Time Series.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment V3</b>\n",
    "\n",
    "Thanks for taking the time to improve the project! It is accepted, and you can move on to the next sprint! Keep up the good work! :)\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Review**\n",
    "\n",
    "Hi, my name is Dmitry and I will be reviewing your project.\n",
    "  \n",
    "You can find my comments in colored markdown cells:\n",
    "  \n",
    "<div class=\"alert alert-success\">\n",
    "  If everything is done successfully.\n",
    "</div>\n",
    "  \n",
    "<div class=\"alert alert-warning\">\n",
    "  If I have some (optional) suggestions, or questions to think about, or general comments.\n",
    "</div>\n",
    "  \n",
    "<div class=\"alert alert-danger\">\n",
    "  If a section requires some corrections. Work can't be accepted with red comments.\n",
    "</div>\n",
    "  \n",
    "Please don't remove my comments, as it will make further review iterations much harder for me.\n",
    "  \n",
    "Feel free to reply to my comments or ask questions using the following template:\n",
    "  \n",
    "<div class=\"alert alert-info\">\n",
    "  For your comments and questions.\n",
    "</div>\n",
    "  \n",
    "First of all, thank you for turning in the project! You did a pretty good job overall, but there are a couple of problems that need to be fixed before the project is accepted. Let me know if you have any questions!\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project description\n",
    "\n",
    "Sweet Lift Taxi company has collected historical data on taxi orders at airports. To attract more drivers during peak hours, we need to predict the amount of taxi orders for the next hour. Build a model for such a prediction.\n",
    "\n",
    "The RMSE metric on the test set should not be more than 48.\n",
    "\n",
    "## Project instructions\n",
    "\n",
    "1. Download the data and resample it by one hour.\n",
    "2. Analyze the data.\n",
    "3. Train different models with different hyperparameters. The test sample should be 10% of the initial dataset. \n",
    "4. Test the data using the test sample and provide a conclusion.\n",
    "\n",
    "## Data description\n",
    "\n",
    "The data is stored in file `taxi.csv`. The number of orders is in the '*num_orders*' column."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparation\n",
    "Initially I will import the necessary libraries and do preliminary observations of the condition of the data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import TimeSeriesSplit\n",
    "\n",
    "\n",
    "from lightgbm import LGBMRegressor\n",
    "from xgboost import XGBRegressor\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n",
      "\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 26496 entries, 2018-03-01 00:00:00 to 2018-08-31 23:50:00\n",
      "Data columns (total 1 columns):\n",
      " #   Column      Non-Null Count  Dtype\n",
      "---  ------      --------------  -----\n",
      " 0   num_orders  26496 non-null  int64\n",
      "dtypes: int64(1)\n",
      "memory usage: 414.0 KB\n",
      "\n"
     ]
    },
    {
     "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>num_orders</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datetime</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:00:00</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:10:00</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:20:00</th>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:30:00</th>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:40:00</th>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     num_orders\n",
       "datetime                       \n",
       "2018-03-01 00:00:00           9\n",
       "2018-03-01 00:10:00          14\n",
       "2018-03-01 00:20:00          28\n",
       "2018-03-01 00:30:00          20\n",
       "2018-03-01 00:40:00          32"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check data, assign datetime column as index, convert to datetime object\n",
    "data = pd.read_csv('/datasets/taxi.csv', index_col=[0], parse_dates=True)\n",
    "\n",
    "#check if data is sorted\n",
    "print(data.index.is_monotonic)\n",
    "print()\n",
    "\n",
    "#check info\n",
    "data.info()\n",
    "print()\n",
    "\n",
    "#check first few rows\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "Good idea to check that the time series is in correct order!\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "num_orders    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check for NaN\n",
    "data.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 26496 entries, 2018-03-01 00:00:00 to 2018-08-31 23:50:00\n",
      "Data columns (total 1 columns):\n",
      " #   Column      Non-Null Count  Dtype\n",
      "---  ------      --------------  -----\n",
      " 0   num_orders  26496 non-null  int64\n",
      "dtypes: int64(1)\n",
      "memory usage: 414.0 KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I see no null values, the datatype is appropriate. The data is ready for analysis. Now I will resample the data by the hour "
   ]
  },
  {
   "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>num_orders</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datetime</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-03-01 00:00:00</th>\n",
       "      <td>124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 01:00:00</th>\n",
       "      <td>85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 02:00:00</th>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 03:00:00</th>\n",
       "      <td>66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-01 04:00:00</th>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     num_orders\n",
       "datetime                       \n",
       "2018-03-01 00:00:00         124\n",
       "2018-03-01 01:00:00          85\n",
       "2018-03-01 02:00:00          71\n",
       "2018-03-01 03:00:00          66\n",
       "2018-03-01 04:00:00          43"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#resample by hour, aggregation = sum\n",
    "data = data.resample('1H').sum()\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "The data was resampled correctly\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 4416 entries, 2018-03-01 00:00:00 to 2018-08-31 23:00:00\n",
      "Freq: H\n",
      "Data columns (total 1 columns):\n",
      " #   Column      Non-Null Count  Dtype\n",
      "---  ------      --------------  -----\n",
      " 0   num_orders  4416 non-null   int64\n",
      "dtypes: int64(1)\n",
      "memory usage: 69.0 KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis\n",
    "Now I will look into the seasonality and any trends present "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x1080 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get different attributes of our time series\n",
    "decomposed = seasonal_decompose(data)\n",
    "\n",
    "#specify fig size for all plots\n",
    "plt.figure(figsize=(10,15))\n",
    "\n",
    "#3 rows, 1 column, image 1\n",
    "plt.subplot(311)\n",
    "decomposed.trend.plot(ax=plt.gca()) #specifying axis to display graph correctly\n",
    "plt.title('Trend')\n",
    "plt.xlabel('Month')\n",
    "plt.ylabel('Number of Orders')\n",
    "\n",
    "#subplot configuration to prevent overlap\n",
    "plt.tight_layout(h_pad=7)\n",
    "\n",
    "#image 2\n",
    "plt.subplot(312)\n",
    "decomposed.seasonal.plot(ax=plt.gca())\n",
    "plt.title('Seasonality')\n",
    "plt.xlabel('Month')\n",
    "plt.ylabel('Number of Orders')\n",
    "\n",
    "#image 3\n",
    "plt.subplot(313)\n",
    "decomposed.seasonal['2018-03-01 00:00:00': '2018-03-05 00:00:00'].plot(ax=plt.gca())\n",
    "plt.title('Seasonal')\n",
    "plt.xlabel('Month')\n",
    "plt.ylabel('Number of Orders')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-warning\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "The x axis label of the last plot is incorrect\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From these plots we can see that there is a positive trend moving into the summer months, and at the start of the day there is a peak in taxi orders, followed by a drop and then an increase into the end of the day "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "Alright, you did a seasonal decompostition and noted some patterns!\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To start I will design a funtion for feature engineering. This extraction the day of the week, month, day, and year. As well as the previous days cycle and a series of 3 lags. These features will be very useful in training our model to predict the number of orders "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>num_orders</th>\n",
       "      <th>hour</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>previous_day</th>\n",
       "      <th>lag_1</th>\n",
       "      <th>lag_2</th>\n",
       "      <th>lag_3</th>\n",
       "      <th>lag_4</th>\n",
       "      <th>lag_5</th>\n",
       "      <th>lag_6</th>\n",
       "      <th>rolling_mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datetime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-03-02 00:00:00</th>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>124.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>65.714286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-02 01:00:00</th>\n",
       "      <td>120</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>85.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>72.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-02 02:00:00</th>\n",
       "      <td>75</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>71.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>79.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-02 03:00:00</th>\n",
       "      <td>64</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>66.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>66.0</td>\n",
       "      <td>83.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-03-02 04:00:00</th>\n",
       "      <td>20</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>43.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>83.714286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     num_orders  hour  month  day  dayofweek  previous_day  \\\n",
       "datetime                                                                     \n",
       "2018-03-02 00:00:00          90     0      3    2          4         124.0   \n",
       "2018-03-02 01:00:00         120     1      3    2          4          85.0   \n",
       "2018-03-02 02:00:00          75     2      3    2          4          71.0   \n",
       "2018-03-02 03:00:00          64     3      3    2          4          66.0   \n",
       "2018-03-02 04:00:00          20     4      3    2          4          43.0   \n",
       "\n",
       "                     lag_1  lag_2  lag_3  lag_4  lag_5  lag_6  rolling_mean  \n",
       "datetime                                                                     \n",
       "2018-03-02 00:00:00   58.0  113.0   66.0   61.0   45.0   73.0     65.714286  \n",
       "2018-03-02 01:00:00   90.0   58.0  113.0   66.0   61.0   45.0     72.285714  \n",
       "2018-03-02 02:00:00  120.0   90.0   58.0  113.0   66.0   61.0     79.000000  \n",
       "2018-03-02 03:00:00   75.0  120.0   90.0   58.0  113.0   66.0     83.285714  \n",
       "2018-03-02 04:00:00   64.0   75.0  120.0   90.0   58.0  113.0     83.714286  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def make_features(data, max_lag):\n",
    "    data = data.copy(deep=True)\n",
    "    data['hour'] = data.index.hour\n",
    "    data['month'] = data.index.month\n",
    "    data['day'] = data.index.day\n",
    "    data['dayofweek'] = data.index.dayofweek\n",
    "    \n",
    "    #seasons last 1 day, so we'll create a column that utilizes the previous cycle\n",
    "    data['previous_day'] = data['num_orders'].shift(24)\n",
    "    \n",
    "    for lag in range(1, max_lag + 1):\n",
    "        data['lag_{}'.format(lag)] = data['num_orders'].shift(lag)\n",
    "    \n",
    "    data['rolling_mean'] = data['num_orders'].shift().rolling(7).mean()\n",
    "    \n",
    "    \n",
    "    data = data.dropna()\n",
    "    return data\n",
    "\n",
    "full_data = make_features(data, 6)\n",
    "full_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-warning\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "Feature extraction looks good: there are calendar features and lag features. One omission is an hour feature. On the other hand, year is not useful as it is constant in our data.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-danger\">\n",
    "<s><b>Reviewer's comment</b>\n",
    "\n",
    "Next hour's number of orders can't be used as a feature: this is data from the future :)\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment V2</b>\n",
    "\n",
    "Ok, great! Now there is an hour feature, and next hour was removed\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 4392 entries, 2018-03-02 00:00:00 to 2018-08-31 23:00:00\n",
      "Freq: H\n",
      "Data columns (total 13 columns):\n",
      " #   Column        Non-Null Count  Dtype  \n",
      "---  ------        --------------  -----  \n",
      " 0   num_orders    4392 non-null   int64  \n",
      " 1   hour          4392 non-null   int64  \n",
      " 2   month         4392 non-null   int64  \n",
      " 3   day           4392 non-null   int64  \n",
      " 4   dayofweek     4392 non-null   int64  \n",
      " 5   previous_day  4392 non-null   float64\n",
      " 6   lag_1         4392 non-null   float64\n",
      " 7   lag_2         4392 non-null   float64\n",
      " 8   lag_3         4392 non-null   float64\n",
      " 9   lag_4         4392 non-null   float64\n",
      " 10  lag_5         4392 non-null   float64\n",
      " 11  lag_6         4392 non-null   float64\n",
      " 12  rolling_mean  4392 non-null   float64\n",
      "dtypes: float64(8), int64(5)\n",
      "memory usage: 480.4 KB\n"
     ]
    }
   ],
   "source": [
    "full_data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that we have our features we are ready to split our data and make a function for our evaluation metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# function to define metric\n",
    "def rmse(true, pred):\n",
    "    return mean_squared_error(true, pred)**0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "features_train size: 3556\n",
      "features_valid size: 396\n",
      "features_test size: 440\n",
      "target_train size: 3556\n",
      "target_valid size: 396\n",
      "target_test size: 440\n"
     ]
    }
   ],
   "source": [
    "#get train and test sets\n",
    "train, test = train_test_split(full_data, shuffle = False, test_size = 0.1)\n",
    "train, valid = train_test_split(train, shuffle = False, test_size = 0.1)\n",
    "\n",
    "features_train = train.drop('num_orders', axis = 1)\n",
    "features_valid = valid.drop('num_orders', axis = 1)\n",
    "target_train = train['num_orders']\n",
    "target_valid = valid['num_orders']\n",
    "features_test = test.drop('num_orders', axis = 1)\n",
    "target_test = test['num_orders']\n",
    "\n",
    "#check size\n",
    "print('features_train size:', features_train.shape[0])\n",
    "print('features_valid size:', features_valid.shape[0])\n",
    "print('features_test size:', features_test.shape[0])\n",
    "print('target_train size:', target_train.shape[0])\n",
    "print('target_valid size:', target_valid.shape[0])\n",
    "print('target_test size:', target_test.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "The data was split into train and test without shuffling\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TimeSeriesSplit(gap=0, max_train_size=None, n_splits=5, test_size=None)\n"
     ]
    }
   ],
   "source": [
    "# time series split\n",
    "tscv = TimeSeriesSplit(n_splits=5)\n",
    "print(tscv)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE for the training set: 35.56636943399032\n"
     ]
    }
   ],
   "source": [
    "lr_model = LinearRegression()\n",
    "\n",
    "lr_model.fit(features_train, target_train)\n",
    "\n",
    "lr_train_prediction = lr_model.predict(features_valid)\n",
    "print(\"RMSE for the training set:\",  rmse(target_valid, lr_train_prediction) )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-danger\">\n",
    "<s><b>Reviewer's comment V2</b>\n",
    "\n",
    "The test set should not be used here.\n",
    "    \n",
    "The first RMSE value is actually for the validation set, not the train set, right?\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Regressor RMSE :  32.73761218793571\n"
     ]
    }
   ],
   "source": [
    "rf = RandomForestRegressor()\n",
    "\n",
    "rf_model = rf.fit(features_train,target_train)\n",
    "\n",
    "rf_prediction = rf_model.predict(features_valid)\n",
    "print(\"Random Forest Regressor RMSE : \", rmse(target_valid, rf_prediction))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest with hyperparameter tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best hyperparameters are: {'max_depth': None, 'min_samples_leaf': 4, 'n_estimators': 100}\n"
     ]
    }
   ],
   "source": [
    "grid = {\n",
    "    \"n_estimators\" : [100, 200],\n",
    "    \"max_depth\" : [None, 2, 4, 6],\n",
    "    \"min_samples_leaf\" : [2, 4]\n",
    "}\n",
    "# define the model \n",
    "regressor = RandomForestRegressor(random_state = 12345)\n",
    "# define the grid search\n",
    "grid_search_rf = GridSearchCV(estimator = regressor, \n",
    "                              param_grid = grid, \n",
    "                              scoring=\"neg_mean_squared_error\", \n",
    "                              cv=tscv)\n",
    "# execute search\n",
    "grid_search_rf.fit(features_train, target_train)\n",
    "# summarize result\n",
    "print('The best hyperparameters are: {}'.format(grid_search_rf.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf_model = RandomForestRegressor(**grid_search_rf.best_params_)\n",
    "\n",
    "rf_model.fit(features_train, target_train)\n",
    "\n",
    "rf_pred = rf_model.predict(features_valid)\n",
    "\n",
    "rf_rmse = rmse(target_valid, rf_pred)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest with Hyperparameters : 32.759969008171\n"
     ]
    }
   ],
   "source": [
    "print(\"Random Forest with Hyperparameters :\", rf_rmse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LightGBM Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best hyperparameters are: {'learning_rate': 0.05, 'n_estimators': 800, 'num_leaves': 5}\n",
      "Fit time: 0.6375446319580078\n",
      "Score time: 0.038527631759643556\n",
      "RMSE: 23.79300765701394\n"
     ]
    }
   ],
   "source": [
    "lgbm_grid = {'learning_rate': [0.05, 0.1],\n",
    "             'n_estimators': [500, 800, 1000],\n",
    "             'num_leaves': [2,3,5]\n",
    "            }\n",
    "# define the model \n",
    "lgbm_regr = LGBMRegressor(random_state = 12345)\n",
    "\n",
    "# define the grid search\n",
    "grid_search_lgbm = GridSearchCV(\n",
    "    estimator = lgbm_regr, \n",
    "    param_grid = lgbm_grid, \n",
    "    scoring = \"neg_mean_squared_error\", \n",
    "    cv = tscv, \n",
    "    n_jobs = 1\n",
    ")\n",
    "# execute search\n",
    "grid_search_lgbm.fit(features_train, target_train)\n",
    "# summarize result\n",
    "print('The best hyperparameters are: {}'.format(grid_search_lgbm.best_params_))\n",
    "\n",
    "results = grid_search_lgbm.cv_results_\n",
    "#get best index\n",
    "best_index = grid_search_lgbm.best_index_\n",
    "\n",
    "#get results\n",
    "fit_time = results['mean_fit_time'][best_index]\n",
    "score_time = results['mean_score_time'][best_index]\n",
    "RMSE = (grid_search_lgbm.best_score_*-1)**0.5\n",
    "\n",
    "print('Fit time:', fit_time)\n",
    "print('Score time:', score_time)\n",
    "print('RMSE:', RMSE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 31.909\n"
     ]
    }
   ],
   "source": [
    "lgbm_model = LGBMRegressor(**grid_search_lgbm.best_params_)\n",
    "lgbm_model.fit(features_train, target_train)\n",
    "    \n",
    "lgbm_pred = lgbm_model.predict(features_valid)\n",
    "    # determine RMSE for LightGBM regressor\n",
    "lgbm_rmse = rmse(target_valid, lgbm_pred)\n",
    "    \n",
    "print('RMSE: {:.3f}'.format(lgbm_rmse))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The best hyperparameters are: {'learning_rate': 0.1, 'max_depth': 4, 'n_estimators': 100}\n"
     ]
    }
   ],
   "source": [
    "xgb_grid = {'learning_rate':  [0.01, 0.1], \n",
    "            'max_depth': [2, 4],\n",
    "            'n_estimators': [50, 100]\n",
    "           }\n",
    "# define the model \n",
    "xgb_regr = XGBRegressor(random_state = 12345)\n",
    "\n",
    "# define the grid search\n",
    "grid_search_xgb = GridSearchCV(\n",
    "    estimator = xgb_regr, \n",
    "    param_grid = xgb_grid, \n",
    "    scoring = \"neg_mean_squared_error\", \n",
    "    cv = tscv, \n",
    "    n_jobs = 1\n",
    ")\n",
    "# execute search\n",
    "grid_search_xgb.fit(features_train, target_train)\n",
    "# summarize result\n",
    "print('The best hyperparameters are: {}'.format(grid_search_xgb.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, enable_categorical=False,\n",
       "             gamma=0, gpu_id=-1, importance_type=None,\n",
       "             interaction_constraints='', learning_rate=0.1, max_delta_step=0,\n",
       "             max_depth=4, min_child_weight=1, missing=nan,\n",
       "             monotone_constraints='()', n_estimators=100, n_jobs=4,\n",
       "             num_parallel_tree=1, predictor='auto', random_state=0, reg_alpha=0,\n",
       "             reg_lambda=1, scale_pos_weight=1, subsample=1, tree_method='exact',\n",
       "             validate_parameters=1, verbosity=None)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb_model = XGBRegressor(**grid_search_xgb.best_params_)\n",
    "\n",
    "xgb_model.fit(features_train, target_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1mRMSE using XGBoost Regressor\u001b[0m\n",
      "RMSE: 32.918\n"
     ]
    }
   ],
   "source": [
    "xgb_pred = xgb_model.predict(features_valid)\n",
    "    \n",
    "xgb_rmse = rmse(target_valid, xgb_pred)\n",
    "print(\"\\033[1m\" + 'RMSE using XGBoost Regressor' + \"\\033[0m\")\n",
    "print('RMSE: {:.3f}'.format(xgb_rmse))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-danger\">\n",
    "<s><b>Reviewer's comment</b>\n",
    "\n",
    "Three problems:\n",
    "    \n",
    "1. The test set should only be used to evaluate the final model after we're done with model selection, hyperparameter tuning, etc. If we use the test set to make any decisions about the models, the final model's test set score will be overly optimistic.\n",
    "    \n",
    "2. Train set score is not useful to compare the models: it only tells us how well the model memorized the train set, but not how well it will generalize to new data. That's why we need to use either a separate validation set or cross-validation to compare the models and tune hyperparameters.\n",
    "\n",
    "3. The default k-fold cross-validation is not appropriate for time series models: due to mixing of train and validation data in all folds but one it causes data leakage. The solution is to use a different split: [TimeSeriesSplit](http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.TimeSeriesSplit.html) (you can just pass an object of this class as the `cv` parameter to `GridSearchCV`, `cross_val_score` and so on)\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-warning\">\n",
    "<b>Reviewer's comment V2</b>\n",
    "\n",
    "The problems were fixed!\n",
    "   \n",
    "Note that if you use cross-validation anyway, you could just take the `best_score_` of the grid search object and use it to compare the models (it contains the cross-validation score of the best model) instead of using a separate validation set. For linear regression you can use `cross_val_score` with a time series split.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final test\n",
    "I will perform the final test using the test dataset on the random forest regressor because it provided me with the best score in training. I will retrain the model on the concatonated training and validation datasets. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3952\n"
     ]
    },
    {
     "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>hour</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>previous_day</th>\n",
       "      <th>lag_1</th>\n",
       "      <th>lag_2</th>\n",
       "      <th>lag_3</th>\n",
       "      <th>lag_4</th>\n",
       "      <th>lag_5</th>\n",
       "      <th>lag_6</th>\n",
       "      <th>rolling_mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datetime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-07-28 04:00:00</th>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>109.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>192.0</td>\n",
       "      <td>122.0</td>\n",
       "      <td>213.0</td>\n",
       "      <td>161.714286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-28 05:00:00</th>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>64.0</td>\n",
       "      <td>215.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>192.0</td>\n",
       "      <td>122.0</td>\n",
       "      <td>173.428571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-28 06:00:00</th>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>78.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>215.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>192.0</td>\n",
       "      <td>156.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-28 07:00:00</th>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>44.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>215.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>144.857143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-28 08:00:00</th>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>28</td>\n",
       "      <td>5</td>\n",
       "      <td>89.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>215.0</td>\n",
       "      <td>132.0</td>\n",
       "      <td>145.0</td>\n",
       "      <td>123.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     hour  month  day  dayofweek  previous_day  lag_1  lag_2  \\\n",
       "datetime                                                                       \n",
       "2018-07-28 04:00:00     4      7   28          5         109.0  132.0  145.0   \n",
       "2018-07-28 05:00:00     5      7   28          5          64.0  215.0  132.0   \n",
       "2018-07-28 06:00:00     6      7   28          5          78.0   91.0  215.0   \n",
       "2018-07-28 07:00:00     7      7   28          5          44.0   44.0   91.0   \n",
       "2018-07-28 08:00:00     8      7   28          5          89.0   39.0   44.0   \n",
       "\n",
       "                     lag_3  lag_4  lag_5  lag_6  rolling_mean  \n",
       "datetime                                                       \n",
       "2018-07-28 04:00:00  195.0  192.0  122.0  213.0    161.714286  \n",
       "2018-07-28 05:00:00  145.0  195.0  192.0  122.0    173.428571  \n",
       "2018-07-28 06:00:00  132.0  145.0  195.0  192.0    156.000000  \n",
       "2018-07-28 07:00:00  215.0  132.0  145.0  195.0    144.857143  \n",
       "2018-07-28 08:00:00   91.0  215.0  132.0  145.0    123.000000  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features_train_test=pd.concat([features_valid, features_train])\n",
    "\n",
    "print(features_train_test.shape[0])\n",
    "\n",
    "features_train_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "target_train_test=pd.concat([target_valid, target_train])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3952\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "datetime\n",
       "2018-07-28 04:00:00    215\n",
       "2018-07-28 05:00:00     91\n",
       "2018-07-28 06:00:00     44\n",
       "2018-07-28 07:00:00     39\n",
       "2018-07-28 08:00:00     73\n",
       "Name: num_orders, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(target_train_test.shape[0])\n",
    "\n",
    "target_train_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Regressor Final RMSE : 44.20187726877742\n"
     ]
    }
   ],
   "source": [
    "rf_model.fit(features_train_test, target_train_test)\n",
    "rf_pred_test = rf_model.predict(features_test)\n",
    "\n",
    "rf_rmse_test = rmse(target_test, rf_pred_test)\n",
    "\n",
    "print(\"Random Forest Regressor Final RMSE :\", rf_rmse_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment</b>\n",
    "\n",
    "The final model was evaluated on the test set.\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "The Random Forest Regressor provided a final score of 44.2on the final test. This is above the threshold of 48 that our client has set. I tested many different models with hyperparameter tuning, however, I was unable to get a test result of 48 or below. I believe I may be able to achieve a better result with a larger test set.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-danger\">\n",
    "<s><b>Reviewer's comment V2</b>\n",
    "\n",
    "Not sure how a larget test set is supposed to help get better results :)\n",
    "    \n",
    "Getting RMSE lower than 48 is a requirement in this project though. One suggestion I have is to refit the final model on the combined train+validation data before evaluating it on the test set. Now you have a gap between train and test data which may be problematic for time series models (especially if there is a growing trend which you observed).\n",
    "\n",
    "</div>\n",
    "\n",
    "Thank you so much, I was absolutely at a loss. That worked! -Johannah "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-success\">\n",
    "<b>Reviewer's comment V3</b>\n",
    "\n",
    "You're welcome! :)\n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Review checklist"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- [x]  Jupyter Notebook is open\n",
    "- [ ]  The code is error-free\n",
    "- [ ]  The cells with the code have been arranged in order of execution\n",
    "- [ ]  The data has been downloaded and prepared\n",
    "- [ ]  The data has been analyzed\n",
    "- [ ]  The model has been trained and hyperparameters have been selected\n",
    "- [ ]  The models have been evaluated. Conclusion has been provided\n",
    "- [ ] *RMSE* for the test set is not more than 48"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}