Data Analysis of FordGo Bikes (SQL).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The Importance of Weekend Trips"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Executive Summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Overview, Recommendations and Impact\n",
    "In this report, I propose a solution to the issue of decreased ridership during the weekends. In this report you will find a dive into the breakdown of ridership by day and the differences in ridership on weekends by subscriber type: customer (24hour/3day pass) or subscriber (30 day/annual pass). \n",
    "\n",
    "Based on these differences, this report proposes the increasing marketing materials on fun weekend get aways, but also substantially proposed creative discounts for future subscription purchases to subscribers who take trips on weekends, delivered via the mobile app. Secondly, it proposes introducing a weekend flex pass that would remove the 30minute ride time cap on weekends and would offer unlimited rides on weekends.\n",
    "\n",
    "Evaluating the potential impact of these proposals, if the company managed to increase ridership levels on weekends to be on par with weekdays, this would correspond over our service, to an additional 200,000 trips."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problem Statement\n",
    "This report finds that there is a significant drop in ridership on weekends as shown by the graph generated below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "Waiting on bqjob_r51ecc6a2fed359f3_0000016d6a9652a5_1 ... (0s) Current status: DONE   "
     ]
    },
    {
     "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": [
    "! bq query --use_legacy_sql=false --format=csv 'SELECT COUNT(*) as num_trips, (EXTRACT(dayofweek FROM start_date)) as day FROM `bigquery-public-data.san_francisco.bikeshare_trips` GROUP BY (EXTRACT(dayofweek FROM start_date))' > daycounts.csv\n",
    "df_week = pd.read_csv('daycounts.csv')\n",
    "df_week = df_week.sort_values(by = ['day'])\n",
    "df_week = df_week.reset_index(drop = True)\n",
    "arr_week = np.array(df_week['num_trips'])\n",
    "\n",
    "\n",
    "#makes graph of rides by weekday\n",
    "N = len(arr_week)\n",
    "bottom = 50000\n",
    "\n",
    "theta = np.linspace(0.0, 2 * np.pi, N, endpoint=False)\n",
    "radii = arr_week\n",
    "width = (2*np.pi) / N\n",
    "\n",
    "ax = plt.subplot(111, polar=True)\n",
    "bars = ax.bar(theta, radii, width=width, bottom=bottom)\n",
    "\n",
    "# Use custom colors and opacity\n",
    "for r, bar in zip(radii, bars):\n",
    "    bar.set_facecolor(plt.cm.jet(r / 300000))\n",
    "    bar.set_alpha(0.8)\n",
    "radians = np.linspace(0, 2 * np.pi, num=7, endpoint=False)\n",
    "ax.set_theta_zero_location('E')\n",
    "    # Set up labels\n",
    "ax.set_xticks(radians)\n",
    "ax.set_xticklabels(['Sa','F', 'Th', 'W', 'Tu', 'M', 'Su'], size=14, weight='bold')\n",
    "ax.set_yticklabels(['0','50', '100', '150'])\n",
    "ax.set_xlabel('Total Number of Bike Trips by Day of Week (In Thousands)')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I also wanted to highlight the most popular stations from which commuters were traveling from to give some background onto how which stations are the most impacted by travel. I've created a measure of popularity that discounts for the amount of time a station has been active."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Waiting on bqjob_r5e92943528d29c1d_0000016d711d2577_1 ... (0s) Current status: DONE   "
     ]
    },
    {
     "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>popularity</th>\n",
       "      <th>station</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>32.651842</td>\n",
       "      <td>San Francisco Caltrain (Townsend at 4th)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25.202156</td>\n",
       "      <td>San Francisco Caltrain 2 (330 Townsend)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>22.010767</td>\n",
       "      <td>Harry Bridges Plaza (Ferry Building)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18.793638</td>\n",
       "      <td>Grant Avenue at Columbus Avenue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18.463645</td>\n",
       "      <td>Embarcadero at Sansome</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   popularity                                   station\n",
       "0   32.651842  San Francisco Caltrain (Townsend at 4th)\n",
       "1   25.202156   San Francisco Caltrain 2 (330 Townsend)\n",
       "2   22.010767      Harry Bridges Plaza (Ferry Building)\n",
       "3   18.793638           Grant Avenue at Columbus Avenue\n",
       "4   18.463645                    Embarcadero at Sansome"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "! bq query --use_legacy_sql=false --format=csv 'SELECT COUNT(*)/DATE_DIFF(CURRENT_DATE(), MAX(installation_date), day) as popularity, max(start_station_name) as station FROM `bigquery-public-data.san_francisco.bikeshare_stations` inner join `bigquery-public-data.san_francisco.bikeshare_trips` on start_station_id = station_id GROUP BY station_id order by popularity desc;' > popularity.csv\n",
    "df_pop = pd.read_csv('popularity.csv')\n",
    "df_pop.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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_trips</th>\n",
       "      <th>day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>51375</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>169937</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>184405</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>180767</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>176908</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>159977</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>60279</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   num_trips  day\n",
       "0      51375    1\n",
       "1     169937    2\n",
       "2     184405    3\n",
       "3     180767    4\n",
       "4     176908    5\n",
       "5     159977    6\n",
       "6      60279    7"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_week\n",
    "#number of trips by day"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you notice here, the number of trips during the weekend is significantly lower than the number of trips on weekdays. This is a big gap in the number of trips taken on these two days. I propose putting incentives in place to increase the amount of weekend trips.\n",
    "\n",
    "Upon further exploration of this phenomenon, I discovered that the breakdown of subscriber type (subscriber: annual or 30day, customer (24 hour or 3 day pass) for trips differs between week days and weekends:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "Waiting on bqjob_r1b5e723e9ac4189b_0000016d6a965acc_1 ... (0s) Current status: DONE   "
     ]
    },
    {
     "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_trips</th>\n",
       "      <th>Weekday</th>\n",
       "      <th>subscriber_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55152</td>\n",
       "      <td>False</td>\n",
       "      <td>Customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>56502</td>\n",
       "      <td>False</td>\n",
       "      <td>Subscriber</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>81657</td>\n",
       "      <td>True</td>\n",
       "      <td>Customer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>790337</td>\n",
       "      <td>True</td>\n",
       "      <td>Subscriber</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   num_trips  Weekday subscriber_type\n",
       "0      55152    False        Customer\n",
       "1      56502    False      Subscriber\n",
       "2      81657     True        Customer\n",
       "3     790337     True      Subscriber"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "! bq query --use_legacy_sql=false --format=csv 'SELECT COUNT(*) as num_trips, (EXTRACT(dayofweek FROM start_date) BETWEEN 2 AND 6) as Weekday, subscriber_type FROM `bigquery-public-data.san_francisco.bikeshare_trips` GROUP BY (EXTRACT(dayofweek FROM start_date) BETWEEN 2 AND 6), subscriber_type' > daycounts_sub.csv\n",
    "df_week_sub = pd.read_csv('daycounts_sub.csv')\n",
    "df_week_sub = df_week_sub.sort_values(by = ['Weekday', 'subscriber_type'])\n",
    "df_week_sub = df_week_sub.reset_index(drop = True)\n",
    "display(df_week_sub)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This lookup table succinctly communicates the difference in number of trips by weekday and subscriber type."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this chart here, you can see a pattern emerge for what kind of people take trips depending on the day of the week. On Weekdays, it's primarily subscribers, but on weekends, it's about 50-50. \n",
    "\n",
    "*Simply put, there is a huge pool of subscribers who seemingly are not as actively using their subscription on weekends as much as they are on weekdays. There is a smaller group of customers on weekdays who could also contribute to trips on weekends.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is for this reason I would recommend that we can increase ridership by targetting commuters and finding ways to get them to ride more on weekends. To do so, we have access to providing discounts, marketing and promotions via the mobile app users use to boost weekend traffic. However, subscribers have already paid for a year in advance or paid for the month. To address the need to mobilize subscribers on weekends requires creative incentive making."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### My proposal is two fold: \n",
    "1) Promote subscriber weekend travel through mobile ads about destinations in their home city explorable on weekends. Provide discounts on the app for monthly subscribers (30 day prepaid) for credits toward next month's subscription fee for taking weekend rides. These discounts would be delivered via the mobile app notification system.\n",
    "\n",
    "2) Introduce a weekend flex pass, marketed heavily toward both weekday and weekend customers. This weekend pass would offer unlimited rides for the weekend only. There is room for debate and further analysis on consumer propensity to purchase passes on what to price the pass, but for the purpose of increasing ridership, this report would advocate for it to be low, but economically sustainable for the company."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Proposal Part 1 would be highly effective. Specifically, targetting commuters during the weekday could result in massive gains to ridership on weekends!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "Waiting on bqjob_r79cfc27b5ef16641_0000016d6a9660a8_1 ... (0s) Current status: DONE   "
     ]
    }
   ],
   "source": [
    "#estimate the number of commuter trips per weekday\n",
    "! bq query --use_legacy_sql=false --format=csv 'SELECT subscriber_type, COUNT(*)/5 as number_of_commuter_trips_per_weekday FROM `bigquery-public-data.san_francisco.bikeshare_trips` WHERE (start_station_id != end_station_id) AND ((EXTRACT(HOUR FROM start_date) BETWEEN 4 and 11) OR (EXTRACT(HOUR FROM start_date) BETWEEN 15 and 21)) AND (EXTRACT(dayofweek FROM start_date) BETWEEN 2 and 6) GROUP BY subscriber_type' > commuterchart.csv\n",
    "df_commuterchart = pd.read_csv('commuterchart.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7fb5abfbd080>"
      ]
     },
     "execution_count": 6,
     "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": [
    "df_commuterchart2= df_commuterchart.set_index('subscriber_type')\n",
    "df_commuterchart2.plot.bar(color = 'green')\n",
    "plt.xlabel('Subscriber Type')\n",
    "plt.ylabel('Count of Commuter Trips Per Weekday')\n",
    "plt.legend()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Total: ~150,000 commuter trips per weekday.\n",
    "The vast majority of commuters in a given weekday are of type subscriber. Proposal 1 hopes to directly target them and increase their presence during the weekend.\n",
    "\n",
    "For proposal part 2, we can see that this pass is particularly impactful given customer ride duration on weekends."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "Waiting on bqjob_rfa1998a4021926_0000016d6a966822_1 ... (0s) Current status: DONE   "
     ]
    }
   ],
   "source": [
    "! bq query --use_legacy_sql=false --format=csv 'SELECT floor((duration_sec/60)/5.00)*5 as bucket_floor, count(*) as counts, MAX(EXTRACT(hour FROM start_date) BETWEEN 2 AND 6) as weekday FROM `bigquery-public-data.san_francisco.bikeshare_trips` group by bucket_floor, (EXTRACT(hour FROM start_date) BETWEEN 2 AND 6) order by bucket_floor' > duration_trip.csv\n",
    "df_duration = pd.read_csv('duration_trip.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_duration['counts'] = np.log(np.array(df_duration['counts']))\n",
    "ax = df_duration.groupby(['bucket_floor','weekday']).mean().unstack().plot.bar()\n",
    "plt.xlabel('Trip Duration (in minutes)')\n",
    "plt.ylabel('Log of Trip Count')\n",
    "arr = np.array([])\n",
    "for i in range(0,251,10):\n",
    "    arr = np.append(arr, [i, ''])\n",
    "L=plt.legend()\n",
    "L.get_texts()[0].set_text('Weekday')\n",
    "L.get_texts()[1].set_text('Weekend')\n",
    "_ = ax.set_xticklabels(arr)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see a clear drop close to the 30 minute mark for ridership. The benefit of a weekend pass is that customers would not be limited to 30 minute rides as they currently are under the 24 hour or 3 day plan. Some customers already spend longer than 30 minutes riding for additional charges, but perhaps removing this charge on weekends would be helpful to increase ridership duration on weekends. This would incentivize longer rides (increased duration of trips) and also promote weekend travel.\n",
    "\n",
    "To quantify the potential upshot from this proposal if both:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Number of Commuter trips = `150,000`\n",
    "\n",
    "Total number of weekend trips (per day): `55,000`\n",
    "\n",
    "If we were able to get the number of weekend trips on a given day to be equal to the number of just commuter trips on a given weekday, by applying these two policy proposals of a weekend pass and discount rates for subscribers who trip on weekends, we would see an increase in ridership of `200,000` \n",
    "\n",
    "100,000 difference per day * 2 weekend days = potentially `200,000` additional total trips.\n",
    "\n",
    "In conclusion, these moderate proposals aim to increase ridership by focusing on improving weekend ridership, a currently massive gap via creative promotion of weekend riding via credits for future subscriptions and a weekend riding pass not capped at 30 minutes.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Appendix Question:\n",
    "\n",
    "A question that has been posed is do we have enough bikes for increased demand (can we handle more trips). Across the board there is rarely a shortage of bikes. This proposal would not burden the system too greatly judging by the availability of bikes across hours.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Waiting on bqjob_r4ce0cdd3cc2ce31_0000016d6a967167_1 ... (0s) Current status: DONE   \n",
      "+-----------------------+------+\n",
      "|      zero_bikes       | hour |\n",
      "+-----------------------+------+\n",
      "|  0.005568870334295351 |    0 |\n",
      "|  0.005793741706456771 |    1 |\n",
      "|  0.005929378869800297 |    2 |\n",
      "|  0.005922113934821312 |    3 |\n",
      "|  0.005850309548545847 |    4 |\n",
      "|  0.006061168710036863 |    5 |\n",
      "|  0.008135347743934457 |    6 |\n",
      "|  0.009372805432640845 |    7 |\n",
      "|   0.01103854455988539 |    8 |\n",
      "|   0.01665494211922165 |    9 |\n",
      "|  0.011365765742764121 |   10 |\n",
      "|  0.007211770829882354 |   11 |\n",
      "|  0.004977557727741395 |   12 |\n",
      "| 0.0038672944001008035 |   13 |\n",
      "|  0.005205108179381264 |   14 |\n",
      "|  0.005517999614381971 |   15 |\n",
      "|  0.007295290355663985 |   16 |\n",
      "|  0.015812084185369005 |   17 |\n",
      "|  0.022024245040427253 |   18 |\n",
      "|  0.011409481925709846 |   19 |\n",
      "|  0.005570904509979374 |   20 |\n",
      "|  0.003279470044843433 |   21 |\n",
      "|  0.003946000812292544 |   22 |\n",
      "|  0.004991617759912132 |   23 |\n",
      "+-----------------------+------+\n"
     ]
    }
   ],
   "source": [
    "! bq query --use_legacy_sql=false 'SELECT * FROM(SELECT COUNT(CASE WHEN (bikes_available = 0 and docks_available >1) THEN 1 END)/COUNT(CASE WHEN (docks_available >1) THEN 1 END) as zero_bikes, EXTRACT(hour FROM time) as hour FROM `bigquery-public-data.san_francisco.bikeshare_status` GROUP BY EXTRACT(hour FROM time)) ORDER BY hour'\n",
    "#zero_bikes is the count of times when 0 bikes are available at stations.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}