mc51/kvb_part3_descriptives.ipynb

## kvb_part3_descriptives.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "+ title: Investigating bike rentals in Cologne - Part 3: Descriptive Statistics\n",
    "+ date: 2018-02-04\n",
    "+ modified: 2018-04-01\n",
    "+ Series: Cologne bike rentals\n",
    "+ tags: python, pandas, matplotlib, kvb, statistics\n",
    "+ Slug: cologne-bike-rentals-statistics\n",
    "+ Category: Python\n",
    "+ Authors: MC \n",
    "+ Summary: In this multi part post, I will take a look at the bike rental system in Cologne. Following up on the previous work we start digging into the data using descriptive statistics. We look at some interesting findings and visualize them using matplotlib and seaborn. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After [gathering]({filename}/kvb_part1.md) and cleaning data on rental bikes in Cologne, we already looked at [map visualizations]({filename}/kvb_part2.md). Here, we will dive into the data more deeply. Using descriptive statistics we will try to find interesting patterns. For this, we will use the [pandas](https://pandas.pydata.org/) library. It is great for manipulating and analyzing data.  Moreover, we will use [matplotlib](https://matplotlib.org/) and [seaborn](https://seaborn.pydata.org) to visualize some of our findings.\n",
    "\n",
    "You can find the data used in this post [here](https://data.world/mc51/kvb-bike-locations-from-2017-02-05-to-03-06-10min-interval). This post was written in a jupyter notebook. You can also find it at [github](https://gist.github.com/mc51/f939d4ae13dede7d370221bd54a2af4b).\n",
    "<br>\n",
    "Let's get started by setting up things:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from geopy.distance import vincenty\n",
    "import warnings\n",
    "warnings.simplefilter('ignore')\n",
    "# define style of plots\n",
    "%matplotlib inline\n",
    "plt.style.use('ggplot')\n",
    "plt.rcParams[\"figure.figsize\"] = (10,8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After setting up the environment, we read in the data and drop some unnecessary columns. Following, we take a first glance at the structure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "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>bike_id</th>\n",
       "      <th>scrape_weekday</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>scrape_datetime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22460</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.933528</td>\n",
       "      <td>6.995009</td>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22134</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.896996</td>\n",
       "      <td>6.980310</td>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21911</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.967579</td>\n",
       "      <td>6.936491</td>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21522</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.975012</td>\n",
       "      <td>7.007270</td>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21914</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.970199</td>\n",
       "      <td>6.900565</td>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   bike_id scrape_weekday        lat       lon     scrape_datetime\n",
       "0    22460            Sun  50.933528  6.995009 2017-02-05 00:00:02\n",
       "1    22134            Sun  50.896996  6.980310 2017-02-05 00:00:02\n",
       "2    21911            Sun  50.967579  6.936491 2017-02-05 00:00:02\n",
       "3    21522            Sun  50.975012  7.007270 2017-02-05 00:00:02\n",
       "4    21914            Sun  50.970199  6.900565 2017-02-05 00:00:02"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "BIKES_COMBINED = \"../data/bikedata_10min-int-observ_Feb-March.csv\"\n",
    "bikes = pd.read_csv(BIKES_COMBINED)\n",
    "bikes.drop(['u_id','bike_name'], axis=1, inplace=True)\n",
    "bikes.scrape_datetime = pd.to_datetime(bikes.scrape_datetime)\n",
    "bikes.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have one observation for all bike locations (lat and lon) every ten minutes. The bikes can be identified by their `bike_id`.\n",
    "If a `bike_id` is missing at one observation point it means that the bike is currently **not** available. This is the case when a bike is in use. Alternatively, the bike (or the GPS) could be broken or under maintenance. This structure has a downside. Because the number of rows per `bike_id` varies, it's impossible to work with averages. Thus, we make it more consistent by adjusting the number of rows / observations for each `bike_id`:"
   ]
  },
  {
   "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>scrape_datetime</th>\n",
       "      <th>bike_id</th>\n",
       "      <th>scrape_weekday</th>\n",
       "      <th>lat</th>\n",
       "      <th>lon</th>\n",
       "      <th>obs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "      <td>21004</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.954703</td>\n",
       "      <td>6.889604</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "      <td>21005</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "      <td>21006</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "      <td>21007</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.909306</td>\n",
       "      <td>6.944799</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-02-05 00:00:02</td>\n",
       "      <td>21008</td>\n",
       "      <td>Sun</td>\n",
       "      <td>50.941107</td>\n",
       "      <td>6.997360</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      scrape_datetime  bike_id scrape_weekday        lat       lon  obs\n",
       "0 2017-02-05 00:00:02    21004            Sun  50.954703  6.889604  1.0\n",
       "1 2017-02-05 00:00:02    21005           None        NaN       NaN  NaN\n",
       "2 2017-02-05 00:00:02    21006           None        NaN       NaN  NaN\n",
       "3 2017-02-05 00:00:02    21007            Sun  50.909306  6.944799  1.0\n",
       "4 2017-02-05 00:00:02    21008            Sun  50.941107  6.997360  1.0"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Same number of rows for each bike_id -> Align number of observations\n",
    "bikes = bikes.assign(obs=1).set_index(['scrape_datetime','bike_id']).\\\n",
    "                            unstack().stack(dropna=False).reset_index()\n",
    "bikes.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In result, we have a row for each `bike_id` for every `scrape_datetime`. If the bike was really observed at this time we have `obs = 1`. Otherwise, we know the bike was not actually available.  \n",
    "If between two (actual) observation points the coordinates have changed, the bike has been used. Accordingly, we can create a dummy variable to indicate this. We do this by grouping the data by `bike_id` and applying a function to every non-empty row in the `lat` column. The function checks if the difference between the current and previous row (i.e. the current and previous observation) is greater than zero:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute dummy (0/1) for location changes:\n",
    "# did coordinates change between two observations? \n",
    "bikes['change'] = bikes[bikes.obs==1].groupby('bike_id')['lat'].\\\n",
    "                        apply(lambda x: abs(x.diff()) > 0).astype(int)\n",
    "bikes.loc[bikes.change.isnull(), \"change\"] = 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, that we know a bike was used let's find out what distance it traveled. Because we only have the start and end point of the journey this will be a rough estimate. We can expect the actual distance to be significantly longer. To achieve this, first we create a new column. It contains the coordinates shifted by one observation period for each bike. Then, we create the index `m` that includes all rows in our data where the bike has been used. Using this, we can compute the distance between the actual coordinates and the shifted coordinates. This is done by applying the `vincenty` function from the `geopy` package to each row in `m`:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute Trip Distance when coords changed \n",
    "bikes['coords'] = bikes[['lat','lon']].values.tolist() \n",
    "bikes['coords_shifted'] =  bikes[bikes.obs==1].groupby('bike_id')['coords'].shift()\n",
    "m = bikes['coords_shifted'].notnull() & bikes['change'] == 1\n",
    "\n",
    "bikes.loc[m, 'trip_distance'] = bikes.loc[m, :].apply(\n",
    "    lambda x: vincenty(x['coords'], x['coords_shifted']).kilometers, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualizing will give us a good impression about the typical distance of trips. We do this by using `seaborn` to draw a histogram:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'Histogram of bike trip distances')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmkAAAH0CAYAAABmYbrSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzt3XtUVXXC//HPAQTFI3jgmHfLa94gbPCWI1CSZeVkarenmjRbXfDJSWtWZTm5pjSaIszCLEu6mE85ZrScx2okCiptJBErNcHUnpx0EA5eCJWR8/390c8zESDghfM9+n6t5VqczT57f/f5kr7bex+OwxhjBAAAAKsE+XsAAAAAqI1IAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaUAzO++88/T444/7exgB5ejRo7rtttsUHR0th8OhTz75pNY6O3fulMPh0GeffXbcbTkcDi1ZsqTex/4we/Zs9erV67Rse9KkSUpOTq73MQB7EWnAKXC8f/h+HQH5+fmaPn16o7b72WefyeFwaOfOnadimAHrnXfe0dKlS7Vy5Urt3r1bF1100Qlva/fu3Zo4ceIpHF1tu3btqjcm63L//ffriy++OK1jOubZZ5/VX//610avHxISoldfffX0DQhAvUL8PQDgbNOuXTt/D6FeVVVVCg0N9fcwaikuLlbnzp1PKs6O6dChwykY0anh9XpljJHT6ZTT6WyWfUZGRjbLfgCcPM6kAc3s15c733vvPQ0aNEjh4eFq27athgwZog0bNmjnzp0aOXKkJKl79+5yOBxKSkqSJBlj9PTTT6tHjx4KDQ1Vz549NW/evBr7KSsr07XXXqvWrVurffv2mjVrlm699dYaZ/ySkpI0ZcoUzZo1Sx07dlTnzp0lSUuXLtXQoUMVGRkpt9utK6+8UkVFRb7nHbu0uHTpUl122WUKDw9X3759lZubq3/+85+64oor1Lp1a/Xv31+ffvrpcV+Pho4lKSlJs2bN0vbt2+VwOHTeeecdd3s7duzQqFGj1KpVK3Xv3l1vvvlmje83dHlzyZIlcjqdevvtt33L3nrrLcXFxally5Y677zzNGPGDP3000/1bqNr166SpIsvvrjGmI9d1nz77bfVt29fhYaGasuWLbUudx57vHTpUvXo0UMtW7ZUcnKyduzYcdxjLy8v1/XXX++b80ceeUS//lCZX5/13bRpky677DK1bdtWrVu3Vr9+/fTGG29I+vlntbq6WpMnT5bD4ZDD4fDt5+abb1a3bt3UqlUrnX/++UpLS6uxr2P7eemll3TuuecqIiJCV199tfbu3VtjPNnZ2Ro5cqTCw8MVGRmpxMREfffdd41+7T/77DONGDFCbdq0UZs2bXTBBRfoww8/PO7rBAQMA+Ck3XrrrWbUqFF1fk+SeeONN3yPzz33XPPYY48ZY4zZvXu3adGihXnyySfN9u3bzebNm82bb75pvvrqK3P06FHz3nvvGUlm3bp1Zvfu3aasrMwYY8zzzz9vWrZsaV588UVTVFRkXnjhBRMWFmZefvll337Gjh1revfubXJycsw333xjJk2aZCIiImqMMzEx0TidTnPnnXeaTZs2ma+++soYY8zixYvNypUrzbZt20xBQYEZO3as6dWrlzly5IgxxpgdO3YYSaZHjx7m3XffNVu3bjXjxo0zHTt2NKNGjTIrVqwwW7duNePHjzddunQxVVVV9b52DR1LWVmZue+++8x5551ndu/ebUpKSurczrExdezY0SxZssR8++235uGHHzYOh8Pk5+fXOx+/fPyXv/zFREZGmuzsbN/3MzMzTdu2bc3rr79uvvvuO5Obm2tiYmLMzTffXO8xFRQUGEnmnXfeqTHmRx991LRq1cokJCSYtWvXmq1bt5oDBw6YRx991PTs2dP3/EcffdSEh4ebESNGmHXr1pl169aZIUOGmNjYWOP1euvd77hx40zPnj3NRx99ZL755htz0003mTZt2tSY81//rMbExJgbb7zRbNq0yXz33Xdm1apVZuXKlcYYY0pKSkxwcLCZN2+e2b17t9m9e7cx5uef29TUVLN+/Xqzfft288Ybb5jWrVubxYsX19hPRESEueGGG8zXX39tPv/8c9OtWzfz+9//3rfO6tWrTVBQkPnDH/5gCgsLzZYtW8zLL79stmzZ0qjX/ujRo8blcpnp06eboqIiU1RUZFasWGHy8vLqfY2AQEKkAafArbfeaoKDg03r1q1r/TlepB37x3zHjh11bvfTTz+t8/tdunQxf/zjH2ssu/fee0337t2NMcYUFRUZSTVio6qqynTp0qVWpPXu3dtUV1cf9/jKysqMJPPZZ58ZY/4TROnp6b511q1bZySZp59+2rfs2PF9/fXX9W67oWMxxtSKmLocG9MjjzxSY/nw4cPNTTfd5HtcV6S99tprZtq0aaZDhw5mw4YNNZ5/7rnnmhdeeKHGstzcXCPJeDyeOsfyww8/GEnm448/rrH80UcfNQ6Hw3z//fe1lv860iSZ4uJi37KtW7caSWb16tV17rO4uNhIMn//+999y44cOWI6dep03EiLiIgwmZmZdW7TGGOCg4OP+/1jpk2bZpKTk2vsx+12m8OHD/uWPfHEE6ZDhw6+x7/97W/NlVdeWe82G3rtPR5Pna8zcKbgcidwigwdOlSFhYW1/hxPbGysLrvsMg0cOFDXXHONnn32Wf3www/Hfc6BAwe0a9cuJSQk1FiemJionTt3qrKyUps3b5YkDRs2zPf9Fi1aKD4+vtb2fvOb3ygoqOZfBYWFhbrmmmvUvXt3tWnTRt26dZMkff/99zXWu+CCC3xfH7vXKzY2ttaykpKSEz6Wpho+fHiNxyNGjPC9HvV55JFHtHTpUq1Zs0ZxcXG+5Xv37tX333+vGTNm+O4bczqdGjNmjCRp27ZtTR5f+/btfa/n8bRr167GJdA+ffrI7XbXeyzHlv/yvr3Q0FANHjz4uPu5//77dfvttyspKUmzZ89WQUFBg2Pzer1KTU1VXFyc3G63nE6nFi5cWOvno1+/fgoLC/M97ty5s/71r3/5Hq9fv16jR4+ucx+Nee1dLpduv/12XXbZZRozZoxSU1O1devWBscPBAoiDThFWrVqpV69etX6czzBwcF6//33lZOTo8GDB+udd95Rnz599Le//a3B/R27P+gY86t7j+papy6tW7eu8biyslKjR4+Ww+HQ4sWLtW7dOuXn58vhcKiqqqrGui1atKi1r7qWeb3ekz6WE9WYbSUnJ6uysrLGfWjSf8b97LPP1gjvjRs3qri4WDExMU0ez69f76Y43rGc6Gs2a9YsFRUV6brrrtM333yjYcOG6ZFHHjnuc9LS0vTEE0/onnvu0erVq1VYWKjbb7+91s/Hr9+E4nA4ao2zvp/Rxr72ixYt0vr163XppZcqNzdXAwcO1Isvvtik1wCwFZEG+JnD4dCQIUM0c+ZM5eXlKTExUZmZmZL+849cdXW1b/2IiAh16dJFubm5NbaTl5en7t27Kzw8XP3795ckrV271vf9o0ePav369Q2OZ8uWLdq7d6/mzJmjiy++WP369VN5efkpDadjGnMsTfXrX2Wxdu1a9evX77jPueSSS7Rq1So9/vjjeuyxx3zL27dvr65du2rr1q11BnjLli3r3F5d89ZUe/furXEDfVFRkcrKyuo9lgEDBkiS1qxZ41tWVVWl/Pz8BvfVo0cPpaSkaPny5frzn/+sF154wfe90NDQWseRl5enyy+/XFOmTNGgQYPUq1cvFRcXN+n4pJ/P4tZ3k39TXvuBAwdqxowZev/99zVlyhS99NJLTR4LYCN+BQfgR2vWrNFHH32k0aNHq2PHjiouLtZXX32lKVOmSJLOPfdcBQUFadWqVbr++usVFhamyMhIPfTQQ7rvvvvUu3dvJSUlKScnRy+88IIyMjIkSb1799bYsWM1depUvfjii2rXrp3S0tJ04MCBBs+unXvuuQoLC9Nzzz2n++67Tzt37tSDDz7YqLNyJ6KhY2mqV155RX379lV8fLyWLFmitWvX1nrna10SExP14YcfasyYMaqqqvLF2pw5czRlyhS1bdtW48aNU4sWLbRlyxa9//779Z6xOXYJ8O9//7sGDBigsLAwuVyuJh1HeHi4Jk+erPT0dBljdM899ygmJqbe38fXq1cv/e53v/PNefv27ZWamqqDBw/Wu4+Kigo98MADmjBhgrp37659+/bpgw8+8EW+9PM7iz/++GONGTNGoaGhcrvdOv/88/XGG2/o448/VufOnfX666/rH//4R5OPcdasWRozZozuvfde3XbbbQoLC9PatWs1fPhwnX/++Q2+9tu2bdOiRYs0duxYde3aVT/++KM+/fRTXXjhhU0aB2ArzqQBfhQZGam1a9fq6quvVu/evXXbbbfppptu0qxZsyT9fDbhiSeeUGpqqjp27Kirr75aknT33Xfrz3/+s+bOnav+/fvrySefVGpqqi/uJCkzM1MDBw7UmDFjlJSUpM6dO+vSSy+t9+zPMW63W0uWLNHq1as1YMAA3X///Xr66adr3bd2qjTmWJoiNTVVL730kmJjY/X666/rtddea/C+rGNGjBih1atX67nnntODDz4oSbrlllu0bNky/e///q+GDBmiwYMHa/bs2b5fV1KXoKAgZWRkaNmyZeratasGDRrU5OPo2LGj7rjjDk2YMEEjRoxQq1at9O677x43lhcvXqy4uDhdddVVSkxMVOfOnXXNNdfUu35ISIjKy8s1ZcoU9evXT5dddpnat2+vpUuX+tZJS0vT+vXr1b17d9/v+Js1a5YSExN19dVXa/jw4SovL9e0adOafIyjR4/WqlWr9I9//ENDhw7VkCFD9Nprr/kumTf02rdu3VrFxcW64YYb1KdPH02YMEEXXXSRnn/++SaPBbCRw5yOaxgArFNdXa2+ffvqd7/7ndLS0vw9HBzH7NmztWTJkhN6YwKAMweXO4EzVF5enkpKSjRo0CAdPHhQ6enp2rlzpyZNmuTvoQEAGoFIA85Q1dXVevzxx7Vt2za1aNFCAwcO1Mcff3xC70gEADQ/LncCAABYiDcOAAAAWIhIAwAAsBCRBgAAYKGAfuPAjz/+6O8h4AS43W6Vlpb6exg4QcxfYGP+AhdzF9g6derU5OdwJg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAuF+HsAJ8Ob90Gz7Cco4fJm2Q8AAMAxnEkDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFgopKEVSktLlZGRoX379snhcCg5OVlXXHGFKioqlJ6err1796pdu3aaPn26nE6njDHKzMzUhg0bFBYWppSUFPXo0UOS9Mknn2jFihWSpPHjxyspKUmStH37dmVkZKiqqkqDBg3S5MmT5XA4Tt9RAwAAWK7BM2nBwcG65ZZblJ6erjlz5ujDDz/Url27lJWVpZiYGM2fP18xMTHKysqSJG3YsEF79uzR/Pnzdccdd+jll1+WJFVUVGj58uWaO3eu5s6dq+XLl6uiokKStGjRIt15552aP3++9uzZo8LCwtN4yAAAAPZrMNJcLpfvTFirVq3UuXNneTwe5efnKzExUZKUmJio/Px8SdKXX36phIQEORwO9enTRz/99JPKy8tVWFio2NhYOZ1OOZ1OxcbGqrCwUOXl5Tp06JD69Okjh8OhhIQE37YAAADOVk26J62kpEQ7duxQr169tH//frlcLkk/h9yBAwckSR6PR2632/ec6OhoeTweeTweRUdH+5ZHRUXVufzY+gAAAGezBu9JO+bw4cNKS0vTpEmTFB4eXu96xphay+q7v8zhcNS5fn2ys7OVnZ0tSUpNTZXT6Wz0c09G+C+iEycvJCSkRsgjsDB/gY35C1zM3dmnUZF29OhRpaWlaeTIkRo6dKgkKTIyUuXl5XK5XCovL1dERISkn8+ElZaW+p5bVlYml8ulqKgobd682bfc4/Gof//+io6OVllZWY31o6Ki6hxHcnKykpOTfY+P3dN2ulX+4nhw8txud42fEQQW5i+wMX+Bi7kLbJ06dWrycxq83GmM0cKFC9W5c2ddddVVvuXx8fHKzc2VJOXm5mrw4MG+5Xl5eTLGqKioSOHh4XK5XIqLi9PGjRtVUVGhiooKbdy4UXFxcXK5XGrVqpWKiopkjFFeXp7i4+ObfCAAAABnkgbPpG3dulV5eXnq1q2b/vjHP0qSbrzxRo0bN07p6enKycmR2+3WjBkzJEmDBg1SQUGBpk2bptDQUKWkpEiSnE6nJkyYoIceekiSNHHiRN/lyttvv10LFixQVVWV4uLiNGjQoNNysAAAAIHCYZpyU5hldr21uFn2E5RwebPs52zBKfvAxvwFNuYvcDF3ge20XO4EAABA8yPSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALBTi7wEEAm/eB826v6CEy5t1fwAAwD6cSQMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYKKShFRYsWKCCggJFRkYqLS1NkrRs2TJ99NFHioiIkCTdeOONuvDCCyVJ7777rnJychQUFKTJkycrLi5OklRYWKjMzEx5vV6NGjVK48aNkySVlJRo3rx5qqioUPfu3XXPPfcoJKTBYQEAAJzRGjyTlpSUpJkzZ9ZafuWVV+qpp57SU0895Qu0Xbt2ac2aNXrmmWf08MMP65VXXpHX65XX69Urr7yimTNnKj09XZ9//rl27dolSVqyZImuvPJKzZ8/X61bt1ZOTs4pPkQAAIDA02Ck9e/fX06ns1Eby8/P10UXXaQWLVronHPOUYcOHbRt2zZt27ZNHTp0UPv27RUSEqKLLrpI+fn5MsZo06ZNGjZsmKSfgzA/P//kjggAAOAMcMLXFT/88EPl5eWpR48e+v3vfy+n0ymPx6PevXv71omKipLH45EkRUdH+5ZHR0eruLhYBw8eVHh4uIKDg2utDwAAcDY7oUgbPXq0Jk6cKEl6++239frrryslJUXGmDrXr2u5w+Fo8n6zs7OVnZ0tSUpNTW30Gb5AE+52+3sIp1VISIjcZ/gxnsmYv8DG/AUu5u7sc0KR1rZtW9/Xo0aN0pNPPinp5zNkZWVlvu95PB5FRUVJUo3lZWVlcrlcatOmjSorK1VdXa3g4OAa69clOTlZycnJvscVFRUnMnzrVZaW+nsIp5Xb7VbpGX6MZzLmL7Axf4GLuQtsnTp1avJzTuhXcJSXl/u+Xrdunbp27SpJio+P15o1a/Tvf/9bJSUl2r17t3r16qWePXtq9+7dKikp0dGjR7VmzRrFx8fL4XBowIAB+uKLLyRJn3zyieLj409kSAAAAGeUBs+kzZs3T5s3b9bBgwd111136brrrtOmTZu0c+dOORwOtWvXTnfccYckqWvXrho+fLhmzJihoKAgTZkyRUFBP3fgbbfdpjlz5sjr9eriiy/2hd1NN92kefPm6a233lL37t11ySWXnMbDBQAACAwOU9+NZAFg11uL/T2E0yIo4XJ/D+G04pR9YGP+AhvzF7iYu8DWbJc7AQAAcHoRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhUL8PQDU5s37oFn3F5RwebPuDwAANIwzaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALBTS0AoLFixQQUGBIiMjlZaWJkmqqKhQenq69u7dq3bt2mn69OlyOp0yxigzM1MbNmxQWFiYUlJS1KNHD0nSJ598ohUrVkiSxo8fr6SkJEnS9u3blZGRoaqqKg0aNEiTJ0+Ww+E4TYcLAAAQGBo8k5aUlKSZM2fWWJaVlaWYmBjNnz9fMTExysrKkiRt2LBBe/bs0fz583XHHXfo5ZdflvRz1C1fvlxz587V3LlztXz5clVUVEiSFi1apDvvvFPz58/Xnj17VFhYeKqPEQAAIOA0GGn9+/eX0+mssSw/P1+JiYmSpMTEROXn50uSvvzySyUkJMjhcKhPnz766aefVF5ersLCQsXGxsrpdMrpdCo2NlaFhYUqLy/XoUOH1KdPHzkcDiUkJPi2BQAAcDY7oXvS9u/fL5fLJUlyuVw6cOCAJMnj8cjtdvvWi46OlsfjkcfjUXR0tG95VFRUncuPrQ8AAHC2a/CetKYwxtRaVt/9ZQ6Ho871jyc7O1vZ2dmSpNTU1Fpn+HBiwn8R1s0hJCSkRswjsDB/gY35C1zM3dnnhCItMjJS5eXlcrlcKi8vV0REhKSfz4SVlpb61isrK5PL5VJUVJQ2b97sW+7xeNS/f39FR0errKysxvpRUVH17jc5OVnJycm+x8fua8PJqfzFnDUHt9td4+cEgYX5C2zMX+Bi7gJbp06dmvycE7rcGR8fr9zcXElSbm6uBg8e7Fuel5cnY4yKiooUHh4ul8uluLg4bdy4URUVFaqoqNDGjRsVFxcnl8ulVq1aqaioSMYY5eXlKT4+/kSGBAAAcEZp8EzavHnztHnzZh08eFB33XWXrrvuOo0bN07p6enKycmR2+3WjBkzJEmDBg1SQUGBpk2bptDQUKWkpEiSnE6nJkyYoIceekiSNHHiRN+lyttvv10LFixQVVWV4uLiNGjQoNN1rAAAAAHDYZp6Y5hFdr212N9DOCMEJVzerPvjlH1gY/4CG/MXuJi7wNZslzsBAABwehFpAAAAFiLSAAAALESkAQAAWIhIAwAAsNAp/cQBBCZv3gfNu8PxNzfv/gAACECcSQMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAuF+HsAOPtU/j1L3oqKZtlXUMLlzbIfAABONc6kAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACwUcjJPnjp1qlq2bKmgoCAFBwcrNTVVFRUVSk9P1969e9WuXTtNnz5dTqdTxhhlZmZqw4YNCgsLU0pKinr06CFJ+uSTT7RixQpJ0vjx45WUlHTSBwYAABDITirSJOnRRx9VRESE73FF+hmXAAAP2ElEQVRWVpZiYmI0btw4ZWVlKSsrSzfffLM2bNigPXv2aP78+SouLtbLL7+suXPnqqKiQsuXL1dqaqok6cEHH1R8fLycTufJDg0AACBgnfLLnfn5+UpMTJQkJSYmKj8/X5L05ZdfKiEhQQ6HQ3369NFPP/2k8vJyFRYWKjY2Vk6nU06nU7GxsSosLDzVwwIAAAgoJ30mbc6cOZKkSy+9VMnJydq/f79cLpckyeVy6cCBA5Ikj8cjt9vte150dLQ8Ho88Ho+io6N9y6OiouTxeE52WAAAAAHtpCLtscceU1RUlPbv36/HH39cnTp1qnddY0ytZQ6Ho85161uenZ2t7OxsSVJqaiqXRANUcFBw881dwWfNs5//L3z0uGbdnz+EhITU+B8uBBbmL3Axd2efk4q0qKgoSVJkZKQGDx6sbdu2KTIyUuXl5XK5XCovL/fdrxYdHa3S0lLfc8vKyuRyuRQVFaXNmzf7lns8HvXv37/O/SUnJys5Odn3uKKi4mSGDz9xOp1n7NxV/uJn/Ezldrtr/LeMwML8BS7mLrAd70RWfU74nrTDhw/r0KFDvq+/+uordevWTfHx8crNzZUk5ebmavDgwZKk+Ph45eXlyRijoqIihYeHy+VyKS4uThs3blRFRYUqKiq0ceNGxcXFneiwAAAAzggnfCZt//79evrppyVJ1dXV+u1vf6u4uDj17NlT6enpysnJkdvt1owZMyRJgwYNUkFBgaZNm6bQ0FClpKRI+vmsyoQJE/TQQw9JkiZOnMhlTAAAcNZzmLpuFgsQu95a7O8h4AScyZc7gxIu9/cQTjsuuQQ25i9wMXeBrVkvdwIAAOD0IdIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALBQiL8HAJxJvHkfNOv+ghIub9b9AQCaD2fSAAAALESkAQAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAsRaQAAABYi0gAAACxEpAEAAFiISAMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQiH+HgCAE+fN+6BZ9xeUcHmz7g8AzmacSQMAALAQkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALBTi7wEACBzevA9U6XTKW1HRLPsLSri8WfYDADbiTBoAAICFiDQAAAALEWkAAAAWItIAAAAsRKQBAABYiEgDAACwEJEGAABgISINAADAQkQaAACAhYg0AAAACxFpAAAAFiLSAAAALMQHrAOwljfvg2bbFx/mDsA2nEkDAACwEJEGAABgISINAADAQkQaAACAhXjjAACoed+kIPFGBQAN40waAACAhYg0AAAACxFpAAAAFiLSAAAALMQbBwDAD3ijAoCGEGkAcBY4FoWVTqe8FRWnfX9EIXDyrIm0wsJCZWZmyuv1atSoURo3bpy/hwQAOEF87ipw8qy4J83r9eqVV17RzJkzlZ6ers8//1y7du3y97AAAAD8xoozadu2bVOHDh3Uvn17SdJFF12k/Px8denSxc8jAwDYjvv7cKayItI8Ho+io6N9j6Ojo1VcXOzHEQEAULfmjsJjmut+wuZE8B6fFZFmjKm1zOFw1FqWnZ2t7OxsSVJqaqq63HDbaR8bTo+2/h4ATgrzF9iYv8DF3J1drLgnLTo6WmVlZb7HZWVlcrlctdZLTk5WamqqUlNT9eCDDzbnEHEKMXeBjfkLbMxf4GLuAtuJzJ8VkdazZ0/t3r1bJSUlOnr0qNasWaP4+Hh/DwsAAMBvrLjcGRwcrNtuu01z5syR1+vVxRdfrK5du/p7WAAAAH5jRaRJ0oUXXqgLL7yw0esnJyefxtHgdGLuAhvzF9iYv8DF3AW2E5k/h6nrrn0AAAD4lRX3pAEAAKAmay53NhYfHxW4SktLlZGRoX379snhcCg5OVlXXHGFv4eFJvB6vXrwwQcVFRXFO80CzE8//aSFCxfqhx9+kMPh0N13360+ffr4e1hopL/97W/KycmRw+FQ165dlZKSotDQUH8PC/VYsGCBCgoKFBkZqbS0NElSRUWF0tPTtXfvXrVr107Tp0+X0+k87nYC6kwaHx8V2IKDg3XLLbcoPT1dc+bM0Ycffsj8BZhVq1apc+fO/h4GTkBmZqbi4uI0b948PfXUU8xjAPF4PHr//feVmpqqtLQ0eb1erVmzxt/DwnEkJSVp5syZNZZlZWUpJiZG8+fPV0xMjLKyshrcTkBF2i8/PiokJMT38VEIDC6XSz169JAktWrVSp07d5bH4/HzqNBYZWVlKigo0KhRo/w9FDRRZWWltmzZoksuuUSSFBISotatW/t5VGgKr9erqqoqVVdXq6qqqs7fJQp79O/fv9ZZsvz8fCUmJkqSEhMTG9UvAXW5k4+POnOUlJRox44d6tWrl7+HgkZ69dVXdfPNN+vQoUP+HgqaqKSkRBEREVqwYIG+//579ejRQ5MmTVLLli39PTQ0QlRUlMaOHau7775boaGhuuCCC3TBBRf4e1hoov379/vi2uVy6cCBAw0+J6DOpDX246Ngt8OHDystLU2TJk1SeHi4v4eDRli/fr0iIyN9Z0IRWKqrq7Vjxw6NHj1af/nLXxQWFtaoSy2wQ0VFhfLz85WRkaEXX3xRhw8fVl5enr+HhWYQUJHW2I+Pgr2OHj2qtLQ0jRw5UkOHDvX3cNBIW7du1ZdffqmpU6dq3rx5+uabbzR//nx/DwuNFB0drejoaPXu3VuSNGzYMO3YscPPo0Jjff311zrnnHMUERGhkJAQDR06VEVFRf4eFpooMjJS5eXlkqTy8nJFREQ0+JyAijQ+PiqwGWO0cOFCde7cWVdddZW/h4Mm+K//+i8tXLhQGRkZuvfeezVw4EBNmzbN38NCI7Vt21bR0dH68ccfJf38j36XLl38PCo0ltvtVnFxsY4cOSJjjL7++mve+BGA4uPjlZubK0nKzc3V4MGDG3xOwP0y24KCAr322mu+j48aP368v4eERvr222/1pz/9Sd26dfNdpr7xxhub9EkT8L9NmzZp5cqV/AqOALNz504tXLhQR48e1TnnnKOUlJQG3/4Peyxbtkxr1qxRcHCwzjvvPN11111q0aKFv4eFesybN0+bN2/WwYMHFRkZqeuuu06DBw9Wenq6SktL5Xa7NWPGjAb/Gwy4SAMAADgbBNTlTgAAgLMFkQYAAGAhIg0AAMBCRBoAAICFiDQAAAALEWkArFRaWqpbbrlFXq/3lG536tSp+uqrryRJK1as0MKFC0/p9gHgVCHSAPjFL2OpLm63W2+88YaCgk7fX1Pjx4/XXXfd1eB6s2fP1kcffXTaxgEAdSHSAFinurra30MAAL8L8fcAAJx9nnvuOZWWlurJJ59UUFCQJk6cqDfffFN33XWX/vrXv+qcc87R1KlT9d///d/6n//5HwUHB2v27Nnq06ePvv76a/34448aMGBAo35rfl5ent566y0dPny41seRLVu2THv27NG0adNUVVWlhQsXqrCwUF6vVx07dtQDDzyg999/X1u2bFFxcbFeffVVJSUlacqUKcrMzNS6detUWVmpDh06aNKkSerXr59vu7t27VJoaKjWrVsnt9utqVOnqmfPnpJ+vpT76quvasuWLTLGaMSIEZoyZYokKScnRytXrtS+ffvUq1cv3XHHHWrXrt1pmAUAtiPSADS7e+65R99++63uvPNOxcbGqqSkRG+++aY2b96s9PR0BQUFad++fbWel5ubq4cffljnnHOOnn/+eS1evPi4nyG6a9cuLVq0SA899JB69+6tpUuXqqysrM51c3NzVVlZqRdeeEEtWrTQzp07FRoaqhtvvFFbt27VyJEjNWrUKN/6PXv21MSJExUeHq5Vq1bpmWeeUUZGhkJDQyVJ69ev13333aeUlBS99dZbWrx4sebMmSOv16snn3xSAwYMUEZGhoKCgrR9+3ZJ0rp16/Tuu+/qgQceUMeOHZWVlaVnn31Wjz/++Mm83AACFJc7AVjj2muvVcuWLX2h82sJCQnq1q2bWrZsqRtuuEFr16497hsLvvjiC/3mN79R//791aJFC11//fW+z439teDgYFVUVGjPnj0KCgpSjx49FB4eXu+2ExIS1KZNGwUHB2vs2LE6evSo7wPMJalv37668MILFRQUpISEBO3cuVOStG3bNnk8Ht1yyy2+Y+3bt68kKTs7W9dcc426dOmi4OBgXXPNNdq5c6f27t3b0EsH4AzEmTQA1oiOjm70991ut6qrq3XgwAG1bdu2zvU9Hk+N57Rs2VJt2rSpc92EhASVlZVp3rx5qqys1MiRI3XDDTcoJKTuvyZXrlypnJwceTweORwOHTp0SAcPHvR9PzIy0vd1aGio/v3vf6u6ulqlpaVq166dgoODa21z7969yszM1Ouvv+5bZoyRx+PhkidwFiLSAFijvrNcx/zyUmVpaamCg4MVERFR7/oul0v//Oc/fY+PHDlSI6R+KSQkRNdee62uvfZalZSU6IknnlCnTp10ySWX1BrXli1b9N577+lPf/qTunTpoqCgIE2ePFnGmAaP0e12q7S0VNXV1bVCze12a/z48Ro5cmSD2wFw5uNyJwC/aNu2rUpKSpr0nE8//VS7du3SkSNHtGzZMg0bNuy4v6Jj2LBhWr9+vb799lsdPXpUb7/9dr0h9c033+j//u//5PV6FR4erpCQEN+2IyMj9a9//cu37qFDh3yB6PV6tXz5clVWVjbqGHr16iWXy6U333xThw8fVlVVlb799ltJ0qWXXqqsrCz98MMPkqTKykqtXbu2UdsFcObhTBoAvxg3bpwWL16sJUuWaPz48Y16TkJCgjIyMvTjjz+qX79+SklJOe76Xbt21ZQpU/Tss8/qyJEjuuqqq+q9pLpv3z4tWrRIHo9HLVu21PDhw31ntK644gplZGRo9erVGjlypCZNmqS4uDj94Q9/UFhYmK688kq53e5GHUNQUJAeeOABLV68WCkpKXI4HBoxYoT69u2rIUOG6PDhw5o3b55KS0sVHh6umJgYDR8+vFHbBnBmcZjGnJ8HAD+bPXt2rXdYAsCZjMudAAAAFuJyJ4CA9umnn+qll16qtbxdu3Z65pln/DAiADg1uNwJAABgIS53AgAAWIhIAwAAsBCRBgAAYCEiDQAAwEJEGgAAgIWINAAAAAv9PxQBjnpTcdQiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e6983828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# seaborn histogram - exclude missings\n",
    "ax = sns.distplot(bikes.trip_distance[bikes.trip_distance.notna()],\n",
    "                  kde=False, bins=100)\n",
    "ax.set_xlim(0,10)\n",
    "ax.set_title(\"Histogram of bike trip distances\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most **trips are rather short**. Typically, they will be under two kilometers. Trips above four kilometers distance are the exception.  \n",
    "As a next step, we can start to look at some descriptives for the usage and distance variables we've just defined. Let's take a look at the usage first:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute some descriptives statistics on bike usage\n",
    "length_observ_period = (bikes.scrape_datetime.dt.dayofyear.max()\\\n",
    "                            - bikes.scrape_datetime.dt.dayofyear.min()) + 1\n",
    "num_bikes = len(bikes.groupby('bike_id'))\n",
    "usage_mean_daily = np.mean(bikes.groupby(bikes.scrape_datetime.dt.dayofyear).change.sum()\n",
    "                           / num_bikes)\n",
    "num_observ_per_bike = bikes[bikes.obs.notna()].groupby(['bike_id']).size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Days in dataset:  41\n",
      "Number of bikes:  1230\n",
      "Daily mean usage per bike:  1.7067023597065238\n"
     ]
    }
   ],
   "source": [
    "print(\"Days in dataset: \", length_observ_period)\n",
    "print(\"Number of bikes: \", num_bikes)\n",
    "print(\"Daily mean usage per bike: \", usage_mean_daily)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "During our observation period of **41 days** we have **1230 different bikes** on the streets. On average a bike is **used 1.7 times a day**. Hence, there are about 2100 trips made each day.\n",
    "<br>\n",
    "Taking advantage of the possibility to directly plot pandas DataFrames and Series with `pyplot` we look at some graphs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'number of bikes')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e69466a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = num_observ_per_bike.hist(bins=int(num_bikes/20))\n",
    "ax.set_title(\"Total observations\")\n",
    "ax.set_xlabel(\"number of observations\")\n",
    "ax.set_ylabel(\"number of bikes\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that there are about 50 bikes which have almost no observations in our data. Also a few others have less than 3000. Above 3000 there is a steady increase. Probably, bikes with very few observations have broken down at some point at the beginning of the period. However, bikes that have been used a lot will also have fewer observations. Consequently, there are quite a few bikes that seem to be used very rarely.<br>\n",
    "Next, I'd like to get an overview over the number of bike rentals during our observation period. A very intelligible way to visualize a single measure over a time period are calendar heatmaps. The [calmap](https://pythonhosted.org/calmap/) library simplifies creating such charts. Here is what we have to do:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x7f55e66b1080>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e6a412e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Calendar Heatmap \n",
    "import calmap\n",
    "\n",
    "# Compute Total usage per day \n",
    "grp = bikes.groupby(bikes.scrape_datetime.dt.date).change.sum()\n",
    "# Convert to DatetimeIndex for plotting\n",
    "grp.index = grp.index.to_datetime()\n",
    "# Plot the calmap and add legend/colorbar\n",
    "fig = plt.figure()\n",
    "cax = calmap.yearplot(grp)\n",
    "cax.set_xlim(5,12)\n",
    "fig.colorbar(cax.get_children()[1], ax=cax, orientation='vertical')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The resulting chart gives away a lot of interesting patterns. First, we see that the number of rentals per day varies greatly. It has a low of about 1000 and a high of 3000. Then, we learn that on weekends rentals are the lowest. In contrast, Fridays seem to be quite popular for renting bikes. There is a lot of variance from week to week even between same weekdays. During all days in the 4th week rentals were very low. As opposed to this, rentals throughout the 2nd and last week were high. If we looked at a longer time period, we would probably see seasonal patterns as well.  \n",
    "Next, we look at differences between weekdays more closely. For that, we will depict the distribution of rentals by weekday using grouped boxplots:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of bikes used per weekday on average:  scrape_weekday\n",
      "Mon    2338.734940\n",
      "Tue    2469.297945\n",
      "Wed    2276.813934\n",
      "Thu    2606.723549\n",
      "Fri    2471.326781\n",
      "Sat    1652.402089\n",
      "Sun    1569.121447\n",
      "dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e69838d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "weekday_index = [\"Mon\",\"Tue\",\"Wed\",\"Thu\",\"Fri\",\"Sat\",\"Sun\"]\n",
    "bikes_weekday = bikes.groupby(['bike_id', \"scrape_weekday\"]).change.sum().unstack()\\\n",
    "                                / length_observ_period * 7 * num_bikes\n",
    "# reorder columns\n",
    "bikes_weekday = bikes_weekday[weekday_index]\n",
    "# draw boxplots with our DataFrame\n",
    "ax = sns.boxplot(data=bikes_weekday)\n",
    "ax.set_title(\"Mean Usage of bikes by weekday\")\n",
    "ax.set_xlabel(\"weekday\")\n",
    "ax.set_ylabel(\"bikes used\")\n",
    "print(\"Total number of bikes used per weekday on average: \", bikes_weekday.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As seen before, Thursdays and Fridays are the most popular rental days. More than 2500 trips are made on average on these days. On the weekends there is a massive drop. Usage is almost 40% lower compared to the weekdays. This indicates that commuting between home and work (or school / university) is the dominant use case for rental bikes.<br>\n",
    "Let's follow up by looking at how rentals change during the course of a day:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'bikes used')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e6642ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "usage_by_hour = bikes.groupby(bikes.scrape_datetime.dt.hour).change.sum()\\\n",
    "                                                        / length_observ_period\n",
    "ax = usage_by_hour.plot()\n",
    "plt.xticks(range(0,24)[::2])\n",
    "ax.set_title(\"Usage of bikes by hour\")\n",
    "ax.set_xlabel(\"time\")\n",
    "ax.set_ylabel(\"bikes used\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The graph aboves shows the usage of bikes by hour of the day for an average day. We see that the usage takes off at about 6 a.m. in the morning. It climbs to its first peak at around 9 and then drops somewhat. This corresponds well with people going to work. Then, at about 11 a.m. usage starts to increase steadily. The maximum is reached at about 6 p.m. Again, this fits the fact that people use the bikes to return back home after work. Thereafter, there is an even decline until the minimum at around 5 a.m. <br>\n",
    "Let's see if there is a difference in this graph between weekdays and weekends:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute Usage by time for weekdays vs weekends\n",
    "usage_by_hour_wd = bikes[bikes.scrape_datetime.dt.weekday <= 5].groupby(\n",
    "                       bikes.scrape_datetime.dt.hour).change.sum() / (length_observ_period * 5/7)\n",
    "usage_by_hour_we = bikes[bikes.scrape_datetime.dt.weekday > 5].groupby(\n",
    "                       bikes.scrape_datetime.dt.hour).change.sum() / (length_observ_period * 2/7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e657dc18>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Draw two plots for comparing week vs. weekend use\n",
    "plt.subplot(2, 1, 1)\n",
    "ax = usage_by_hour_wd.plot()\n",
    "plt.xticks(range(0,24)[::2])\n",
    "ax.set_title(\"Usage of bikes by hour\\nWeekdays\")\n",
    "ax.set_xlabel(\"time\")\n",
    "ax.set_ylabel(\"bikes used\")\n",
    "\n",
    "plt.subplot(2, 1, 2)\n",
    "ax = usage_by_hour_we.plot()\n",
    "plt.xticks(range(0,24)[::2])\n",
    "ax.set_title(\"Weekends\")\n",
    "ax.set_xlabel(\"time\")\n",
    "ax.set_ylabel(\"bikes used\")\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A few interesting points can be taken from this comparison. On the weekends, bike rentals in the morning begin to increase two hours later. Also, instead of the two peaks during weekdays there is only one peak. This speaks for the fact that commuting to work does not play a major role on weekends. Moreover, there is an increase of night time rentals between 11 p.m. and 2 a.m. This is consistent with a leisure oriented use of the bikes.<br>\n",
    "Moving on, we investigate the traveled distance by first looking at some descriptives:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute descriptive stats on distance \n",
    "distance_total = bikes.trip_distance.sum()\n",
    "distance_by_bike = bikes.trip_distance.sum() / num_bikes\n",
    "distance_by_day = distance_total / length_observ_period \n",
    "distance_by_hour = bikes.groupby(bikes.scrape_datetime.dt.hour).trip_distance.sum()\\\n",
    "                                                            / length_observ_period \n",
    "distance_by_weekday  = bikes.groupby(['scrape_weekday']).trip_distance.sum()\\\n",
    "                                                       / length_observ_period * 7"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total distance:  120799.78858713032\n",
      "Total Distance per bike:  98.21121023343927\n",
      "Total distance per day:  2946.3363070031783\n",
      "Average distance per hour:  scrape_datetime\n",
      "0      61.618007\n",
      "1      45.893735\n",
      "2      42.977367\n",
      "3      29.953201\n",
      "4      27.588448\n",
      "5      15.331579\n",
      "6      33.689674\n",
      "7      78.382511\n",
      "8     154.956016\n",
      "9     174.225709\n",
      "10    125.791852\n",
      "11    117.863162\n",
      "12    134.439276\n",
      "13    163.887926\n",
      "14    173.059349\n",
      "15    189.615456\n",
      "16    219.522643\n",
      "17    264.317294\n",
      "18    264.098703\n",
      "19    197.830509\n",
      "20    150.793323\n",
      "21    108.055639\n",
      "22     92.281110\n",
      "23     80.163819\n",
      "Name: trip_distance, dtype: float64\n",
      "Average distance per weekday:  scrape_weekday\n",
      "Fri    3445.663328\n",
      "Mon    2988.295342\n",
      "Sat    2141.913185\n",
      "Sun    2172.674696\n",
      "Thu    3564.889144\n",
      "Tue    3249.941301\n",
      "Wed    3060.977152\n",
      "Name: trip_distance, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(\"Total distance: \", distance_total)\n",
    "print(\"Total Distance per bike: \", distance_by_bike)\n",
    "print(\"Total distance per day: \", distance_by_day)\n",
    "print(\"Average distance per hour: \", distance_by_hour)\n",
    "print(\"Average distance per weekday: \", distance_by_weekday)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we look at the traveled distance during the 40 day period at hand. All bikes taken together have been rode for more than 120.799km. Traveled by car, this would sum up to about 15 tons of CO² emitted. On average, each bike was used for a total of about 100km or 2,5km a day. In general, there are no big surprises here. The overall picture is analogous to what we've seen for the usage.<br>\n",
    "What might be interesting, though, is calculating the distance per trip. So let's give it a shot: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f55e64fbe10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.figure(figsize=(8,10))\n",
    "plt.subplot(2, 1, 1)\n",
    "ax = usage_by_hour.plot()\n",
    "plt.xticks(range(0,24)[::2])\n",
    "ax.set_title(\"Usage of bikes by time\")\n",
    "ax.set_xlabel(\"time\")\n",
    "ax.set_ylabel(\"bikes used\")\n",
    "\n",
    "distance_per_trip_by_hour = distance_by_hour / usage_by_hour\n",
    "plt.subplot(2, 1, 2)\n",
    "ax = distance_per_trip_by_hour.plot()\n",
    "plt.xticks(range(0,24)[::2])\n",
    "ax.set_title(\"Average trip distance by time\")\n",
    "ax.set_xlabel(\"time\")\n",
    "ax.set_ylabel(\"av trip distance\")\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above, we compare the average trip distance with the average number of rentals by hour. A striking finding is the increase in trip distance observed between 22 p.m. and 4 a.m. While not many people rent bikes during this timespan those who do tend to ride for above average distances.\n",
    "<br>\n",
    "<br>\n",
    "This concludes the first statistical analysis of the data. But there is so much more that we can look at. Consequently, in a follow up post I'll start applying machine learning to the data. Especially, in combination with supplemental data sources there are much more insights to be gained!\n",
    "<br><br>"
   ]
  }
 ],
 "metadata": {
  "_draft": {
   "nbviewer_url": "https://gist.github.com/f939d4ae13dede7d370221bd54a2af4b"
  },
  "gist": {
   "data": {
    "description": "Descriptive statistics and visualization of the Cologne Rental Bike location data",
    "public": true
   },
   "id": "f939d4ae13dede7d370221bd54a2af4b"
  },
  "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": "3.6.4"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}