCFFG5N+6U8nRRufkhd9wILPWPk3eDF2eClg24gYnB9rFjsCeR8jI2JoRAzts+KytY7LzMxayZ0BKxoDHF16UX6aALgH2EfS8pJWAPYF7q0RbTyL1EkAWOT6gxr7HbwYx2BmZq3kzsCSrVfZrYVntrD+z4GV80WAU4GdiwsjYhIwCngYeAi4NCImk6KNH87XJ5yU6wE4DThP0r2kEYhqjgGG5osWHwUOb9VRmpnZYnEcsXU5jiM2M2s9xxGbmZlZVb6bwLqc2a8+yR8v/3yzm9Ht7HfoHc1ugpm1E48MNJikYyQ9Jml2Kf5X0saS7s7n7R+TNLJJbbtbUsUhojbW9Y98nv9xSSMk9W1E3WZm1rE8MtB4RwJ7RMQzhbLzSaE6twBI2qIpLWslST0iotaFfwdGxERJywJnALcAO3ZM68zMrFE8MtBAki4mpfGNlnScpBF50RrAc6X1ImJ6Xr+HpLNzBO80Sf+vUNcPcyzw1NJdAJIGS3owr3tz6da//Cv9LEkPS3pC0mdyeS9J1+X1rwd6Feq/SNJESTMlnVYonyXpFEn3ASdKmlRYtqGkR8qPOyLeA34IrC1py7zuQbk9UyRdko+1h6RR+W6F6ZKOy+tuIOlv+VgnSVp/8f4lzMysNTwy0EARcbikz5NuydursOhc4C5JfyfdU395RLwOfJsU/ztM0nLA/TmvfxNgH+CTEfGWpFVyPb8HvhcR4yWdDvwUODYvWzoitpG0Zy7fDTiCdP//IEmDgI++2IGTIuI1ST2AsZIGRcS0vOydiNgeQNJukgZHxBRSFPGoKsc+P9+OuImk94DhwHYR8b6k3wAHAjOBNSNiYK67b978auDMiLhZUk8qdFKLCYT9VnUCoZlZI3lkoANExOWkrP0bgJ2AB/OX/+7AN/L9+Q8BqwIbkr7IL4+It/L2r+W8/74RMT5XewWwQ2E3xXjhAfn5DsBVuY5pwLTC+l/Nv/onA5sDmxWWXV94filwaO40DAeuqXGopYTCXUlhQxPyse1KGjF5GlhP0gW50zRX0oqkDsLNuZ3vlI67qJhAuFJvJxCamTWSRwY6SEQ8D1wGXCZpBjCQ9OX5vYi4s7hu/qJsbQBEpXhhKtUjaV3S5ETDImK2pFEsHBdcjAq+kTTScBfwSET8p9LOc2dhC+AxYDXgioj4cYX1tgQ+BxwFfJUFIxtmZtYkHhnoAJI+L2mZ/PxjpBGAf5Nm5zuisGwjpZjfMcC3JC2fy1eJiDnA7NL1AMDBwHhqu4c0PI+kgcCgXL4S6Qt/jqTVgT2qVRAR7+R2XgRcXuX4liFdQPivPAIxFthP0mql9ktaR2k646Ui4kbSDIlbR8Rc4DlJ++R1lysdt5mZdQyPDHSM3UmxvO/k1ydExIuSLiUN6U+SJOAVYJ+IuEMpn39iPv9+O/DfwDeBi/OX5dOkc/i1XARcLmkaMIUUI0xETJU0mXQO/2ng/hbquZo0s+GY8nJJ7wLLAX8D9s71PyrpZGCMpKWA90kjAW/n9pQ6oaWRg4OBS/J1EO8D++d2mZlZB3AcsbVI0vFAn4j4SbPbAo4jNjNrC9WII/bIgNUk6WZgfWCXZrfFzMzahzsDVlNE7NvsNpR75T9PcsmVn2t2MzqN/3fwnS2vZGZWgy8g7MYkhaQrC6+XlvSKpNtaWc+sfHFgXetImpf/DpD0tqTJSjHND0v6ZluOxczM2s4jA93bm8BASb0i4m3gs6S7HDrSUxGxFYCk9YCbJC2VsxnMzKwDeGTA/gJ8IT8/ALi2tCDfEvinHGf8YE4xRNKqksbkX/SXsCBsqGIMcb0NiYingR8AxzTiwMzMrD7uDNh1wNdyDPAgUhJiyWnA5IgYRLq18fe5/KfAffkX/WhgbQBJm7IghngwKQDpwFa2ZxIpjnkhkg5Tmkth4rw33mtllWZmVotPE3RzETFN0gDSqMDtZYu3B76S17srjwj0IcUcfzmX/1nS7Lx+MYYY0sRIL7eySapUGBEjgZEA66zbx/fDmpk1kDsDBunX/TmkeRNWLZRX+mKOsr9FokoMcStsRYo0NjOzDuLTBAZpzoTTS1MrFxTjjHcCXs3xwcXyPYCV8/oVY4jrbUQeoTgHuKCtB2JmZq3nkQEjIp4Dzquw6FQWxBm/RYpDhnQtwbV51sPxwP/leqrFED9bY/fr52jknsAbwAW+k8DMrGM5jti6HMcRm5m1Xq04Yp8mMDMz6+Z8msC6nOdnP8mpf3Ac8alfdQyxmTWGRwa6EEnH5Nje2ZJOzGWnSvp3Dvl5XNJFhSmCG7HPATm2+GeFsn6S3pc0opV1zWvEOmZm1ljuDHQtRwJ7RsTKEXFmofzcHPKzGbAFsGOD9/s0sFfh9f7AzAbvw8zMmsSdgS5C0sXAesBoScdV+VW+LOmq/Nl5m+9KmiBpqqQbJS2fy/eXNCOX35PLekg6O68/TdL/K9T7NvCYpNKFJ8OBPxTato6ksXm7sZJKiYTrSnog1/mzQn1IOqGwr9Ma8iaZmVmbuDPQRUTE4cDzwM7kL/uC4yRNAV4AnoiIKbn8pogYFhFbkoJ8vp3LTwE+l8u/lMu+DcyJiGHAMOC7ktYt7KMUW/wJUszw84VlI4Df59jiq4Hzc/l5wEW5zhdLK0vaHdgQ2AYYDAyRtEOt4y/GEb8113HEZmaN5M7AkqF0mmA1YAVJX8vlAyXdK2k6KSRo81x+PzBK0neB0kRCuwPfyJ2Kh0hJhBsW9nEHaVbDA4Dry/a/LXBNfn4lKcYYYDsWTHx0ZWH93fNjMgvmIijuaxERMTIihkbE0OVXWrbWqmZm1kq+m2AJEhHvS7qDNHfAdcAoYJ+ImCrpEFLcMBFxuKRPkmYrnCJpMClK+HsRsdAl6jkVkIh4T9IjwH+ROhVfrNWUKs8/qhY4IyIuaeUhmplZO/DIwBJEaXagTwNP5aIVgRckLUNh9kBJ60fEQxFxCvAqsBZwJ3BEXhdJG0laoWwX/wv8KCL+U1b+d6A0GnEgcF9+fn9ZecmdwLck9c77WrMUYWxmZh3PIwNLhuMkHQQsA0wDfpPLf0Ia8n8WmE7qHACcLWlD0i/0scDUvN0AYFLuVLwC7FPcSUTMpPJdBMcAl0k6IW93aC7/PnCNpO8DNxbqGZOnO34gz244DziIOmc4/PjKG/oeezOzBnIcsXU5jiM2M2s9xxGbmZlZVT5NYF3Ok68/xR63fKXZzehwf9n7xpZXMjNrA48MtKCOCOAZkr7UUj016t9J0m1t3HYfSZuVlR2fY4lLoULfyOV3F0KDFktu8xxJkyX9Q9I9kvZqeUszM+uMPDLQsiOBPSLimbLycyPinHwh3L2SVouID0sLJS0dER+0c9v2AW4DHs37PJyUBbBNRMyV1IeyiwAb6N6I2CvvdzDwJ0lvR8TYdtqfmZm1E48M1FBPBHBEPAZ8APTLv75/KWk88H1Ju+Zfz9MlXSZpuVzv5/Ov9/uALxf2d6qk4wuvZ5Tu85f0jRzdO1XSlZI+TUoPPDuPUKwP/DdwZETMzW2bExFXVDiui3Ka38xiFLCkMyU9mvdzTi5bJLq4wnswBTgdODpv018p/nhCfmyXy3fMbZ2S35cVc/kP83s0VdKZlfZhZmbtxyMDNeRwns+TIoArDoPn8J4PSbfUAfSNiB0l9QSeBHaNiCck/Z50H//FwG+BXYB/smiaX6V9bA6cBGwXEa9KWiUiXpM0GrgtIv6Yv1hXjIinatcGwEl5+x7AWEmDgOeAfYFNIiIk9c3rlqKL/10oq2QScEJ+fh5p5OQ+pXkK7gQ2BY4HjoqI+3PGwDuS9iCNXnwyIt6StEqV9+Aw4DCAnv171XGIZmZWL48MtN1xStG95wDDY8E9mqUv942BZyLiifz6ClIy4Ca5/Mm8zVV17GsX4I8R8SpARLxWYR1ROe2vkq9KmkSKA96cNNvhXOAd4FJJXwbeyutWii6uRIXnuwEj8vszGlgpd1buB34l6RhSp+mDvO7lEfFWjWNbKI542ZWWq/MwzcysHh4ZaLtzI+KcCuVv5r+qsKyk2pf2ByzcQetZqKvmF32+RuBNSetFxNPV1lOafOh4YFhEzJY0CugZER9I2gbYlZQaeDSwS5Xo4kq2Ik2GRD6GbSPi7bJ1zpT0Z2BP4EFJu9VzbGZm1r48MtB+HgcGSNogvz4YGJ/L183n+CFN/FMyC9gaQNLWQGnWwLGkX/Or5mWlofQ3WJAqCHAGcKGklfJ6K+Xh9aKVSB2WOZJWB/bI6/YG+kTE7cCxpNkEq0UXLySfZvgJcGEuGkO+fiAvL9Y1PSLOAiaSRknGkKKJS9MrVzxNYGZm7ccjA+0kIt6RdChwg6SlgQnAxRHxbv6C/rOkV0k5/gPzZjeyYObACcATua6Zkn4BjJc0nzS8fwhpMqLf5mH3/YCLgN7ABEnvA++T5hMotmuqpMmkWOGnSUP3kDoVt+RrHQQcl8srRRfvCHwm17M8KUb4mMKdBMeQOiXTSJ+xe4DDgWMl7UyaAvlR4C/5/RgMTJT0HnA76ULIqjbsu77vuTczayDHEVuX4zhiM7PWk+OIzczMrBqfJrAu58nXX2DPm3/e7GZ0mNv3PbnZTTCzJZxHBroIVY9FjsJFiiiFI4VaET0s6RBVCFSqto4WjmN+UtJNKotFNjOzrsOdga7jSGDPiFg5IoopfdNJtwKW7EeOJ25n50bE4IjYkJStcJek/h2wXzMza7AWOwOSvp9vUZOk30maJGn3jmicJaodi/wnYO+83nrAHBakISLpgBz1O0PSWYXyQyU9oRSdvF2hvGKUcC0RcT3pFsGv5zqGSBov6RFJd0paI5cfowVxx9flst6SLs9tnCap+01HaGbWZPWMDHwrZ93vDvQHDgWcH9+BIuJw4HlSLPLsssVzgX9JGkjKLPgo3ljSx4GzSAmGg4FhSjMdrgGcRuoEfJaUQFhSihIeBnwFuLTOZk4CNpG0DHABsF9EDAEuA36R1zkR2CoiBpFuNYSUTzAnIrbI5XdVqlzSYUrzKUx8b+6blVYxM7M2qucCwlKS3p6k2Nipkmql61nHu450quBzpATBQ3P5MODuiHgFQNLVpEhkysqvBzbK5bsBmxX+iUtRwi0pbbAxKTfhr7mOHsALedk04GpJfyKNaJT299Fpjogo7+yUykcCIwH6bLCm74c1M2ugejoDj0gaQ0rD+3H+YviwhW2sY90KnA1MzLHEpfK2RCJXjBKuo/+3FSlVUMDMiNi2wjpfIHVGvgT8RGkCJscRm5k1WT2nCb5NGt4dlieTWZYFvzytE8hf3D9iwXB8yUPAjpL6Kc1QeAApEvkhYCdJq+Zh/f0L21SMEq4ln+ffHbgW+AfQX9K2edkykjaXtBSwVkSMA34I9CWlJZbvb+VWHLqZmTVAPSMDQTqnvBdpzvoVWDCBjnUSEXFdhbIXJP0YGEf6BX57RNwC6fZA4AHSEP4kFsxIWC1KuNxxkg4ifR5mkCY1Kp122A84X1KfXMevSdHKV+Uyka5LeF3Sz/P+ZpBiik8Dbqp1rBv2XcP33puZNVCLccSSLiKdFtglIjbNv9zG5AvMzDqc44jNzFqvVhxxPSMDn4yIrfOkNORpb5dtaAvNWuHJ11/hCzdd1OxmNNSfv3xEs5tgZt1YPdcMvJ/PNwek+9DxBYRLJEknSZqZ7/efIumTNdYdlU8HtHYfTjI0M+tk6hkZOB+4GVhNaRrd/QCfsF3C5Av+9gK2ztMK9yNdLNrezo2Ic3IbhpOSDLcoXX9gZmbtr+bIQL4C/BnS1d9nkC422ycibuiAtlnHWgN4NSLeBYiIVyPieUmn5CTCGZJGVsqYaE3iYC3lSYZmZtYxanYGIuJD4H8j4vGIuDAiRkTEYx3UNutYY4C1ckTxbyTtmMtHRMSwiBgI9CKNHnykDYmDLZkEbLKYx2JmZq1QzzUDYyR9xamDS7aImAcMAQ4jzW1wvaRDgJ0lPSRpOinWePOyTYuJg1NIp5A+kZeVEgcPAj6osykVP2cLxRHPmVf/gZmZWYvquWbgB6R7yT+Q9A45MS4iVmrXllmHi4j5wN3A3fnL//8Bg4ChEfGvnE1QnjHR2sTBlpSSDMvbVogjXseJhWZmDdTiyEBErBgRS0XEshGxUn7tjsASRtLGkjYsFA0mpQkCvCqpN+ni0XKtTRys1YZikqGZmXWQFkcGJO1QqTwi7ml8c6yJegMXSOpLGtL/J+mUwevAdGAWMKF8o4h4r5WJg+VVVE0yNDOzjlFPAuGthZc9gW2ARyJil/ZsmFk1TiA0M2u9xUogjIgvllW2FvA/DWqbmZmZNVk9FxCWe4509bhZU/xz9mvs9cerO3Sft+13YIfuz8ysI7V4AaGkCySdnx8jgHuBqe3ftM4vh+o8Jmm2pBNz2amSQtIGhfWOy2UVh2eq1P1RbG896zQj2lfSpS3tI7fx4+3ZDjMzWzz1jAwUT85+AFwbEfe3U3u6miOBPSLimbLy6cDXgJ/n1/sBj3ZAezos2ldSj4j4Th2rHkK6MPD5VtY9v61tMzOz1qkndKhvRFyRH1dHxP2Svt/uLevkJF0MrAeMzr/8i7/i/wTsnddbD5hDCvIpbXuApOk54vesQvmhOQFwPLBdoby/pBtzLPAESR8tq6Y82rc1kcGSeku6PLdxWr7lD0nzJJ0u6SFgW0l3l0Y78rL/lTRJ0tjc5v2AoaTgoSmSeknaVdLkXPdlkpbL289Sij6+D9i/df8aZma2OOrpDHyzQtkhDW5HlxMRh5N+7e4MzC5bPBf4l6SBwAHA9aUFecj8LFKa32BgmKR98pfzaaROwGeB4vD7eaRf/cOArwCX1tnMScAman1k8E+AORGxRS6/K5evAMyIiE9GxH1l+1oBmBQRWwPjgZ9GxB9JI0sHRsRg0syXo4DhEbEFaWSqOHfvOxGxfUS0OI+BmZk1TtXTBJIOIP2qXFfS6MKiFYH/tHfDlgDXkU4VfA7YFTg0lw8D7i4N3Uu6mpTSR1n59cBGuXw3YLPCPforSVqxjjaUNihGBgP0IE06BQsig/9EGtEo7e9rpUoiotTZmQ/cWGVfH7Kg03MVcFOFdTYGnomIJ/LrK4CjSLkEFLZf9ECkw0i5B/Tqt2q11czMrA1qXTPwd9IXRj/gfwvlb5C+QKy2W4GzgYkRMbfwRV5rjodqoQ9LAdtGxNvFwgoBPuVK0b6tjQxWlba804pz+ZW2b6nBb1atrBBH3Hf99RxHbGbWQFVPE0TEsxFxd0RsGxHjC49JEVHvpDPdVv7i/hELhuNLHgJ2lNRPUg/SaYTxuXwnSavmYf3iefMxwNGlF5IGt7T/smjf1kYGl+9v5ToOeSkWxBV/HSidRniDNJoE8DgwoHCnxcH52M3MrInqubXwU/mitXmS3pM0X9LcjmhcVxcR10XEpLKyF4AfA+NIt2hOiohbcvmpwAPA30jn+0uOAYbmi/kepfp0wMeVbi0EDiJH+0bEe6Qv6rMkTQWmAJ8mnS64SmlSosnkyGDSXRAr5wscp5Kui2jJm8Dmkh4hXQ9xei4fBVysNKOhSKdLbsj7/BC4uI66zcysHdUTRzyRdP74BtKV4d8ANoiIk9q/edZVSJoXETUnImoUxxGbmbWeFieOGCAi/lm49/tySX9vaAvNzMysaerpDLwlaVlgiqT/IV1UuEL7Nsu6mo4aFQD45+w5fOmPt7a8YoOM3u+LLa9kZtaF1ZMzcHBe72jSeeG1SPe6WyejFHl8ZeH10pJekXRbG+ubJalfhfIvKccvt7HeefnvAElv5xCixyQ9LKlSroWZmbWjemYtfFZSL2CNiDitA9pkbfcmMFBSr3w3w2eBfzd6JxExGhjd4or1eSoitoKP0hpvkrRURFzeoPrNzKwF9dxN8EXS1ed35NeDy0KIrHP5Cyk7ANJti9eWFkjaRtLf8y/xv0vaOJf3kHROIX74e4X6vpcjhqdL2iSvX5wgaZTSJFZ/l/R0jiAu7e+EfCfKNEktdiQj4mngB6S7J8zMrIPUc5rgVGAb4HWAiJgCDGivBtliuw74mqSewCBSfkHJ48AO+Zf4KcAvc/lhwLosiCUuzg/8ao4Yvgg4vso+1wC2B/YCzgSQtDuwIemzMxgYImmHKtsXTQI2KS+UdJikiZImvjd3Th3VmJlZveq5gPCDiJhTR9qddQIRMU3SANKowO1li/sAV0jakJQQuEwu3w24uBQmFRGvFbYpxQo/Any5ym7/FBEfAo9KWj2X7Z4fk/Pr3qTOwT0tHELFD9rCCYQbOoHQzKyB6ukMzJD0daBH/hI5hhRVbJ3XaOAcYCegGOT/M2BcROybOwx35/Jq8cMA7+a/86n+eXm38FyFv2dExCWtaTgpQvmxVm5jZmaLoeppgsJV6U8Bm5P+h38taUa+Y9u9ZbY4LgNOj4jpZeV9WHBB4SGF8jHA4ZKWBpC0SgPacCfwLUm9c51rSlqt1ga5g3IOaYZFMzPrILVGBoZIWgcYToqjLU5WtDzwTns2zNouIp4jTXtc7n9Ipwl+wIJpiSFNibwRME3S+8BvgRGL2YYxkjYFHsinmOaRIpJfLlt1fUmTgZ6keQwu8J0EZmYdq2ocsaRjSHPNr8fCt6cJiIhYr/2bZ7YoxxGbmbVerTjiWrMWnh8RmwKXRcR6hce67giYmZktOeoJHTqiIxpiVq+nZs9j3xvva3nFBrn5K9t32L7MzJqhnpwBMzMzW4ItcZ0BSfMlTZE0Q9INkpZfjLrullTx/EoL2/WVdGTh9VI5pW9GTvKbIGndvOx2SX3b2say/V4qabNG1FWh7lbfTirpdEm7tUd7zMyscZa4zgDwdkQMjoiBwHvA4cWFknp0QBv6AkcWXg8HPg4MiogtgH1ZkOi4Z0S83oidRsR3IuLRRtRVoe5Pt2GbUyLib+3RHjMza5wlsTNQdC+wgaSdJI2TdA0wXVJPSZfnX+mTJe0MIKmXpOtylv71QK9SRaWZ9vLz/SSNys9Xl3SzpKn58WlSJO/6eYTibFJc7ws5pY+IeC4iZuftZ0nqJ+nwvP4USc9IGpeX7y7pgTw/wA2l+/YrKY1k5LkGRhVGIo5rYZtzJd2jNHPgMEk3SXpS0s/Lj1/SGnnd0ujLZ6rtL5ftVzjO07ToPAf9Jf01l18i6VlVninxozjid+e+XuOf3MzMWqueBMIuKQfo7EGeYImUkT8wIp6R9F8AEbFF/lIaI2kj0q2Ub0XEIEmDSDn5LTkfGJ9T/XqQYndPzPsanNvyCeA+SZ8BxgJXRcTkYiURcTFwsaRlSBkAv8pfiicDu0XEm5J+RJrI5/QW2jQYWDOPjlDHaYj3ImIHSd8HbgGGAK8BT0k6NyL+U1j368CdEfGLfLzLt2J/r0bE1vkUyvHAd4CfAndFxBmSPk+aJ2ERxTjildffxHHEZmYNtCSODPSSNAWYCPwf8Ltc/nBEPJOfbw9cCRARjwPPkkJ3dgCuyuXTgGl17G8X0iQ+RMT8iFhkFp0cArQx8GPgQ2CspF2r1Hce6cvxVuBTwGbA/fmYvgmsU0ebngbWk3RB/oKd28L6pVkopwMzI+KFiHg317NW2boTgEMlnQpsERFvtGJ/xXkOBuTn25MmVyIi7gBm13F8ZmbWQEviyMDbpV/kJUoJeG8Wi2psX+1XZ7G8Z2sblb9c/wL8RdJLwD6kUYJiOw8hfdkfXWjnXyPigFbua7akLYHPAUcBXwW+VWOT0twCH7LwPAMfUvYZiYh7lGYf/AJwpaSzI+L3de6v0jwHngHLzKzJlsSRgXrcAxwIkE8PrA38o6x8IGkK4JKXJG0qaSnSBYAlY0mnF8jnzlcixequWFpB0taSPp6fL5XrfbbYIElDSEPnB5WuLQAeBLaTtEFeZ/nc3pry6YWlIuJG4CfA1i2+I3VSiqh+OSJ+Sxp12Xox93cfqfNQmvZ45Ua11czM6rMkjgzU4zek8/PTgQ+AQyLiXUkXAZdLmgZMAR4ubHMicBvwL2AG6doAgO8DIyV9m/SL94iIeEDS/ZJmkEYDxgK/lbRc3uZhFs3+PxpYBRiXRzImRsR38mjBtYVtTwaeaOH41szHUers/bjFd6R+OwEnKM1hMA/4xmLu7zTS8Q0HxgMvkDpTVa2/cm8HAZmZNVDVuQnMOkLu5MyPiA8kbQtcVH6ap5znJjAzaz3VmJugu44MWOexNvCHPKrwHvDdljZ4+vV3GX7TPxvWgOu/vEHD6jIz64rcGeiCJN0MrFtW/KOIuLPGNhcC25UVn9fs6YIj4klgq2a2wcysu3NnoB1J+hjwa2AY6Ur6WcCxEdHSOf+aImLfXP+pwLyIOKeObY6q0c7VSRcDrgUsA8yKiD1b06acLfD1iPhNa7YzM7Pm6653E7Q7pasAbwbujoj1I2Iz4L+B1ZvYpmqdv9NJtzBumdt5Yhuq78vCEcxmZtZFuDPQfnYG3s/JggBExBRSEuHZheje4QBKkcm3ldaVNCLfSVCK8j1L0sP5schJbknrS7pD0iOS7i3E/Y6S9CuleOOzqrR1DeC5Qjun5W2vlLR3YR9XS/qSpM1zO6YoRTdvyKIRzEg6QWlSpmmSTstlAyQ9rjSp0oxc52757osnJW1TqYELxRHPea3FN9/MzOrnzkD7GUhK2iv3ZVJ875bAbsDZktaoo765EbEN6ZbEX1dYPhL4XkSU8gqKw/UbkSKN/6tK3RcCv1Oav+GkUiYCcClwKICkPsCngdtJkz+dl6/6H0rqSJwIPJUniTohZwZsSIqBHgwMyWFFABuQkhYHAZuQIo63z+3+70oNjIiRETE0IoYu12eVKodhZmZt4WsGOt72wLURMZ8UZDSedE1BS5HB1xb+nltcoDR50aeBG3JGAcByhVVuyPurKCLulLQe8HnSfA6TJQ2MiPGSLpS0GqkTc2O+BfAB4CSlORduiognC/st2T0/SnMw9CZ1Dv4PeCYipue2zwTGRkTk3IcBLbwPZmbWYO4MtJ+ZwH4VyqvF737AwiM15ZHHUeU5ebvXa9yf/2aV8gUVRrwGXANck09X7ADcSJrD4UDga+SI4Yi4RtJDpEjiOyV9hzQ/QZGAMyLikoUKpQEsGnlcjEP2Z9LMrIP5NEH7uQtYTtJH981LGkaaiGd4ji7uT/rSfZgUT7yZpOXykHz5REbDC38fKC6IiLnAM5L2z/tRniugLpJ2kbR8fr4isD7pFzzAKODYvJ+ZeZ31gKcj4nzSJEeDKItgBu4EvpVHLZC0Zh5hMDOzTsa/wtpJHvbeF/i1pBOBd8i3FpKGzKeSfuH/MCJeBJD0B9JMiU+yYHi9ZLn8a3wpoNLERQcCF0k6mXR74HV5H/UYAoyQVBqduDQiJuTjeEnSY8CfCusPBw7KkcQvAqdHxGvFCOZ83cCmwAP5FMI84CBSZPNiWa/vcg4KMjNrIMcRdwGSZgFDI+LVJux7edLUxltXmp65GRxHbGbWeo4jtjaRtBtwGfCrztIRAHj59fe58OaXGlLXUfs2LfbBzKzTcGegC4iIAY2oR9KhpFkWi+6vlk4YEX8jzR1gZmZLMF9A2E1IWpUFHYGPAf3z829KWuwIYUmrS7pN0lRJj0q6PZd/XNIfF7d+MzNrPx4Z6CYi4j+k8J+65jSQtHREfNCKXZQijc/L2w/K+32eyrdYmplZJ+GRgW6uGIMs6VRJIyWNAX4vqb+kG3Ok8ARJ5bMeFlWLNB6Q7zAgRxBPyY9XJP00ly8SW1yhnR/FEc+b6zhiM7NG8siAlRsCbB8Rb0u6Bjg3Iu6TtDYpO2DTKttdCFwv6Wjgb8DleVTgIxHxHQBJ6+S6RpXFFgsYLWmHiLinbNuRpMhl1t5gS98CY2bWQO4MWLnREfF2fr4bKQiptGwlSStGxBvlG1WLNC5fT1JP4Abg6Ih4VtL3qBxbfE/5tmZm1j7cGbByxejipYBtC52DmqpEGpdP1nQxaT6Dv+XXFWOLzcys47gzYLWMAY4GSlMSD87TMC9C0i7AgxHxVoVI49I6RwErRsSZheI7gZ9Jujoi5klakzT188vVGrVa32WcD2Bm1kDuDFgtxwAXSppG+qzcQ5q+uJKKkcZ5YqKS44H3JU3Jry+OiIurxBZX7QyYmVljOY7YuhzHEZuZtZ7jiG2JMmf2B/zl+sWfpmGP4f0a0Bozs67PnQFrldZGGpuZWefXtNAhSR+TdJ2kp0rxtZI2amD9p0o6vgH1VIzZbWUdfSUdubhtaWEfX8pTJbdH3afnSYuIiMsjYnDZY5GOgKShks5vj/aYmVljNWVkQOlKsZuBKyLia7lsMLA68EST2lQtfrdizG4r9QWOBBZ7DoBqImI0MLqd6j6lDdtMBHxi38ysC2jWyMDOpNvHLi4V5FvW7pN0tqQZkqZLGg4LR+bm1yMkHZKfz5J0lqSH82OD8p1JWl/SHZIekXSvpE1y+ShJv5I0DjirSlurxexeKWnvwj6uzr/ON8/tmJLjdTcEzgTWz2Wl2/QWieDN0b2P59jeGbnO3STdL+lJSdtUe0MlHSJpRH6+f95+qqSq4T15mz9JulXSM5KOlvQDSZMlPShplcL7tF9+fmYeIZkm6Zxq+9OiMceXSbpb0tOSjim04Sf5mP8q6dpqozkqxBHPnfufaodkZmZt0KxrBgayaBgNwJdJk+lsCfQDJtT6MiuYGxHbSPoG8Gtgr7LlI4HDI+JJSZ8k/ULfJS/bCNgtIuZXqbtazO6lwHHALZL6AJ8GvgmcC5wXEVdLWhboAZwIDIyIwQCqEsFLui9/A2B/4DBgAvB1YHvgS8B/A/vU8X6cAnwuIv4tqW8L6w4EtgJ6Av8EfhQRW0k6Fyi9n+R2rwLsC2wSEVGou579bULqBK4I/EPSRaR/56/k/S8NTKLy52KhOOIN1x/sW2DMzBqos01UtD1wbUTMj4iXgPHAsDq2u7bwd9viAkm9SV/UNyjd334J6dd+yQ01OgJExJ3AesBvSV9okyX1j4jxwAaSVgMOAG7MpxkeAP5b0o+Adaqk9+3OggjeSbneDfOyZyJiekR8CMwExka6/3M6MKCO9wLgflLu/3dJnZFaxkXEGxHxCjAHuDWXV9rfXOAd4FJJXwbeasX+/hwR70bEq6QMgdVJ/963RMTbOeL41irbmplZO2pWZ2AmKaSmnCqUAZSCbEp6li2PKs/J271edsFbcbKdN2lBRLwWEddExMGkX+s75EVXAgcChwKX53WvIf2Kfxu4UymZr1wpgrfUng0i4nd52buF9T4svP6QOkdyIuJw4GRgLWCKpFVrrF73/nJnZxvgRtIIxR2t2F9xP/Nz3dX+vc3MrAM16zTBXcAvJX03In4LIGkYMBsYLukKYBXSl+4JwDKkCXOWI3UEdgXuK9Q3nHRefjjpl/lHImJuPh++f0TckC9eHBQRU+tpqGrH7I4CHgZejIiZef31gKcj4vz8fBAwlTQ8XlIxgree9tTZ5vUj4iHgIUlfJH1JL/aJ9jzKsnxE3C7pQdJphWr7q8d9wCWSziB9Fr9AGoGpqc/KSzsjwMysgZrSGcjnm/cFfq10O9w7wCzgWNKsdVNJv/B/GBEvAkj6AzANeJIFM9yVLCfpIdIowAEVdnkgcJGkk0kdi+vyPupRMWY3H8dLkh4D/lRYfzhwkKT3gReB0yPitXwR4AzgLxFxgipH8FY9XdFKZ+cLFwWMpf5jbcmKpGskeua6j6uxvx1bqizHFY/O6z9LuvtgToPaamZmderyccSSZgFD87nojt738qRz61tHhL/E2kBS7zw6sjxp7oPDImJSrW02GzA4rj55TJv3udV3VmvztmZmXZVqxBF3tgsIuwylEJ7HgQvcEVgsI/OFnZNIF2HW7AiYmVnjdfk44ogY0Ih61MqY3Yj4G7B2I/ZdL0mjSBc8ble26KN25nV2JF353wt4kHSh3w/LtnkmIvZdjLbsTwpkepF018G2hVM6vwH+r2yq4ooi4uttbYOZmTVGl+8MNEpEXE6+I6CTG1fHPAAnRMQf88WSxwKnknIO3mtgO74NHBkR4yQdDpxDulZia9Itg5XuFqmLqqdBmplZO/BpgiaTtIKkP+f0vhmShks6RSmdcIakkflLvXy7IZLGK6Uq3ilpjfJ1IjmX9Ot9j7zdRTnJb6YWJB/uKunmQt2flXRTfn6AUhrkDEln5bJTSF/4FyslKo4kJSzuDIwAjgbWVuXUxy9Kekgp5fBvklbP5afmYx0D/L5x77CZmbXEnYHm+zzwfERsGREDSUP6IyJiWH7di7JERUnLABcA+0XEEOAy4Bc19lEKNgI4KV9AMgjYUWmuhbuATSX1z+scClwu6eOkmOZdSMmQwyTtExGnk678PzAiTsgBSUeQ8geeiIh7SB2E7+X2Hc+CeRnuAz4VEVuR7uoonr4YAuxd6dSBCnHEs99wHLGZWSO5M9B804HdlOZX+Ey+GHHn/Ot5OumLePOybTYmxQj/NV98dzLwiRr7KI4sfFXSJNLtmZsDm+WEwytJw/x9SSmOfyGlP94dEa/kYfurWRC4tJA8t8QM4DctpD5+ghTGNJ2UIVE8ttFVEhuJiJERMTQihq68Yq0MJTMzay1fM9BkEfGEpCHAnsAZeZj8KNLtkv+SdCqLJi4KmBkR21KfrYCxktYl/UofFhGz88WGpbovJ8UBv0OKaP6g0umJFnyYHx+lPlZY5wLgVxExWtJOpOsZSlpMgzQzs8bzyECT5aH4tyLiKtJFeFvnRa/mX9j7VdjsH0B/SdvmOpaRVD56gJJjSL/K7wBWIn3hzsnn6vcorZsnX3qeNMowKhc/RDqV0E9SD1Kg0/iWjiki5gLP5DsOSu3YMi/uA/w7P/9mS3WZmVn788hA821BSvD7kBRJfAQp9386KZVxQvkGEfGe0pTC5yvNmLg0aXbBmXmVsyX9BFiedGvhzvlOgqmSJuf1niZNMFR0NdA/Ih7N+3lB0o+BcaTRiNsj4pY6j6ta6uOppNMH/85tW7fO+j6yfL+lHRxkZtZAXT6B0BpH0ghgcmHSpE5p6NChMXHixGY3w8ysS6mVQOiRAQPSh4R0CuG/mt2Wlrz/4vu88D//bnnFCtb44ZoNbo2ZWdfnzoABkG8BNDOzbqjdLiCUdIykxyTNVpqZsBQsE5I2KKx3XC6rOHRRpe5D8pB2Xevk/f5b0hRJT0q6SdJmbT22Ott4aUv7yG38eIP2d7rSfAkNJ+n2fMtha7Y5XNI32qM9ZmbWWO05MnAksEdEPFNWPh34GvDz/Ho/4NF2bEfJuRFxDoCk4cBdkraIiFcavSNJPSLiO3Wsegjp3vznW1n3IlMdR8Qp9bewdSJizzZsc3F7tMXMzBqvXUYGJF0MrAeMzr/8i7/i/wTsnddbjzR//SuFbReJv83lh0p6QtJ4ChP1SOov6Ual+N4Jkson8VlERFwPjAG+nuuoGO2bRzcelTRN0nW5rLeky3Mbp0n6Si6fl3+dPwRsK+nu0mhHXva/kiZJGpvbvB8wFLg6j1j0UooFnpzrvkzScnn7WUoRxfcB+1d5z0flOpF0ZqHd59T4dxqlFE88TtLTknbM+31MKYOgtN6sfHvhItHJ1faXR2OOz8/vVgpVejj/G34mly8v6Q95u+uVgpbqHiEyM7PGaJeRgYg4XNLngZ0pi9Ilzab3L0kDSZ2C60nxt6V77s8ixdLOBsZI2od0v/tpuXwO6Va3ybm+80i/+u+TtDZwJ7BpHc2cBGyiBdG+e0fEK/kL7hfAt4ATgXUj4t3CMPlPgDkRsUVu88q5fAVgRukXuhbO61kBmBQR/6WU6//TiDha0tHA8RExUVJP0v39u+Ygot+TbjP8da7jnYjYvqWDkrQKsC+wSUREHcP7K5NSDr9ECh3aDvgOMEHS4JwsWFKKTv5C3lefVuxv6YjYRtKewE+B3UijR7MjYlD+PEypsi2SDgMOA1izry8CNDNrpGaFDl1HOlWwD3Bzobxa/O0nC+XvkToQJbsBI5Rib0cDK0lasY42lL6ta0X7TiP9cj8IKM2itxtwYamSiJidn84nZfNX8mGhzVeRJvkptzFpWuEn8usrWDj69/pFN6loLilF8FJJXwbeamH9W3Mc8XTgpYiYnucamEmamrioUnRyvfu7Kf99pFDv9qTPAhExg/R+V1SMI151BccRm5k1UrM6A7cCB5PmvJ9bKK8Vf1stEGEpYNuIGJwfa0bEG3W0YSvgMRZE+5a23yIids/rfIH0xT8EeETS0nn9Sm15p9K5/FYcS0vRv3VF9eZO1Dakjsk+pOTBWt7Nfz8sPC+9XmjkKHdUhpA6BWdIOqUV+yvVPb9Qb2vjjs3MrB00pTOQJ6P5EYvOtFct/vYhYCdJq+Zh/eJ58zGkKXMBkDS4pf3n8/y7A9dSJdpX0lLAWhExjjSzXl+gd4X9rUzLlmJBrPDXSTP3AbwBlEYxHgcGaMGdFgdTR/RvhWPrDfSJiNuBY0mzDTaEKkQnL+b+7gO+muvejJTGaGZmHaxpOQMRcV2Fsqrxt0oT9jwAvEA6398jb3YMcKGkaaTjuQc4vMIuj8vD/SuQruDfpXQngSpH+z4BXJXLRLou4XVJP8/7m0H6lXsaC4bAq3kT2Fwp2GcOMDyXjwIulvQ2aabAQ0lRvUuTYojbckX+isAt+RoEAce1oY5qKkUnL87+fgNckf/tJpNOE8xpaaNlPraMw4PMzBrIccQdQNK8iOjd7HZ0Nnn0Z5mIeEfS+sBYYKN8XUhVjiM2M2s9OY7YOqnlgXH51I+AI1rqCAC8/9JbvPTrR2qus/qxDlQ0M6uXOwMdoJGjApIupJCzkJ0XEZfX2OYkFs0nuCEiyq/ZWNy2HQuMjIiadzAU1nuDlLVgZmZN5M5AFxMRR7Vhm1+w6MWa7eFY0q2TLd3OWO96ZmbWAZp1a6F1cRXSCH8KfJw07D8ur3ORpImSZko6LZcdU2G93SU9oJTQeEO+Q8HMzDqIOwPWVqU0wi0jYiDpDozngZ0jYue8zkn5YpVBpFtGB0XE+cX1JPUjBT3tFhFbAxOBH3T0wZiZdWfuDFhbVUojLPdVSZNItw1uDlSaxfFTufz+nAD5TWCd8pUkHZZHGSa+9ubs8sVmZrYYfM2AtUmeP2EIsCcpjXBMcbmkdYHjgWERMVtp4qOeFaoS8NeIOKCF/Y0ERgJsudZmvh/WzKyBPDJgbVIpjZCFExVXIoUtzZG0OrBHYfPieg8C25WSF/NMhht1wCGYmVnmkQFrq0pphNsCf5H0Qr4eYDJpwqOngfsL244sW+8Q4FrlKZtJ1xA8gZmZdQgnEFqX4wRCM7PWq5VA6NMEZmZm3ZxPE1iX88HLc3l5xJia66x29O41l5uZ2QIeGbC6SBqVZ3dsaZ1nchDRE5J+L8nTC5qZdXLuDFijnRARWwIbk/IFxklatsltMjOzGtwZ6MYqRAoPl3SKpAn59UhJqrDdEEnjJT0i6U5Ja5SvE8m5wIvk2wqrxBPvKunmQt2flXRT+x21mZmVc2egeyuPFL4DGBERw/LrXsBexQ3ydMMXAPtFxBDgMmpPgjQJ2CQ/XySeGLgL2FRS/7zOocAiMzAWEwj/M69S2KGZmbWVOwPdW6VI4Z0lPSRpOrALKUa4aGNgIPDXHB98MvCJGvsojiwsEk8c6d7WK4GDJPUlZxWUVxIRIyNiaEQMXbV3n7Ycq5mZVeG7CbqxKpHCRwFDI+Jfkk5l0QhhATMjYts6d7MVMLaFeOLLgVuBd4AbIuKDxTkuMzNrHY8MdGNVIoUBXs3TCFe6e+AfQH9J2+Y6lpFUPnqAkmOANUinH6rGE0fE86SZDE8GRjXo8MzMrE4eGejeKkUK70M6fTALmFC+QUS8l28xPF9SH9Jn6Nek2GFyfT8BlifNO7BzRLwHTK0RTwxwNdA/Ih5t5AGamVnLHEdsnYKkEcDkiPhdS+s6jtjMrPVqxRG7M2BNJ+kR0imEz0bEu3Ws/wbpdEVn1A94tdmNqKCztgs6b9s6a7ug87bN7Wq9jmzbOhHRv9ICdwasy5E0sVrvttk6a9s6a7ug87ats7YLOm/b3K7W6yxt8wWEZmZm3Zw7A2ZmZt2cOwPWFY1sdgNq6Kxt66ztgs7bts7aLui8bXO7Wq9TtM3XDJiZmXVzHhkwMzPr5twZMDMz6+bcGbCmk/R5Sf+Q9E9JJ1ZYLknn5+XTJG3d0raSVpH0V0lP5r8rd1S7JK0laZykx/J0zd8vbHOqpH9LmpIfe3ZUu/KyWZKm531PLJQv9vu1OG2TtHHhPZkiaa6kY/OyjnjPNpH0gKR3JR1fz7Yd+J5VbFsn+JzVes+a/Tmr9p41+3N2YP7cT5P0d0lbtrRto96zFkWEH3407QH0AJ4C1gOWBaaSZjMsrrMnaSZDAZ8CHmppW+B/gBPz8xOBszqwXWsAW+fnKwJPFNp1KnB8M96vvGwW0K9CvYv1fjWibWX1vEgKSOmo92w1YBhpOu7j69m2A9+zam1r9uesYrs6yeesatua/Dn7NLByfr4HHfD/snofHhmwZtsG+GdEPB1pDoPrgL3L1tkb+H0kDwJ9Ja3RwrZ7A1fk51eQ5lzokHZFxAsRMQkgIt4AHgPWbOX+G96uFupd3PerkW3bFXgqIp5tQxva1K6IeDkiJpDm6Kh32w55z6q1rdmfsxrvWS1Nfc/KNONz9veImJ1fPsiC6d/b+3PWIncGrNnWBP5VeP0ci/4Prdo6tbZdPSJegPQ/TdIvhY5q10ckDSBN4/xQofjoPEx4WRuG/Ba3XQGMkfSIpMMK6yzu+9WItpV8Dbi2rKy937O2bNtR71mLmvQ5q6XZn7N6NPtz9m3SKFlL2zbiPWuROwPWbKpQVn6/a7V16tm2rRanXWlhmgb6RuDYiJibiy8C1gcGAy8A/9vB7douIrYmDVEeJWmHVu6/PduGpGWBLwE3FJZ3xHvWHtt2SP1N/JzV0uzPWe0Kmvw5k7QzqTPwo9Zu217cGbBmew5Yq/D6E8Dzda5Ta9uXSsPP+e/LHdguJC1D+h/01RFxU2mFiHgpIuZHxIfAb0nDgx3Wrogo/X0ZuLmw/8V9vxa7bdkewKSIeKlU0EHvWVu27aj3rKomf86q6gSfs5Y07XMmaRBwKbB3RPynjm0b8Z61yJ0Ba7YJwIaS1s299a8Bo8vWGQ18Q8mngDl5uKzWtqOBb+bn3wRu6ah2SRLwO+CxiPhVcYOy8+P7AjM6sF0rSFoxt2MFYPfC/hf3/VqsthWWH0DZ0G0HvWdt2baj3rOKOsHnrFq7OsPnrCVN+ZxJWhu4CTg4Ip6oc9tGvGcta4+rEv3wozUP0hXmT5Cupj0plx0OHJ6fC7gwL58ODK21bS5fFRgLPJn/rtJR7QK2Jw3xTQOm5MeeedmVed1ppP/I1+jAdq1Hukp5KjCz0e9XA/4tlwf+A/Qpq7Mj3rOPkX6dzQVez89Xau/P2OK0rRN8zqq1qzN8zmr9ezbzc3YpMLvw7zWx1raNfM9aejiO2MzMrJvzaQIzM7Nuzp0BMzOzbs6dATMzs27OnQEzM7Nuzp0BMzOzbs6dATMzs27OnQEz61QkHaM0Le/VrdxugKSvt1e7Cvu5XVLf9t5PYX99JR3ZUfuz7sk5A2bWqUh6HNgjIp5p5XY7kaag3auV2/WIiPmt2aajSOpBiqm9LSIGNrs9tuTyyICZdRqSLiYl2I2WdFKePW6CpMmS9s7rDJB0r6RJ+fHpvPmZwGckTZF0nKRDJI0o1H1b7jAgaZ6k0yU9BGwr6SBJD+dtL8lfwtXaOEtSv9yOxyVdKmmGpKsl7SbpfklPStomr3+qpCsl3ZXLv5vLJensvO10ScNz+U6Sxkm6hpSIdyawfm7b2ZJ6Sxqbj3162fvymKTfSpopaYykXnnZBpL+Jmlq3m79XH5Cfn+nSTqtUf+O1gW1R6yhH3744UdbH8AsoB/wS+CgXNaXFNW6AilOtmcu35Ac6QrsRPoFXarnEGBE4fVtwE75eQBfzc83BW4FlsmvfwN8o472DQA+ALYg/bB6BLiMFLm8N/CnvP6ppGjeXnm7fwEfB74C/BXoAawO/B+wRj6ON4F18/YDgBmF/S/NgmjdfsA/8z5L7Rmcl/2h8P49BOybn/fM7+HuwMi87VL5/dmh2f/+fjTnsXQLfQUzs2bZHfiSpOPz657A2qTZ3EZIGgzMBzZqQ93zSbP9AewKDAEmpLl/6EX9M8M9ExHTASTNBMZGREiaTvpyLrklIt4G3pY0jjQj3vbAtZFOUbwkaTwwjJSn/3BUP00i4JdK0wJ/SJr3fvVCe6bk548AA/KkQWtGxM0AEfFObu/upPd4cl6/N6lzdU+dx25LEHcGzKyzEvCViPjHQoXSqcBLwJakX7TvVNn+AxY+Fdqz8PydWHCdgIArIuLHbWjju4XnHxZef8jC/38tvzgrqDyHfcmbNZYdCPQHhkTE+5JmseDYiu2ZT+rYVNuPgDMi4pIa+7JuwtcMmFlndSfwvTxVL5K2yuV9gBcizTt/MGmYHeANYMXC9rOAwZKWkrQW1eenHwvsJ2m1vJ9VJK3T0COBvSX1lLQq6TTABNIv8OGSekjqD+wAPFxh2/Lj6gO8nDsCOwM12xoRc4HnJO0DIGk5ScuT3t9vSeqdy9csvQfW/bgzYGad1c+AZYBpkmbk15DO6X9T0oOkUwSlX9HTgA/yRXLHAfcDz5AuwjsHmFRpJxHxKHAyMEbSNNJ5/DUqrbsYHgb+DDwI/Cwingduzm2eCtwF/DAiXqzQvv8A9+cLDc8GrgaGSppIGiV4vI79Hwwck4/v78DHImIMcA3wQD6t8UcW7nRYN+JbC83M2lE+rTEvIs5pdlvMqvHIgJmZWTfnkQEzswpyBsFyZcUHl+4eMFuSuDNgZmbWzfk0gZmZWTfnzoCZmVk3586AmZlZN+fOgJmZWTf3/wEoVJ5WCxwFxwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_features(X_train.columns, ideal_model.feature_importances_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "9a408725",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0000000000000002"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(ideal_model.feature_importances_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "64f35191",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "216605"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.ProductSize.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "07d029a9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Medium            64342\n",
       "Large / Medium    51297\n",
       "Small             27057\n",
       "Mini              25721\n",
       "Large             21396\n",
       "Compact            6280\n",
       "Name: ProductSize, dtype: int64"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.ProductSize.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "e5c89026",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None or Unspecified    77111\n",
       "Yes                     3985\n",
       "Name: Turbocharged, dtype: int64"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Turbocharged.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "db8ec6e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None or Unspecified    85074\n",
       "Manual                  9678\n",
       "Hydraulic               7580\n",
       "Name: Thumb, dtype: int64"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Thumb.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1b4525f5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}