NYC Taxi Trip Duration LR DT FE.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problem Statement\n",
    "\n",
    "At some point or the other almost each one of us has used an Ola or Uber for taking a ride. \n",
    "\n",
    "Ride hailing services are services that use online-enabled platforms to connect between passengers and local drivers using their personal vehicles. In most cases they are a comfortable method for door-to-door transport. Usually they are cheaper than using licensed taxicabs. Examples of ride hailing services include Uber and Lyft.\n",
    "\n",
    "<img src=\"2560px-Ride_hailing_services_Wikivoyage_banner.jpg\">\n",
    "\n",
    "To improve the efficiency of taxi dispatching systems for such services, it is important to be able to predict how long a driver will have his taxi occupied. If a dispatcher knew approximately when a taxi driver would be ending their current ride, they would be better able to identify which driver to assign to each pickup request.\n",
    "\n",
    "In this competition, we are challenged to build a model that predicts the total ride duration of taxi trips in New York City."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Exploratory Data Analysis\n",
    "Let's check the data files! According to the data description we should find the following columns:\n",
    "\n",
    " - **id** - a unique identifier for each trip\n",
    " - **vendor_id** - a code indicating the provider associated with the trip record\n",
    " - **pickup_datetime** - date and time when the meter was engaged\n",
    " - **dropoff_datetime** - date and time when the meter was disengaged\n",
    " - **passenger_count** - the number of passengers in the vehicle (driver entered value)\n",
    " - **pickup_longitude** - the longitude where the meter was engaged\n",
    " - **pickup_latitude** - the latitude where the meter was engaged\n",
    " - **dropoff_longitude** - the longitude where the meter was disengaged\n",
    " - **dropoff_latitude** - the latitude where the meter was disengaged\n",
    " - **store_and_fwd_flag** - This flag indicates whether the trip record was held in vehicle memory before sending to the vendor because the vehicle did not have a connection to the server (Y=store and forward; N=not a store and forward trip)\n",
    " - **trip_duration** - (target) duration of the trip in seconds\n",
    "\n",
    "Here, we have 2 variables dropoff_datetime and store_and_fwd_flag which are not available before the trip starts and hence will not be used as features to the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.1 Load Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from datetime import timedelta\n",
    "import datetime as dt\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('nyc_taxi_final.zip')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### File structure and content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We have 729322 rows.\n",
      "We have 11 columns\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "id                              id0889885\n",
       "vendor_id                               1\n",
       "pickup_datetime       2016-03-11 23:35:37\n",
       "dropoff_datetime      2016-03-11 23:53:57\n",
       "passenger_count                         2\n",
       "pickup_longitude                 -73.9883\n",
       "pickup_latitude                   40.7317\n",
       "dropoff_longitude                -73.9948\n",
       "dropoff_latitude                  40.6949\n",
       "store_and_fwd_flag                      N\n",
       "trip_duration                        1100\n",
       "Name: 1, dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('We have {} rows.'.format(df.shape[0]))\n",
    "print('We have {} columns'.format(df.shape[1]))\n",
    "df.iloc[1,:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At first glance, we can see the types of each variable and what they look like. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Missing Values\n",
    "Knowing about missing values is important because they indicate how much we don’t know about our data. Making inferences based on just a few cases is often unwise. In addition, many modelling procedures break down when missing values are involved and the corresponding rows will either have to be removed completely or the values need to be estimated somehow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                    0\n",
       "vendor_id             0\n",
       "pickup_datetime       0\n",
       "dropoff_datetime      0\n",
       "passenger_count       0\n",
       "pickup_longitude      0\n",
       "pickup_latitude       0\n",
       "dropoff_longitude     0\n",
       "dropoff_latitude      0\n",
       "store_and_fwd_flag    0\n",
       "trip_duration         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(pd.isnull(df))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fortunately, in this dataset we do not have any missing values which is great."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reformatting features & Checking consistency\n",
    "There are a variety of features within the dataset and it is important to convert them into the right format such that we can analyse them easily. This would include converting datetime features and string features. \n",
    "\n",
    "Also, one important thing is never to take assumptions without backing it with data. Here, as you can see the trip duration can also be calculated pick up and drop off datetime. We will check whether the given duration is consistent with the calculated trip duration "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 12)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# converting strings to datetime features\n",
    "df['pickup_datetime'] = pd.to_datetime(df.pickup_datetime)\n",
    "df['dropoff_datetime'] = pd.to_datetime(df.dropoff_datetime)\n",
    "\n",
    "# Converting yes/no flag to 1 and 0\n",
    "df['store_and_fwd_flag'] = 1 * (df.store_and_fwd_flag.values == 'Y')\n",
    "\n",
    "df['check_trip_duration'] = (df['dropoff_datetime'] - df['pickup_datetime']).map(lambda x: x.total_seconds())\n",
    "\n",
    "duration_difference = df[np.abs(df['check_trip_duration'].values  - df['trip_duration'].values) > 1]\n",
    "duration_difference.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This implies that there is no inconsistency in data wrt the drop location and trip duration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Target Exploration\n",
    "In this section we will take a look at the trip duration which is the target variable. It is crucial to understand it in detail as this is what we are trying to predict accurately."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    202.589444\n",
       "mean       0.264508\n",
       "std        1.073507\n",
       "min        0.000278\n",
       "25%        0.110278\n",
       "50%        0.184167\n",
       "75%        0.298611\n",
       "max      538.815556\n",
       "Name: trip_duration, dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['trip_duration'].describe()/3600 # Trip duration in hours"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Woah! There is a trip with duration of 979 hours. This is a huge outlier and might create problems at the prediction stage. One idea is to log transform the trip duration before prediction to visualise it better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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['log_trip_duration'] = np.log(df['trip_duration'].values + 1)\n",
    "sns.distplot(df['log_trip_duration'], kde = False, bins = 200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We find:\n",
    "\n",
    "1. The majority of rides follow a rather smooth distribution that looks almost log-normal with a peak just around exp(6.5)  i.e. about 17 minutes.\n",
    "2. There are several suspiciously short rides with less than 10 seconds duration.\n",
    "3. As discussed earlier, there are a few huge outliers near 12."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Univariate Visualization\n",
    "First of all, let us look at some of the binary features. Looking at each feature might uncover some insight that might be useful at later modelling stages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Frequency')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Binary Features\n",
    "plt.figure(figsize=(22, 6))\n",
    "#fig, axs = plt.subplot(ncols=2)\n",
    "\n",
    "# Passenger Count\n",
    "plt.subplot(131)\n",
    "sns.countplot(df['passenger_count'])\n",
    "plt.xlabel('Passenger Count')\n",
    "plt.ylabel('Frequency')\n",
    "\n",
    "\n",
    "# vendor_id\n",
    "plt.subplot(132)\n",
    "sns.countplot(df['vendor_id'])\n",
    "plt.xlabel('vendor_id')\n",
    "plt.ylabel('Frequency')\n",
    "\n",
    "# store_and_fwd_flag\n",
    "plt.subplot(133)\n",
    "sns.countplot(df['store_and_fwd_flag'])\n",
    "plt.xlabel('store_and_fwd_flag')\n",
    "plt.ylabel('Frequency')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observations:\n",
    "1. Most of the trips involve only 1 passenger. There are trips with 7-9 passengers but they are very low in number.\n",
    "2. Vendor 2 has more number of trips as compared to vendor 1\n",
    "3. The store_and_fwd_flag values, indicating whether the trip data was sent immediately to the vendor (“0”) or held in the memory of the taxi because there was no connection to the server (“1”), show that there was almost no storing taking place"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, we will delve into the datetime features to understand the trend of number of hourly/monthly/daily taxi trips "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Timestamp('2016-01-01 00:01:14'), Timestamp('2016-06-30 23:59:37'))"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['pickup_datetime'].min(), df['pickup_datetime'].max()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clearly, These trips are for first 6 months of 2016. To look at trends, we first need to extract week days and hour of day from the pickup date."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['day_of_week'] = df['pickup_datetime'].dt.weekday\n",
    "df['hour_of_day'] = df['pickup_datetime'].dt.hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Total number of pickups')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Datetime features\n",
    "plt.figure(figsize=(22, 6))\n",
    "\n",
    "# Passenger Count\n",
    "plt.subplot(121)\n",
    "sns.countplot(df['day_of_week'])\n",
    "plt.xlabel('Week Day')\n",
    "plt.ylabel('Total Number of pickups')\n",
    "\n",
    "# vendor_id\n",
    "plt.subplot(122)\n",
    "sns.countplot(df['hour_of_day'])\n",
    "plt.xlabel('Hour of Day')\n",
    "plt.ylabel('Total number of pickups')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Number of pickups for weekends is much lower than week days with a peak on Thursday (4). Note that here weekday is a decimal number, where 0 is Sunday and 6 is Saturday.\n",
    "* Number of pickups as expected is highest in late evenings. However, it is much lower during the morning peak hours."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Lattitude & Longitude"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets look at the geospatial or location features to check consistency. They should not vary much as we are only considering trips within New York city."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "f, axes = plt.subplots(2,2,figsize=(10, 10), sharex=False, sharey = False)\n",
    "sns.despine(left=True)\n",
    "sns.distplot(df['pickup_latitude'].values, label = 'pickup_latitude',color=\"b\",bins = 100, ax=axes[0,0])\n",
    "sns.distplot(df['pickup_longitude'].values, label = 'pickup_longitude',color=\"r\",bins =100, ax=axes[1,0])\n",
    "sns.distplot(df['dropoff_latitude'].values, label = 'dropoff_latitude',color=\"b\",bins =100, ax=axes[0,1])\n",
    "sns.distplot(df['dropoff_longitude'].values, label = 'dropoff_longitude',color=\"r\",bins =100, ax=axes[1,1])\n",
    "plt.setp(axes, yticks=[])\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Findings - (Here, red represents pickup and dropoff Longitudes & blue represents pickup & dropoff lattitudes)\n",
    "\n",
    "1. From the plot above it is clear that pick and drop latitude are centered around 40 to 41, and longitude are situated around -74 to -73. \n",
    "2. Some extreme co-ordinates has squeezed the plot such that we see a spike here\n",
    "3. A good idea is to remove these outliers and look at the distribution more closely"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = df.loc[(df.pickup_latitude > 40.6) & (df.pickup_latitude < 40.9)]\n",
    "df = df.loc[(df.dropoff_latitude>40.6) & (df.dropoff_latitude < 40.9)]\n",
    "df = df.loc[(df.dropoff_longitude > -74.05) & (df.dropoff_longitude < -73.7)]\n",
    "df = df.loc[(df.pickup_longitude > -74.05) & (df.pickup_longitude < -73.7)]\n",
    "df_data_new = df.copy()\n",
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "f, axes = plt.subplots(2,2,figsize=(10, 10), sharex=False, sharey = False)#\n",
    "sns.despine(left=True)\n",
    "sns.distplot(df_data_new['pickup_latitude'].values, label = 'pickup_latitude',color=\"b\",bins = 100, ax=axes[0,0])\n",
    "sns.distplot(df_data_new['pickup_longitude'].values, label = 'pickup_longitude',color=\"r\",bins =100, ax=axes[0,1])\n",
    "sns.distplot(df_data_new['dropoff_latitude'].values, label = 'dropoff_latitude',color=\"b\",bins =100, ax=axes[1, 0])\n",
    "sns.distplot(df_data_new['dropoff_longitude'].values, label = 'dropoff_longitude',color=\"r\",bins =100, ax=axes[1, 1])\n",
    "plt.setp(axes, yticks=[])\n",
    "plt.tight_layout()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* We have a much better view of the distribution of coordinates instead of spikes. And we see that most trips are concentrated between these lat long only with a few significant clusters.\n",
    "* These clusters are represented by the numerous peaks in the lattitude and longitude histograms\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bivariate Relations with Target\n",
    "Now that we have gone through all the basic features one by one. Let us start looking at their relation with the target. This will help us in selecting and extracting features at the modelling stage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'vendor_id', 'pickup_datetime', 'dropoff_datetime',\n",
       "       'passenger_count', 'pickup_longitude', 'pickup_latitude',\n",
       "       'dropoff_longitude', 'dropoff_latitude', 'store_and_fwd_flag',\n",
       "       'trip_duration', 'check_trip_duration', 'log_trip_duration',\n",
       "       'day_of_week', 'hour_of_day'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Trip Duration vs Weekday\n",
    "\n",
    "For different week days, the trip durations can vary as different week days might have different traffic densities especially the weekends might have a much different patterns as compared to working days. Weekday is taken as a decimal number, where 0 - Sunday and 6 is Saturday."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "summary_wdays_avg_duration = pd.DataFrame(df.groupby(['day_of_week'])['trip_duration'].median())\n",
    "summary_wdays_avg_duration.reset_index(inplace = True)\n",
    "summary_wdays_avg_duration['unit']=1\n",
    "\n",
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "sns.set_context(\"poster\")\n",
    "sns.tsplot(data=summary_wdays_avg_duration, time=\"day_of_week\", unit = \"unit\", value=\"trip_duration\")\n",
    "sns.despine(bottom = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "summary_hourly_avg_duration = pd.DataFrame(df.groupby(['hour_of_day'])['trip_duration'].median())\n",
    "summary_hourly_avg_duration.reset_index(inplace = True)\n",
    "summary_hourly_avg_duration['unit']=1\n",
    "\n",
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "sns.set_context(\"poster\")\n",
    "sns.tsplot(data=summary_hourly_avg_duration, time=\"hour_of_day\", unit = \"unit\", value=\"trip_duration\")\n",
    "sns.despine(bottom = False)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Trip durations are definitely shorter for late night and early morning hours that can be attributed to low traffic density\n",
    "* It follows a similar pattern when compared to number of pickups indicating a correlation between number of pickups and trip duration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### vendor_id vs Trip Duration\n",
    "\n",
    "Let's check how the trip duration varies for different vendors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(22, 6))\n",
    "sns.boxplot(x=\"vendor_id\", y=\"trip_duration\", data=df)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Woah! This did not came out as expected. The only thing I can see from this boxplot is that for vendor 2, there are a number of outliers exceeding 24 hours while vendor 1 does not have such long trips.\n",
    "\n",
    "There could be 2 solutions to this:\n",
    "1. Remove the huge outliers and plot again \n",
    "2. Look at median trip duration for both vendors on hourly basis\n",
    "\n",
    "Let's try the first technique now and check trips below 50000 seconds only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(22, 6))\n",
    "df_sub = df[df['trip_duration'] < 50000]\n",
    "sns.boxplot(x=\"vendor_id\", y=\"trip_duration\", data=df_sub)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, we were in a false perception earlier that vendor 1 had more outliers. Since the median is just around 600 seconds, we observe that vendor 2 has many more outliers as compared to vendor 1. Next, to confirm this, we will quickly look at the mean wrt day of week for both vendors using tsplot (time series plot) from seaborn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mean Trip Duration Vendor Wise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "summary_wdays_avg_duration = pd.DataFrame(df.groupby(['vendor_id','day_of_week'])['trip_duration'].mean())\n",
    "summary_wdays_avg_duration.reset_index(inplace = True)\n",
    "summary_wdays_avg_duration['unit']=1\n",
    "\n",
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "sns.set_context(\"poster\")\n",
    "sns.tsplot(data=summary_wdays_avg_duration, time=\"day_of_week\", unit = \"unit\", condition=\"vendor_id\", value=\"trip_duration\")\n",
    "sns.despine(bottom = False)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Median Trip Duration Vendor Wise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "summary_wdays_avg_duration = pd.DataFrame(df.groupby(['vendor_id','day_of_week'])['trip_duration'].median())\n",
    "summary_wdays_avg_duration.reset_index(inplace = True)\n",
    "summary_wdays_avg_duration['unit']=1\n",
    "\n",
    "sns.set(style=\"white\", palette=\"muted\", color_codes=True)\n",
    "sns.set_context(\"poster\")\n",
    "sns.tsplot(data=summary_wdays_avg_duration, time=\"day_of_week\", unit = \"unit\", condition=\"vendor_id\", value=\"trip_duration\")\n",
    "sns.despine(bottom = False)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Median trip duration does not vary much as can be seen from the above plot for different vendors. It emphasises the importance of looking at the correct measure for central tendency for analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Trip Duration vs Passenger Count\n",
    "\n",
    "Again as we are aware, there are a large number of outliers for trip duration and we will not be able to observe the differences. For this, we have taken a cutoff of 10000 seconds and used a boxplot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    515243\n",
       "2    104576\n",
       "5     38776\n",
       "3     29561\n",
       "6     24035\n",
       "4     13972\n",
       "0        31\n",
       "9         1\n",
       "Name: passenger_count, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.passenger_count.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.passenger_count.value_counts()\n",
    "plt.figure(figsize=(22, 6))\n",
    "df_sub = df[df['trip_duration'] < 10000]\n",
    "sns.boxplot(x=\"passenger_count\", y=\"trip_duration\", data=df_sub)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* The boxplot clearly shows that there not much of a difference in distribution for the most frequently occuring passenger count values - 1, 2, 3.\n",
    "* Another key observation is that the number of outliers are reduced for higher passenger counts but that only comes down to the individual frequencies of each passenger count."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualise most frequently occuring Pickup points on the lattitude-longitude Map\n",
    "\n",
    "Here, we try to visualise the most frequently occuring pickup points on the map and check how it is distributed spatially."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "rgb = np.zeros((3000, 3500, 3), dtype=np.uint8)\n",
    "rgb[..., 0] = 0\n",
    "rgb[..., 1] = 0\n",
    "rgb[..., 2] = 0\n",
    "df_data_new['pick_lat_new'] = list(map(int, (df['pickup_latitude'] - (40.6000))*10000))\n",
    "df_data_new['drop_lat_new'] = list(map(int, (df['dropoff_latitude'] - (40.6000))*10000))\n",
    "df_data_new['pick_lon_new'] = list(map(int, (df['pickup_longitude'] - (-74.050))*10000))\n",
    "df_data_new['drop_lon_new'] = list(map(int,(df['dropoff_longitude'] - (-74.050))*10000))\n",
    "\n",
    "summary_plot = pd.DataFrame(df_data_new.groupby(['pick_lat_new', 'pick_lon_new'])['id'].count())\n",
    "\n",
    "summary_plot.reset_index(inplace = True)\n",
    "summary_plot.head(120)\n",
    "lat_list = summary_plot['pick_lat_new'].unique()\n",
    "for i in lat_list:\n",
    "    lon_list = summary_plot.loc[summary_plot['pick_lat_new']==i]['pick_lon_new'].tolist()\n",
    "    unit = summary_plot.loc[summary_plot['pick_lat_new']==i]['id'].tolist()\n",
    "    for j in lon_list:\n",
    "        a = unit[lon_list.index(j)]\n",
    "        if (a//25) >0:\n",
    "            rgb[i][j][0] = 255\n",
    "            rgb[i,j, 1] = 0\n",
    "            rgb[i,j, 2] = 0\n",
    "        elif (a//10)>0:\n",
    "            rgb[i,j, 0] = 0\n",
    "            rgb[i,j, 1] = 255\n",
    "            rgb[i,j, 2] = 0\n",
    "        else:\n",
    "            rgb[i,j, 0] = 255\n",
    "            rgb[i,j, 1] = 255\n",
    "            rgb[i,j, 2] = 255\n",
    "fig, ax = plt.subplots(nrows=1,ncols=1,figsize=(14,20))\n",
    "ax.imshow(rgb, cmap = 'hot')\n",
    "ax.set_axis_off() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Findings - From the heatmap kind of image above -\n",
    "\n",
    "* White points - 1-10 trips have white as pickup point\n",
    "* Green points - 10-25 trips have green as pickup point\n",
    "* Red points - More than 25 trips have red as pickup point\n",
    "\n",
    "As expected there are a few small clusters for hot pickup points as displayed by red in the above plot. Most pickup points have less than 10 trips and distributed all over the city.\n",
    "\n",
    "If you go and have a look at an actual map of New York City, red and green points are mostly concentrated around the Manhatten Area"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Correlation Heatmap\n",
    "\n",
    "Let us quickly look at the correlation heatmap to check the correlations amongst all features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "df = df.drop(['id', 'pickup_datetime', 'dropoff_datetime',\n",
    "       'passenger_count','check_trip_duration','log_trip_duration'],\n",
    "        axis=1)\n",
    "corr = df.apply(lambda x: pd.factorize(x)[0]).corr()\n",
    "ax = sns.heatmap(corr, xticklabels=corr.columns, yticklabels=corr.columns, \n",
    "                 linewidths=.2, cmap=\"YlGnBu\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusions\n",
    "1. The majority of rides follow a rather smooth distribution that looks almost log-normal with a peak just around exp(6.5)  i.e. about 17 minutes.\n",
    "2. There are several suspiciously short rides with less than 10 seconds duration.\n",
    "3. As discussed earlier, there are a few huge outliers near 12.\n",
    "4. Most of the trips involve only 1 passenger. There are trips with 7-9 passengers but they are very low in number.\n",
    "5. Vendor 2 has more number of trips as compared to vendor 1\n",
    "6. Number of pickups for weekends is much lower than week days with a peak on Thursday (4). Note that here weekday is a decimal number, where 0 is Sunday and 6 is Saturday.\n",
    "7. Number of pickups as expected is highest in late evenings. However, it is much lower during the morning peak hours.\n",
    "8. We see that most trips are concentrated between these lat long only with a few significant clusters. These clusters are represented by the numerous peaks in the lattitude and longitude histograms\n",
    "9. Trip durations are definitely shorter for late night and early morning hours that can be attributed to low traffic density\n",
    "10. It follows a similar pattern when compared to number of pickups indicating a correlation between number of pickups and trip duration\n",
    "11. Median trip duration does not vary much as can be seen from the above plot for different vendors.\n",
    "12. The boxplot clearly shows that there not much of a difference in distribution for the most frequently occuring passenger count values - 1, 2, 3.\n",
    "13. Another key observation is that the number of outliers are reduced for higher passenger counts but that only comes down to the individual frequencies of each passenger count.\n",
    "14. From the correlation heatmap we see that the lattitude and longitude features have higher correlation with the target as compared to the other features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NYC Taxi Trip Duration Feature Engineering & Model Building\n",
    "Now, that we have seen the Exploration and have a good understanding of data shape and structure. We also looked at firstly the basic models such as decision tree and linear regression and later on ensemble methods such as random forest and XGBoost (Gradient Boosting).\n",
    "\n",
    "But the model is as good as the training data. Can we engineer new features to improve performance? Let's find out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Dictionary\n",
    "It is always a good idea to have the data dictionary handy.\n",
    "\n",
    " - **id** - a unique identifier for each trip\n",
    " - **vendor_id** - a code indicating the provider associated with the trip record\n",
    " - **pickup_datetime** - date and time when the meter was engaged\n",
    " - **dropoff_datetime** - date and time when the meter was disengaged\n",
    " - **passenger_count** - the number of passengers in the vehicle (driver entered value)\n",
    " - **pickup_longitude** - the longitude where the meter was engaged\n",
    " - **pickup_latitude** - the latitude where the meter was engaged\n",
    " - **dropoff_longitude** - the longitude where the meter was disengaged\n",
    " - **dropoff_latitude** - the latitude where the meter was disengaged\n",
    " - **store_and_fwd_flag** - This flag indicates whether the trip record was held in vehicle memory before sending to the vendor because the vehicle did not have a connection to the server (Y=store and forward; N=not a store and forward trip)\n",
    " - **trip_duration** - duration of the trip in seconds\n",
    "\n",
    "Here dropoff_datetime and trip_duration are only available for the train set as that represents the target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Libraries\n",
    "We will load libraries required to build models and validation sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np \n",
    "import pandas as pd\n",
    "import datetime as dt\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "from sklearn.linear_model import LinearRegression \n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "from sklearn.linear_model import Ridge, Lasso\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.neighbors import KNeighborsRegressor\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('nyc_taxi_final.zip')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing & Feature Extraction\n",
    "As is clear from the previous modules, we can only feed numeric features as input to our models. So our next task is to convert the features in numeric form. It is time to jump into getting our data ready for feeding into the model but before that it is important to use the variables to do some feature engineering as t\n",
    "\n",
    "Some of my ideas to create new variables and the reasons are as follows\n",
    "\n",
    "* Difference between pickup and dropoff latitude - will give an idea about the distance covered which could be predictive\n",
    "* Difference between pickup and dropoff longitude - same reason as above\n",
    "* Haversine distance between pickup and dropoff co-ordinates - to capture the actual distance travelled\n",
    "* Pickup minute - since pickup hour is an important variable, the minute of pickup might well have been predictive\n",
    "* Pickup day of year - same reason as above\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DateTime Conversion\n",
    "The datetime features from csv files are read as strings and in order to easily extract features like day of week, month, year etc. we need to convert it into datetime format of python."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# converting strings to datetime features\n",
    "df['pickup_datetime'] = pd.to_datetime(df.pickup_datetime)\n",
    "df['dropoff_datetime'] = pd.to_datetime(df.dropoff_datetime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Log transform the Y values\n",
    "df_y = np.log1p(df['trip_duration'])\n",
    "\n",
    "# Add some datetime features\n",
    "df.loc[:, 'pickup_weekday'] = df['pickup_datetime'].dt.weekday\n",
    "df.loc[:, 'pickup_hour_weekofyear'] = df['pickup_datetime'].dt.weekofyear\n",
    "df.loc[:, 'pickup_hour'] = df['pickup_datetime'].dt.hour\n",
    "df.loc[:, 'pickup_minute'] = df['pickup_datetime'].dt.minute\n",
    "df.loc[:, 'pickup_dt'] = (df['pickup_datetime'] - df['pickup_datetime'].min()).dt.total_seconds()\n",
    "df.loc[:, 'pickup_week_hour'] = df['pickup_weekday'] * 24 + df['pickup_hour']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Distance Features\n",
    "As discussed earlier, distance features must be important and must be included here\n",
    "\n",
    "#### Eucledian Distance\n",
    "Let's Calculate the Eucledian distance between pickup and drop off location to get some idea on how far the pickup and dropoff points are since this would definitely impact the trip duration even though we know that cars can't fly"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "#displacement\n",
    "y_dist= df['pickup_longitude'] - df['dropoff_longitude']\n",
    "x_dist = df['pickup_latitude'] - df['dropoff_latitude']\n",
    "\n",
    "#square distance\n",
    "df['dist_sq'] = (y_dist ** 2) + (x_dist ** 2)\n",
    "\n",
    "#distance\n",
    "df['dist_sqrt'] = df['dist_sq'] ** 0.5\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Haversine Distance\n",
    "Let's calculate the distance (km) between pickup and dropoff points. The haversine formula determines the great-circle distance between two points on a sphere given their longitudes and latitudes. We will also calculate the approximate angle at which the dropoff location lies wrt the pickup location. pd.DataFrame.apply() would be too slow so the haversine function is rewritten to handle arrays.\n",
    "\n",
    "<img src=\"haversine_dist.svg.png\" width=200 height=200>\n",
    "\n",
    "Haversine direction represents the information of angle of the line connecting the dropoff and pickup point over the surface of earth wrt equator."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "def haversine_array(lat1, lng1, lat2, lng2):\n",
    "    lat1, lng1, lat2, lng2 = map(np.radians, (lat1, lng1, lat2, lng2))\n",
    "    AVG_EARTH_RADIUS = 6371  # in km\n",
    "    lat = lat2 - lat1\n",
    "    lng = lng2 - lng1\n",
    "    d = np.sin(lat * 0.5) ** 2 + np.cos(lat1) * np.cos(lat2) * np.sin(lng * 0.5) ** 2\n",
    "    h = 2 * AVG_EARTH_RADIUS * np.arcsin(np.sqrt(d))\n",
    "    return h\n",
    "\n",
    "def direction_array(lat1, lng1, lat2, lng2):\n",
    "    AVG_EARTH_RADIUS = 6371  # in km\n",
    "    lng_delta_rad = np.radians(lng2 - lng1)\n",
    "    lat1, lng1, lat2, lng2 = map(np.radians, (lat1, lng1, lat2, lng2))\n",
    "    y = np.sin(lng_delta_rad) * np.cos(lat2)\n",
    "    x = np.cos(lat1) * np.sin(lat2) - np.sin(lat1) * np.cos(lat2) * np.cos(lng_delta_rad)\n",
    "    return np.degrees(np.arctan2(y, x))\n",
    "\n",
    "\n",
    "df['haversine_distance'] = haversine_array(df['pickup_latitude'].values, \n",
    "                                                     df['pickup_longitude'].values, \n",
    "                                                     df['dropoff_latitude'].values, \n",
    "                                                     df['dropoff_longitude'].values)\n",
    "\n",
    "\n",
    "df['direction'] = direction_array(df['pickup_latitude'].values, \n",
    "                                          df['pickup_longitude'].values, \n",
    "                                          df['dropoff_latitude'].values, \n",
    "                                          df['dropoff_longitude'].values)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fastest route by road\n",
    "Sometimes, adding external information can be crucial for improving the model. Here we will use data extracted from The Open Source Routing Machine or OSRM for each trip in our original dataset. OSRM is a C++ implementation of a high-performance routing engine for shortest paths in road networks. This will give us a very good estimate of distances between pickup and dropoff Points\n",
    "\n",
    "Source: http://project-osrm.org/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>id</th>\n",
       "      <th>total_distance</th>\n",
       "      <th>total_travel_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>id2875421</td>\n",
       "      <td>2009.1</td>\n",
       "      <td>164.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>id2377394</td>\n",
       "      <td>2513.2</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>id3504673</td>\n",
       "      <td>1779.4</td>\n",
       "      <td>235.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>id2181028</td>\n",
       "      <td>1614.9</td>\n",
       "      <td>140.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>id0801584</td>\n",
       "      <td>1393.5</td>\n",
       "      <td>189.4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          id  total_distance  total_travel_time\n",
       "0  id2875421          2009.1              164.9\n",
       "1  id2377394          2513.2              332.0\n",
       "2  id3504673          1779.4              235.8\n",
       "3  id2181028          1614.9              140.1\n",
       "4  id0801584          1393.5              189.4"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fr1 = pd.read_csv('osrm/fastest_routes_train_part_1.zip',\n",
    "                  usecols=['id', 'total_distance', 'total_travel_time'])\n",
    "fr2 = pd.read_csv('osrm/fastest_routes_train_part_2.zip',\n",
    "                  usecols=['id', 'total_distance', 'total_travel_time'])\n",
    "\n",
    "df_street_info = pd.concat((fr1, fr2))\n",
    "df = df.merge(df_street_info, how='left', on='id')\n",
    "\n",
    "df_street_info.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Binning\n",
    "The lattitude and longitude could be a bit noisy and it might be a good idea to bin them and create new features after rounding their values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Binned Coordinates ###\n",
    "df['pickup_latitude_round3'] = np.round(df['pickup_latitude'],3)\n",
    "df['pickup_longitude_round3'] = np.round(df['pickup_longitude'],3)\n",
    "\n",
    "df['dropoff_latitude_round3'] = np.round(df['dropoff_latitude'],3)\n",
    "df['dropoff_longitude_round3'] = np.round(df['dropoff_longitude'],3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Other Features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### One Hot Encoding\n",
    "Here, Vendor ID can be converted to one hot encoding or frequency encoding since in the raw data it has values 1 and 2 without any inherent order. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    390481\n",
       "1    338841\n",
       "Name: vendor_id, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.vendor_id.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, there is not much difference in the frequencies of both and that might not make for an important feature. so we will just convert it to 0 and 1 by subtracting 1 from it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['vendor_id'] = df['vendor_id'] - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                          0\n",
       "vendor_id                   0\n",
       "pickup_datetime             0\n",
       "dropoff_datetime            0\n",
       "passenger_count             0\n",
       "pickup_longitude            0\n",
       "pickup_latitude             0\n",
       "dropoff_longitude           0\n",
       "dropoff_latitude            0\n",
       "store_and_fwd_flag          0\n",
       "trip_duration               0\n",
       "pickup_weekday              0\n",
       "pickup_hour_weekofyear      0\n",
       "pickup_hour                 0\n",
       "pickup_minute               0\n",
       "pickup_dt                   0\n",
       "pickup_week_hour            0\n",
       "dist_sq                     0\n",
       "dist_sqrt                   0\n",
       "haversine_distance          0\n",
       "direction                   0\n",
       "total_distance              1\n",
       "total_travel_time           1\n",
       "pickup_latitude_round3      0\n",
       "pickup_longitude_round3     0\n",
       "dropoff_latitude_round3     0\n",
       "dropoff_longitude_round3    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(pd.isnull(df))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For a route, the total distance and travel time are not available. Let's impute that with 0\n",
    "df.fillna(0, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we go on to build a model, we must drop the variables that should not be fed as features to the algorithms. We will drop\n",
    "* id - Uniquely represents a sample in the train set\n",
    "* pickup_datetime - Since we have extracted the datetime features, there is no need to keep the datetime column\n",
    "* dropoff_datetime - If this is used to create features, it would be a leakage and we will get perfect model performance. Why? The time gap between dropoff_datetime and pickup_datetime is essentially what we are trying to predict\n",
    "* trip_duration - This is the target variable so needs to be dropped\n",
    "* store_and_fwd_flag - This variable is not available before the start of the trip and should not be used for modelling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(['id', 'pickup_datetime', 'dropoff_datetime', 'trip_duration','store_and_fwd_flag'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Building"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, before we go on to build the model, let us look at the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>vendor_id</th>\n",
       "      <th>passenger_count</th>\n",
       "      <th>pickup_longitude</th>\n",
       "      <th>pickup_latitude</th>\n",
       "      <th>dropoff_longitude</th>\n",
       "      <th>dropoff_latitude</th>\n",
       "      <th>pickup_weekday</th>\n",
       "      <th>pickup_hour_weekofyear</th>\n",
       "      <th>pickup_hour</th>\n",
       "      <th>pickup_minute</th>\n",
       "      <th>...</th>\n",
       "      <th>dist_sq</th>\n",
       "      <th>dist_sqrt</th>\n",
       "      <th>haversine_distance</th>\n",
       "      <th>direction</th>\n",
       "      <th>total_distance</th>\n",
       "      <th>total_travel_time</th>\n",
       "      <th>pickup_latitude_round3</th>\n",
       "      <th>pickup_longitude_round3</th>\n",
       "      <th>dropoff_latitude_round3</th>\n",
       "      <th>dropoff_longitude_round3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-73.953918</td>\n",
       "      <td>40.778873</td>\n",
       "      <td>-73.963875</td>\n",
       "      <td>40.771164</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>16</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000159</td>\n",
       "      <td>0.012592</td>\n",
       "      <td>1.199073</td>\n",
       "      <td>-135.634530</td>\n",
       "      <td>1630.9</td>\n",
       "      <td>172.5</td>\n",
       "      <td>40.779</td>\n",
       "      <td>-73.954</td>\n",
       "      <td>40.771</td>\n",
       "      <td>-73.964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>-73.988312</td>\n",
       "      <td>40.731743</td>\n",
       "      <td>-73.994751</td>\n",
       "      <td>40.694931</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>23</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001397</td>\n",
       "      <td>0.037371</td>\n",
       "      <td>4.129111</td>\n",
       "      <td>-172.445217</td>\n",
       "      <td>5428.7</td>\n",
       "      <td>581.8</td>\n",
       "      <td>40.732</td>\n",
       "      <td>-73.988</td>\n",
       "      <td>40.695</td>\n",
       "      <td>-73.995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>-73.997314</td>\n",
       "      <td>40.721458</td>\n",
       "      <td>-73.948029</td>\n",
       "      <td>40.774918</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>17</td>\n",
       "      <td>59</td>\n",
       "      <td>...</td>\n",
       "      <td>0.005287</td>\n",
       "      <td>0.072712</td>\n",
       "      <td>7.250753</td>\n",
       "      <td>34.916093</td>\n",
       "      <td>9327.8</td>\n",
       "      <td>748.9</td>\n",
       "      <td>40.721</td>\n",
       "      <td>-73.997</td>\n",
       "      <td>40.775</td>\n",
       "      <td>-73.948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>-73.961670</td>\n",
       "      <td>40.759720</td>\n",
       "      <td>-73.956779</td>\n",
       "      <td>40.780628</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>44</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000461</td>\n",
       "      <td>0.021473</td>\n",
       "      <td>2.361097</td>\n",
       "      <td>10.043567</td>\n",
       "      <td>8022.7</td>\n",
       "      <td>612.2</td>\n",
       "      <td>40.760</td>\n",
       "      <td>-73.962</td>\n",
       "      <td>40.781</td>\n",
       "      <td>-73.957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>-74.017120</td>\n",
       "      <td>40.708469</td>\n",
       "      <td>-73.988182</td>\n",
       "      <td>40.740631</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>42</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001872</td>\n",
       "      <td>0.043264</td>\n",
       "      <td>4.328534</td>\n",
       "      <td>34.280582</td>\n",
       "      <td>5468.7</td>\n",
       "      <td>645.0</td>\n",
       "      <td>40.708</td>\n",
       "      <td>-74.017</td>\n",
       "      <td>40.741</td>\n",
       "      <td>-73.988</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   vendor_id  passenger_count  pickup_longitude  pickup_latitude  \\\n",
       "0          1                1        -73.953918        40.778873   \n",
       "1          0                2        -73.988312        40.731743   \n",
       "2          1                2        -73.997314        40.721458   \n",
       "3          1                6        -73.961670        40.759720   \n",
       "4          0                1        -74.017120        40.708469   \n",
       "\n",
       "   dropoff_longitude  dropoff_latitude  pickup_weekday  \\\n",
       "0         -73.963875         40.771164               0   \n",
       "1         -73.994751         40.694931               4   \n",
       "2         -73.948029         40.774918               6   \n",
       "3         -73.956779         40.780628               1   \n",
       "4         -73.988182         40.740631               2   \n",
       "\n",
       "   pickup_hour_weekofyear  pickup_hour  pickup_minute  \\\n",
       "0                       9           16             40   \n",
       "1                      10           23             35   \n",
       "2                       7           17             59   \n",
       "3                       1            9             44   \n",
       "4                       7            6             42   \n",
       "\n",
       "             ...              dist_sq  dist_sqrt  haversine_distance  \\\n",
       "0            ...             0.000159   0.012592            1.199073   \n",
       "1            ...             0.001397   0.037371            4.129111   \n",
       "2            ...             0.005287   0.072712            7.250753   \n",
       "3            ...             0.000461   0.021473            2.361097   \n",
       "4            ...             0.001872   0.043264            4.328534   \n",
       "\n",
       "    direction  total_distance  total_travel_time  pickup_latitude_round3  \\\n",
       "0 -135.634530          1630.9              172.5                  40.779   \n",
       "1 -172.445217          5428.7              581.8                  40.732   \n",
       "2   34.916093          9327.8              748.9                  40.721   \n",
       "3   10.043567          8022.7              612.2                  40.760   \n",
       "4   34.280582          5468.7              645.0                  40.708   \n",
       "\n",
       "   pickup_longitude_round3  dropoff_latitude_round3  dropoff_longitude_round3  \n",
       "0                  -73.954                   40.771                   -73.964  \n",
       "1                  -73.988                   40.695                   -73.995  \n",
       "2                  -73.997                   40.775                   -73.948  \n",
       "3                  -73.962                   40.781                   -73.957  \n",
       "4                  -74.017                   40.741                   -73.988  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have all numerical data types in our dataset now. Time to delve into model building. A very simple baseline could just be the mean of the values in the train set. Let's check the performance on that."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Defining Metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test Train Split\n",
    "\n",
    "We have all numbers in our dataset now. Time to delve into model building. But before that, we need to finalise a validation strategy to create the train and test sets. Here, we will do a random split and keep one third of the data in test set and remaining two third of data in the train set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Splitting the data into Train and Validation set\n",
    "from sklearn.model_selection import train_test_split \n",
    "xtrain, xtest, ytrain, ytest = train_test_split(df,df_y,test_size=1/3, random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mean Prediction\n",
    "Before we go on to try any machine learning model, let us look at the performance of a basic model that just says the mean of trip duration in the train set is the prediction for all the trips in the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7986672307875027"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_pred = np.repeat(ytrain.mean(),len(ytest))\n",
    "\n",
    "sqrt(mean_squared_error(ytest, mean_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross validation\n",
    "Cross Validation is one of the most important concepts in any type of data modelling. It simply says, try to leave a sample on which you do not train the model and test the model on this sample before finalizing the model.\n",
    "\n",
    "we divide the entire population into k equal samples. Now we train models on k-1 samples and validate on 1 sample. Then, at the second iteration we train the model with a different sample held as validation. \n",
    "\n",
    "In k iterations, we have basically built model on each sample and held each of them as validation. This is a way to reduce the selection bias and reduce the variance in prediction power."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cv_score(ml_model, rstate = 11,cols = df.columns):\n",
    "    i = 1\n",
    "    cv_scores = []\n",
    "    df1 = df.copy()\n",
    "    df1 = df[cols]\n",
    "    \n",
    "    kf = KFold(n_splits=5,random_state=rstate,shuffle=True)\n",
    "    for train_index,test_index in kf.split(df1,df_y):\n",
    "        print('\\n{} of kfold {}'.format(i,kf.n_splits))\n",
    "        xtr,xvl = df1.loc[train_index],df1.loc[test_index]\n",
    "        ytr,yvl = df_y[train_index],df_y[test_index]\n",
    "\n",
    "        model = ml_model\n",
    "        model.fit(xtr, ytr)\n",
    "        train_val = model.predict(xtr)\n",
    "        pred_val = model.predict(xvl)\n",
    "        rmse_score_train = sqrt(mean_squared_error(ytr, train_val))\n",
    "        rmse_score = sqrt(mean_squared_error(yvl, pred_val))\n",
    "        sufix = \"\"\n",
    "        msg = \"\"\n",
    "        #msg += \"Train RMSE: {:.5f} \".format(rmse_score_train)\n",
    "        msg += \"Valid RMSE: {:.5f}\".format(rmse_score)\n",
    "        print(\"{}\".format(msg))\n",
    "        # Save scores\n",
    "        cv_scores.append(rmse_score)\n",
    "        i+=1\n",
    "    return cv_scores"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Linear Regression\n",
    "Lets begin by using the simplest regression algorithm Linear regression to check the performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "Valid RMSE: 0.54881\n",
      "\n",
      "2 of kfold 5\n",
      "Valid RMSE: 0.54975\n",
      "\n",
      "3 of kfold 5\n",
      "Valid RMSE: 0.54520\n",
      "\n",
      "4 of kfold 5\n",
      "Valid RMSE: 0.56114\n",
      "\n",
      "5 of kfold 5\n",
      "Valid RMSE: 0.54677\n"
     ]
    }
   ],
   "source": [
    "linreg_scores = cv_score(LinearRegression())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can already see that the performance of even linear regression has improved a lot. This demonstrates the power of feature engineering. Let's try decision tree once more and check performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "1 of kfold 5\n",
      "Valid RMSE: 0.42677\n",
      "\n",
      "2 of kfold 5\n",
      "Valid RMSE: 0.42935\n",
      "\n",
      "3 of kfold 5\n",
      "Valid RMSE: 0.41977\n",
      "\n",
      "4 of kfold 5\n",
      "Valid RMSE: 0.42661\n",
      "\n",
      "5 of kfold 5\n",
      "Valid RMSE: 0.42153\n"
     ]
    }
   ],
   "source": [
    "dtree_scores = cv_score(DecisionTreeRegressor(min_samples_leaf=25, min_samples_split=25))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "results_df = pd.DataFrame({'linear_regression':linreg_scores, 'dtree': dtree_scores})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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": [
    "results_df.plot(y=[\"linear_regression\", \"dtree\"], kind=\"bar\", legend=False)\n",
    "plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Woah! That's a lot of improvement. The reason for this could be the non linear relationship between the trip duration values and the location coordinates of pickup and dropoff points.\n",
    "\n",
    "<img src=\"download_dow.png\" width=500 height=500>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Tree Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(criterion='mse', max_depth=None, max_features=None,\n",
       "           max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "           min_impurity_split=None, min_samples_leaf=25,\n",
       "           min_samples_split=25, min_weight_fraction_leaf=0.0,\n",
       "           presort=False, random_state=None, splitter='best')"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dtree = DecisionTreeRegressor(min_samples_leaf=25, min_samples_split=25)\n",
    "dtree.fit(xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "decision_tree = tree.export_graphviz(dtree,out_file='tree.dot',feature_names=xtrain.columns,max_depth=2,filled=True)\n",
    "!dot -Tpng tree.dot -o tree.png"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let us view the decision tree till depth 2 and find out the features at the root and the first node.\n",
    "\n",
    "<img src=\"tree.png\" width=500 height=500>\n",
    "\n",
    "As is clear from the above decision tree the extra features added are adding a lot of value to our decision tree learning indicating that the additional features carry good value and are very important to the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at this, seems like a good idea to try more advanced decision tree based techniques which we will look at in the next section."
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}