notebooks/02_EDA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook does EDA on the augmented Rossmann dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Navigate to the project's root directory:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/opt/rossmann-tf\n"
     ]
    }
   ],
   "source": [
    "%cd .."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import KBinsDiscretizer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = pd.read_parquet('./data/train_clean.parquet')\n",
    "test_df = pd.read_parquet('./data/test_clean.parquet')"
   ]
  },
  {
   "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>index</th>\n",
       "      <th>Store</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>Date</th>\n",
       "      <th>Sales</th>\n",
       "      <th>Customers</th>\n",
       "      <th>Open</th>\n",
       "      <th>Promo</th>\n",
       "      <th>StateHoliday</th>\n",
       "      <th>SchoolHoliday</th>\n",
       "      <th>...</th>\n",
       "      <th>AfterStateHoliday</th>\n",
       "      <th>BeforeStateHoliday</th>\n",
       "      <th>AfterPromo</th>\n",
       "      <th>BeforePromo</th>\n",
       "      <th>SchoolHoliday_bw</th>\n",
       "      <th>StateHoliday_bw</th>\n",
       "      <th>Promo_bw</th>\n",
       "      <th>SchoolHoliday_fw</th>\n",
       "      <th>StateHoliday_fw</th>\n",
       "      <th>Promo_fw</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>2015-07-31</td>\n",
       "      <td>5263</td>\n",
       "      <td>555</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2015-07-31</td>\n",
       "      <td>6064</td>\n",
       "      <td>625</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>67</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>2015-07-31</td>\n",
       "      <td>8314</td>\n",
       "      <td>821</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>2015-07-31</td>\n",
       "      <td>13995</td>\n",
       "      <td>1498</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>67</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2015-07-31</td>\n",
       "      <td>4822</td>\n",
       "      <td>559</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>57</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 93 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   index  Store  DayOfWeek       Date  Sales  Customers  Open  Promo  \\\n",
       "0      0      1          5 2015-07-31   5263        555     1      1   \n",
       "1      1      2          5 2015-07-31   6064        625     1      1   \n",
       "2      2      3          5 2015-07-31   8314        821     1      1   \n",
       "3      3      4          5 2015-07-31  13995       1498     1      1   \n",
       "4      4      5          5 2015-07-31   4822        559     1      1   \n",
       "\n",
       "   StateHoliday  SchoolHoliday  ...  AfterStateHoliday  BeforeStateHoliday  \\\n",
       "0         False              1  ...                 57                   0   \n",
       "1         False              1  ...                 67                   0   \n",
       "2         False              1  ...                 57                   0   \n",
       "3         False              1  ...                 67                   0   \n",
       "4         False              1  ...                 57                   0   \n",
       "\n",
       "   AfterPromo  BeforePromo  SchoolHoliday_bw  StateHoliday_bw  Promo_bw  \\\n",
       "0           0            0               5.0              0.0       5.0   \n",
       "1           0            0               5.0              0.0       5.0   \n",
       "2           0            0               5.0              0.0       5.0   \n",
       "3           0            0               5.0              0.0       5.0   \n",
       "4           0            0               5.0              0.0       5.0   \n",
       "\n",
       "   SchoolHoliday_fw  StateHoliday_fw  Promo_fw  \n",
       "0               7.0              0.0       5.0  \n",
       "1               1.0              0.0       1.0  \n",
       "2               5.0              0.0       5.0  \n",
       "3               1.0              0.0       1.0  \n",
       "4               1.0              0.0       1.0  \n",
       "\n",
       "[5 rows x 93 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>index</th>\n",
       "      <th>Id</th>\n",
       "      <th>Store</th>\n",
       "      <th>DayOfWeek</th>\n",
       "      <th>Date</th>\n",
       "      <th>Open</th>\n",
       "      <th>Promo</th>\n",
       "      <th>StateHoliday</th>\n",
       "      <th>SchoolHoliday</th>\n",
       "      <th>Year</th>\n",
       "      <th>...</th>\n",
       "      <th>AfterStateHoliday</th>\n",
       "      <th>BeforeStateHoliday</th>\n",
       "      <th>AfterPromo</th>\n",
       "      <th>BeforePromo</th>\n",
       "      <th>SchoolHoliday_bw</th>\n",
       "      <th>StateHoliday_bw</th>\n",
       "      <th>Promo_bw</th>\n",
       "      <th>SchoolHoliday_fw</th>\n",
       "      <th>StateHoliday_fw</th>\n",
       "      <th>Promo_fw</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>...</td>\n",
       "      <td>105</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>...</td>\n",
       "      <td>105</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>...</td>\n",
       "      <td>115</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>...</td>\n",
       "      <td>115</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>2015-09-17</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>...</td>\n",
       "      <td>105</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 92 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   index  Id  Store  DayOfWeek       Date  Open  Promo  StateHoliday  \\\n",
       "0      0   1      1          4 2015-09-17   1.0      1         False   \n",
       "1      1   2      3          4 2015-09-17   1.0      1         False   \n",
       "2      2   3      7          4 2015-09-17   1.0      1         False   \n",
       "3      3   4      8          4 2015-09-17   1.0      1         False   \n",
       "4      4   5      9          4 2015-09-17   1.0      1         False   \n",
       "\n",
       "   SchoolHoliday  Year  ...  AfterStateHoliday  BeforeStateHoliday  \\\n",
       "0              0  2015  ...                105                   0   \n",
       "1              0  2015  ...                105                   0   \n",
       "2              0  2015  ...                115                   0   \n",
       "3              0  2015  ...                115                   0   \n",
       "4              0  2015  ...                105                   0   \n",
       "\n",
       "   AfterPromo  BeforePromo  SchoolHoliday_bw  StateHoliday_bw  Promo_bw  \\\n",
       "0           0            0               0.0              0.0       4.0   \n",
       "1           0            0               0.0              0.0       4.0   \n",
       "2           0            0               0.0              0.0       4.0   \n",
       "3           0            0               0.0              0.0       4.0   \n",
       "4           0            0               0.0              0.0       4.0   \n",
       "\n",
       "   SchoolHoliday_fw  StateHoliday_fw  Promo_fw  \n",
       "0               0.0              0.0       1.0  \n",
       "1               0.0              0.0       1.0  \n",
       "2               0.0              0.0       1.0  \n",
       "3               0.0              0.0       1.0  \n",
       "4               0.0              0.0       1.0  \n",
       "\n",
       "[5 rows x 92 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Column Names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check column names that are in `train_df` but not in `test_df`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Customers', 'Sales'}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(train_df.columns).difference(set(test_df.columns))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What is in `test_df` but not in `train_df`?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Id'}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(test_df.columns).difference(set(train_df.columns))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I guess if you knew the number of customers then you could give a good approximation of the sales...\n",
    "\n",
    "Note that `Sales` is the variable you are trying to predict."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(844338, 93)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(41088, 92)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature types"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create lists to isolate the continuous and categorical variable:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from https://github.com/fastai/course-v3/blob/master/nbs/dl1/lesson6-rossmann.ipynb\n",
    "cat_vars = ['Store', 'DayOfWeek', 'Year', 'Month', 'Day', 'StateHoliday', 'CompetitionMonthsOpen',\n",
    "    'Promo2Weeks', 'StoreType', 'Assortment', 'PromoInterval', 'CompetitionOpenSinceYear', 'Promo2SinceYear',\n",
    "    'State', 'Week', 'Events', 'Promo_fw', 'Promo_bw', 'StateHoliday_fw', 'StateHoliday_bw',\n",
    "    'SchoolHoliday_fw', 'SchoolHoliday_bw']\n",
    "\n",
    "cont_vars = ['CompetitionDistance', 'Max_TemperatureC', 'Mean_TemperatureC', 'Min_TemperatureC',\n",
    "   'Max_Humidity', 'Mean_Humidity', 'Min_Humidity', 'Max_Wind_SpeedKm_h', \n",
    "   'Mean_Wind_SpeedKm_h', 'CloudCover', 'trend', 'trend_DE',\n",
    "   'AfterStateHoliday', 'BeforeStateHoliday', 'Promo', 'SchoolHoliday']\n",
    "\n",
    "target = ['Sales']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# my modifications\n",
    "# based on https://en.wikipedia.org/wiki/Okta, CloudCover is categorical\n",
    "cat_vars = ['Store', 'DayOfWeek', 'Year', 'Month', 'Day', 'StateHoliday', 'CompetitionMonthsOpen',\n",
    "    'Promo2Weeks', 'StoreType', 'Assortment', 'PromoInterval', 'CompetitionOpenSinceYear', 'Promo2SinceYear',\n",
    "    'State', 'Week', 'Events', 'Promo_fw', 'Promo_bw', 'StateHoliday_fw', 'StateHoliday_bw',\n",
    "    'SchoolHoliday_fw', 'SchoolHoliday_bw', 'Promo', 'SchoolHoliday', 'CloudCover']\n",
    "\n",
    "cont_vars = ['CompetitionDistance', 'Max_TemperatureC', 'Mean_TemperatureC', 'Min_TemperatureC',\n",
    "   'Max_Humidity', 'Mean_Humidity', 'Min_Humidity', 'Max_Wind_SpeedKm_h', \n",
    "   'Mean_Wind_SpeedKm_h', 'trend', 'trend_DE',\n",
    "   'AfterStateHoliday', 'BeforeStateHoliday']\n",
    "\n",
    "target = ['Sales']\n",
    "\n",
    "cat_vars = sorted(cat_vars)\n",
    "cont_vars = sorted(cont_vars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What are the variables not used as features?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'AfterPromo',\n",
       " 'AfterSchoolHoliday',\n",
       " 'BeforePromo',\n",
       " 'BeforeSchoolHoliday',\n",
       " 'CompetitionDaysOpen',\n",
       " 'CompetitionOpenSince',\n",
       " 'CompetitionOpenSinceMonth',\n",
       " 'Customers',\n",
       " 'Date',\n",
       " 'Date_DE',\n",
       " 'Day_DE',\n",
       " 'Dayofweek',\n",
       " 'Dayofweek_DE',\n",
       " 'Dayofyear',\n",
       " 'Dayofyear_DE',\n",
       " 'Dew_PointC',\n",
       " 'Elapsed',\n",
       " 'Elapsed_DE',\n",
       " 'Is_month_end',\n",
       " 'Is_month_end_DE',\n",
       " 'Is_month_start',\n",
       " 'Is_month_start_DE',\n",
       " 'Is_quarter_end',\n",
       " 'Is_quarter_end_DE',\n",
       " 'Is_quarter_start',\n",
       " 'Is_quarter_start_DE',\n",
       " 'Is_year_end',\n",
       " 'Is_year_end_DE',\n",
       " 'Is_year_start',\n",
       " 'Is_year_start_DE',\n",
       " 'Max_Gust_SpeedKm_h',\n",
       " 'Max_Sea_Level_PressurehPa',\n",
       " 'Max_VisibilityKm',\n",
       " 'MeanDew_PointC',\n",
       " 'Mean_Sea_Level_PressurehPa',\n",
       " 'Mean_VisibilityKm',\n",
       " 'Min_DewpointC',\n",
       " 'Min_Sea_Level_PressurehPa',\n",
       " 'Min_VisibilitykM',\n",
       " 'Month_DE',\n",
       " 'Open',\n",
       " 'Precipitationmm',\n",
       " 'Promo2',\n",
       " 'Promo2Days',\n",
       " 'Promo2Since',\n",
       " 'Promo2SinceWeek',\n",
       " 'StateName',\n",
       " 'State_DE',\n",
       " 'WindDirDegrees',\n",
       " 'file',\n",
       " 'file_DE',\n",
       " 'index',\n",
       " 'week',\n",
       " 'week_DE'}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(train_df.columns).difference(set(cat_vars + cont_vars + target))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Continuous variables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Are there missing values in the continuous variables?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AfterStateHoliday         0\n",
       "BeforeStateHoliday        0\n",
       "CompetitionDistance    2186\n",
       "Max_Humidity              0\n",
       "Max_TemperatureC          0\n",
       "Max_Wind_SpeedKm_h        0\n",
       "Mean_Humidity             0\n",
       "Mean_TemperatureC         0\n",
       "Mean_Wind_SpeedKm_h       0\n",
       "Min_Humidity              0\n",
       "Min_TemperatureC          0\n",
       "trend                     0\n",
       "trend_DE                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[cont_vars].isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AfterStateHoliday       0\n",
       "BeforeStateHoliday      0\n",
       "CompetitionDistance    96\n",
       "Max_Humidity            0\n",
       "Max_TemperatureC        0\n",
       "Max_Wind_SpeedKm_h      0\n",
       "Mean_Humidity           0\n",
       "Mean_TemperatureC       0\n",
       "Mean_Wind_SpeedKm_h     0\n",
       "Min_Humidity            0\n",
       "Min_TemperatureC        0\n",
       "trend                   0\n",
       "trend_DE                0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df[cont_vars].isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Following Jeremy Howard's [approach](https://github.com/fastai/course-v3/blob/master/nbs/dl1/lesson6-rossmann.ipynb), we will replace the missing `CompetitionDistance` with the median and create a boolean variable to indicate that it is missing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "median_dist = train_df['CompetitionDistance'].median()\n",
    "\n",
    "train_df['CompetitionDistance_na'] = train_df['CompetitionDistance'].isna()\n",
    "train_df['CompetitionDistance'] = np.where(train_df['CompetitionDistance'].isna(), \n",
    "                                           median_dist, \n",
    "                                           train_df['CompetitionDistance'])\n",
    "\n",
    "test_df['CompetitionDistance_na'] = test_df['CompetitionDistance'].isna()\n",
    "test_df['CompetitionDistance'] = np.where(test_df['CompetitionDistance'].isna(), \n",
    "                                          median_dist, \n",
    "                                          test_df['CompetitionDistance'])\n",
    "\n",
    "# treat the missing competition distance column as categorical\n",
    "cat_vars = cat_vars + ['CompetitionDistance_na']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "View the distribution of the continuous variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(15,15))\n",
    "axes[-1][-1].axis('off')\n",
    "axes[-1][-2].axis('off')\n",
    "axes[-1][-3].axis('off')\n",
    "\n",
    "for i in range(4):\n",
    "    for j in range(4):\n",
    "        idx = i*4 + j\n",
    "        if idx > len(cont_vars) - 1:\n",
    "            break\n",
    "        var = cont_vars[i*4 + j]\n",
    "        \n",
    "        train_df[var].hist(ax=axes[i,j])\n",
    "        axes[i,j].set_title(var)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create list of variables to do log scaling:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "log_scale_vars = ['CompetitionDistance', 'Max_Wind_SpeedKm_h', 'Mean_Wind_SpeedKm_h']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(nrows=2, ncols=2, figsize=(7,7))\n",
    "axes[-1][-1].axis('off')\n",
    "\n",
    "for i in range(2):\n",
    "    for j in range(2):\n",
    "        idx = i*2 + j\n",
    "        if idx > len(log_scale_vars) - 1:\n",
    "            break\n",
    "        var = log_scale_vars[i*2 + j]\n",
    "        \n",
    "        train_df[var].apply(lambda val: np.log(val+1/1000)).hist(ax=axes[i,j])\n",
    "        axes[i,j].set_title(var)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create list of variables to do quantile bucketing and figure out the bin widths:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "quantile_buckets = ['AfterStateHoliday', 'BeforeStateHoliday', 'Max_Humidity']\n",
    "n_quantiles = [10, 10, 2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Buckets for AfterStateHoliday:\n",
      "Total buckets: 10\n",
      "Bin edges: [array([  0.,   4.,   8.,  15.,  22.,  30.,  42.,  53.,  68.,  84., 135.])]\n",
      "2.0    100039\n",
      "7.0     88771\n",
      "9.0     88233\n",
      "5.0     87381\n",
      "4.0     85110\n",
      "1.0     85109\n",
      "8.0     81476\n",
      "3.0     81039\n",
      "6.0     80103\n",
      "0.0     67077\n",
      "Name: AfterStateHoliday_bucket, dtype: int64\n",
      "Total observations: 844338\n",
      "\n",
      "\n",
      "Buckets for BeforeStateHoliday:\n",
      "Total buckets: 10\n",
      "Bin edges: [array([-195.,  -85.,  -68.,  -52.,  -39.,  -28.,  -20.,  -11.,   -6.,\n",
      "         -2.,    0.])]\n",
      "9.0    98344\n",
      "6.0    94996\n",
      "1.0    87815\n",
      "3.0    85653\n",
      "2.0    83740\n",
      "5.0    82477\n",
      "7.0    82262\n",
      "0.0    79848\n",
      "4.0    75311\n",
      "8.0    73892\n",
      "Name: BeforeStateHoliday_bucket, dtype: int64\n",
      "Total observations: 844338\n",
      "\n",
      "\n",
      "Buckets for Max_Humidity:\n",
      "Total buckets: 2\n",
      "Bin edges: [array([ 44.,  94., 100.])]\n",
      "1.0    501606\n",
      "0.0    342732\n",
      "Name: Max_Humidity_bucket, dtype: int64\n",
      "Total observations: 844338\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for col, n in zip(quantile_buckets, n_quantiles):\n",
    "    print('Buckets for {}:'.format(col))\n",
    "    \n",
    "    est = KBinsDiscretizer(n_bins=n, encode='ordinal', strategy='quantile')\n",
    "    bucket_name = col + '_bucket'\n",
    "    cat_vars = cat_vars + [bucket_name]\n",
    "    \n",
    "    train_vals = train_df[col].values.reshape(-1,1)\n",
    "    train_df[bucket_name] = est.fit_transform(train_vals).reshape(-1)\n",
    "    \n",
    "    test_vals = test_df[col].values.reshape(-1, 1)\n",
    "    test_df[bucket_name] = est.transform(test_vals).reshape(-1)\n",
    "    \n",
    "    results = train_df[bucket_name].value_counts()\n",
    "    n_buckets, = results.shape\n",
    "    \n",
    "    print('Total buckets: {}'.format(n_buckets))\n",
    "    print('Bin edges: {}'.format(est.bin_edges_))\n",
    "    print(results)\n",
    "    print('Total observations: {}'.format(results.sum()))\n",
    "    print('\\n')\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What would z-scoring the continuous variables look like?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "z_score_train_df = (train_df[cont_vars] - train_df[cont_vars].mean())/train_df[cont_vars].std(ddof=0)\n",
    "\n",
    "fig, axes = plt.subplots(nrows=4, ncols=4, figsize=(15,15))\n",
    "axes[-1][-1].axis('off')\n",
    "axes[-1][-2].axis('off')\n",
    "axes[-1][-3].axis('off')\n",
    "\n",
    "for i in range(4):\n",
    "    for j in range(4):\n",
    "        idx = i*4 + j\n",
    "        if idx > len(cont_vars) - 1:\n",
    "            break\n",
    "        var = cont_vars[i*4 + j]\n",
    "        \n",
    "        z_score_train_df[var].hist(ax=axes[i,j])\n",
    "        axes[i,j].set_title(var)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert all the continuos variables to float32:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df[cont_vars] = train_df[cont_vars].astype(np.float32)\n",
    "test_df[cont_vars] = test_df[cont_vars].astype(np.float32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Categorical variables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Set `CloudCover` to a string:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['CloudCover'] = train_df['CloudCover'].astype(np.str)\n",
    "test_df['CloudCover'] = test_df['CloudCover'].astype(np.str)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check for missing values in the categorical variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Assortment                        0\n",
       "CloudCover                        0\n",
       "CompetitionMonthsOpen             0\n",
       "CompetitionOpenSinceYear          0\n",
       "Day                               0\n",
       "DayOfWeek                         0\n",
       "Events                       203057\n",
       "Month                             0\n",
       "Promo                             0\n",
       "Promo2SinceYear                   0\n",
       "Promo2Weeks                       0\n",
       "PromoInterval                423292\n",
       "Promo_bw                          0\n",
       "Promo_fw                          0\n",
       "SchoolHoliday                     0\n",
       "SchoolHoliday_bw                  0\n",
       "SchoolHoliday_fw                  0\n",
       "State                             0\n",
       "StateHoliday                      0\n",
       "StateHoliday_bw                   0\n",
       "StateHoliday_fw                   0\n",
       "Store                             0\n",
       "StoreType                         0\n",
       "Week                              0\n",
       "Year                              0\n",
       "CompetitionDistance_na            0\n",
       "AfterStateHoliday_bucket          0\n",
       "BeforeStateHoliday_bucket         0\n",
       "Max_Humidity_bucket               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[cat_vars].isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Assortment                       0\n",
       "CloudCover                       0\n",
       "CompetitionMonthsOpen            0\n",
       "CompetitionOpenSinceYear         0\n",
       "Day                              0\n",
       "DayOfWeek                        0\n",
       "Events                       10155\n",
       "Month                            0\n",
       "Promo                            0\n",
       "Promo2SinceYear                  0\n",
       "Promo2Weeks                      0\n",
       "PromoInterval                17232\n",
       "Promo_bw                         0\n",
       "Promo_fw                         0\n",
       "SchoolHoliday                    0\n",
       "SchoolHoliday_bw                 0\n",
       "SchoolHoliday_fw                 0\n",
       "State                            0\n",
       "StateHoliday                     0\n",
       "StateHoliday_bw                  0\n",
       "StateHoliday_fw                  0\n",
       "Store                            0\n",
       "StoreType                        0\n",
       "Week                             0\n",
       "Year                             0\n",
       "CompetitionDistance_na           0\n",
       "AfterStateHoliday_bucket         0\n",
       "BeforeStateHoliday_bucket        0\n",
       "Max_Humidity_bucket              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df[cat_vars].isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Assortment                    object\n",
       "CloudCover                    object\n",
       "CompetitionMonthsOpen          int64\n",
       "CompetitionOpenSinceYear       int32\n",
       "Day                            int64\n",
       "DayOfWeek                      int64\n",
       "Events                        object\n",
       "Month                          int64\n",
       "Promo                          int64\n",
       "Promo2SinceYear                int32\n",
       "Promo2Weeks                    int64\n",
       "PromoInterval                 object\n",
       "Promo_bw                     float64\n",
       "Promo_fw                     float64\n",
       "SchoolHoliday                  int64\n",
       "SchoolHoliday_bw             float64\n",
       "SchoolHoliday_fw             float64\n",
       "State                         object\n",
       "StateHoliday                    bool\n",
       "StateHoliday_bw              float64\n",
       "StateHoliday_fw              float64\n",
       "Store                          int64\n",
       "StoreType                     object\n",
       "Week                           int64\n",
       "Year                           int64\n",
       "CompetitionDistance_na          bool\n",
       "AfterStateHoliday_bucket     float64\n",
       "BeforeStateHoliday_bucket    float64\n",
       "Max_Humidity_bucket          float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[cat_vars].dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fill missing categorical variables with '-1' (these are string columns):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df[cat_vars] = train_df[cat_vars].fillna('-1')\n",
    "test_df[cat_vars] = test_df[cat_vars].fillna('-1')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert the float64, bool and int64 columns into int32:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "float_columns = train_df[cat_vars]\\\n",
    "    .dtypes\\\n",
    "    .where(lambda val: (val == np.float64) | (val == np.bool) | (val == np.int64))\\\n",
    "    .dropna()\\\n",
    "    .index\\\n",
    "    .tolist()\n",
    "\n",
    "for column in float_columns:\n",
    "    train_df[column] = train_df[column].astype(np.int32)\n",
    "    test_df[column] = train_df[column].astype(np.int32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Assortment                   object\n",
       "CloudCover                   object\n",
       "CompetitionMonthsOpen         int32\n",
       "CompetitionOpenSinceYear      int32\n",
       "Day                           int32\n",
       "DayOfWeek                     int32\n",
       "Events                       object\n",
       "Month                         int32\n",
       "Promo                         int32\n",
       "Promo2SinceYear               int32\n",
       "Promo2Weeks                   int32\n",
       "PromoInterval                object\n",
       "Promo_bw                      int32\n",
       "Promo_fw                      int32\n",
       "SchoolHoliday                 int32\n",
       "SchoolHoliday_bw              int32\n",
       "SchoolHoliday_fw              int32\n",
       "State                        object\n",
       "StateHoliday                  int32\n",
       "StateHoliday_bw               int32\n",
       "StateHoliday_fw               int32\n",
       "Store                         int32\n",
       "StoreType                    object\n",
       "Week                          int32\n",
       "Year                          int32\n",
       "CompetitionDistance_na        int32\n",
       "AfterStateHoliday_bucket      int32\n",
       "BeforeStateHoliday_bucket     int32\n",
       "Max_Humidity_bucket           int32\n",
       "dtype: object"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[cat_vars].dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Assortment                   object\n",
       "CloudCover                   object\n",
       "CompetitionMonthsOpen         int32\n",
       "CompetitionOpenSinceYear      int32\n",
       "Day                           int32\n",
       "DayOfWeek                     int32\n",
       "Events                       object\n",
       "Month                         int32\n",
       "Promo                         int32\n",
       "Promo2SinceYear               int32\n",
       "Promo2Weeks                   int32\n",
       "PromoInterval                object\n",
       "Promo_bw                      int32\n",
       "Promo_fw                      int32\n",
       "SchoolHoliday                 int32\n",
       "SchoolHoliday_bw              int32\n",
       "SchoolHoliday_fw              int32\n",
       "State                        object\n",
       "StateHoliday                  int32\n",
       "StateHoliday_bw               int32\n",
       "StateHoliday_fw               int32\n",
       "Store                         int32\n",
       "StoreType                    object\n",
       "Week                          int32\n",
       "Year                          int32\n",
       "CompetitionDistance_na        int32\n",
       "AfterStateHoliday_bucket      int32\n",
       "BeforeStateHoliday_bucket     int32\n",
       "Max_Humidity_bucket           int32\n",
       "dtype: object"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df[cat_vars].dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the vocabulary for the object columns:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vocab for Assortment:\n",
      "['a' 'c' 'b']\n",
      "\n",
      "\n",
      "vocab for CloudCover:\n",
      "['1.0' '4.0' '2.0' '6.0' '5.0' 'nan' '3.0' '8.0' '7.0' '0.0']\n",
      "\n",
      "\n",
      "vocab for Events:\n",
      "['Fog' '-1' 'Rain' 'Rain-Thunderstorm' 'Fog-Rain' 'Rain-Hail-Thunderstorm'\n",
      " 'Fog-Rain-Thunderstorm' 'Thunderstorm' 'Rain-Hail' 'Fog-Thunderstorm'\n",
      " 'Rain-Snow' 'Fog-Rain-Hail-Thunderstorm' 'Snow' 'Rain-Snow-Hail'\n",
      " 'Rain-Snow-Hail-Thunderstorm' 'Rain-Snow-Thunderstorm' 'Fog-Rain-Snow'\n",
      " 'Fog-Snow' 'Snow-Hail' 'Fog-Rain-Snow-Hail' 'Fog-Rain-Hail'\n",
      " 'Fog-Snow-Hail']\n",
      "\n",
      "\n",
      "vocab for PromoInterval:\n",
      "['-1' 'Jan,Apr,Jul,Oct' 'Feb,May,Aug,Nov' 'Mar,Jun,Sept,Dec']\n",
      "\n",
      "\n",
      "vocab for State:\n",
      "['HE' 'TH' 'NW' 'BE' 'SN' 'SH' 'HB,NI' 'BY' 'BW' 'RP' 'ST' 'HH']\n",
      "\n",
      "\n",
      "vocab for StoreType:\n",
      "['c' 'a' 'd' 'b']\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "obj_columns = train_df[cat_vars]\\\n",
    "    .dtypes.where(lambda val: val == np.object)\\\n",
    "    .dropna()\\\n",
    "    .index\\\n",
    "    .tolist()\n",
    "\n",
    "for column in obj_columns:\n",
    "    print('vocab for {}:'.format(column))\n",
    "    print(train_df[column].unique())\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check that test set has no new vocabulary:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in obj_columns:\n",
    "    train_vocab = train_df[column].unique()\n",
    "    test_vocab = test_df[column].unique()\n",
    "    \n",
    "    assert all([vocab in train_vocab for vocab in test_vocab])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Target Variable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`Sales` is a float, not an integer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['Sales'] = train_df['Sales'].astype(np.float32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "View distribution of sales:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7b9b3fe2e8>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "train_df['Sales'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    844338.000000\n",
       "mean       6955.958496\n",
       "std        3103.815674\n",
       "min          46.000000\n",
       "25%        4859.000000\n",
       "50%        6369.000000\n",
       "75%        8360.000000\n",
       "max       41551.000000\n",
       "Name: Sales, dtype: float64"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Sales'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Save"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Save all edits to train and test sets to a new parquet file:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# you need date to split into train and validation set\n",
    "train_cols = ['Date'] + cont_vars + cat_vars + ['Sales']\n",
    "# you need Id to make submission to kaggle later\n",
    "test_cols = ['Id'] + cont_vars + cat_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df[train_cols].to_parquet('./data/train_model_features.parquet', index=False)\n",
    "test_df[test_cols].to_parquet('./data/test_model_features.parquet', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Print the updated list of continous and categorical variables for reference:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['AfterStateHoliday',\n",
       " 'BeforeStateHoliday',\n",
       " 'CompetitionDistance',\n",
       " 'Max_Humidity',\n",
       " 'Max_TemperatureC',\n",
       " 'Max_Wind_SpeedKm_h',\n",
       " 'Mean_Humidity',\n",
       " 'Mean_TemperatureC',\n",
       " 'Mean_Wind_SpeedKm_h',\n",
       " 'Min_Humidity',\n",
       " 'Min_TemperatureC',\n",
       " 'trend',\n",
       " 'trend_DE']"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cont_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Assortment',\n",
       " 'CloudCover',\n",
       " 'CompetitionMonthsOpen',\n",
       " 'CompetitionOpenSinceYear',\n",
       " 'Day',\n",
       " 'DayOfWeek',\n",
       " 'Events',\n",
       " 'Month',\n",
       " 'Promo',\n",
       " 'Promo2SinceYear',\n",
       " 'Promo2Weeks',\n",
       " 'PromoInterval',\n",
       " 'Promo_bw',\n",
       " 'Promo_fw',\n",
       " 'SchoolHoliday',\n",
       " 'SchoolHoliday_bw',\n",
       " 'SchoolHoliday_fw',\n",
       " 'State',\n",
       " 'StateHoliday',\n",
       " 'StateHoliday_bw',\n",
       " 'StateHoliday_fw',\n",
       " 'Store',\n",
       " 'StoreType',\n",
       " 'Week',\n",
       " 'Year',\n",
       " 'CompetitionDistance_na',\n",
       " 'AfterStateHoliday_bucket',\n",
       " 'BeforeStateHoliday_bucket',\n",
       " 'Max_Humidity_bucket']"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_vars"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.8"
  },
  "toc": {
   "nav_menu": {
    "height": "275px",
    "width": "646px"
   },
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}