{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# Big Mart Sales Prediction\n",
    "By Patrick Kong\n",
    "\n",
    "[Website](pkong97.github.io) | [Github](https://github.com/pkong97) | [LinkedIn](https://www.linkedin.com/in/pkong97/)\n",
    "\n",
    "\n",
    "## <u>Contents:</u>\n",
    "\n",
    "### <a href=\"#sec 1\"><u>Section 1. Introduction</u></a>\n",
    "1. Background\n",
    "2. Questions\n",
    "\n",
    "### <a href=\"#sec 2\"><u>Section 2. Data</u></a>\n",
    "1. Data source\n",
    "2. Data cleaning\n",
    "3. Replace missing and impossible values\n",
    "\n",
    "### <a href=\"#sec 3\"><u>Section 3. Exploratory Data Analysis</u></a>\n",
    "1. Correlation matrix and heatmap\n",
    "2. Important variables\n",
    "\n",
    "### <a href=\"#sec 4\"><u>Section 4. Model Fitting</u></a>\n",
    "1. Change categorical variables into numerical variables\n",
    "2. Mean-based model\n",
    "3. Linear regression model\n",
    "4. Ridge regression model\n",
    "5. Decision tree model\n",
    "6. Random frest model\n",
    "\n",
    "### <a href=\"#sec 5\"><u>Section 5. Conclusion</u></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <div id=\"sec 1\"><u>Section 1. Introduction</u></div>\n",
    "\n",
    "**1. Background**\n",
    "\n",
    "The data scientists at BigMart have collected 2013 sales data for 1559 products across 10 stores in different cities. Also, certain attributes of each product and store have been defined. The aim is to build a predictive model and predict the sales of each product at a particular outlet.\n",
    "\n",
    "Using this model, BigMart will try to understand the properties of products and outlets which play a key role in increasing sales.\n",
    "\n",
    "**2. Questions**\n",
    "\n",
    "Based on the data, I would like to answer the following questions:\n",
    "* Location type: What impact does the tier (ie. size of the city) of a location have on an item's sales?\n",
    "* Outlet size and type: What impact does the size and type of the outlet have on an item's sales?\n",
    "* Product visibility: How does the placement of an item in store affect its sales?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <div id=\"sec 2\"><u>Section 2. Data</u></div>\n",
    "\n",
    "**1. Data source**\n",
    "\n",
    "The for this case study was given by [Analytics Vidhya](https://datahack.analyticsvidhya.com/contest/practice-problem-big-mart-sales-iii/#ProblemStatement).\n",
    "\n",
    "The train.csv dataset will be used to train a regression model which will be tested on the test.csv dataset.\n",
    "\n",
    "**Preview of the training dataset**\n",
    "\n",
    "The dataset contains items that BigMart sells with information about its qualities, sales, and where it is sold."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load appropriate libraries\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": false
   },
   "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>Item_Identifier</th>\n",
       "      <th>Item_Weight</th>\n",
       "      <th>Item_Fat_Content</th>\n",
       "      <th>Item_Visibility</th>\n",
       "      <th>Item_Type</th>\n",
       "      <th>Item_MRP</th>\n",
       "      <th>Outlet_Identifier</th>\n",
       "      <th>Outlet_Establishment_Year</th>\n",
       "      <th>Outlet_Size</th>\n",
       "      <th>Outlet_Location_Type</th>\n",
       "      <th>Outlet_Type</th>\n",
       "      <th>Item_Outlet_Sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FDA15</td>\n",
       "      <td>9.300</td>\n",
       "      <td>Low Fat</td>\n",
       "      <td>0.016047</td>\n",
       "      <td>Dairy</td>\n",
       "      <td>249.8092</td>\n",
       "      <td>OUT049</td>\n",
       "      <td>1999</td>\n",
       "      <td>Medium</td>\n",
       "      <td>Tier 1</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>3735.1380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DRC01</td>\n",
       "      <td>5.920</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.019278</td>\n",
       "      <td>Soft Drinks</td>\n",
       "      <td>48.2692</td>\n",
       "      <td>OUT018</td>\n",
       "      <td>2009</td>\n",
       "      <td>Medium</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Supermarket Type2</td>\n",
       "      <td>443.4228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>FDN15</td>\n",
       "      <td>17.500</td>\n",
       "      <td>Low Fat</td>\n",
       "      <td>0.016760</td>\n",
       "      <td>Meat</td>\n",
       "      <td>141.6180</td>\n",
       "      <td>OUT049</td>\n",
       "      <td>1999</td>\n",
       "      <td>Medium</td>\n",
       "      <td>Tier 1</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>2097.2700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>FDX07</td>\n",
       "      <td>19.200</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>Fruits and Vegetables</td>\n",
       "      <td>182.0950</td>\n",
       "      <td>OUT010</td>\n",
       "      <td>1998</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Grocery Store</td>\n",
       "      <td>732.3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NCD19</td>\n",
       "      <td>8.930</td>\n",
       "      <td>Low Fat</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>Household</td>\n",
       "      <td>53.8614</td>\n",
       "      <td>OUT013</td>\n",
       "      <td>1987</td>\n",
       "      <td>High</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>994.7052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>FDP36</td>\n",
       "      <td>10.395</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>Baking Goods</td>\n",
       "      <td>51.4008</td>\n",
       "      <td>OUT018</td>\n",
       "      <td>2009</td>\n",
       "      <td>Medium</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Supermarket Type2</td>\n",
       "      <td>556.6088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>FDO10</td>\n",
       "      <td>13.650</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.012741</td>\n",
       "      <td>Snack Foods</td>\n",
       "      <td>57.6588</td>\n",
       "      <td>OUT013</td>\n",
       "      <td>1987</td>\n",
       "      <td>High</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>343.5528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>FDP10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Low Fat</td>\n",
       "      <td>0.127470</td>\n",
       "      <td>Snack Foods</td>\n",
       "      <td>107.7622</td>\n",
       "      <td>OUT027</td>\n",
       "      <td>1985</td>\n",
       "      <td>Medium</td>\n",
       "      <td>Tier 3</td>\n",
       "      <td>Supermarket Type3</td>\n",
       "      <td>4022.7636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>FDH17</td>\n",
       "      <td>16.200</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.016687</td>\n",
       "      <td>Frozen Foods</td>\n",
       "      <td>96.9726</td>\n",
       "      <td>OUT045</td>\n",
       "      <td>2002</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tier 2</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>1076.5986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>FDU28</td>\n",
       "      <td>19.200</td>\n",
       "      <td>Regular</td>\n",
       "      <td>0.094450</td>\n",
       "      <td>Frozen Foods</td>\n",
       "      <td>187.8214</td>\n",
       "      <td>OUT017</td>\n",
       "      <td>2007</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tier 2</td>\n",
       "      <td>Supermarket Type1</td>\n",
       "      <td>4710.5350</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Item_Identifier  Item_Weight Item_Fat_Content  Item_Visibility  \\\n",
       "0           FDA15        9.300          Low Fat         0.016047   \n",
       "1           DRC01        5.920          Regular         0.019278   \n",
       "2           FDN15       17.500          Low Fat         0.016760   \n",
       "3           FDX07       19.200          Regular         0.000000   \n",
       "4           NCD19        8.930          Low Fat         0.000000   \n",
       "5           FDP36       10.395          Regular         0.000000   \n",
       "6           FDO10       13.650          Regular         0.012741   \n",
       "7           FDP10          NaN          Low Fat         0.127470   \n",
       "8           FDH17       16.200          Regular         0.016687   \n",
       "9           FDU28       19.200          Regular         0.094450   \n",
       "\n",
       "               Item_Type  Item_MRP Outlet_Identifier  \\\n",
       "0                  Dairy  249.8092            OUT049   \n",
       "1            Soft Drinks   48.2692            OUT018   \n",
       "2                   Meat  141.6180            OUT049   \n",
       "3  Fruits and Vegetables  182.0950            OUT010   \n",
       "4              Household   53.8614            OUT013   \n",
       "5           Baking Goods   51.4008            OUT018   \n",
       "6            Snack Foods   57.6588            OUT013   \n",
       "7            Snack Foods  107.7622            OUT027   \n",
       "8           Frozen Foods   96.9726            OUT045   \n",
       "9           Frozen Foods  187.8214            OUT017   \n",
       "\n",
       "   Outlet_Establishment_Year Outlet_Size Outlet_Location_Type  \\\n",
       "0                       1999      Medium               Tier 1   \n",
       "1                       2009      Medium               Tier 3   \n",
       "2                       1999      Medium               Tier 1   \n",
       "3                       1998         NaN               Tier 3   \n",
       "4                       1987        High               Tier 3   \n",
       "5                       2009      Medium               Tier 3   \n",
       "6                       1987        High               Tier 3   \n",
       "7                       1985      Medium               Tier 3   \n",
       "8                       2002         NaN               Tier 2   \n",
       "9                       2007         NaN               Tier 2   \n",
       "\n",
       "         Outlet_Type  Item_Outlet_Sales  \n",
       "0  Supermarket Type1          3735.1380  \n",
       "1  Supermarket Type2           443.4228  \n",
       "2  Supermarket Type1          2097.2700  \n",
       "3      Grocery Store           732.3800  \n",
       "4  Supermarket Type1           994.7052  \n",
       "5  Supermarket Type2           556.6088  \n",
       "6  Supermarket Type1           343.5528  \n",
       "7  Supermarket Type3          4022.7636  \n",
       "8  Supermarket Type1          1076.5986  \n",
       "9  Supermarket Type1          4710.5350  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.read_csv(\"train.csv\")\n",
    "train.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2. Data cleaning**\n",
    "\n",
    "There are many missing values in the columns \"Item_Weight\" and \"Outlet_Size\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item_Identifier                 0\n",
       "Item_Weight                  1463\n",
       "Item_Fat_Content                0\n",
       "Item_Visibility                 0\n",
       "Item_Type                       0\n",
       "Item_MRP                        0\n",
       "Outlet_Identifier               0\n",
       "Outlet_Establishment_Year       0\n",
       "Outlet_Size                  2410\n",
       "Outlet_Location_Type            0\n",
       "Outlet_Type                     0\n",
       "Item_Outlet_Sales               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.apply(lambda x: sum(x.isnull()), axis = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\"Item_Visibility\" has a minimum value of 0 which is impossible - an item cannot be invisible when sold in a store.\n",
    "\"Outlet_Establishment_Year\" ranges from 1985 to 2009 which can be converted to how old the store is instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Item_Weight</th>\n",
       "      <th>Item_Visibility</th>\n",
       "      <th>Item_MRP</th>\n",
       "      <th>Outlet_Establishment_Year</th>\n",
       "      <th>Item_Outlet_Sales</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>7060.000000</td>\n",
       "      <td>8523.000000</td>\n",
       "      <td>8523.000000</td>\n",
       "      <td>8523.000000</td>\n",
       "      <td>8523.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>12.857645</td>\n",
       "      <td>0.066132</td>\n",
       "      <td>140.992782</td>\n",
       "      <td>1997.831867</td>\n",
       "      <td>2181.288914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.643456</td>\n",
       "      <td>0.051598</td>\n",
       "      <td>62.275067</td>\n",
       "      <td>8.371760</td>\n",
       "      <td>1706.499616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.555000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.290000</td>\n",
       "      <td>1985.000000</td>\n",
       "      <td>33.290000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>8.773750</td>\n",
       "      <td>0.026989</td>\n",
       "      <td>93.826500</td>\n",
       "      <td>1987.000000</td>\n",
       "      <td>834.247400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>12.600000</td>\n",
       "      <td>0.053931</td>\n",
       "      <td>143.012800</td>\n",
       "      <td>1999.000000</td>\n",
       "      <td>1794.331000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16.850000</td>\n",
       "      <td>0.094585</td>\n",
       "      <td>185.643700</td>\n",
       "      <td>2004.000000</td>\n",
       "      <td>3101.296400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>21.350000</td>\n",
       "      <td>0.328391</td>\n",
       "      <td>266.888400</td>\n",
       "      <td>2009.000000</td>\n",
       "      <td>13086.964800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Item_Weight  Item_Visibility     Item_MRP  Outlet_Establishment_Year  \\\n",
       "count  7060.000000      8523.000000  8523.000000                8523.000000   \n",
       "mean     12.857645         0.066132   140.992782                1997.831867   \n",
       "std       4.643456         0.051598    62.275067                   8.371760   \n",
       "min       4.555000         0.000000    31.290000                1985.000000   \n",
       "25%       8.773750         0.026989    93.826500                1987.000000   \n",
       "50%      12.600000         0.053931   143.012800                1999.000000   \n",
       "75%      16.850000         0.094585   185.643700                2004.000000   \n",
       "max      21.350000         0.328391   266.888400                2009.000000   \n",
       "\n",
       "       Item_Outlet_Sales  \n",
       "count        8523.000000  \n",
       "mean         2181.288914  \n",
       "std          1706.499616  \n",
       "min            33.290000  \n",
       "25%           834.247400  \n",
       "50%          1794.331000  \n",
       "75%          3101.296400  \n",
       "max         13086.964800  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of 0 values after modification: 0\n"
     ]
    }
   ],
   "source": [
    "#Fix impossible minimum values of 0 in item visibility\n",
    "avg_vis = train.pivot_table(values='Item_Visibility', index='Item_Identifier', aggfunc=np.mean)\n",
    "\n",
    "#Replace 0 values with mean visibility of that product\n",
    "for i in range(0, len(train['Item_Visibility'])):\n",
    "    curr_item = train.loc[i, ('Item_Identifier')]\n",
    "    if train.loc[i, ('Item_Visibility')] == 0:\n",
    "        train.loc[i, ('Item_Visibility')] = avg_vis['Item_Visibility'][curr_item]\n",
    "\n",
    "print('Number of 0 values after modification: ' + str(sum(train['Item_Visibility'] ==0)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The \"Item_Fat_Content\" variable has \"Low Fat\" additionally labelled as \"LF\" and \"low fat\". \"Regular\" is also mislabelled as \"reg\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Low Fat    5089\n",
       "Regular    2889\n",
       "LF          316\n",
       "reg         117\n",
       "low fat     112\n",
       "Name: Item_Fat_Content, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['Item_Fat_Content'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**3. Replace missing and impossible values**\n",
    "\n",
    "We replace the missing values in the \"Item_Weights\" column with the mean of that item type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of missing values after modification: 0\n"
     ]
    }
   ],
   "source": [
    "# Replace missing values in Item_Weights\n",
    "mean_weights = train.pivot_table(values = 'Item_Weight', index = ['Item_Type'], aggfunc = np.mean)\n",
    "for i in range(0, len(train['Item_Type'])):\n",
    "    curr_item = train.loc[i, ('Item_Type')]\n",
    "    if train['Item_Weight'].isna()[i]:\n",
    "        train.loc[i, ('Item_Weight')] = mean_weights['Item_Weight'][curr_item]\n",
    "\n",
    "print('Number of missing values after modification: ' + str(sum(train['Item_Weight'].isnull())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We replace the missing values in the \"Outlet_Size\" column according to the most frequent outlet type.\n",
    "\n",
    "Based on the pivot table below, if the outlet type is:\n",
    "* Supermarket Type 2 or 3, the outlet size is 'Medium'\n",
    "* Supermarket Type 1, the outlet size is 'Small'\n",
    "* Grocerhy Store, the outlet size is 'Small'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Outlet_Type</th>\n",
       "      <th>Grocery Store</th>\n",
       "      <th>Supermarket Type1</th>\n",
       "      <th>Supermarket Type2</th>\n",
       "      <th>Supermarket Type3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Outlet_Size</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>High</th>\n",
       "      <td>NaN</td>\n",
       "      <td>932.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Medium</th>\n",
       "      <td>NaN</td>\n",
       "      <td>930.0</td>\n",
       "      <td>928.0</td>\n",
       "      <td>935.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Small</th>\n",
       "      <td>528.0</td>\n",
       "      <td>1860.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Outlet_Type  Grocery Store  Supermarket Type1  Supermarket Type2  \\\n",
       "Outlet_Size                                                        \n",
       "High                   NaN              932.0                NaN   \n",
       "Medium                 NaN              930.0              928.0   \n",
       "Small                528.0             1860.0                NaN   \n",
       "\n",
       "Outlet_Type  Supermarket Type3  \n",
       "Outlet_Size                     \n",
       "High                       NaN  \n",
       "Medium                   935.0  \n",
       "Small                      NaN  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.pivot_table(values='Outlet_Location_Type', index = 'Outlet_Size', columns=['Outlet_Type'], aggfunc='count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of missing values after modification: 0\n"
     ]
    }
   ],
   "source": [
    "# Replace missing values in 'Outlet_Size' according to mode of 'Outlet_Type'\n",
    "\n",
    "for i in range(0, len(train['Outlet_Type'])):\n",
    "    curr_type = train.loc[i, ('Outlet_Type')]\n",
    "    if train['Outlet_Size'].isna()[i]:\n",
    "        if curr_type == 'Supermarket Type(2|3)':\n",
    "            train.loc[i, ('Outlet_Size')] = 'Medium'\n",
    "        else:\n",
    "            train.loc[i, ('Outlet_Size')] = 'Small'\n",
    "\n",
    "print('Number of missing values after modification: ' + str(sum(train['Outlet_Size'].isnull())))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We replace the following:\n",
    "* 'LF' and 'low fat' with 'Low Fat'\n",
    "* 'reg' with 'Regular'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Low Fat    5517\n",
       "Regular    3006\n",
       "Name: Item_Fat_Content, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fix mislabellings in fat content\n",
    "train['Item_Fat_Content'] = train['Item_Fat_Content'].replace({'LF':'Low Fat', 'low fat':'Low Fat', 'reg':'Regular'})\n",
    "train['Item_Fat_Content'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We adjust the outlet establishment year to represent the age of the outlet in years."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Adjust outlet establishment year to age\n",
    "train['Outlet_Age'] = 2013 - train['Outlet_Establishment_Year']\n",
    "train['Outlet_Age'].describe()\n",
    "train.drop(['Outlet_Establishment_Year'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item_Identifier         0\n",
       "Item_Weight             0\n",
       "Item_Fat_Content        0\n",
       "Item_Visibility         0\n",
       "Item_Type               0\n",
       "Item_MRP                0\n",
       "Outlet_Identifier       0\n",
       "Outlet_Size             0\n",
       "Outlet_Location_Type    0\n",
       "Outlet_Type             0\n",
       "Item_Outlet_Sales       0\n",
       "Outlet_Age              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# No more missing values\n",
    "train.apply(lambda x: sum(x.isnull()), axis = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clean Test Data**\n",
    "\n",
    "We apply all the data-cleaning procedures on the training dataset to the test dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of missing values after modification: 0\n"
     ]
    }
   ],
   "source": [
    "test = pd.read_csv('test.csv')\n",
    "test['Item_Fat_Content'] = test['Item_Fat_Content'].replace({'LF':'Low Fat', 'low fat':'Low Fat', 'reg':'Regular'})\n",
    "\n",
    "avg_vis2 = test.pivot_table(values='Item_Visibility', index='Item_Identifier')\n",
    "\n",
    "#Replace 0 values with mean visibility of that product\n",
    "for i in range(0, len(test['Item_Visibility'])):\n",
    "    curr_item = test.loc[i, ('Item_Identifier')]\n",
    "    if test.loc[i, ('Item_Visibility')] == 0:\n",
    "        test.loc[i, ('Item_Visibility')] = avg_vis2['Item_Visibility'][curr_item]\n",
    "\n",
    "mean_weights = test.pivot_table(values = 'Item_Weight', index = ['Item_Type'], aggfunc = np.mean)\n",
    "for i in range(0, len(test['Item_Type'])):\n",
    "    curr_item = test.loc[i, ('Item_Type')]\n",
    "    if test['Item_Weight'].isna()[i]:\n",
    "        test.loc[i, ('Item_Weight')] = mean_weights['Item_Weight'][curr_item]\n",
    "        \n",
    "\n",
    "for i in range(0, len(test['Outlet_Type'])):\n",
    "    curr_type = test.loc[i, ('Outlet_Type')]\n",
    "    if test['Outlet_Size'].isna()[i]:\n",
    "        if curr_type == 'Supermarket Type(2|3)':\n",
    "            test.loc[i, ('Outlet_Size')] = 'Medium'\n",
    "        else:\n",
    "            test.loc[i, ('Outlet_Size')] = 'Small'\n",
    "\n",
    "print('Number of missing values after modification: ' + str(sum(test['Outlet_Size'].isnull())))\n",
    "\n",
    "#Adjust outlet establishment year to age\n",
    "test['Outlet_Age'] = 2013 - test['Outlet_Establishment_Year']\n",
    "test.drop(['Outlet_Establishment_Year'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <div id=\"sec 3\"><u>Section 3. Exploratory Data Analysis</u></div>\n",
    "\n",
    "**1. Correlation Matrix and Heatmap**\n",
    "\n",
    "The variables \"Item_MRP\", \"Outlet_Type\", and \"Outlet_Age\" have the strongest positive correlations with our predictor \"Item_Outlet_Sales\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>Item_Weight</th>\n",
       "      <th>Item_Visibility</th>\n",
       "      <th>Item_MRP</th>\n",
       "      <th>Item_Outlet_Sales</th>\n",
       "      <th>Outlet_Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Item_Weight</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.018053</td>\n",
       "      <td>0.025821</td>\n",
       "      <td>0.012088</td>\n",
       "      <td>0.008376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Item_Visibility</th>\n",
       "      <td>-0.018053</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.004525</td>\n",
       "      <td>-0.128449</td>\n",
       "      <td>0.075175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Item_MRP</th>\n",
       "      <td>0.025821</td>\n",
       "      <td>-0.004525</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.567574</td>\n",
       "      <td>-0.005020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Item_Outlet_Sales</th>\n",
       "      <td>0.012088</td>\n",
       "      <td>-0.128449</td>\n",
       "      <td>0.567574</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.049135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Outlet_Age</th>\n",
       "      <td>0.008376</td>\n",
       "      <td>0.075175</td>\n",
       "      <td>-0.005020</td>\n",
       "      <td>0.049135</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Item_Weight  Item_Visibility  Item_MRP  Item_Outlet_Sales  \\\n",
       "Item_Weight           1.000000        -0.018053  0.025821           0.012088   \n",
       "Item_Visibility      -0.018053         1.000000 -0.004525          -0.128449   \n",
       "Item_MRP              0.025821        -0.004525  1.000000           0.567574   \n",
       "Item_Outlet_Sales     0.012088        -0.128449  0.567574           1.000000   \n",
       "Outlet_Age            0.008376         0.075175 -0.005020           0.049135   \n",
       "\n",
       "                   Outlet_Age  \n",
       "Item_Weight          0.008376  \n",
       "Item_Visibility      0.075175  \n",
       "Item_MRP            -0.005020  \n",
       "Item_Outlet_Sales    0.049135  \n",
       "Outlet_Age           1.000000  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_corr = train.corr()\n",
    "train_corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Correlation Heatmap of Item Variables')]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "plt.figure(figsize = (15,10))\n",
    "sns.heatmap(train_corr, \n",
    "            xticklabels=train_corr.columns.values,\n",
    "            yticklabels=train_corr.columns.values).set(title='Correlation Heatmap of Item Variables')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2. Important variables**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Supermarket Type 3 has the highest average sales followed by supermarket type 1, supermarket type 2, and lastly, grocery stores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Item_Outlet_Sales</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Outlet_Type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Grocery Store</th>\n",
       "      <td>339.828500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supermarket Type1</th>\n",
       "      <td>2316.181148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supermarket Type2</th>\n",
       "      <td>1995.498739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supermarket Type3</th>\n",
       "      <td>3694.038558</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Item_Outlet_Sales\n",
       "Outlet_Type                         \n",
       "Grocery Store             339.828500\n",
       "Supermarket Type1        2316.181148\n",
       "Supermarket Type2        1995.498739\n",
       "Supermarket Type3        3694.038558"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outlet_type_sales = train.pivot_table(index=\"Outlet_Type\", values=\"Item_Outlet_Sales\", aggfunc=np.mean)\n",
    "outlet_type_sales"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\"Item_MRP\" has the strongest correlation with \"Item_Outlet_sales\" at 0.57. Here we see visualize the positive trend between \"Item_MRP\" and \"Item_Outlet_Sales\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x2587fc11a90>]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "train.plot(x='Item_MRP', y='Item_Outlet_Sales', \n",
    "           kind='scatter', \n",
    "           title='Item Sales vs. Item Max Retail Price', \n",
    "           xlabel='Item Max Retail Price', ylabel=\"Item Sales\", \n",
    "           grid=True,\n",
    "           figsize=(15,15))\n",
    "\n",
    "# Add trendline\n",
    "z = np.polyfit(train['Item_MRP'], train['Item_Outlet_Sales'], 1)\n",
    "p = np.poly1d(z)\n",
    "plt.plot(train['Item_MRP'], p(train['Item_MRP']), color=\"red\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* \"Outlet_Age\" has a weak correlation with \"Item_Outlet_Sales\" at 0.049. Sales do not seem to vary much depending on the outlet age. \n",
    "* There also seems to be an outlier in that outlets of age 15 have much lower average sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Average Item Sales of Outlets Based on Age'}, xlabel='Outlet Age (Years)', ylabel='Average Item Sales ($)'>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg_sales_age = train.pivot_table(index='Outlet_Age', values='Item_Outlet_Sales', aggfunc=np.mean)\n",
    "avg_sales_age.plot(\n",
    "           kind='bar', \n",
    "           title='Average Item Sales of Outlets Based on Age', \n",
    "           xlabel='Outlet Age (Years)', ylabel=\"Average Item Sales ($)\", \n",
    "           grid=True,\n",
    "           figsize=(7,7),\n",
    "           legend=False)\n",
    " "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that all the outlets of age 15 years are grocery stores. We know from above that grocery stores have the least amount of average sales. We will keep these values as is."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Grocery Store    555\n",
       "Name: Outlet_Type, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train.Outlet_Age == 15].Outlet_Type.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <div id=\"sec 4\"><u>Section 4. Model Fitting</u></div>\n",
    "\n",
    "**1. Change categorical variables into numerical variables**\n",
    "\n",
    "Scikit-learn only accepts numerical variables, therefore we convert all the categorical and nominal variables into numeric types."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Item_Identifier          object\n",
       "Item_Weight             float64\n",
       "Item_Fat_Content          int32\n",
       "Item_Visibility         float64\n",
       "Item_Type                 int32\n",
       "Item_MRP                float64\n",
       "Outlet_Identifier        object\n",
       "Outlet_Size               int32\n",
       "Outlet_Location_Type      int32\n",
       "Outlet_Type               int32\n",
       "Item_Outlet_Sales       float64\n",
       "Outlet_Age                int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Transform all categorical variables into numerics\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "var_mod = ['Item_Fat_Content','Item_Type', 'Outlet_Size',\n",
    "          'Outlet_Location_Type', 'Outlet_Type']\n",
    "le = LabelEncoder()\n",
    "for i in var_mod:\n",
    "    train[i] = le.fit_transform(train[i])\n",
    "    test[i] = le.fit_transform(test[i])\n",
    "train.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2. Mean-based model**\n",
    "\n",
    "First we fit a mean-based model to get a baseline prediction.\n",
    "\n",
    "Uploading this dataset to Analytics Vidhya's solution checker gives the following score:\n",
    "\n",
    "Public LB Score = 1773.83 (lower is better)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Baseline model - Mean based:\n",
    "mean_sales = train['Item_Outlet_Sales'].mean()\n",
    "\n",
    "model1 = test.copy()\n",
    "model1['Item_Outlet_Sales'] = mean_sales\n",
    "\n",
    "#Export baseline model\n",
    "model1.to_csv(\"mean-model.csv\", index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression, Ridge, Lasso\n",
    "from sklearn.model_selection import KFold #For K-fold cross validation\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "target = 'Item_Outlet_Sales'\n",
    "IDcol = ['Item_Identifier', 'Outlet_Identifier']\n",
    "\n",
    "def regression_model(model, train, test, predictors, target, IDcol, filename):\n",
    "    model.fit(train[predictors], train[target])\n",
    "    \n",
    "    #Predict training set\n",
    "    predictions = model.predict(train[predictors])\n",
    "    \n",
    "    #Cross validation\n",
    "    cv_score = cross_val_score(model, train[predictors],\n",
    "                                               train[target], cv = 20,\n",
    "                                               scoring = 'neg_mean_squared_error')\n",
    "    cv_score = np.sqrt(np.abs(cv_score))\n",
    "    print(\"CV Score = \" + str(np.mean(cv_score)))\n",
    "    \n",
    "    #Predict on test data:\n",
    "    test[target] = model.predict(test[predictors])\n",
    "    \n",
    "    #Export submission file:\n",
    "    IDcol.append(target)\n",
    "    submission = pd.DataFrame({x: test[x] for x in IDcol})\n",
    "    submission.to_csv(filename, index = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**3. Linear regression model**\n",
    "<p>Public LB Score = 1271.42 </p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CV Score = 1205.6279644382307\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Model Coefficients'}>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictors = [x for x in train.columns if x not in [target]+IDcol]\n",
    "lm1 =LinearRegression(normalize = True)\n",
    "regression_model(lm1, train, test, predictors, target, IDcol, 'lm1.csv')\n",
    "coef1 = pd.Series(lm1.coef_, predictors).sort_values()\n",
    "coef1.plot(kind='bar', title='Model Coefficients')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**4. Ridge Regression model**\n",
    "<p>Public LB Score = 1273.24</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CV Score = 1207.268600984918\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Model Coefficients'}>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAFoCAYAAAC4+ecUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8+yak3AAAACXBIWXMAAAsTAAALEwEAmpwYAAArKklEQVR4nO3deZgdVZ3/8feHAIIIBMaIbIGAAQQVhIiijguboCjiiILKqgQEVMbHJYz6Qx0FHUEFRSEKIzKDGBAkKsomiwooCUaWAEOIQBKjbAIRlfX7++OcW6l0bqc7pG+dSu7n9Tz95N5T1V3f9FLfOrsiAjMzM4CVSgdgZmbt4aRgZmYVJwUzM6s4KZiZWcVJwczMKk4KZmZWcVKwFZKkTSWFpJWHce7Bkn7dRFxdrr2PpDmS/ibp5ZK2lDRD0gJJH5Z0mqTPDOPr/FzSQU3EbCu2If9gzHpN0t3ABsAGEfFArfz3wHbAuIi4u0hwKY5Vgf8A3kuK837gl8DnRyCuE4GjI+KifK0zgCsjYrul+SIRsecyxkG+/sHAByLitSPx9Wz545qCtcUfgf07byS9FHhuuXAWcT7wNuA9wNrAtsB0YJcR+NqbALcu4b1Zo5wUrC3OBg6svT8I+H79BElrS/q+pPsl3SPp05JWysdGSTpR0gOSZgNv6fK5Z0iaL2mepC9IGjVUUJJ2BXYD9o6IGyLiqYh4JCJOjYgz8jkbSJoq6SFJsyQdVvv8lSRNknSXpAclTZG0rqTnSPobMAr4Qz7+S+CNwDdzc9IWkr4n6Qu1r7d3bl56NH/OHrn8KkkfqJ13qKTbJP1V0iWSNqkdC0lHSLpT0sOSTlXyYuA0YKd8/Yfz+W+WNDM3ac2T9LGhvm+2/HJSsLa4HlhL0ovzzXo/4H8GnPMN0pP6ZsDrSUnkkHzsMGAv4OXABOCdAz73e8BTwIvyObsDH2BouwK/i4g5SzjnXGAuqWnpncDxknbOxz4EvD3HuwHwV+DUiHg8Ip6Xz9k2IjaPiJ2BX5Gak54XEf9Xv4ikHUmJ8uPAaOB1wN0Dg5G0N6m56x3AmPw1fzDgtL2AVwAvA94FvCkibgOOAK7L1x+dzz0DODwi1gReQmo6sxWUk4K1Sae2sBtwGzCvc6CWKI6NiAW5Lf8k4IB8yruAr0fEnIh4CDih9rnrAW8GjomIxyLiPuBr+esN5V+A+YMdlLQx8BrgkxHxz4iYAXyXhbWeI4BPRcTciHgc+CzwzuF0gHfxfuDMiLgsIp6JiHkRcXuX844AToiI2yLiKeB4YLt6bQH4UkQ8HBH3AleS+m4G8ySwtaS1IuKvEXHjs4jdlhNOCtYmZ5Pa7Q9mQNMR8HxgFeCeWtk9wIb59QbAnAHHOjbJnzs/N5c8DJwOvGAYMT0IrL+E4xsAD0XEgkHi2gS4sHbd24CngfWGce2BNgbuGsZ5mwAn1675EKBaTAB/rr3+O/A8BvdvpKR6j6SrJe20VFHbcsVJwVojIu4hdTi/GbhgwOEHSE+s9afdsSysTcwn3TTrxzrmAI8Dz4+I0fljrYjYZhhhXQ7sKGmjQY7/CVhX0pqDxDUH2LN23dERsVpEzFvsKw1tDrD5MM87fMA1V4+Ia4fxuYstm5z7UvYmJdEfA1OWJmhbvjgpWNu8H9g5Ih6rF0bE06Sb0RclrZmbQj7Kwn6HKcCHJW0kaR1gUu1z5wOXAidJWit3/m4u6fVDBRMRlwOXkZ72d5C0cr7+EZIOzX0N1wInSFpN0svy/6ET12k55k0AJI3Jbf7PxhnAIZJ2yf+HDSVt1eW804BjJW2Tr7m2pH2HeY2/ABvlYbhIWlXSeyWtHRFPAo8CzzzL+G054KRgrRIRd0XEtEEOfwh4DJgN/Bo4BzgzH/sOcAnwB+BGFq9pHAisCswkdfaez5KbhereCVwM/BB4BLiF1Jl9eT6+P7ApqdZwIXBcTiYAJwNTgUslLSB1qL9ymNddRET8jtSx/rUcx9UsWnPqnHch8GXgXEmP5niHO4/hl6QhsX+W1JkzcgBwd/5aR5Dma9gKSt5kx8zMOlxTMDOzipOCmZlVRiQpSDpT0n2SbqmVrSvpsjxr8rLc+UeeOXlKnvl5k6Tta59zUD7/TnlxLzOzxo1UTeF7wB4DyiYBV0TEeOAKFo4G2RMYnz8mAt+GlESA40idcDsCx3USiZmZNWNEkkJEXEOaIFO3N3BWfn0Waap/p/z7kVwPjJa0PvAm4LKIeCgi/koaBjgw0ZiZWQ/1cuns9fL4cEizJzszODdk0Zmnc3PZYOWLkTSRVMtgjTXW2GGrrboN1TYzs8FMnz79gYgYM7C8kf0UIiIkjdjY14iYDEwGmDBhQkybNtiwdjMz60bSPd3Kezn66C+5WYj87325fB6LLkewUS4brNzMzBrSy6QwlbQmPvnfi2rlB+ZRSK8CHsnNTJcAu0taJ3cw757LzMysISPSfCTpB8AbgOdLmksaRfQlYIqk95NWjXxXPv1i0oJns0irMx4CEBEPSfpP4IZ83ufzEshmZtaQ5X6ZC/cpmJktPUnTI2LCwHLPaDYzs4qTgpmZVZwUzMys0sg8BTMzW7JNJ/1sRL7O3V96yzJ9vmsKZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVnFScHMzCpOCmZmVnFSMDOzipOCmZlVnBTMzKzipGBmZhUnBTMzqzgpmJlZxUnBzMwqTgpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWcVJwczMKj1PCpLulnSzpBmSpuWydSVdJunO/O86uVySTpE0S9JNkrbvdXxmZrZQUzWFN0bEdhExIb+fBFwREeOBK/J7gD2B8fljIvDthuIzMzPKNR/tDZyVX58FvL1W/v1IrgdGS1q/QHxmZn2piaQQwKWSpkuamMvWi4j5+fWfgfXy6w2BObXPnZvLzMysASs3cI3XRsQ8SS8ALpN0e/1gRISkWJovmJPLRICxY8eOXKRmZn2u5zWFiJiX/70PuBDYEfhLp1ko/3tfPn0esHHt0zfKZQO/5uSImBARE8aMGdPL8M3M+kpPk4KkNSSt2XkN7A7cAkwFDsqnHQRclF9PBQ7Mo5BeBTxSa2YyM7Me63Xz0XrAhZI61zonIn4h6QZgiqT3A/cA78rnXwy8GZgF/B04pMfxmZlZTU+TQkTMBrbtUv4gsEuX8gCO6mVMZmY2OM9oNjOzipOCmZlVnBTMzKzipGBmZhUnBTMzqzgpmJlZxUnBzMwqTgpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWcVJwczMKk4KZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVnFScHMzCpOCmZmVnFSMDOzipOCmZlVnBTMzKzipGBmZpWVSwdgZta0TSf9bES+zt1fesuIfJ02aV1NQdIeku6QNEvSpNLxmJn1k1YlBUmjgFOBPYGtgf0lbV02KjOz/tG25qMdgVkRMRtA0rnA3sDMolGZ2bMyUs00sGI21bSRIqJ0DBVJ7wT2iIgP5PcHAK+MiKMHnDcRmAgwduzYHe65554lft02/mK2MSZwW6tZv5A0PSImDCxvVfPRcEXE5IiYEBETxowZUzocM7MVRtuSwjxg49r7jXKZmZk1oG1J4QZgvKRxklYF9gOmFo7JzKxvtKqjOSKeknQ0cAkwCjgzIm4tHJaZWd9oVVIAiIiLgYtLx2Fm1o/a1nxkZmYFOSmYmVnFScHMzCpOCmZmVmldR7OV5ZnIZv3NNQUzM6s4KZiZWcVJwczMKk4KZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOreEZzIZ45bGZt5JqCmZlVnBTMzKzipGBmZhUnBTMzqzgpmJlZxUnBzMwqTgpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWaVnSUHSZyXNkzQjf7y5duxYSbMk3SHpTbXyPXLZLEmTehWbmZl11+uls78WESfWCyRtDewHbANsAFwuaYt8+FRgN2AucIOkqRExs8cxmplZVmI/hb2BcyPiceCPkmYBO+ZjsyJiNoCkc/O5TgpmZg3pdZ/C0ZJuknSmpHVy2YbAnNo5c3PZYOVmZtaQZUoKki6XdEuXj72BbwObA9sB84GTlj3c6roTJU2TNO3+++8fqS9rZtb3lqn5KCJ2Hc55kr4D/DS/nQdsXDu8US5jCeUDrzsZmAwwYcKEWIqQzcxsCXo5+mj92tt9gFvy66nAfpKeI2kcMB74HXADMF7SOEmrkjqjp/YqPjMzW1wvO5r/S9J2QAB3A4cDRMStkqaQOpCfAo6KiKcBJB0NXAKMAs6MiFt7GJ+ZmQ3Qs6QQEQcs4dgXgS92Kb8YuLhXMZmZ2ZJ5RrOZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVnFScHMzCpOCmZmVnFSMDOzipOCmZlVnBTMzKzipGBmZhUnBTMzqzgpmJlZxUnBzMwqTgpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWcVJwczMKk4KZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVllmZKCpH0l3SrpGUkTBhw7VtIsSXdIelOtfI9cNkvSpFr5OEm/zeU/lLTqssRmZmZLb1lrCrcA7wCuqRdK2hrYD9gG2AP4lqRRkkYBpwJ7AlsD++dzAb4MfC0iXgT8FXj/MsZmZmZLaZmSQkTcFhF3dDm0N3BuRDweEX8EZgE75o9ZETE7Ip4AzgX2liRgZ+D8/PlnAW9fltjMzGzp9apPYUNgTu393Fw2WPm/AA9HxFMDyruSNFHSNEnT7r///hEN3Mysn6081AmSLgde2OXQpyLiopEPaWgRMRmYDDBhwoQoEYOZ2YpoyKQQEbs+i687D9i49n6jXMYg5Q8CoyWtnGsL9fPNzKwhvWo+mgrsJ+k5ksYB44HfATcA4/NIo1VJndFTIyKAK4F35s8/CChSCzEz62fLOiR1H0lzgZ2An0m6BCAibgWmADOBXwBHRcTTuRZwNHAJcBswJZ8L8Engo5JmkfoYzliW2MzMbOkN2Xy0JBFxIXDhIMe+CHyxS/nFwMVdymeTRieZmVkhntFsZmYVJwUzM6s4KZiZWcVJwczMKk4KZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVnFScHMzCpOCmZmVnFSMDOzipOCmZlVnBTMzKzipGBmZhUnBTMzqzgpmJlZxUnBzMwqTgpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWcVJwczMKsuUFCTtK+lWSc9ImlAr31TSPyTNyB+n1Y7tIOlmSbMknSJJuXxdSZdJujP/u86yxGZmZktvWWsKtwDvAK7pcuyuiNgufxxRK/82cBgwPn/skcsnAVdExHjgivzezMwatExJISJui4g7hnu+pPWBtSLi+ogI4PvA2/PhvYGz8uuzauVmZtaQXvYpjJP0e0lXS/rXXLYhMLd2ztxcBrBeRMzPr/8MrNfD2MzMrIuVhzpB0uXAC7sc+lREXDTIp80HxkbEg5J2AH4saZvhBhURISmWENNEYCLA2LFjh/tlzcxsCEMmhYjYdWm/aEQ8DjyeX0+XdBewBTAP2Kh26ka5DOAvktaPiPm5mem+JXz9ycBkgAkTJgyaPMzMbOn0pPlI0hhJo/LrzUgdyrNz89Cjkl6VRx0dCHRqG1OBg/Lrg2rlZmbWkGUdkrqPpLnATsDPJF2SD70OuEnSDOB84IiIeCgfOxL4LjALuAv4eS7/ErCbpDuBXfN7MzNr0JDNR0sSERcCF3Yp/xHwo0E+Zxrwki7lDwK7LEs8Zma2bDyj2czMKk4KZmZWcVIwM7OKk4KZmVWcFMzMrOKkYGZmFScFMzOrOCmYmVnFScHMzCpOCmZmVnFSMDOzipOCmZlVnBTMzKzipGBmZpVlWjp7eXH3l95SOgQzs+WCawpmZlZxUjAzs4qTgpmZVZwUzMys4qRgZmYVJwUzM6s4KZiZWcVJwczMKk4KZmZWUUSUjmGZSLofuGcEvtTzgQdG4OuMtDbG5ZiGxzENXxvjWtFj2iQixgwsXO6TwkiRNC0iJpSOY6A2xuWYhscxDV8b4+rXmNx8ZGZmFScFMzOrOCksNLl0AINoY1yOaXgc0/C1Ma6+jMl9CmZmVnFNwczMKk4KZmZWcVIwM7NK3ycFSW+V1Lrvg6TVJW1ZOg5bepLGDafMbGlIem4T12ndzbCAdwN3SvovSVuVDgZSogJmAL/I77eTNLVwTM+V9BlJ38nvx0vayzF19aMuZec3HkVNm75XklaTdIykb0o6XFJr9oqX9OXhlDVJ0qslzQRuz++3lfStXl2v75NCRLwPeDlwF/A9SddJmihpzYJhfRbYEXgYICJmAKWfNP8beBzYKb+fB3yhXDhAy2KStJWkfwPWlvSO2sfBwGql4sra9L06C5gA3AzsCZxUKI5udutStmfjUSzqa8CbgAcBIuIPwOt6dbHWZOiSIuJRSecDqwPHAPsAH5d0SkR8o0BIT0bEI5IWCbNAHHWbR8S7Je0PEBF/14AAHRNbAnsBo4G31soXAIeVCKimTd+rrSPipQCSzgB+VyiOiqQPAkcCm0m6qXZoTeA3ZaJaKCLmDPhxPd2ra/V9UpC0N3Aw8CLg+8COEXFfbr+bCZRICrdKeg8wStJ44MPAtQXiqHtC0urk5CRpc9KTZ0mtiikiLgIukrRTRFxXKo5BtOl79WTnRUQ8Vf7ZAoBzgJ8DJwCTauULIuKhMiFV5kh6NRCSVgE+AtzWq4v1/eQ1SWcBZ0TENV2O7RIRVxSI6bnAp4DdAQGXAP8ZEf9sOpZaTLvnmLYGLgVeAxwcEVcVjGk34NNtiinHNYZUM9iU2oNXRBxaMKbWfK8kPQ081nlLqqH/Pb+OiFir6ZjqJI0C1mPRn929BeN5PnAysCupyf8S4CMR8WBPruekoC9HxCeHKitB0lqkP5IFpWMBkPQvwKtIf7zXR0TxZYVbGtO1wK+A6dSq+RHRrQO6MW38XrWNpKNJfXp/AZ7JxRERLysWVMOcFKQbI2L7AWU3lfwlkPQK4ExSeybAI8ChETG9YExXACdFxMW1sskRMbFgTNt3KX4EuCcinmo6ng5JMyJiu1LXH4ykdwCvJTUh/ToiLiwc0iIkjQaOiogvFoxhFvDKXj2FPxuSNiPVFF5F+tldB/x7RMzuxfX6dvSRpA9KuhnYStJNtY8/AjcN9fk9dgZwZERsGhGbAkeRRo+UNA74pKTjamWl15r/FnA9aZGw75D+WM4D7sjNXaX8VNKbC15/MXkI4xGkET+3AIdLOrVQLBtLmizpp5I+IGkNSScBdwIvKBFTzRzSg0WbnANMAdYHNiD9jv+gVxfr25qCpLWBdWhhx5Kk30fEyweULVajaTimG0nDZE8BNgbeB1xZOKYLgM9ExK35/dbA54FPABc0/bQuaQHpSU7AGqSO3CdpQVu5pNuBF0f+g1easHlrRLy4QCxXAleTkvge+WMG6en3z03HU5dHQ20J/IxaR3xEfLVgTIu1XEj6Q0Rs24vr9fPoo4iIuyUdNfCApHULJ4arJZ1OehoI0gS7qzrNJRFxY4GYlJtkjszj7n9NSqolbdFJCAARMVPSVhExu8SIlogoObdlKLOAsSzcunbjXFbCuhHx2fz6Ekn7Au+NiGeW8DlNuTd/rJo/2uDnkiYB57LwfnCxpHUBRvpe1c81hZ9GxF65uajzdNcREbFZodA6T1KDiYjYubFgMkmHR8Tptfc7kNp/S46o+SHwEOmPBdIfyxhSLebXEfGKQnG1rq9D0tXAK0hzAoJU65uW4yIi3tZgLH8A3sDCv7kr6+9L19QhjQCMiL+XjgMg36MGM+L3qr5NCm0maVRE9GxyytKQtFae3Ldut+Ml/4DzuPsjSZ2nkCYZnUqq9j83Iv5WKK7rge1J7fcALyW1468NfDAiLi0Q0+uXdDwirm4wlrtJI3u6VedKP5DtROrTe15EjJW0LXB4RBxZKqam9W1SGORprlKoiQYASbNJ6+ecGRE9m6QyzFhaW6MaSNLGwH4R8ZXCcbSqryPH8CHgfyLir01fe3ki6bfAO4GpnX49SbdExEsKxjSdlKjOiYiHe329fu5TWNJ6KwE03kRTsy2wH3BG7hA8Ezg3Ih5tOpCI2Cv/W3rtpa7yRLF9gf1JIzPaMMyyVX0d2XrADXnAwJnAJVHoibDND2T5+o0tKTFM7wYOAaZJmkYaiXhpr35+fVtTWF7kav85pPV0zifNbG6sg1DSJsDDEfFIfv9G4O3A3cCpEfFEU7HUYloTeAfwHmAL4ALg3RGxUdOxdDNIX8fzgQMo29ch0iz5Q0jDiaeQZvPf1XAcz5Ca0zqT5wbWPos9kCmtgfZV4JvAK0lLSkyIiP1KxdSRHxD3Ar5NSlT/DZzsjuYRImnniPhlntCzmIi4oEBMK0daC2YU8BbSH++mwNnA/wL/ChwfEVs0GNNvgX0i4k+StgMuJw3jfRlp4b4PNBVLLaZ/kDpMP026yYak2W1pyhqkr+NbwD8p2NeRY9uW9Hu1B6mD91XAZRHxiQZjOIbURPMIKXFeWPJ7UjdgSQmRlgT5cOnOb0kvI/3c3kxa5uJ/Sb9fB4x0c2Q/J4XPRcRxkrpNCosSo2o6cxFyn8KVpKe4awecc0pEfLjBmKox0pJOBJ6JiE/kp5YZA8dPNxTTMaTmtTVIw3Z/SLqxtSIptImkoyPim5I+AhxIejr/LvDjiHgy/xzvjIjNC8S2GennuDdpqOzxkZaJL0bSayLiN0OVNRTLpRGxe+5TeJjUr/CjiHi8ds4FEdH1wfZZX7dfk0IbdSatSXpei56cbo6FyxzfCBwbEZfk96WXA+ncVPYHxgPHkZ46/69QPFMi4l15pvxif1iFEmjnQeNzpIEL93Q558WlBjRI2ob0MzwA+ERETCkRRy2ebsveFJk4WvvZbRY9WtKi63X7PSkoLRJ2HLU1YYDPR4G1TyTNJbVndhUFZlVKOpk0vX4+8DZSJ+qTktYHfhIRpZe6AEDSS0jJ4d0R8aJCMawfEfNzP8xiut2QG4ip6Ez4bgbUEOaQmpB+FhH/KBjTTsCrSfupfK12aC1S82lPZg8PEdNs4GODHe9VE7eTgnQZcA3wP7novcAbImLXArHMJ3UidR2iEhGfazaiqnPy3aTEMCUi5uXylwMv6NQa2kjSdRGx09Bn9uTamwDjI+Ly3MewchRY7VbSU6RlqRc7RKGlN3JH803ARcCjDKhVFXr4eT1pAt0RwGm1QwtIDz93FojpQdL3aLD5HD1p4nZS6DIGud5k0nAsrXuqG66SN+DBqMsaUg1d9zBgImk5h82VNko6LSJ2KRBLke/Bkkj6LEvYSbDEw0+HpE1K1Oi6KXU/6Od5Ch2XStqPNDwP0qiIUk+/wxrELmmdFk5CKr0HcTelnniOIi0j8VuAiLhTUunVP1sjFq57tESSjo2IE3oczkDPkTSZxTdIKjFMtsiklr5NClp0RctjSMM+AUYBf2MJbXk9NNwnyStIyyi0SX9XORf1eEQ80ZkAJWllyn1/zhvOSYVuwEPZlzT8uUnnkZqPvkv5SWsHDOekka6l921SiBauaLkUY6FbsantcqDU9+lqSf8BrK60DeaRwE9KBBIRxw/z1BI34KGU+Pk9FRHfLnDdxUTELcM8dURr6f28yc5W+d/tu32Ujm8IbXwqb/wPWNKXhygb1pNWD0wC7ictiHc4cDFpol2btfFBo8Tv+U8kHSlpfUnrdj4KxLE0RvT71LcdzcpbSar7MtVFp9oPpUQHlIbYy1rSS5biyWakYmrVVqp5tMhvSTOYrwV+Gy1ZfnkobRzkUKKTXN2XqY42T4wc6Z9dPzcfTcz/vrF0LM9Ciae63YBPDijbs1PWZEKQ9EFSk8xmkupbp65JuiGXMo60bMSrgWOBHfJN5jfAb0pPzBpCG2sKw+oPGUnR0oUfhzCiP7u+rSl0KO369IuIWCDp06QO3P+MiN8XjmsUaWXL+giIe/OxxnaGq9+AgfrCaWuSbnTvayKOATG1divVOklrkNarOQYYFxGjykY0OEn/sRT9DyN1zXHAh1h8pE9jG/50iWkV4IPA63LRVcDpEfFkwZgaraU7KeTmBkmvBb4AfAX4fxHxyoIxfYg0y/ovpM1IIFVhSyyT0OobcP65jY+I/1ZazGzNiFjSTlW9jGUDUi3h1aRdzgCmA9cD15Uc/97SG/AfSOv53MzC3/NGN/zpEtN3gVWAs3LRAcDTUWDhx1pMjTaTOiksXG/oBODmiDin9IQfSbOAV5ZYamNJ2nQDzvEcR1oCesuI2CLflM+LiNcUiucZ4EbSMgnnRYFlxQfT0hvwb0s+fHUj6Q8Dl7ToVtZQLEVq6X3bp1AzT9LppDbzL0t6DuVHZc0h753bFvUbMGkd91VJS4MUuQFn+wAvJ92IibS8d8mhxq8BdspxfVRp28nr8se0qK1uWcA/I+KUgtfv5uT8e3UpaQtVoPgmO09L2jzyHhN5naZS8xXOAX5Ow7V01xSk55LWlr85zzxdH3hpFNhHtxbTGaSb789Y9I+l8TVhajHNIN+AY+E2haVXSf1dROyohatJrkFqpikWU52kTYG3kjZq2Sgiis36lvQe0kqyrbkB59r5AaSn4HozaclNdnYhPfTMJnXgbgIcEhHdRik2GVdjtfS+rSkob0hPmvhxVS5bl/QHM61gaAD35o9V80cbPBERISmg6kQtbUqu5Y3O6w0dCnynZEB5/kunX+E1pB3zrmfRRdZKeCnpBrwztRswZbed3RfYrE3NbBFxRV6rastcdEfhGl7jtfS+rSloOdiQXtLzcjDF91aQ9DHSk+ZupOrsoaSNxL9ROK7dSFtMirTv8GUFY3kA+BOpueg3wLXR4NapS5L7qbZu0w1Y0o+BiRFxXwtieR/pfnj2gPJOR/M5ZSJrvpbez0nhhRHx59JxdKO0N8DZQGcm5QPAgVHbDL6ENt2A20jS2pH3sh7ivMbXGWrTDbhD0lWkbV1vYNEmrcZHRCltO7vLwAewXCO+JiJ2aDqmWgyNNpP2bfMRMEPSLaTOnB8N54+5QZOBj3baMSW9gdQs8uqCMZGTQPFEoEUXM6w/1RTbI4B04eH+DpVYZ2g0cLuk4jfgmuMKXnugVbrVyCPisTx3oaRGm0n7OSlsSNqcez/gBEnXk/b7vSgK7gCVrVHv2IqIq0q14bfxBhwtXMxwKZWYPdymGzCQhsNq0c2InktapbiE1SWtERGP1QvzaLai/XoRcWKupT9K6lf4f72spfdt81GdpFVJSzbsB7wRuCIi3lswngtJwyw77ZvvA3aIiH1KxdRGks6OiAOGKmubUusMdbsBR4Hd4GrxtGkzoo+Rlq4/ojPJMI8eOxW4KiK+0nRMpZQej98KufNtJnAbKRu/uGxEHAqMAS7IH2NyWTGSzh5OWcO2qb9R2regWNvvUiixouxhwPnA6bloQ+DHTccxwFGkETSPQtqMCCiyGVFEnEja+vIaSQ8qLW54NfDTUglB0gJJj9b+fbT+vlfX7efmIyRtTKod7A+sQWo+eltE3F4yrki7qn24ZAxdtOYGLOlYoLNfQeePQ8ATpP6Ytmt8oTfauRtcmzYjIiJOA07rTIDsVouSdFBEnLXYJ/cmniLNpH2bFCRdS3pamgIcFhHTC4eEpK9HxDGSfkKXP45CozJadwPOI3dOkHRCRBxbIoYl0RDrDEXDC89lrboBZ1erJZsR1Q3RpPYRFq6L1Iimm0n7tk9B0uuAX8UQ34Amhw9K2iEipkt6fbfjhdepad0NOP8MFxMR1zQdS53auc7QfwEPAweSEtaRwMyI+FTBmFYC3s+iw5yLTj4cisrs8bBIH1RO6DdFxNY9uV6/JoXhKtEpKOkjEXHyUGUNx9S6G3CuUXWsRmoemV5ymQRo7UJvrbsBt/H3fChN3g/qtXSgs1lTVUvv1UOak8IQ2vBkUCqOAddv5Q24LvcRfT0i/q1wHG1cZ6h1N+A2/p4PpdD9oNFaet/2KSyFxrKmpP2B9wDjJE2tHVoTKLp3QUS8tf6+cwMuE82g5lJ+5Bi0c52hg4CBCeDgLmU91+bfc0njYsBCcwPKSuzs9/NuNfVe1dKdFIbW5PDBa4H5wPOBk2rlC4Cbun5GOcVvwJK+wcKkvRK1ZbQLa81Cby29Abf59/xHpN0X684nj7SLiKMbjwg+Xntd1dLp0UOGk8LQGhs+mCfN3ENak79VWnoDnsnCGbAPAz+IiJJ7NHfcQlpWog3rDLXuBtzG33Ol1W23AdaW9I7aobVIN+Jimq6l931SaOPwQUmvAr5BehJflXTje6zUmj5Za27AefTF8aQJfffm4rHAmXnxsGL76Wajack6Qy29AXeWTlnsEOXWrtoS2Iv0s6vfhBcAhxWIZ0l6Wkvv+6RAmtV5Bml89DNLPrUx3yRNqjuPtI76gcAWJQJp6Q34K6Tmj3GdMeWS1gJOzB8fKRBTXWvWGWrjDbiNa1dFxEXARZJ2iojrSsdT13Qtve9HH7V0+OC0iJig2prppUZlSPoa6Qb8711uwP+IiMZvwJLuBLYYOMdE0ijg9ogY33RMA7VtnSEbHkmrkYbubkOt2Sgiii0zo7RXc72W/sde1tJdU2jnPrF/z4v0zciTjuZTbp2qvRhwA46IR/Mv6u2UeSqPbpMOI+Jp5Z3hSqov9AZsTpo5fxppwTVrt7NJv9dvAj4PvJe0JlrjStXSvSBeGj54GPAlUkfcSaSn4JIOIP1sjgYeAzYGSo29H/QGTLllEmZKOnBgodLuWUXXrcpas9CbLbUXRcRnSH14ZwFvAUq1JHyF9GAxLiK2z3M6NiP1e/TsHuWaQouGD9Y8QNoT+Z/A53KzyHMKxTJT0oER8f16YeEb8FHABZIOJQ3Ng9T3sjrQhuXF27jOkA1P5+n7YaUdEP9MuYRepJbupNCu4YMdV5A2AOrsBLU6qXmrxM5rrbsBR8Q84JWSdmbh6q0XR8QVJeLpopULvdmwTJa0DvBpYCrwPOAzhWIp0kzqjuYW7RNbi2lGRGw3VFmTBtyAZ7boBtw6bVxnyJZM0vER8R/59W7Rgv3HlfbVvmCQWvq7enWPclJo54qkvwE+1OnslrQD8M2IaM1YcxtcG9cZsiWrr8NUYhHMQWLakLTJ1j/oUkvPNeaRv26/JwVo3/BBSa8AzgX+RHrSfCHw7mjBng82tOVxobd+18ak0NF0Lb3vk4JatE/sgLhWIc2yBLijBbN0bQi1dYZeC/yqdmhN4JnSv1M2OElzga+SHsL+Pb+uRMRXu33eisgdzS3cpjAnhA8CnZURr5J0uhND67VunSEbtu+QkvfA133HNYU8o7lTvc/DB2/szCQuFNN3gVVYuO3fAcDTEfGBUjGZWbM7MZbimkI7hw++IiK2rb3/pdIWj9ZibVxnyEbcvoCTwgpuEmn44M3A4aTx7qWHDz4tafOIuAtA0mbA04VjsiG0caE3G3FN7q9ShJNCGvp5MqkdEWjF8MGPA1dKmk36JdwEOKRgPGaWrPDt7e5TaOnwQUnPoTb6CNgrIn5UMCSzvteGe0Ov9W1NoaXbFFYi4nFqI1byEtZOCmZlNbYTYyl9mxRY/oYPrvBtmWaltXEnxqb1ffPR8kLSvRExtnQcZiuyPMrvDNLAk2onxpLL3jStb2sKbRw+KOlmBo9pvYbDMetH/4yIU0oHUZJrCi2S12AaVKRN2M2sRyS9BxhPu3ZibFTf1hTaaLg3fUnXecVUs554KWkFgZ1Z2HwU+X1fcFJYPq029Clm9iy0cSfGRnmP5uWT2/zMeqOzE2Pfck3BzGyh0cDtklqzE2PTnBSWT56zYNYbx5UOoDQ3H7WQpC8PUXZAg+GY9Y08H+FuYJX8+gagb0YegZNCW+3WpWzPzouIuKXBWMz6Rt6J8Xzg9Fy0IfDjYgEV4OajFpH0QdJ+DptJqi+1sSbwmzJRmfWV1u3E2DQnhXY5B/g5aROPSbXyBRFRfJE+sz7weEQ8IaVuu7wTY1+N9nPzUYtExCMRcXdE7A9sDOycJ7StlBfqMrPeGrgT43mU34mxUV7mooUkHQdMALaMiC0kbQCcFxGvKRya2QpN0kqknRh3J43yu6QFOzE2ykmhhSTNAF4O3NjZ0EPSTRHxsqKBma3guu262IKdGBvl5qN2eiJStg4ASWsUjsesXxzUpezgpoMoyR3N7TRF0unA6DxE7lBqe0ib2chq+06MTXLzUUvlTq56u+ZlhUMyW2HlZevH0WXkH3BTRDxVJLACnBTMzKzi5qMWqe0GJxYdG11sNzizftDGnRhLcU3BzMwqHn3UQpLOHk6ZmdlIc1Jop23qb/JU+x0KxWJmfcRJoUUkHZvbNl8m6dH8sQD4C3BR4fDMrA+4T6GFJJ0QEceWjsPM+o+TQgtJel238oi4pulYzKy/OCm0kKT6qoyrkdZ3nx4ROxcKycz6hOcptFBEvLX+XtLGwNfLRGNm/cQdzcuHucCLSwdhZis+1xRaSNI3WDi7ciXyMtrlIjKzfuGk0E4zgVH59cPADyLCezSbWc85KbRInqR2PGmp7Htz8VjgTEm/i4gniwVnZn3BfQrt8hVgXWBcRGwfEdsDmwGjgRNLBmZm/cFDUltE0p3AFjHghyJpFHB7RIwvE5mZ9QvXFNolBiaEXPg03Zf1NTMbUU4K7TJT0oEDCyW9D7i9QDxm1mfcfNQikjYELgD+AUzPxROA1YF9ImJeqdjMrD84KbSQpJ1ZuHz2zIi4omQ8ZtY/nBTMzKziPgUzM6s4KZiZWcVJwczMKk4KZmZWcVIwM7PK/wfv+FhIb4xRmgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "predictors = [x for x in train.columns if x not in [target]+IDcol]\n",
    "lm2 = Ridge(alpha=0.05,normalize=True)\n",
    "regression_model(lm2, train, test, predictors, target, IDcol, 'lm2.csv')\n",
    "coef2 = pd.Series(lm2.coef_, predictors).sort_values()\n",
    "coef2.plot(kind='bar', title='Model Coefficients')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**5. Decision tree model**\n",
    "<p>Public LB Score = 1167.33 (Rank 965)</p>\n",
    "\n",
    "We see that \"Item_MRP\", \"Outlet_Type\", and \"Outlet_Age\" are the most important features for the Decision tree model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CV Score = 1088.7693871484626\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Feature Importances'}>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAFoCAYAAABHdwCzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8+yak3AAAACXBIWXMAAAsTAAALEwEAmpwYAAApjElEQVR4nO3df/ylc53/8cfToKRBrdkSkxlCFIWJfuyWVVoi+rEVhWRLkWL7VktJ0m76qZ/aKNpWKyGrKVqs0Ca1huTHxJrs+JVqVJgkP5/fP97XmTnzmTPzOR/OOdd1rp732+1zc851XZ9zvZzPZ56f93lf7+v9lm0iImL8rVJ3ARERMRgJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQY6UkLZR0r6Q/dH09ZQCv+ZJB1djH+Y6S9PVRnW9lJO0n6Yd11xHtlECPfrzc9uO7vn5ZZzGSVq3z/I/UuNYd4yOBHo+IpLUlnSjpdkm3SfonSdOqfRtL+r6k30q6Q9K/S1qn2ncy8FTgO1Vr/72SdpB064TXX9KKr1rYZ0j6uqS7gf1Wdv4+arekgyTdIGmxpA9XNf9I0t2STpO0enXsDpJulfS+6v9loaQ3THgf/k3SIkk3STpC0irVvv0kXSLp05J+C3wT+BLwvOr//c7quF0l/bQ69y2Sjup6/VlVvW+UdHNVw/u79k+ravtF9f9yuaSZ1b6nSzpf0u8kXS/ptV3f9zJJ86vvuU3Su/v80UeDJdDjkfpX4EHgacDWwEuBN1f7BBwDPAXYHJgJHAVgex/gZpa2+j/e5/n2AM4A1gH+fZLz9+NvgW2B5wLvBU4A9q5qfSawV9exTwbWBdYH3gicIGmzat/ngbWBjYAXAfsCb+r63u2BG4EnVa//NuDS6v99neqYe6rvWwfYFThQ0ism1PtXwGbAi4EjJW1ebX9XVevLgLWA/YE/SloTOB84BfhLYE/gi5K2qL7vROCttqdX/7/fn/wti6ZLoEc/zpJ0Z/V1lqQnUQLkUNv32P4N8GlKaGB7ge3zbd9nexFwLCXsHo1LbZ9l+2FKcK3w/H36uO27bV8LXAOcZ/tG23cB36P8kej2ger/52LgbOC11SeCPYHDbS+2vRD4FLBP1/f90vbnbT9o+95ehdi+yPbVth+2fRXwDZZ/vz5k+17bPwN+Bjyr2v5m4Ajb17v4me3fArsBC21/tTr3T4FvAa+pvu8BYAtJa9n+ve0rpvDeRUOlTy/68Qrb/9V5Imk7YDXgdkmdzasAt1T7nwR8FvhrYHq17/ePsoZbuh5vuLLz9+nXXY/v7fH8yV3Pf2/7nq7nN1E+faxb1XHThH3rr6DuniRtD3yU0lJeHXgMcPqEw37V9fiPwOOrxzOBX/R42Q2B7TvdOpVVgZOrx68GjgA+Kukq4DDbl05WazRbWujxSNwC3Aesa3ud6mst28+o9n8EMLCl7bUoXQ3q+v6JU3zeAzyu86Rq+c6YcEz390x2/kF7QtWF0fFU4JfAHZSW7oYT9t22grp7PYfSLTIXmGl7bUo/u3oc18stwMYr2H5x1/uzTtXNcyCA7cts70HpjjkLOK3P80WDJdBjymzfDpwHfErSWpJWqS4qdroJpgN/AO6StD7wngkv8WtKn3PH/wKPrS4OrkZpOT7mUZx/GD4kaXVJf03pzjjd9kOUIPxnSdMlbUjp017ZEMlfAxt0LrpWpgO/s/2n6tPP66dQ11eAD0vaRMVWkv4C+C6wqaR9JK1WfT1H0ubV/8cbJK1t+wHgbuDhKZwzGiqBHo/UvpTugfmU7pQzgPWqfR8CtgHuovQ3nznhe48Bjqj65N9d9VsfRAmn2ygt9ltZuZWdf9B+VZ3jl5QLsm+zfV217x2Uem8EfkhpbZ+0ktf6PnAt8CtJd1TbDgKOlrQYOJKptZaPrY4/jxLMJwJr2F5MuVC8Z1X3r4CPsfQP5T7AwmrU0NuANxBjT1ngImLFJO0AfN32BjWXEjGptNAjIloigR4R0RLpcomIaIm00CMiWiKBHhHRErXdKbruuut61qxZdZ0+ImIsXX755XfYnnjjHVBjoM+aNYt58+bVdfqIiLEk6aYV7UuXS0RESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJRq8pOuuwswf2Wgs/uuvAXisioonSQo+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREn0FuqSdJV0vaYGkw3rs30/SIklXVl9vHnypERGxMpPOtihpGnAcsBNwK3CZpLm250849Ju2Dx5CjRER0Yd+WujbAQts32j7fuBUYI/hlhUREVPVT6CvD9zS9fzWattEr5Z0laQzJM0cSHUREdG3QV0U/Q4wy/ZWwPnA13odJOkASfMkzVu0aNGATh0REdBfoN8GdLe4N6i2LWH7t7bvq55+Bdi21wvZPsH2HNtzZsyY8UjqjYiIFegn0C8DNpE0W9LqwJ7A3O4DJK3X9XR34OeDKzEiIvox6SgX2w9KOhg4F5gGnGT7WklHA/NszwXeKWl34EHgd8B+Q6w5IiJ66GuRaNvnAOdM2HZk1+PDgcMHW1pERExF7hSNiGiJBHpEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREAj0ioiUS6BERLZFAj4hoiQR6RERLJNAjIloigR4R0RIJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREgn0iIiW6CvQJe0s6XpJCyQdtpLjXi3JkuYMrsSIiOjHpIEuaRpwHLALsAWwl6Qtehw3HTgE+Mmgi4yIiMn100LfDlhg+0bb9wOnAnv0OO7DwMeAPw2wvoiI6FM/gb4+cEvX81urbUtI2gaYafvsAdYWERFT8KgvikpaBTgW+H99HHuApHmS5i1atOjRnjoiIrr0E+i3ATO7nm9QbeuYDjwTuEjSQuC5wNxeF0Ztn2B7ju05M2bMeORVR0TEcvoJ9MuATSTNlrQ6sCcwt7PT9l2217U9y/Ys4MfA7rbnDaXiiIjoadJAt/0gcDBwLvBz4DTb10o6WtLuwy4wIiL6s2o/B9k+BzhnwrYjV3DsDo++rIiImKrcKRoR0RIJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREAj0ioiUS6BERLZFAj4hoiQR6RERLJNAjIlpi1boLGEezDjt7IK+z8KO7DuR1IiIgLfSIiNZIoEdEtEQCPSKiJRLoEREt0VegS9pZ0vWSFkg6rMf+t0m6WtKVkn4oaYvBlxoRESszaaBLmgYcB+wCbAHs1SOwT7G9pe1nAx8Hjh10oRERsXL9tNC3AxbYvtH2/cCpwB7dB9i+u+vpmoAHV2JERPSjn3Ho6wO3dD2/Fdh+4kGS3g68C1gd2HEg1UVERN8GdlHU9nG2Nwb+ETii1zGSDpA0T9K8RYsWDerUERFBf4F+GzCz6/kG1bYVORV4Ra8dtk+wPcf2nBkzZvRdZERETK6fQL8M2ETSbEmrA3sCc7sPkLRJ19NdgRsGV2JERPRj0j502w9KOhg4F5gGnGT7WklHA/NszwUOlvQS4AHg98Abh1l0REQsr6/JuWyfA5wzYduRXY8PGXBdERExRblTNCKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREAj0ioiUS6BERLZFAj4hoiQR6RERLJNAjIloigR4R0RIJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREn0FuqSdJV0vaYGkw3rsf5ek+ZKuknSBpA0HX2pERKzMpIEuaRpwHLALsAWwl6QtJhz2U2CO7a2AM4CPD7rQiIhYuX5a6NsBC2zfaPt+4FRgj+4DbF9o+4/V0x8DGwy2zIiImEw/gb4+cEvX81urbSvy98D3eu2QdICkeZLmLVq0qP8qIyJiUgO9KCppb2AO8Ile+22fYHuO7TkzZswY5KkjIv7srdrHMbcBM7ueb1BtW4aklwDvB15k+77BlBcREf3qp4V+GbCJpNmSVgf2BOZ2HyBpa+B4YHfbvxl8mRERMZlJA932g8DBwLnAz4HTbF8r6WhJu1eHfQJ4PHC6pCslzV3By0VExJD00+WC7XOAcyZsO7Lr8UsGXFdERExR7hSNiGiJBHpEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREAj0ioiUS6BERLZFAj4hoiQR6RERLJNAjIloigR4R0RIJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJVatu4AYjFmHnT2Q11n40V0H8joRMXppoUdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREv0FeiSdpZ0vaQFkg7rsf+Fkq6Q9KCkvxt8mRERMZlJA13SNOA4YBdgC2AvSVtMOOxmYD/glEEXGBER/elnHPp2wALbNwJIOhXYA5jfOcD2wmrfw0OoMSIi+tBPl8v6wC1dz2+ttk2ZpAMkzZM0b9GiRY/kJSIiYgVGelHU9gm259ieM2PGjFGeOiKi9foJ9NuAmV3PN6i2RUREg/QT6JcBm0iaLWl1YE9g7nDLioiIqZo00G0/CBwMnAv8HDjN9rWSjpa0O4Ck50i6FXgNcLyka4dZdERELK+v2RZtnwOcM2HbkV2PL6N0xURERE1yp2hEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREAj0ioiUS6BERLZFAj4hoiQR6RERLJNAjIloigR4R0RIJ9IiIlkigR0S0RAI9IqIlEugRES2RQI+IaIkEekRESyTQIyJaIoEeEdESCfSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJBHpEREsk0CMiWiKBHhHREgn0iIiWSKBHRLREX4EuaWdJ10taIOmwHvsfI+mb1f6fSJo18EojImKlJg10SdOA44BdgC2AvSRtMeGwvwd+b/tpwKeBjw260IiIWLlV+zhmO2CB7RsBJJ0K7AHM7zpmD+Co6vEZwBckybYHWGuMmVmHnT2w11r40V0H8jqpKdpMk2WupL8Ddrb95ur5PsD2tg/uOuaa6phbq+e/qI65Y8JrHQAcUD3dDLh+QP8f6wJ3THrUaKWm/qSm/jWxrtTUn0HWtKHtGb129NNCHxjbJwAnDPp1Jc2zPWfQr/topKb+pKb+NbGu1NSfUdXUz0XR24CZXc83qLb1PEbSqsDawG8HUWBERPSnn0C/DNhE0mxJqwN7AnMnHDMXeGP1+O+A76f/PCJitCbtcrH9oKSDgXOBacBJtq+VdDQwz/Zc4ETgZEkLgN9RQn+UBt6NMwCpqT+pqX9NrCs19WckNU16UTQiIsZD7hSNiGiJBHpEREsk0CMiWiKBPgSSHld3DU0n6eWSGvX7J2l2P9si+iVpDUmbjep8jfoH1Q9Jj5V0qKQvSHprNe69ESQ9X9J84Lrq+bMkfbHmmh4n6QOSvlw930TSbnXWVHkdcIOkj0t6et3FVL7VY9sZI6+iSxN/fpKWm6up17ZRauj79HLgSuA/q+fPljRxyPdAjV2gA18D5gBXUyYM+1S95Szj08DfUt1UZftnwAtrrQi+CtwHPK96fhvwT/WVU9jeG9ga+AXwr5IulXSApOmjrkXS0yW9Glhb0qu6vvYDHjvqeiZo4s9vpx7bdhl5Fctq4vt0FGUurDsBbF8JDPUTX2Nat1Owhe0tASSdCPxPzfUsw/Ytkro3PVRXLZWNbb9O0l4Atv+oCQXWxfbdks4A1gAOBV4JvEfS52x/foSlbAbsBqwDvLxr+2LgLSOso5fG/PwkHQgcBGwk6aquXdOBS+qoqUtj3qcuD9i+a0IZQx0nPo6B/kDnQXXTU521THSLpOcDlrQacAjw85prul/SGlS/SJI2prRkaiVpD2A/4GnAvwHb2f5Ndf1hPjCyQLf9beDbkp5n+9JRnbdPTfr5nQJ8DzgG6F4XYbHt39VT0hJNep86rpX0emCapE2AdwI/GuYJx+7GIkkPAfd0nlJad3+sHtv2WjXWti7wWeAllO6sc4FDbNc2r42knYAjKHPZnwe8ANjP9kV11VTV9TXgRNs/6LHvxbYvqKGmGZQW+Sy6Gju29x91LV01NfXnNw14Esu+TzfXWM9LgffToPepapy8H3gpJZ/OBT5s+09DO+e4BXpMnaS/AJ5L+aX68cRpjesg6WO2/3GybSOu6UfAfwOX09VVZrvXxdKRadrPr5oK5Cjg18DD1Wbb3qq2omje+9QhaS3K+7N46OdqS6BLWgd4u+1/rrGGjSgt9OdSPvpdCvxDZ3GQmmrapsfmu4CbbD846no6JF1he5sJ266qMxQkXWn72XWdf0UkvQr4K8rv1A9t/0fN9SygrHfQmBlVJV0AfMr2OV3bTrB9wEq+bdg1PQc4iXKNAcq/u/1tXz6sc47dKBdJMyWdIOm7kt4saU1JnwJuAP6y5vJOAU4D1gOeApwOfKPWiuCLwI8pkwN9mfJH5nTg+upj6khJOlDS1cDTJV3V9fV/wFWTff+QfVfSy2quYRnVsNe3UUZ1XQO8VdJx9VbFLZRwapLZwD9K+mDXtrrnRD8ROMj2LNuzgLdTRuMMzdi10CVdCFxMCaadq68rKS3hX9VYWs8WpqSf2X5WjTWdCXzA9rXV8y2Ao4H3AmeOukUqaW3gCTTowpqkxZTWr4A1KRfTHqAZ12WuAzbvTEetcjPWtbY3r7GmEykjg86m68Kj7WNrrOkKyhDBz1HWZtgbuHDip8AR1/RT21tP2LbcJ9NBGsdRLk+0fVT1+FxJrwHeYPvhlXzPqHxP0mHAqZSAeB1wjqQnAtQUWJt2wryqYb6kp9u+saYRQra9UNLbJ+6Q9MQ63iPbIx/7PgULgKcCN1XPZ1bb6nRz9bV69dUEqroQD6ruH/ghpeFQp4slHU/5lN7Jg4s63aC2rxj0Ccexhf4zYAdK6wngwu7ndQ6fqroNVsS2NxpZMRVJ36TMUX9qtel1wAxKC+aHtp8z4nq+a3u36r3qtIo7anmPOpp4vUHSxcBzKPdbmNIKnVfVhe3d66irqu1xtv9Y1/m7SXqr7eO7nm9LuaZW5wilC1ey27Z3HPg5xzDQF1KurPdqXtYaCE1Ujc09iHJRDcoNIMdRPio/zvYf6qqtaST9GNiG0l8NsCWl33pt4EDb59VQ04tWtt/2xaOqpUPS8yj9w4+3/VRJzwLeavugGmpZq7pB7Ym99tfcwJtme6Q3Fo5doDeZpMspv+in2L6z5nJ6kjQT2NP2J2o6/0r7D4fxMbRfTbveUNXwDuDrtn8/6nOviKSfUJaanNvpI5Z0je1n1lBLkz/x3UiZH+gk2yO5wXDs+tCbHAiU7ow3AfMkzaNc0T7PNf/VrG6YeQ2wF2X0TZ3D3lY2946BgX8MnYKmXW+AcvPOZdVFv5OAc+v+fYLmTHFhe7fqv02cFfNZlOU4T6wuZp8EnGr77mGdcOxa6JIepnwM7tw0MPEvcp2BACwZibAb8C+UX/SvAp8d5cc/lUmuXgW8HtgUOBN4ne0NRlXDuFnB9YZ1gX2o4XpDV12i3G34JspQvNMod9n+oqZ6zgCOBb4AbE+Z4mKO7VGvJYykDYE7bd9VPf8b4BXAQuA42/ePuqZeqq6zUyjzBZ1BuWN04Be3xzHQD6V83LuL8g/vP5rUDyxpK8o/vJdRbvX9d0r/9T6j/Mgu6V7KhbQjKGFkSTfWfY1B0o62v1/dLLMc22eOuqaOFVxv+CLwJ2q+3lD1U7+JMkz3QsrNa+fbfm8NtXRPcSHKrfbvrKO/uur+eaXtX0p6NvBflCGxW1Emx3pzDTWt6jLP1DRgV8rPbRZwMiUP/hr4iO1NB37ucQv0juquzD2BPShDuj7iMj1lHbWcZ/ulVR/6nZR+9G/Zvq/rmDNt9wyxIdV0KOX9WZMybOqblACoO9A/ZPuDknrdYOE6RyU0iaSDbX9B0iHAvpRPpF8BzrL9QPUp8AbbG9dQ2wtsXzLZthHVsuTeD0mfBB62/d7q/bly4n0hI6rpCtvbVH3oF1I+Tf1owjGfs/3OgZ97XAMdQNIzKKG1D/Be26fVVEfnB7iRa7zNv5euP3x7AZsAH6R8qvnfWgtrEEmn2X5tdQfrcv8gag6FD1Euqt3U45jNR3WxrVdtk20bUS1Xe+l02lcAh9s+t3pey1QSnRuKJD1+1J/qxi7QJ7TMb6F0u5xt+94aa7oRePeK9tfZjdBN0jMpwf4620+ruZa/oPxxWTJHCXC0a5gfRNJ6tm+v+mOX0ytMR1BTLQG5MtVwxedT5q7/dNeutSjdHiO/I1rSZylTbdwO7E65sP2ApPWA79ge+e3/km6lXGPoyUO8o3YcA/1hypwf3wbuZkKLaphv1kpq+m1Vz4rGxje2G0HSpbafN/mRAz/v+cAPgK9Xm94A7GD7JaOupVsV6pvY/q+qT31Vj2CWvB51PEiZFnq5XdQ0HUF1YW8HytwyX+ratZgSnjfUUJMoF6/XA06zfVu1fWvgLzut9RHXdDtlQETPoVG2PzS0c49hoB/FSlb9GOabtSJNbE31Sz3mmxjReZcbt9z98bkOkt4CHECZXmJjlUUJvmT7xTXUUsvPpR+SNqzjU8ujMcqGS515MHbj0L10HpeVknS47WOGXM6S043oPMNQ11/08yTtSRmCB2Xk0shbUxO8nXJr/U8AbN8gqe4ZPJvoMZJOYPmFQGofMrwSo1wbtq88kPQED/iGsbEL9Cl4DWX40ijs089BdXVvNImWndnwUMpQLoBpwB9YybWIEbjP9v2dG2YkrUp9f/BO7+egETdcOk6ndLl8hfrXzO3XKH+O/X6iu4Ay1cTAtDnQR9Zqtn1Nn4fWvYJ8LyP9dOFmz2x4saT3AWuoLP12EPCdOgqx/ZE+Dx1lw6XjQdv/MuJzjo0pjMcf+L+9sVvgYgqaeHFg5DVJ+tgk2/r6dDEokp5e/XebXl+jrKWHw4BFlMm53gqcQ7kxq8nq6O77jqSDJK0n6YmdrxrqmIomdosOPA/G7qJov5p4UamOiyUrGDNc21JvqpYFU++pRWuZuqEapfQTyp2hPwJ+4oZMCzuZmn6nek0T7TpvWtMka9RKeuYUPkmPxDB+dm3ucumrD3LERtZKkHQgpctgI0ndS7tNpwRXLVyt8Wj7b+qqoYfZlFvpnw8cDmxbhdYlwCV13bDWp5G3PN3MibB2AiYuML5LZ1vTwrwy8J/d2LbQJc0G3sHyV9rrnPC/Ma0ENXCpt24qK039p+3Fko6gXBz6sO2f1lwaktakzL9xKDDb9rR6K1oxSe+bQn/7oM65GnAg8MJq00XA8bYfGGUdVS1LGi5A92Rl0yl/jPcedU3dqvlcnsSyGXVztW/gK3SNc6D/jDJnytWUBS+Aeib876qpUd0bXTX8FeVmma+qTKw03fbKVlcaRU1X2d6qqu2fgE8AR9revoZankJpnT+fsjoQwOWUxbUvrXPMdUMbLl8BVgO+Vm3aB3jI9UyE1diGi8pc9h8Efs3SjPIw82CcA/0ndfzj76XJrQSVVdDnAJvZ3rQKr9Ntv6Cumqq6OvNdHANcbfuUGm9yehi4gnI7++luyJSr0NiGy3ILn/faNmpNa7hIWgBsP8rpLMa5D/2zVVidx7Irj9exwMUpwPdoYCsBeCWwNSWwcJlmtAlDB29TWUB3J+Bjkh5DfaOuXgA8j/JevUtlmcNLq6957po1swZ/sv25Gs/fy0OSNnY1H3s1v1Kt49G7Gy6U9QdWp0wrUWfD5RaqtV9HZZxb6MdQPur9gmU/ztR6t1oDWwn/Y3s7LZ29b01KN0Ld3UCPo8ztfXV1R+Z6wJauYd3OiSTNAl5OWbhhA9u13T8g6fWUWTKb0HDp1PRiSmjeSLmwtyHwJtsrWxR52DVdSdVw8dJl8Wrt7pR0IuUPzNks+7Mb2nxT49xCfw2wUcM+HjexlXBa1RJep5qrZH/gy3UVo2pRX8pNVhdV255I+YWfV2NdT2dpP/oLKCvL/JhlJ6Gqw5aUhsuOdDVcqHGpPtsXVPPcbFZtur7mTzEA99u2JMOSC9t1u7n6Wr36GrpxbqGfBRxg+zd119LRxFZCVcNOlCXMRFmT8vwaa2ncor6S7gB+SeliuQT4kYewPNgjUfXDbtGEhoukvSmZcfKE7Z2LoqfUUxlIejflk8xOlK7P/SmLtX++rpo6JD0ewCOYG32cA/0iyjJTl7Hsx5k6r/43snujSSQ92fav6q6jm6S1Xa1JOclxI583pUkNF5Xl3l48MZiq3/Mf2N62nsqW1NGYhktVzzMpcxV17qK9A9jXXQuRD9o4d7l8sO4CemhM94aWnQSr+692bfNpV66UdA3lQvK3+gnSYZtCDXXMm7IOcJ2kJjRcVuvVyrR9TzU2vVZVgNca4hOcALyrc21B0g6UPHj+sE44toFu+2ItuxjB4ygz9tVZ0yerVsLdlP7FI+tqJbi5k2CtT1lceE/gGEk/pqx5+m3XuOpUn+qYD6RJDZc1JK1p+57ujdWoqZH0EU/U4IYLwJrdF4ptXzTsvv1x7nJpzGIETSbpZNv7TLatDpJWp9yevSfwN8AFtt9Qb1UrVse8KdV5l2u4uJ5VlN5NmRr2bZ2braoRQccBF9n+xKhrajJJ/0EZLty55rA3sK3tVw7rnOM82+LbKaMR7oayGAFQy2IEkhZLurvrv3d3P6+jpi7P6H6iMsd3rX2dHdWFvvnAzyk/x83rrWhSI2+hVw2XM4Djq03rA2eNug4on0ApSy3+QNJvVSY1uxj4bt1hLunkfraN2P7ADODM6mtGtW1oxrbLhQYtRtDE7g1JhwOdub07f1QE3E/p26uNpJmUVvlewJqULpfdbV9XZ119qGPCt0atomT7S8CXOjen9fqkIOmNtr+23DcPV+MaLi6rEb1zlOcc50C/WA1ZjKCjSd0b1WiMYyQdY/vwUZ9/RST9iNLKPA14i+3Lay5pCU0yb4pHPAlWpTENl26TdPkcwtJ5XoaqiQ0XSZ+xfaik79DjZzXMC9rj3Ie+CvD3LDtMqbYbZqqaluljrf7xXWV7ixpremGv7bZ/MOpaYEk9/+1JfvFqGiLYxHlTPg7cCexL+WNzEDDf9vvrqmkyqmFOniY1XCRta/tySS/qtX+Yv0/jHOiH2P7sZNtGVMuSVgLQWRhhSSuhzl+0qpXQ8VjKx/fL654iYTJ1XIBUgyZ862hiw2UyNf3sGtVwgXoyapwDvddUtSNvGUw4f2NaCStS9V9/xvar665lZWpq5TVx3pTGNFz6VdPPrnENlzoyauz60CXtBbwemC1pbteu6UDdMxt+r1dLoc5WQg+30vzRJFBPP3Hj5k0B3ghMDO/9emwbGUmzPWHCuQnbRr4ilu2XT6hnJvCZUddRnbu2jBq7QKes+Xg7sC7wqa7ti4Gren7H6Lyn6/GSVgI1BoKkz7M0HFehayrdhqvjJp7GTPjW8IbLtygrTHU7g2pUie2DR17R8upsuNSWUWMX6NUNDTdR5q9ulCa1ErrMZ+kdtHcC37Bd25qiU1DHEMFrKLfa1z5vCg1suKjMSPkMYG1Jr+ratRalAVObJjVc6syosQv0rlt9l9tF/bf6TlRbK6EaYfMRyo0MN1ebnwqcpDKJ2MjXf+zW0CGC69CQeVMa2nDZDNiN8j51N14WA2+po6AujWu4SHou8HlKBqxOqe+eYWbU2AV6E2/i6WhSK4GyRud0yiLHi6HMRQ58svo6pKa6Os6iDBH8Dl1DBGvWmHlTmthwsf1t4NuSnmf70lGfv5eGN1y+QLmB7nTKOgn7ApsO84RjO8qliVTWFu1uJfxfXa0ESTcAm04c762yCvl1tjepo66uOho3RBCaM29Kk0l6LGUo5TPo6mqxPdTb2ldQy6cpDZd/6NFwudd2bQ0XSfNsz1HXmggZ5TIGGtpKcK+bd2w/pGpVl5p9Vs1ZExZYdsI3YGPKHa1fokxIFUudDFwH/C1wNPAGynw8ddiNCQ0X23dXjavrqPeT6B9VJqC7srpB7HaGPH/WOE/O1SSfoITAbNvbVGNPN6L0NX6ypprmS9p34kaVVWeaMGfKlpR+149SLvp9ivreq47GTPjWcE+z/QFKf/DXgF2Buj5trbDhQv1TJOxDydiDgXuAmcBQ7/9IC30wmthKeDtwpqT9KUMnofTjrUFZ3b5ujRki2KWR86Y0UOcT550qq/L8ivr+8M2XtK/tf+ve2JCGyx2UtU7/BHyo6u58zDBPmEAfjMZ1b9i+Ddhe0o4snYnuHNsX1FFPD00aItjRuAnfGuoESU8AjgDmAo8HPlBTLU1uuFxAWcyls8rTGpQuxqGtWJSLogOgsu7jmStoJby2jmFvTadmrgk7dvOmjJKkj9h+X/V4J9e8Zme3CQ2X+U1ouEi60vazJ9s20HMm0B89SetTJrC/lx6thKq1HF3qmIluMuM4b8oodc9NUscEXONG0iXAOzoX+iVtC3zB9tDuLUigD1ATWwlN1rQhgk2c8K1JEuhTI+k5wKnALymf+J4MvM5DXAMggR61UIPWhO2aN+WvgP/u2jUdeLiOmppI0q3AsZRw+ofq8RK2j+31fX/OJK1GucMW4PphD2HORdGoS5OWVmvcvCkN9WXKH7mJj6OHKswPBDozsF4k6fhhhnpa6FGLzp2inS6NaojgFZ076mJ8qYbVpppI0leA1Vi6HN8+wEO23zysc6aFHnVpzBDBJs6bMuZeA/zZBzrwHNvP6nr+fZVlDocmgR51OYwyRPBq4K2UMfK1DBFs8oRvY6qOueyb6CFJG9v+BYCkjYCHhnnCBHrU5R3VcMAlIZ4hgq2RftziPcCFkm6k/JHbEHjTME+YPvSoRYYItld+jktJegxdo1yA3Wx/a1jnSws9RqrhS6vFYNSx2lQj2b6PrpFS1XS/CfRojQwRHHMNXW1qXAz1+kK6XCJiSqqRGidSLmgvWW2qzmkbxoWkm20/dVivnxZ6jFSGCLbCn2x/ru4imkrS1az4d/xJQz13WugRMRWSXg9sQoNWm2qSao6iFXJZAHwo0kKPiKnaknLX444s7XJx9fzPXr+BLenSQc+8mECPiKlq4mpT4+ixkx8yNVlTNCKmqrPaVDw6A+/vTgs9IqZqHeA6SY1ZbSqKBHpETNUH6y6gJQY+Jj1dLhExJdV484XAatXjy4CMcJlA0scm2bbPoM+ZQI+IKalWmzoDOL7atD5wVm0FNddOPbbt0nlg+5pBnzBdLhExVU1abapxJB1Imd9/I0nd01lMBy4Z5rkT6BExVffZvl8qXcDValO5Q3GpU4DvURb5OKxr+2LbQ52ALl0uETFVE1ebOp2aVptqItt32V5oey9gJrBjdbPRKtXEZkOTW/8jYkokrUJZbeqllJEa59a12lSTSfogMAfYzPamkp4CnG77BUM7ZwI9Iqai18pSWW1qeZKuBLamLH6+dbXtqmEuhJ4ul4iYqjf22LbfqIsYA/e7tJgNIGnNYZ8wF0Ujoi9ZbWrKTpN0PLBONdRzf7rW0B2GdLlERF+qaWFn02P0BnCV7QdrKazBqovG3dcazh/q+RLoERHtkC6XiOhLVpvqT9f7JJZ9v4b+PqWFHhHREhnlEhExBJJO7mfbICXQIyKG4xndT6opErYd5gkT6BERAyTp8KoffStJd1dfi4FfA98e6rnThx4RMXiSjrF9+EjPmUCPiBg8SS/std32D4Z2zgR6RMTgSeqegfKxlDnkL7e947DOmXHoERFDYPvl3c8lzQQ+M8xz5qJoRMRo3ApsPswTpIUeETEEkj7P0jtFV6GaSneY50ygR0QMx3xgWvX4TuAbtrOmaETEuKhuIPoIZbrcm6vNTwVOkvQ/th8Y1rnThx4RMVifAJ4IzLa9je1tgI2AdYBPDvPEGbYYETFAkm4ANvWEcJU0DbjO9ibDOnda6BERg+WJYV5tfIje0w8PTAI9ImKw5kvad+JGSXsD1w3zxOlyiYgYIEnrA2cC9wKXV5vnAGsAr7R929DOnUCPiBg8STuydArd+bYvGPo5E+gREe2QPvSIiJZIoEdEtEQCPSKiJRLoEREtkUCPiGiJ/w8LR84lDXkmIAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "lm3 = DecisionTreeRegressor(max_depth=15, min_samples_leaf = 100)\n",
    "regression_model(lm3, train, test, predictors, target, IDcol, 'lm3.csv')\n",
    "coef3 = pd.Series(lm3.feature_importances_, predictors).sort_values(ascending=False)\n",
    "coef3.plot(kind = 'bar', title = 'Feature Importances')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**6. Random forest model**\n",
    "<p>Public LB Score = 1154.16 (Rank 657 out of 2149)</p>\n",
    "\n",
    "We see that \"Item_MRP\", \"Outlet_Type\", and \"Outlet_Age\" are the most important features for the Random Forest Model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CV Score = 1082.8669304067994\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Feature Importances'}>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "lm4 = RandomForestRegressor(max_depth = 15, min_samples_leaf = 100)\n",
    "regression_model(lm4, train, test, predictors, target, IDcol, 'lm4.csv')\n",
    "coef4 = pd.Series(lm4.feature_importances_, predictors).sort_values(ascending = False)\n",
    "coef4.plot(kind = 'bar', title = 'Feature Importances')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### <div id=\"sec 5\"><u>Section 5. Conclusion</u></div>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The random forest model produced the best prediction score.\n",
    "\n",
    "The overall consensus from model fitting and exploratory data analysis is that:\n",
    "* item price, outlet type, and outlet age in descending order are the best predictors of an item's sales\n",
    "* item visibility has almost no effect on the item's sales"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.1 64-bit",
   "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": "3.9.1"
  },
  "vscode": {
   "interpreter": {
    "hash": "c0480a40de60fe0b4be044949303dbe98ce3610184aeb132e77bccbcbbb8df2a"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}