Baseline_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### IMPORT DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_riders = pd.read_csv('data/Riders.csv')\n",
    "df_sample_submission = pd.read_csv('data/SampleSubmission.csv')\n",
    "df_test = pd.read_csv('data/Test.csv')\n",
    "df_train = pd.read_csv('data/Train.csv')\n",
    "df_variable_definitions = pd.read_csv('data/VariableDefinitions.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### UTILITY FUNCTIONS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "\n",
    "def display_all(df):\n",
    "    with pd.option_context(\"display.max_rows\", 1000, \"display.max_columns\", 1000): \n",
    "        display(df)\n",
    "        \n",
    "def convert_dt(df,cols):\n",
    "    for i in cols:\n",
    "        df[i] = pd.to_datetime(df[i])\n",
    "    return df\n",
    "\n",
    "def time_of_day(df,cols):\n",
    "    for col in cols:\n",
    "        df = df.assign(\n",
    "                    col=pd.cut(\n",
    "                        df[col].dt.hour,\n",
    "                        [0, 6, 12, 18, 24],\n",
    "                        labels=[\"Night\", \"Morning\", \"Afternoon\", \"Evening\"],\n",
    "                    )\n",
    "                )\n",
    "    return df\n",
    "\n",
    "def f(x):\n",
    "    if (x > 0) and (x <= 6):\n",
    "        return 'Night'\n",
    "    elif (x > 6) and (x <= 12 ):\n",
    "        return 'Morning'\n",
    "    elif (x > 12) and (x <= 18):\n",
    "        return'Noon'\n",
    "    elif (x > 18) and (x <= 24):\n",
    "        return 'Eve'\n",
    "    \n",
    "def rmse(x,y): return math.sqrt(((x-y)**2).mean())\n",
    "\n",
    "def print_score(m):\n",
    "    res = [rmse(m.predict(X_train), y_train), rmse(m.predict(X_test), y_test),\n",
    "                m.score(X_train, y_train), m.score(X_test, y_test)]\n",
    "    if hasattr(m, 'oob_score_'): res.append(m.oob_score_)\n",
    "    print(res)\n",
    "    \n",
    "def rf_feat_importance(m, df):\n",
    "    return pd.DataFrame({'cols':df.columns, 'imp':m.feature_importances_}\n",
    "                       ).sort_values('imp', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### COMBINE TRAINING AND TEST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "disjoint_cols = ['Arrival at Destination - Time',\n",
    " 'Arrival at Destination - Weekday (Mo = 1)',\n",
    " 'Arrival at Destination - Day of Month']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train_new = df_train.drop(disjoint_cols,axis=1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train_new[\"col_type\"] = 'train'\n",
    "df_test[\"col_type\"] = 'test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.concat([df_train_new, df_test], ignore_index=True,sort=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(df_riders,how='left',on='Rider Id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Arrival at Pickup - Day of Month', 'Arrival at Pickup - Time',\n",
       "       'Arrival at Pickup - Weekday (Mo = 1)', 'Confirmation - Day of Month',\n",
       "       'Confirmation - Time', 'Confirmation - Weekday (Mo = 1)',\n",
       "       'Destination Lat', 'Destination Long', 'Distance (KM)', 'Order No',\n",
       "       'Personal or Business', 'Pickup - Day of Month', 'Pickup - Time',\n",
       "       'Pickup - Weekday (Mo = 1)', 'Pickup Lat', 'Pickup Long',\n",
       "       'Placement - Day of Month', 'Placement - Time',\n",
       "       'Placement - Weekday (Mo = 1)', 'Platform Type',\n",
       "       'Precipitation in millimeters', 'Rider Id', 'Temperature',\n",
       "       'Time from Pickup to Arrival', 'User Id', 'Vehicle Type', 'col_type',\n",
       "       'No_Of_Orders', 'Age', 'Average_Rating', 'No_of_Ratings'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 165,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### CLEANING DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Time</td>\n",
       "      <td>10:04:47 AM</td>\n",
       "      <td>11:40:22 AM</td>\n",
       "      <td>12:49:34 PM</td>\n",
       "      <td>9:37:56 AM</td>\n",
       "      <td>10:03:53 AM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Time</td>\n",
       "      <td>9:40:10 AM</td>\n",
       "      <td>11:23:21 AM</td>\n",
       "      <td>12:42:44 PM</td>\n",
       "      <td>9:26:05 AM</td>\n",
       "      <td>9:56:18 AM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Destination Lat</td>\n",
       "      <td>-1.30041</td>\n",
       "      <td>-1.295</td>\n",
       "      <td>-1.30092</td>\n",
       "      <td>-1.25715</td>\n",
       "      <td>-1.29504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Destination Long</td>\n",
       "      <td>36.8297</td>\n",
       "      <td>36.8144</td>\n",
       "      <td>36.8282</td>\n",
       "      <td>36.7951</td>\n",
       "      <td>36.8098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Distance (KM)</td>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Order No</td>\n",
       "      <td>Order_No_4211</td>\n",
       "      <td>Order_No_25375</td>\n",
       "      <td>Order_No_1899</td>\n",
       "      <td>Order_No_9336</td>\n",
       "      <td>Order_No_27883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Personal or Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Time</td>\n",
       "      <td>10:27:30 AM</td>\n",
       "      <td>11:44:09 AM</td>\n",
       "      <td>12:53:03 PM</td>\n",
       "      <td>9:43:06 AM</td>\n",
       "      <td>10:05:23 AM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup Lat</td>\n",
       "      <td>-1.31775</td>\n",
       "      <td>-1.35145</td>\n",
       "      <td>-1.30828</td>\n",
       "      <td>-1.2813</td>\n",
       "      <td>-1.2666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup Long</td>\n",
       "      <td>36.8304</td>\n",
       "      <td>36.8993</td>\n",
       "      <td>36.8434</td>\n",
       "      <td>36.8324</td>\n",
       "      <td>36.7921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Time</td>\n",
       "      <td>9:35:46 AM</td>\n",
       "      <td>11:16:16 AM</td>\n",
       "      <td>12:39:25 PM</td>\n",
       "      <td>9:25:34 AM</td>\n",
       "      <td>9:55:18 AM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Platform Type</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Precipitation in millimeters</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Rider Id</td>\n",
       "      <td>Rider_Id_432</td>\n",
       "      <td>Rider_Id_856</td>\n",
       "      <td>Rider_Id_155</td>\n",
       "      <td>Rider_Id_855</td>\n",
       "      <td>Rider_Id_770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Temperature</td>\n",
       "      <td>20.4</td>\n",
       "      <td>26.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19.2</td>\n",
       "      <td>15.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Time from Pickup to Arrival</td>\n",
       "      <td>745</td>\n",
       "      <td>1993</td>\n",
       "      <td>455</td>\n",
       "      <td>1341</td>\n",
       "      <td>1214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>User Id</td>\n",
       "      <td>User_Id_633</td>\n",
       "      <td>User_Id_2285</td>\n",
       "      <td>User_Id_265</td>\n",
       "      <td>User_Id_1402</td>\n",
       "      <td>User_Id_1737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Vehicle Type</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>col_type</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  0               1  \\\n",
       "Arrival at Pickup - Day of Month                  9              12   \n",
       "Arrival at Pickup - Time                10:04:47 AM     11:40:22 AM   \n",
       "Arrival at Pickup - Weekday (Mo = 1)              5               5   \n",
       "Confirmation - Day of Month                       9              12   \n",
       "Confirmation - Time                      9:40:10 AM     11:23:21 AM   \n",
       "Confirmation - Weekday (Mo = 1)                   5               5   \n",
       "Destination Lat                            -1.30041          -1.295   \n",
       "Destination Long                            36.8297         36.8144   \n",
       "Distance (KM)                                     4              16   \n",
       "Order No                              Order_No_4211  Order_No_25375   \n",
       "Personal or Business                       Business        Personal   \n",
       "Pickup - Day of Month                             9              12   \n",
       "Pickup - Time                           10:27:30 AM     11:44:09 AM   \n",
       "Pickup - Weekday (Mo = 1)                         5               5   \n",
       "Pickup Lat                                 -1.31775        -1.35145   \n",
       "Pickup Long                                 36.8304         36.8993   \n",
       "Placement - Day of Month                          9              12   \n",
       "Placement - Time                         9:35:46 AM     11:16:16 AM   \n",
       "Placement - Weekday (Mo = 1)                      5               5   \n",
       "Platform Type                                     3               3   \n",
       "Precipitation in millimeters                    NaN             NaN   \n",
       "Rider Id                               Rider_Id_432    Rider_Id_856   \n",
       "Temperature                                    20.4            26.4   \n",
       "Time from Pickup to Arrival                     745            1993   \n",
       "User Id                                 User_Id_633    User_Id_2285   \n",
       "Vehicle Type                                   Bike            Bike   \n",
       "col_type                                      train           train   \n",
       "\n",
       "                                                  2              3  \\\n",
       "Arrival at Pickup - Day of Month                 30             15   \n",
       "Arrival at Pickup - Time                12:49:34 PM     9:37:56 AM   \n",
       "Arrival at Pickup - Weekday (Mo = 1)              2              5   \n",
       "Confirmation - Day of Month                      30             15   \n",
       "Confirmation - Time                     12:42:44 PM     9:26:05 AM   \n",
       "Confirmation - Weekday (Mo = 1)                   2              5   \n",
       "Destination Lat                            -1.30092       -1.25715   \n",
       "Destination Long                            36.8282        36.7951   \n",
       "Distance (KM)                                     3              9   \n",
       "Order No                              Order_No_1899  Order_No_9336   \n",
       "Personal or Business                       Business       Business   \n",
       "Pickup - Day of Month                            30             15   \n",
       "Pickup - Time                           12:53:03 PM     9:43:06 AM   \n",
       "Pickup - Weekday (Mo = 1)                         2              5   \n",
       "Pickup Lat                                 -1.30828        -1.2813   \n",
       "Pickup Long                                 36.8434        36.8324   \n",
       "Placement - Day of Month                         30             15   \n",
       "Placement - Time                        12:39:25 PM     9:25:34 AM   \n",
       "Placement - Weekday (Mo = 1)                      2              5   \n",
       "Platform Type                                     3              3   \n",
       "Precipitation in millimeters                    NaN            NaN   \n",
       "Rider Id                               Rider_Id_155   Rider_Id_855   \n",
       "Temperature                                     NaN           19.2   \n",
       "Time from Pickup to Arrival                     455           1341   \n",
       "User Id                                 User_Id_265   User_Id_1402   \n",
       "Vehicle Type                                   Bike           Bike   \n",
       "col_type                                      train          train   \n",
       "\n",
       "                                                   4  \n",
       "Arrival at Pickup - Day of Month                  13  \n",
       "Arrival at Pickup - Time                 10:03:53 AM  \n",
       "Arrival at Pickup - Weekday (Mo = 1)               1  \n",
       "Confirmation - Day of Month                       13  \n",
       "Confirmation - Time                       9:56:18 AM  \n",
       "Confirmation - Weekday (Mo = 1)                    1  \n",
       "Destination Lat                             -1.29504  \n",
       "Destination Long                             36.8098  \n",
       "Distance (KM)                                      9  \n",
       "Order No                              Order_No_27883  \n",
       "Personal or Business                        Personal  \n",
       "Pickup - Day of Month                             13  \n",
       "Pickup - Time                            10:05:23 AM  \n",
       "Pickup - Weekday (Mo = 1)                          1  \n",
       "Pickup Lat                                   -1.2666  \n",
       "Pickup Long                                  36.7921  \n",
       "Placement - Day of Month                          13  \n",
       "Placement - Time                          9:55:18 AM  \n",
       "Placement - Weekday (Mo = 1)                       1  \n",
       "Platform Type                                      1  \n",
       "Precipitation in millimeters                     NaN  \n",
       "Rider Id                                Rider_Id_770  \n",
       "Temperature                                     15.4  \n",
       "Time from Pickup to Arrival                     1214  \n",
       "User Id                                 User_Id_1737  \n",
       "Vehicle Type                                    Bike  \n",
       "col_type                                       train  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Arrival at Pickup - Day of Month',\n",
       " 'Arrival at Pickup - Time',\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)',\n",
       " 'Confirmation - Day of Month',\n",
       " 'Confirmation - Time',\n",
       " 'Confirmation - Weekday (Mo = 1)',\n",
       " 'Destination Lat',\n",
       " 'Destination Long',\n",
       " 'Distance (KM)',\n",
       " 'Order No',\n",
       " 'Personal or Business',\n",
       " 'Pickup - Day of Month',\n",
       " 'Pickup - Time',\n",
       " 'Pickup - Weekday (Mo = 1)',\n",
       " 'Pickup Lat',\n",
       " 'Pickup Long',\n",
       " 'Placement - Day of Month',\n",
       " 'Placement - Time',\n",
       " 'Placement - Weekday (Mo = 1)',\n",
       " 'Platform Type',\n",
       " 'Precipitation in millimeters',\n",
       " 'Rider Id',\n",
       " 'Temperature',\n",
       " 'Time from Pickup to Arrival',\n",
       " 'User Id',\n",
       " 'Vehicle Type',\n",
       " 'col_type']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_cols = df.columns.to_list();df_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Arrival at Pickup - Day of Month           31\n",
       "Arrival at Pickup - Time                19189\n",
       "Arrival at Pickup - Weekday (Mo = 1)        7\n",
       "Confirmation - Day of Month                31\n",
       "Confirmation - Time                     19180\n",
       "Confirmation - Weekday (Mo = 1)             7\n",
       "Destination Lat                          6246\n",
       "Destination Long                         6203\n",
       "Distance (KM)                              46\n",
       "Order No                                28269\n",
       "Personal or Business                        2\n",
       "Pickup - Day of Month                      31\n",
       "Pickup - Time                           19161\n",
       "Pickup - Weekday (Mo = 1)                   7\n",
       "Pickup Lat                               4393\n",
       "Pickup Long                              4377\n",
       "Placement - Day of Month                   31\n",
       "Placement - Time                        19062\n",
       "Placement - Weekday (Mo = 1)                7\n",
       "Platform Type                               4\n",
       "Precipitation in millimeters               56\n",
       "Rider Id                                  952\n",
       "Temperature                               188\n",
       "Time from Pickup to Arrival              4067\n",
       "User Id                                  3782\n",
       "Vehicle Type                                1\n",
       "col_type                                    2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4, 14, 4, 6)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "time_of_day_cols = [\"Arrival at Pickup - Time\",\"Confirmation - Time\",\"Pickup - Time\",\"Placement - Time\"]\n",
    "cols_dummied = [\"Arrival at Pickup - Day of Month\",\"Arrival at Pickup - Weekday (Mo = 1)\",\n",
    "                \"Confirmation - Day of Month\",\"Confirmation - Weekday (Mo = 1)\",\"Pickup - Day of Month\"\n",
    "                ,\"Personal or Business\",\"Pickup - Weekday (Mo = 1)\",\"Placement - Day of Month\",\n",
    "                \"Placement - Weekday (Mo = 1)\",\"Platform Type\"] + time_of_day_cols\n",
    "to_drop = [\"Vehicle Type\",\"User Id\",\"Rider Id\",\"Precipitation in millimeters\"]\n",
    "to_int = [\"Time from Pickup to Arrival\",\"Temperature\",\"Pickup Long\",\n",
    "          \"Pickup Lat\",\"Destination Long\",\"Destination Lat\"]\n",
    "len(time_of_day_cols),len(cols_dummied),len(to_drop),len(to_int)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cleaning data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Timestamp('2019-10-27 10:04:47').hour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[time_of_day_cols] = df[time_of_day_cols].apply(pd.to_datetime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in time_of_day_cols:\n",
    "    df[col] = df[col].apply(lambda x: x.hour)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in time_of_day_cols:\n",
    "    df[col] = df[col].apply(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(to_drop,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Arrival at Pickup - Day of Month           0\n",
       "Arrival at Pickup - Time                   1\n",
       "Arrival at Pickup - Weekday (Mo = 1)       0\n",
       "Confirmation - Day of Month                0\n",
       "Confirmation - Time                        1\n",
       "Confirmation - Weekday (Mo = 1)            0\n",
       "Destination Lat                            0\n",
       "Destination Long                           0\n",
       "Distance (KM)                              0\n",
       "Order No                                   0\n",
       "Personal or Business                       0\n",
       "Pickup - Day of Month                      0\n",
       "Pickup - Time                              1\n",
       "Pickup - Weekday (Mo = 1)                  0\n",
       "Pickup Lat                                 0\n",
       "Pickup Long                                0\n",
       "Placement - Day of Month                   0\n",
       "Placement - Time                           1\n",
       "Placement - Weekday (Mo = 1)               0\n",
       "Platform Type                              0\n",
       "Temperature                             5803\n",
       "Time from Pickup to Arrival             7068\n",
       "col_type                                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>Arrival at Pickup - Day of Month</th>\n",
       "      <th>Arrival at Pickup - Time</th>\n",
       "      <th>Arrival at Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Confirmation - Day of Month</th>\n",
       "      <th>Confirmation - Time</th>\n",
       "      <th>Confirmation - Weekday (Mo = 1)</th>\n",
       "      <th>Destination Lat</th>\n",
       "      <th>Destination Long</th>\n",
       "      <th>Distance (KM)</th>\n",
       "      <th>Order No</th>\n",
       "      <th>...</th>\n",
       "      <th>Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Pickup Lat</th>\n",
       "      <th>Pickup Long</th>\n",
       "      <th>Placement - Day of Month</th>\n",
       "      <th>Placement - Time</th>\n",
       "      <th>Placement - Weekday (Mo = 1)</th>\n",
       "      <th>Platform Type</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Time from Pickup to Arrival</th>\n",
       "      <th>col_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>11859</td>\n",
       "      <td>22</td>\n",
       "      <td>None</td>\n",
       "      <td>3</td>\n",
       "      <td>22</td>\n",
       "      <td>None</td>\n",
       "      <td>3</td>\n",
       "      <td>-1.301809</td>\n",
       "      <td>36.75643</td>\n",
       "      <td>2</td>\n",
       "      <td>Order_No_952</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>-1.303971</td>\n",
       "      <td>36.768817</td>\n",
       "      <td>22</td>\n",
       "      <td>None</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>18.6</td>\n",
       "      <td>5.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Arrival at Pickup - Day of Month Arrival at Pickup - Time  \\\n",
       "11859                                22                     None   \n",
       "\n",
       "       Arrival at Pickup - Weekday (Mo = 1)  Confirmation - Day of Month  \\\n",
       "11859                                     3                           22   \n",
       "\n",
       "      Confirmation - Time  Confirmation - Weekday (Mo = 1)  Destination Lat  \\\n",
       "11859                None                                3        -1.301809   \n",
       "\n",
       "       Destination Long  Distance (KM)      Order No  ...  \\\n",
       "11859          36.75643              2  Order_No_952  ...   \n",
       "\n",
       "      Pickup - Weekday (Mo = 1)  Pickup Lat Pickup Long  \\\n",
       "11859                         3   -1.303971   36.768817   \n",
       "\n",
       "       Placement - Day of Month  Placement - Time  \\\n",
       "11859                        22              None   \n",
       "\n",
       "       Placement - Weekday (Mo = 1)  Platform Type Temperature  \\\n",
       "11859                             3              3        18.6   \n",
       "\n",
       "       Time from Pickup to Arrival  col_type  \n",
       "11859                          5.0     train  \n",
       "\n",
       "[1 rows x 23 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Arrival at Pickup - Time'].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in time_of_day_cols:\n",
    "    df[col] = df[col].fillna(df[col].mode()[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Arrival at Pickup - Day of Month           0\n",
       "Arrival at Pickup - Time                   0\n",
       "Arrival at Pickup - Weekday (Mo = 1)       0\n",
       "Confirmation - Day of Month                0\n",
       "Confirmation - Time                        0\n",
       "Confirmation - Weekday (Mo = 1)            0\n",
       "Destination Lat                            0\n",
       "Destination Long                           0\n",
       "Distance (KM)                              0\n",
       "Order No                                   0\n",
       "Personal or Business                       0\n",
       "Pickup - Day of Month                      0\n",
       "Pickup - Time                              0\n",
       "Pickup - Weekday (Mo = 1)                  0\n",
       "Pickup Lat                                 0\n",
       "Pickup Long                                0\n",
       "Placement - Day of Month                   0\n",
       "Placement - Time                           0\n",
       "Placement - Weekday (Mo = 1)               0\n",
       "Platform Type                              0\n",
       "Temperature                             5803\n",
       "Time from Pickup to Arrival             7068\n",
       "col_type                                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    22466.000000\n",
       "mean        23.255689\n",
       "std          3.609541\n",
       "min         11.200000\n",
       "25%         20.600000\n",
       "50%         23.400000\n",
       "75%         26.000000\n",
       "max         32.100000\n",
       "Name: Temperature, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Temperature.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/pandas/core/ops/__init__.py:1115: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  result = method(y)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Arrival at Pickup - Day of Month</th>\n",
       "      <th>Arrival at Pickup - Time</th>\n",
       "      <th>Arrival at Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Confirmation - Day of Month</th>\n",
       "      <th>Confirmation - Time</th>\n",
       "      <th>Confirmation - Weekday (Mo = 1)</th>\n",
       "      <th>Destination Lat</th>\n",
       "      <th>Destination Long</th>\n",
       "      <th>Distance (KM)</th>\n",
       "      <th>Order No</th>\n",
       "      <th>...</th>\n",
       "      <th>Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Pickup Lat</th>\n",
       "      <th>Pickup Long</th>\n",
       "      <th>Placement - Day of Month</th>\n",
       "      <th>Placement - Time</th>\n",
       "      <th>Placement - Weekday (Mo = 1)</th>\n",
       "      <th>Platform Type</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Time from Pickup to Arrival</th>\n",
       "      <th>col_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>0 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [Arrival at Pickup - Day of Month, Arrival at Pickup - Time, Arrival at Pickup - Weekday (Mo = 1), Confirmation - Day of Month, Confirmation - Time, Confirmation - Weekday (Mo = 1), Destination Lat, Destination Long, Distance (KM), Order No, Personal or Business, Pickup - Day of Month, Pickup - Time, Pickup - Weekday (Mo = 1), Pickup Lat, Pickup Long, Placement - Day of Month, Placement - Time, Placement - Weekday (Mo = 1), Platform Type, Temperature, Time from Pickup to Arrival, col_type]\n",
       "Index: []\n",
       "\n",
       "[0 rows x 23 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[df['Temperature'].isnull() & df['col_type'] == 'test']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Temperature'] = df['Temperature'].fillna(df['Temperature'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Arrival at Pickup - Day of Month           0\n",
       "Arrival at Pickup - Time                   0\n",
       "Arrival at Pickup - Weekday (Mo = 1)       0\n",
       "Confirmation - Day of Month                0\n",
       "Confirmation - Time                        0\n",
       "Confirmation - Weekday (Mo = 1)            0\n",
       "Destination Lat                            0\n",
       "Destination Long                           0\n",
       "Distance (KM)                              0\n",
       "Order No                                   0\n",
       "Personal or Business                       0\n",
       "Pickup - Day of Month                      0\n",
       "Pickup - Time                              0\n",
       "Pickup - Weekday (Mo = 1)                  0\n",
       "Pickup Lat                                 0\n",
       "Pickup Long                                0\n",
       "Placement - Day of Month                   0\n",
       "Placement - Time                           0\n",
       "Placement - Weekday (Mo = 1)               0\n",
       "Platform Type                              0\n",
       "Temperature                                0\n",
       "Time from Pickup to Arrival             7068\n",
       "col_type                                   0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Arrival at Pickup - Day of Month</th>\n",
       "      <th>Arrival at Pickup - Time</th>\n",
       "      <th>Arrival at Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Confirmation - Day of Month</th>\n",
       "      <th>Confirmation - Time</th>\n",
       "      <th>Confirmation - Weekday (Mo = 1)</th>\n",
       "      <th>Destination Lat</th>\n",
       "      <th>Destination Long</th>\n",
       "      <th>Distance (KM)</th>\n",
       "      <th>Order No</th>\n",
       "      <th>...</th>\n",
       "      <th>Pickup - Weekday (Mo = 1)</th>\n",
       "      <th>Pickup Lat</th>\n",
       "      <th>Pickup Long</th>\n",
       "      <th>Placement - Day of Month</th>\n",
       "      <th>Placement - Time</th>\n",
       "      <th>Placement - Weekday (Mo = 1)</th>\n",
       "      <th>Platform Type</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Time from Pickup to Arrival</th>\n",
       "      <th>col_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>4</td>\n",
       "      <td>Order_No_4211</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.317755</td>\n",
       "      <td>36.830370</td>\n",
       "      <td>9</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>20.400000</td>\n",
       "      <td>745.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.295004</td>\n",
       "      <td>36.814358</td>\n",
       "      <td>16</td>\n",
       "      <td>Order_No_25375</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.351453</td>\n",
       "      <td>36.899315</td>\n",
       "      <td>12</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>26.400000</td>\n",
       "      <td>1993.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>Morning</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>Morning</td>\n",
       "      <td>2</td>\n",
       "      <td>-1.300921</td>\n",
       "      <td>36.828195</td>\n",
       "      <td>3</td>\n",
       "      <td>Order_No_1899</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>-1.308284</td>\n",
       "      <td>36.843419</td>\n",
       "      <td>30</td>\n",
       "      <td>Morning</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>23.255689</td>\n",
       "      <td>455.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>15</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.257147</td>\n",
       "      <td>36.795063</td>\n",
       "      <td>9</td>\n",
       "      <td>Order_No_9336</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.281301</td>\n",
       "      <td>36.832396</td>\n",
       "      <td>15</td>\n",
       "      <td>Morning</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>1341.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>13</td>\n",
       "      <td>Morning</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>Morning</td>\n",
       "      <td>1</td>\n",
       "      <td>-1.295041</td>\n",
       "      <td>36.809817</td>\n",
       "      <td>9</td>\n",
       "      <td>Order_No_27883</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>-1.266597</td>\n",
       "      <td>36.792118</td>\n",
       "      <td>13</td>\n",
       "      <td>Morning</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>15.400000</td>\n",
       "      <td>1214.0</td>\n",
       "      <td>train</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Arrival at Pickup - Day of Month Arrival at Pickup - Time  \\\n",
       "0                                 9                  Morning   \n",
       "1                                12                  Morning   \n",
       "2                                30                  Morning   \n",
       "3                                15                  Morning   \n",
       "4                                13                  Morning   \n",
       "\n",
       "   Arrival at Pickup - Weekday (Mo = 1)  Confirmation - Day of Month  \\\n",
       "0                                     5                            9   \n",
       "1                                     5                           12   \n",
       "2                                     2                           30   \n",
       "3                                     5                           15   \n",
       "4                                     1                           13   \n",
       "\n",
       "  Confirmation - Time  Confirmation - Weekday (Mo = 1)  Destination Lat  \\\n",
       "0             Morning                                5        -1.300406   \n",
       "1             Morning                                5        -1.295004   \n",
       "2             Morning                                2        -1.300921   \n",
       "3             Morning                                5        -1.257147   \n",
       "4             Morning                                1        -1.295041   \n",
       "\n",
       "   Destination Long  Distance (KM)        Order No  ...  \\\n",
       "0         36.829741              4   Order_No_4211  ...   \n",
       "1         36.814358             16  Order_No_25375  ...   \n",
       "2         36.828195              3   Order_No_1899  ...   \n",
       "3         36.795063              9   Order_No_9336  ...   \n",
       "4         36.809817              9  Order_No_27883  ...   \n",
       "\n",
       "  Pickup - Weekday (Mo = 1)  Pickup Lat Pickup Long  Placement - Day of Month  \\\n",
       "0                         5   -1.317755   36.830370                         9   \n",
       "1                         5   -1.351453   36.899315                        12   \n",
       "2                         2   -1.308284   36.843419                        30   \n",
       "3                         5   -1.281301   36.832396                        15   \n",
       "4                         1   -1.266597   36.792118                        13   \n",
       "\n",
       "   Placement - Time  Placement - Weekday (Mo = 1)  Platform Type Temperature  \\\n",
       "0           Morning                             5              3   20.400000   \n",
       "1           Morning                             5              3   26.400000   \n",
       "2           Morning                             2              3   23.255689   \n",
       "3           Morning                             5              3   19.200000   \n",
       "4           Morning                             1              1   15.400000   \n",
       "\n",
       "   Time from Pickup to Arrival  col_type  \n",
       "0                        745.0     train  \n",
       "1                       1993.0     train  \n",
       "2                        455.0     train  \n",
       "3                       1341.0     train  \n",
       "4                       1214.0     train  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[to_int] = df[to_int].apply(pd.to_numeric, errors='coerce')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ready = pd.get_dummies(data=df, columns=cols_dummied)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Destination Lat', 'Destination Long', 'Distance (KM)', 'Order No',\n",
       "       'Pickup Lat', 'Pickup Long', 'Temperature',\n",
       "       'Time from Pickup to Arrival', 'col_type',\n",
       "       'Arrival at Pickup - Day of Month_1',\n",
       "       ...\n",
       "       'Confirmation - Time_Morning', 'Confirmation - Time_Night',\n",
       "       'Confirmation - Time_Noon', 'Pickup - Time_Eve',\n",
       "       'Pickup - Time_Morning', 'Pickup - Time_Noon', 'Placement - Time_Eve',\n",
       "       'Placement - Time_Morning', 'Placement - Time_Night',\n",
       "       'Placement - Time_Noon'],\n",
       "      dtype='object', length=181)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ready.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get train and test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_train = df_ready[df_ready['col_type'] == 'train'].drop(['col_type','Order No'],axis=1)\n",
    "df_test = df_ready[df_ready['col_type'] == 'test'].drop(['col_type',\"Time from Pickup to Arrival\"],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_test = df_ready[df_ready['col_type'] == 'test'].drop(['col_type',\"Time from Pickup to Arrival\"],axis=1)"
   ]
  },
  {
   "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>Destination Lat</th>\n",
       "      <th>Destination Long</th>\n",
       "      <th>Distance (KM)</th>\n",
       "      <th>Pickup Lat</th>\n",
       "      <th>Pickup Long</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Time from Pickup to Arrival</th>\n",
       "      <th>Arrival at Pickup - Day of Month_1</th>\n",
       "      <th>Arrival at Pickup - Day of Month_2</th>\n",
       "      <th>Arrival at Pickup - Day of Month_3</th>\n",
       "      <th>...</th>\n",
       "      <th>Confirmation - Time_Morning</th>\n",
       "      <th>Confirmation - Time_Night</th>\n",
       "      <th>Confirmation - Time_Noon</th>\n",
       "      <th>Pickup - Time_Eve</th>\n",
       "      <th>Pickup - Time_Morning</th>\n",
       "      <th>Pickup - Time_Noon</th>\n",
       "      <th>Placement - Time_Eve</th>\n",
       "      <th>Placement - Time_Morning</th>\n",
       "      <th>Placement - Time_Night</th>\n",
       "      <th>Placement - Time_Noon</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>4</td>\n",
       "      <td>-1.317755</td>\n",
       "      <td>36.830370</td>\n",
       "      <td>20.400000</td>\n",
       "      <td>745.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>-1.295004</td>\n",
       "      <td>36.814358</td>\n",
       "      <td>16</td>\n",
       "      <td>-1.351453</td>\n",
       "      <td>36.899315</td>\n",
       "      <td>26.400000</td>\n",
       "      <td>1993.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>-1.300921</td>\n",
       "      <td>36.828195</td>\n",
       "      <td>3</td>\n",
       "      <td>-1.308284</td>\n",
       "      <td>36.843419</td>\n",
       "      <td>23.255689</td>\n",
       "      <td>455.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>-1.257147</td>\n",
       "      <td>36.795063</td>\n",
       "      <td>9</td>\n",
       "      <td>-1.281301</td>\n",
       "      <td>36.832396</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>1341.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>-1.295041</td>\n",
       "      <td>36.809817</td>\n",
       "      <td>9</td>\n",
       "      <td>-1.266597</td>\n",
       "      <td>36.792118</td>\n",
       "      <td>15.400000</td>\n",
       "      <td>1214.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 179 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Destination Lat  Destination Long  Distance (KM)  Pickup Lat  Pickup Long  \\\n",
       "0        -1.300406         36.829741              4   -1.317755    36.830370   \n",
       "1        -1.295004         36.814358             16   -1.351453    36.899315   \n",
       "2        -1.300921         36.828195              3   -1.308284    36.843419   \n",
       "3        -1.257147         36.795063              9   -1.281301    36.832396   \n",
       "4        -1.295041         36.809817              9   -1.266597    36.792118   \n",
       "\n",
       "   Temperature  Time from Pickup to Arrival  \\\n",
       "0    20.400000                        745.0   \n",
       "1    26.400000                       1993.0   \n",
       "2    23.255689                        455.0   \n",
       "3    19.200000                       1341.0   \n",
       "4    15.400000                       1214.0   \n",
       "\n",
       "   Arrival at Pickup - Day of Month_1  Arrival at Pickup - Day of Month_2  \\\n",
       "0                                   0                                   0   \n",
       "1                                   0                                   0   \n",
       "2                                   0                                   0   \n",
       "3                                   0                                   0   \n",
       "4                                   0                                   0   \n",
       "\n",
       "   Arrival at Pickup - Day of Month_3  ...  Confirmation - Time_Morning  \\\n",
       "0                                   0  ...                            1   \n",
       "1                                   0  ...                            1   \n",
       "2                                   0  ...                            1   \n",
       "3                                   0  ...                            1   \n",
       "4                                   0  ...                            1   \n",
       "\n",
       "   Confirmation - Time_Night  Confirmation - Time_Noon  Pickup - Time_Eve  \\\n",
       "0                          0                         0                  0   \n",
       "1                          0                         0                  0   \n",
       "2                          0                         0                  0   \n",
       "3                          0                         0                  0   \n",
       "4                          0                         0                  0   \n",
       "\n",
       "   Pickup - Time_Morning  Pickup - Time_Noon  Placement - Time_Eve  \\\n",
       "0                      1                   0                     0   \n",
       "1                      1                   0                     0   \n",
       "2                      1                   0                     0   \n",
       "3                      1                   0                     0   \n",
       "4                      1                   0                     0   \n",
       "\n",
       "   Placement - Time_Morning  Placement - Time_Night  Placement - Time_Noon  \n",
       "0                         1                       0                      0  \n",
       "1                         1                       0                      0  \n",
       "2                         1                       0                      0  \n",
       "3                         1                       0                      0  \n",
       "4                         1                       0                      0  \n",
       "\n",
       "[5 rows x 179 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    df_train.drop(\"Time from Pickup to Arrival\",axis=1), df_train['Time from Pickup to Arrival'], test_size=0.05, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[430.54590849479905, 837.146495509579, 0.8091825121080671, 0.3231760114802076, 0.3707763074458793]\n"
     ]
    }
   ],
   "source": [
    "m = RandomForestRegressor(n_estimators=200, min_samples_leaf=2, max_features=0.5, n_jobs=-1, oob_score=True)\n",
    "m.fit(X_train, y_train)\n",
    "print_score(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Predicting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "order_no = df_test[\"Order No\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = m.predict(df_test.drop(\"Order No\",axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1505.15825595, 1198.70520238, 1185.20826984, ..., 1456.29316558,\n",
       "       2769.70013582, 1755.8066057 ])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7068"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(order_no)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission[\"Time from Pickup to Arrival\"] = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission[\"Time from Pickup to Arrival\"] = y\n",
    "submission = submission.set_index('Order_No')\n",
    "submission.to_csv('rf.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submission_01.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "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>cols</th>\n",
       "      <th>imp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Distance (KM)</td>\n",
       "      <td>0.374793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Destination Long</td>\n",
       "      <td>0.094618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Destination Lat</td>\n",
       "      <td>0.093042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Pickup Lat</td>\n",
       "      <td>0.073878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Pickup Long</td>\n",
       "      <td>0.073221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Temperature</td>\n",
       "      <td>0.068506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>173</td>\n",
       "      <td>Pickup - Time_Noon</td>\n",
       "      <td>0.004035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>177</td>\n",
       "      <td>Placement - Time_Noon</td>\n",
       "      <td>0.003799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>166</td>\n",
       "      <td>Arrival at Pickup - Time_Noon</td>\n",
       "      <td>0.003735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>172</td>\n",
       "      <td>Pickup - Time_Morning</td>\n",
       "      <td>0.003725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>175</td>\n",
       "      <td>Placement - Time_Morning</td>\n",
       "      <td>0.003622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>170</td>\n",
       "      <td>Confirmation - Time_Noon</td>\n",
       "      <td>0.003375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>154</td>\n",
       "      <td>Placement - Weekday (Mo = 1)_2</td>\n",
       "      <td>0.003324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>165</td>\n",
       "      <td>Arrival at Pickup - Time_Morning</td>\n",
       "      <td>0.003299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>116</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)_2</td>\n",
       "      <td>0.003248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>168</td>\n",
       "      <td>Confirmation - Time_Morning</td>\n",
       "      <td>0.003184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)_2</td>\n",
       "      <td>0.003137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)_2</td>\n",
       "      <td>0.003107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)_5</td>\n",
       "      <td>0.003089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)_5</td>\n",
       "      <td>0.003030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)_4</td>\n",
       "      <td>0.003016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>113</td>\n",
       "      <td>Personal or Business_Business</td>\n",
       "      <td>0.003010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>156</td>\n",
       "      <td>Placement - Weekday (Mo = 1)_4</td>\n",
       "      <td>0.002998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>114</td>\n",
       "      <td>Personal or Business_Personal</td>\n",
       "      <td>0.002960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>118</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)_4</td>\n",
       "      <td>0.002925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>157</td>\n",
       "      <td>Placement - Weekday (Mo = 1)_5</td>\n",
       "      <td>0.002920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)_4</td>\n",
       "      <td>0.002902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>119</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)_5</td>\n",
       "      <td>0.002886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>155</td>\n",
       "      <td>Placement - Weekday (Mo = 1)_3</td>\n",
       "      <td>0.002781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)_3</td>\n",
       "      <td>0.002775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)_3</td>\n",
       "      <td>0.002771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>162</td>\n",
       "      <td>Platform Type_3</td>\n",
       "      <td>0.002749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>117</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)_3</td>\n",
       "      <td>0.002675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)_1</td>\n",
       "      <td>0.002614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>115</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)_1</td>\n",
       "      <td>0.002610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>153</td>\n",
       "      <td>Placement - Weekday (Mo = 1)_1</td>\n",
       "      <td>0.002577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)_1</td>\n",
       "      <td>0.002529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>160</td>\n",
       "      <td>Platform Type_1</td>\n",
       "      <td>0.002050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>89</td>\n",
       "      <td>Pickup - Day of Month_8</td>\n",
       "      <td>0.001580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>Arrival at Pickup - Day of Month_30</td>\n",
       "      <td>0.001473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>Confirmation - Day of Month_8</td>\n",
       "      <td>0.001457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>111</td>\n",
       "      <td>Pickup - Day of Month_30</td>\n",
       "      <td>0.001421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>161</td>\n",
       "      <td>Platform Type_2</td>\n",
       "      <td>0.001401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>Confirmation - Day of Month_30</td>\n",
       "      <td>0.001346</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>Arrival at Pickup - Day of Month_8</td>\n",
       "      <td>0.001342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>151</td>\n",
       "      <td>Placement - Day of Month_30</td>\n",
       "      <td>0.001330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>129</td>\n",
       "      <td>Placement - Day of Month_8</td>\n",
       "      <td>0.001316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>Confirmation - Day of Month_7</td>\n",
       "      <td>0.001292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>Arrival at Pickup - Day of Month_14</td>\n",
       "      <td>0.001271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>Pickup - Day of Month_15</td>\n",
       "      <td>0.001236</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       cols       imp\n",
       "2                             Distance (KM)  0.374793\n",
       "1                          Destination Long  0.094618\n",
       "0                           Destination Lat  0.093042\n",
       "3                                Pickup Lat  0.073878\n",
       "4                               Pickup Long  0.073221\n",
       "5                               Temperature  0.068506\n",
       "173                      Pickup - Time_Noon  0.004035\n",
       "177                   Placement - Time_Noon  0.003799\n",
       "166           Arrival at Pickup - Time_Noon  0.003735\n",
       "172                   Pickup - Time_Morning  0.003725\n",
       "175                Placement - Time_Morning  0.003622\n",
       "170                Confirmation - Time_Noon  0.003375\n",
       "154          Placement - Weekday (Mo = 1)_2  0.003324\n",
       "165        Arrival at Pickup - Time_Morning  0.003299\n",
       "116             Pickup - Weekday (Mo = 1)_2  0.003248\n",
       "168             Confirmation - Time_Morning  0.003184\n",
       "38   Arrival at Pickup - Weekday (Mo = 1)_2  0.003137\n",
       "76        Confirmation - Weekday (Mo = 1)_2  0.003107\n",
       "41   Arrival at Pickup - Weekday (Mo = 1)_5  0.003089\n",
       "79        Confirmation - Weekday (Mo = 1)_5  0.003030\n",
       "40   Arrival at Pickup - Weekday (Mo = 1)_4  0.003016\n",
       "113           Personal or Business_Business  0.003010\n",
       "156          Placement - Weekday (Mo = 1)_4  0.002998\n",
       "114           Personal or Business_Personal  0.002960\n",
       "118             Pickup - Weekday (Mo = 1)_4  0.002925\n",
       "157          Placement - Weekday (Mo = 1)_5  0.002920\n",
       "78        Confirmation - Weekday (Mo = 1)_4  0.002902\n",
       "119             Pickup - Weekday (Mo = 1)_5  0.002886\n",
       "155          Placement - Weekday (Mo = 1)_3  0.002781\n",
       "77        Confirmation - Weekday (Mo = 1)_3  0.002775\n",
       "39   Arrival at Pickup - Weekday (Mo = 1)_3  0.002771\n",
       "162                         Platform Type_3  0.002749\n",
       "117             Pickup - Weekday (Mo = 1)_3  0.002675\n",
       "37   Arrival at Pickup - Weekday (Mo = 1)_1  0.002614\n",
       "115             Pickup - Weekday (Mo = 1)_1  0.002610\n",
       "153          Placement - Weekday (Mo = 1)_1  0.002577\n",
       "75        Confirmation - Weekday (Mo = 1)_1  0.002529\n",
       "160                         Platform Type_1  0.002050\n",
       "89                  Pickup - Day of Month_8  0.001580\n",
       "35      Arrival at Pickup - Day of Month_30  0.001473\n",
       "51            Confirmation - Day of Month_8  0.001457\n",
       "111                Pickup - Day of Month_30  0.001421\n",
       "161                         Platform Type_2  0.001401\n",
       "73           Confirmation - Day of Month_30  0.001346\n",
       "13       Arrival at Pickup - Day of Month_8  0.001342\n",
       "151             Placement - Day of Month_30  0.001330\n",
       "129              Placement - Day of Month_8  0.001316\n",
       "50            Confirmation - Day of Month_7  0.001292\n",
       "19      Arrival at Pickup - Day of Month_14  0.001271\n",
       "96                 Pickup - Day of Month_15  0.001236"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fi = rf_feat_importance(m,X_train);fi[:50]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fi.plot('cols', 'imp', figsize=(10,6), legend=False);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_fi(fi): return fi.plot('cols', 'imp', 'barh', figsize=(12,7), legend=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Order No', 'User Id', 'Vehicle Type', 'Platform Type',\n",
       "       'Personal or Business', 'Placement - Day of Month',\n",
       "       'Placement - Weekday (Mo = 1)', 'Placement - Time',\n",
       "       'Confirmation - Day of Month', 'Confirmation - Weekday (Mo = 1)',\n",
       "       'Confirmation - Time', 'Arrival at Pickup - Day of Month',\n",
       "       'Arrival at Pickup - Weekday (Mo = 1)', 'Arrival at Pickup - Time',\n",
       "       'Pickup - Day of Month', 'Pickup - Weekday (Mo = 1)', 'Pickup - Time',\n",
       "       'Arrival at Destination - Day of Month',\n",
       "       'Arrival at Destination - Weekday (Mo = 1)',\n",
       "       'Arrival at Destination - Time', 'Distance (KM)', 'Temperature',\n",
       "       'Precipitation in millimeters', 'Pickup Lat', 'Pickup Long',\n",
       "       'Destination Lat', 'Destination Long', 'Rider Id',\n",
       "       'Time from Pickup to Arrival', 'No_Of_Orders', 'Age', 'Average_Rating',\n",
       "       'No_of_Ratings'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rel_feat = ['Distance (KM)','Destination Long','Destination Lat','Pickup Lat', 'Pickup Long','Pickup time','Temperature']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_fi(fi[:30]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### XGBOOST"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import xgboost "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "xgb = xgboost.XGBRegressor(n_estimators=200, learning_rate=0.09, gamma=0, subsample=0.5,\n",
    "                           colsample_bytree=1, max_depth=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/xgboost/core.py:587: FutureWarning: Series.base is deprecated and will be removed in a future version\n",
      "  if getattr(data, 'base', None) is not None and \\\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[17:32:19] WARNING: src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "XGBRegressor(base_score=0.5, booster='gbtree', colsample_bylevel=1,\n",
       "             colsample_bynode=1, colsample_bytree=1, gamma=0,\n",
       "             importance_type='gain', learning_rate=0.09, max_delta_step=0,\n",
       "             max_depth=7, min_child_weight=1, missing=None, n_estimators=200,\n",
       "             n_jobs=1, nthread=None, objective='reg:linear', random_state=0,\n",
       "             reg_alpha=0, reg_lambda=1, scale_pos_weight=1, seed=None,\n",
       "             silent=None, subsample=0.5, verbosity=1)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[586.1097490588558, 831.7712881230683, 0.646379746035527, 0.3318397030562984]\n"
     ]
    }
   ],
   "source": [
    "print_score(xgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = xgb.predict(df_test.drop(\"Order No\",axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1527.0703, 1228.0945, 1168.4033, ..., 1463.0334, 2850.5103,\n",
       "       2101.073 ], dtype=float32)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission[\"Time from Pickup to Arrival\"] = y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = submission.set_index('Order_No')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submission_xgb.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Distance (KM)': 16527635.78130719,\n",
       " 'Destination Long': 2169926.536926603,\n",
       " 'Temperature': 1922496.519928159,\n",
       " 'Arrival at Pickup - Day of Month_9': 2036861.5693552631,\n",
       " 'Destination Lat': 1888660.6449150054,\n",
       " 'Pickup Lat': 2333949.0609059674,\n",
       " 'Arrival at Pickup - Day of Month_2': 2271802.279282759,\n",
       " 'Pickup Long': 2364764.9801419913,\n",
       " 'Arrival at Pickup - Day of Month_10': 1740901.2862354433,\n",
       " 'Platform Type_1': 1891895.2863517855,\n",
       " 'Arrival at Pickup - Day of Month_22': 1682944.5281417912,\n",
       " 'Platform Type_3': 1965846.7116166665,\n",
       " 'Arrival at Pickup - Day of Month_30': 2404475.0744121494,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_2': 2103237.7397648804,\n",
       " 'Personal or Business_Business': 1819000.7074840001,\n",
       " 'Arrival at Pickup - Day of Month_15': 2328060.518561797,\n",
       " 'Arrival at Pickup - Day of Month_31': 1726036.6419866667,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_1': 2335180.9789506844,\n",
       " 'Platform Type_4': 1716478.9849999999,\n",
       " 'Arrival at Pickup - Day of Month_8': 2336441.6984633664,\n",
       " 'Arrival at Pickup - Day of Month_6': 1807044.2766428574,\n",
       " 'Arrival at Pickup - Day of Month_16': 1858483.7177936507,\n",
       " 'Arrival at Pickup - Day of Month_23': 2117248.4083773587,\n",
       " 'Arrival at Pickup - Day of Month_24': 1856091.8298560602,\n",
       " 'Arrival at Pickup - Day of Month_4': 1946321.9844683548,\n",
       " 'Arrival at Pickup - Day of Month_12': 2112845.9972712765,\n",
       " 'Arrival at Pickup - Day of Month_27': 2320946.4208035716,\n",
       " 'Arrival at Pickup - Day of Month_3': 1828928.9633333334,\n",
       " 'Arrival at Pickup - Day of Month_5': 1608215.8844117646,\n",
       " 'Arrival at Pickup - Day of Month_20': 2120270.4113621954,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_6': 2164711.6864942526,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_7': 1777526.0129999998,\n",
       " 'Placement - Time_Noon': 1707761.3667857142,\n",
       " 'Pickup - Time_Noon': 3157123.654761905,\n",
       " 'Arrival at Pickup - Day of Month_14': 1818335.4889803922,\n",
       " 'Arrival at Pickup - Day of Month_17': 2025237.8616287676,\n",
       " 'Arrival at Pickup - Time_Eve': 1305098.7929166667,\n",
       " 'Arrival at Pickup - Day of Month_21': 2357725.6450471906,\n",
       " 'Pickup - Time_Morning': 1899193.772005,\n",
       " 'Arrival at Pickup - Day of Month_19': 1911305.5699032259,\n",
       " 'Arrival at Pickup - Day of Month_26': 1960356.2046206894,\n",
       " 'Pickup - Time_Eve': 2508894.3981250003,\n",
       " 'Arrival at Pickup - Day of Month_29': 1994046.5958656715,\n",
       " 'Confirmation - Time_Noon': 2736182.5442857146,\n",
       " 'Arrival at Pickup - Day of Month_13': 2080919.8821630438,\n",
       " 'Arrival at Pickup - Day of Month_28': 2064504.6733354428,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_4': 1941100.9082591461,\n",
       " 'Arrival at Pickup - Day of Month_11': 2067956.4983333333,\n",
       " 'Arrival at Pickup - Day of Month_1': 2145019.402711864,\n",
       " 'Arrival at Pickup - Time_Noon': 2803654.835571429,\n",
       " 'Arrival at Pickup - Time_Morning': 1851998.1374001736,\n",
       " 'Arrival at Pickup - Day of Month_7': 1844616.5258666663,\n",
       " 'Arrival at Pickup - Day of Month_25': 1909115.258444156,\n",
       " 'Arrival at Pickup - Day of Month_18': 1746606.2305445783,\n",
       " 'Confirmation - Time_Morning': 1948617.3381228065,\n",
       " 'Platform Type_2': 1934438.4309568624,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_5': 1933102.9979314678,\n",
       " 'Arrival at Pickup - Weekday (Mo = 1)_3': 1850850.671727839,\n",
       " 'Placement - Time_Morning': 2353364.550709091,\n",
       " 'Placement - Time_Eve': 1634053.8341666665,\n",
       " 'Placement - Weekday (Mo = 1)_2': 2832769.8740000003,\n",
       " 'Confirmation - Time_Eve': 2026056.72,\n",
       " 'Placement - Day of Month_17': 2716615.25,\n",
       " 'Placement - Day of Month_18': 2236782.4166666665,\n",
       " 'Confirmation - Time_Night': 346347.594,\n",
       " 'Placement - Weekday (Mo = 1)_4': 1406895.435}"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xgb.get_booster().get_score(importance_type=\"gain\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### NEW"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "rel_feat = ['Distance (KM)','Destination Long','Destination Lat','Pickup Lat', 'Pickup Long','Pickup - Time','Temperature','col_type','Time from Pickup to Arrival']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf = df_rf[rel_feat]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Distance (KM)', 'Destination Long', 'Destination Lat', 'Pickup Lat',\n",
       "       'Pickup Long', 'Pickup - Time', 'Temperature', 'col_type',\n",
       "       'Time from Pickup to Arrival'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rf.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Distance (KM)                    int64\n",
       "Destination Long               float64\n",
       "Destination Lat                float64\n",
       "Pickup Lat                     float64\n",
       "Pickup Long                    float64\n",
       "Pickup - Time                   object\n",
       "Temperature                    float64\n",
       "col_type                        object\n",
       "Time from Pickup to Arrival    float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rf.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df_rf['Pickup - Time'] = pd.to_datetime(df_rf['Pickup - Time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Distance (KM)                           int64\n",
       "Destination Long                      float64\n",
       "Destination Lat                       float64\n",
       "Pickup Lat                            float64\n",
       "Pickup Long                           float64\n",
       "Pickup - Time                  datetime64[ns]\n",
       "Temperature                           float64\n",
       "col_type                               object\n",
       "Time from Pickup to Arrival           float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rf.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df_rf['Pickup - Time'] = df_rf['Pickup - Time'].apply(lambda x: x.hour)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df_rf['Pickup - Time'] = df_rf['Pickup - Time'].apply(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf = pd.get_dummies(data=df_rf, columns=['Pickup - Time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Distance (KM)', 'Destination Long', 'Destination Lat', 'Pickup Lat',\n",
       "       'Pickup Long', 'Temperature', 'col_type', 'Time from Pickup to Arrival',\n",
       "       'Pickup - Time_Eve', 'Pickup - Time_Morning', 'Pickup - Time_Noon'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rf.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rf['Temperature'] = df_rf['Temperature'].fillna(df_rf['Temperature'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Distance (KM)                     0\n",
       "Destination Long                  0\n",
       "Destination Lat                   0\n",
       "Pickup Lat                        0\n",
       "Pickup Long                       0\n",
       "Temperature                       0\n",
       "col_type                          0\n",
       "Time from Pickup to Arrival    7068\n",
       "Pickup - Time_Eve                 0\n",
       "Pickup - Time_Morning             0\n",
       "Pickup - Time_Noon                0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_rf.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = df_rf[df_rf['col_type'] == 'train'].drop(['col_type'],axis=1)\n",
    "test = df_rf[df_rf['col_type'] == 'test'].drop(['col_type',\"Time from Pickup to Arrival\"],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    train.drop(\"Time from Pickup to Arrival\",axis=1), train['Time from Pickup to Arrival'], test_size=0.05, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[485.08175941586194, 838.4574115993895, 0.757780485461041, 0.32105462852962297, 0.3346913691549285]\n"
     ]
    }
   ],
   "source": [
    "m = RandomForestRegressor(n_estimators=40, min_samples_leaf=2, max_features=0.5, n_jobs=-1, oob_score=True)\n",
    "m.fit(X_train, y_train)\n",
    "print_score(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "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>cols</th>\n",
       "      <th>imp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Distance (KM)</td>\n",
       "      <td>0.387286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Destination Lat</td>\n",
       "      <td>0.129442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Destination Long</td>\n",
       "      <td>0.129184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Temperature</td>\n",
       "      <td>0.124282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Pickup Lat</td>\n",
       "      <td>0.103531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Pickup Long</td>\n",
       "      <td>0.100208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Pickup - Time_Noon</td>\n",
       "      <td>0.013001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Pickup - Time_Morning</td>\n",
       "      <td>0.012514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Pickup - Time_Eve</td>\n",
       "      <td>0.000551</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    cols       imp\n",
       "0          Distance (KM)  0.387286\n",
       "2        Destination Lat  0.129442\n",
       "1       Destination Long  0.129184\n",
       "5            Temperature  0.124282\n",
       "3             Pickup Lat  0.103531\n",
       "4            Pickup Long  0.100208\n",
       "8     Pickup - Time_Noon  0.013001\n",
       "7  Pickup - Time_Morning  0.012514\n",
       "6      Pickup - Time_Eve  0.000551"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fi = rf_feat_importance(m,X_train);fi[:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11e8b0da0>"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_fi(fi[:30])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "import scipy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.cluster import hierarchy as hc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = np.round(scipy.stats.spearmanr(X_train).correlation, 4)\n",
    "corr_condensed = hc.distance.squareform(1-corr)\n",
    "z = hc.linkage(corr_condensed, method='average')\n",
    "fig = plt.figure(figsize=(16,10))\n",
    "dendrogram = hc.dendrogram(z, labels=X_train.columns, orientation='left', leaf_font_size=16)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = m.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission[\"Time from Pickup to Arrival\"] = y\n",
    "submission = submission.set_index('Order_No')\n",
    "submission.to_csv('rf.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost\n",
    "xgb = xgboost.XGBRegressor(n_estimators=50, learning_rate=0.09, gamma=0, subsample=0.5,\n",
    "                           colsample_bytree=1, max_depth=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/xgboost/core.py:587: FutureWarning: Series.base is deprecated and will be removed in a future version\n",
      "  if getattr(data, 'base', None) is not None and \\\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[09:59:11] WARNING: /usr/local/miniconda/conda-bld/xgboost_1572315027083/work/src/objective/regression_obj.cu:152: reg:linear is now deprecated in favor of reg:squarederror.\n",
      "[709.0256520313436, 830.272596621326, 0.4825085997637185, 0.33424532594503165]\n"
     ]
    }
   ],
   "source": [
    "xgb.fit(X_train,y_train)\n",
    "print_score(xgb)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = xgb.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()\n",
    "submission[\"Time from Pickup to Arrival\"] = y\n",
    "submission = submission.set_index('Order_No')\n",
    "submission.to_csv('xgb.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### LightGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/socrates/anaconda3/lib/python3.7/site-packages/lightgbm/__init__.py:46: UserWarning: Starting from version 2.2.1, the library file in distribution wheels for macOS is built by the Apple Clang (Xcode_8.3.3) compiler.\n",
      "This means that in case of installing LightGBM from PyPI via the ``pip install lightgbm`` command, you don't need to install the gcc compiler anymore.\n",
      "Instead of that, you need to install the OpenMP library, which is required for running LightGBM on the system with the Apple Clang compiler.\n",
      "You can install the OpenMP library by the following command: ``brew install libomp``.\n",
      "  \"You can install the OpenMP library by the following command: ``brew install libomp``.\", UserWarning)\n"
     ]
    }
   ],
   "source": [
    "import lightgbm as lgb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "hyper_params = {\n",
    "    'task': 'train',\n",
    "    'boosting_type': 'gbdt',\n",
    "    'objective': 'regression',\n",
    "    'metric': 'rmse',\n",
    "    'learning_rate': 0.09,\n",
    "    'feature_fraction': 0.9,\n",
    "    'bagging_fraction': 0.7,\n",
    "    'bagging_freq': 10,\n",
    "    'verbose': 1,\n",
    "    \"max_depth\": 6,\n",
    "    \"max_bin\": 512,\n",
    "    \"num_iterations\": 100000,\n",
    "    \"n_estimators\": 1000\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "gbm = lgb.LGBMRegressor(**hyper_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1]\tvalid_0's l1: 737.771\tvalid_0's rmse: 989.115\n",
      "Training until validation scores don't improve for 100 rounds.\n",
      "[2]\tvalid_0's l1: 713.388\tvalid_0's rmse: 964.968\n",
      "[3]\tvalid_0's l1: 691.625\tvalid_0's rmse: 944.042\n",
      "[4]\tvalid_0's l1: 672.431\tvalid_0's rmse: 926.813\n",
      "[5]\tvalid_0's l1: 656.334\tvalid_0's rmse: 912.234\n",
      "[6]\tvalid_0's l1: 642.366\tvalid_0's rmse: 899.479\n",
      "[7]\tvalid_0's l1: 630.046\tvalid_0's rmse: 888.606\n",
      "[8]\tvalid_0's l1: 619.303\tvalid_0's rmse: 879.584\n",
      "[9]\tvalid_0's l1: 609.916\tvalid_0's rmse: 871.525\n",
      "[10]\tvalid_0's l1: 602.3\tvalid_0's rmse: 864.951\n",
      "[11]\tvalid_0's l1: 594.609\tvalid_0's rmse: 858.575\n",
      "[12]\tvalid_0's l1: 588.329\tvalid_0's rmse: 853.802\n",
      "[13]\tvalid_0's l1: 582.844\tvalid_0's rmse: 849.969\n",
      "[14]\tvalid_0's l1: 577.895\tvalid_0's rmse: 846.38\n",
      "[15]\tvalid_0's l1: 573.631\tvalid_0's rmse: 843.436\n",
      "[16]\tvalid_0's l1: 570.403\tvalid_0's rmse: 841.041\n",
      "[17]\tvalid_0's l1: 567.069\tvalid_0's rmse: 838.679\n",
      "[18]\tvalid_0's l1: 564.617\tvalid_0's rmse: 837.355\n",
      "[19]\tvalid_0's l1: 562.488\tvalid_0's rmse: 835.893\n",
      "[20]\tvalid_0's l1: 561.202\tvalid_0's rmse: 834.509\n",
      "[21]\tvalid_0's l1: 559.612\tvalid_0's rmse: 833.731\n",
      "[22]\tvalid_0's l1: 558.23\tvalid_0's rmse: 833.212\n",
      "[23]\tvalid_0's l1: 556.464\tvalid_0's rmse: 832.266\n",
      "[24]\tvalid_0's l1: 555.176\tvalid_0's rmse: 831.493\n",
      "[25]\tvalid_0's l1: 553.785\tvalid_0's rmse: 830.841\n",
      "[26]\tvalid_0's l1: 553.074\tvalid_0's rmse: 830.404\n",
      "[27]\tvalid_0's l1: 552.093\tvalid_0's rmse: 829.99\n",
      "[28]\tvalid_0's l1: 551.261\tvalid_0's rmse: 829.483\n",
      "[29]\tvalid_0's l1: 550.604\tvalid_0's rmse: 829.149\n",
      "[30]\tvalid_0's l1: 549.534\tvalid_0's rmse: 828.534\n",
      "[31]\tvalid_0's l1: 548.724\tvalid_0's rmse: 828.031\n",
      "[32]\tvalid_0's l1: 547.948\tvalid_0's rmse: 828.135\n",
      "[33]\tvalid_0's l1: 547.015\tvalid_0's rmse: 827.575\n",
      "[34]\tvalid_0's l1: 546.556\tvalid_0's rmse: 827.683\n",
      "[35]\tvalid_0's l1: 546.265\tvalid_0's rmse: 827.525\n",
      "[36]\tvalid_0's l1: 545.464\tvalid_0's rmse: 826.88\n",
      "[37]\tvalid_0's l1: 545.121\tvalid_0's rmse: 826.835\n",
      "[38]\tvalid_0's l1: 544.687\tvalid_0's rmse: 826.54\n",
      "[39]\tvalid_0's l1: 544.348\tvalid_0's rmse: 826.547\n",
      "[40]\tvalid_0's l1: 543.944\tvalid_0's rmse: 826.29\n",
      "[41]\tvalid_0's l1: 543.447\tvalid_0's rmse: 826.038\n",
      "[42]\tvalid_0's l1: 543.2\tvalid_0's rmse: 825.74\n",
      "[43]\tvalid_0's l1: 543.28\tvalid_0's rmse: 826.015\n",
      "[44]\tvalid_0's l1: 543.212\tvalid_0's rmse: 825.841\n",
      "[45]\tvalid_0's l1: 542.887\tvalid_0's rmse: 825.59\n",
      "[46]\tvalid_0's l1: 542.562\tvalid_0's rmse: 825.187\n",
      "[47]\tvalid_0's l1: 542.144\tvalid_0's rmse: 824.782\n",
      "[48]\tvalid_0's l1: 542.009\tvalid_0's rmse: 824.929\n",
      "[49]\tvalid_0's l1: 541.972\tvalid_0's rmse: 825.023\n",
      "[50]\tvalid_0's l1: 541.733\tvalid_0's rmse: 824.981\n",
      "[51]\tvalid_0's l1: 541.45\tvalid_0's rmse: 824.542\n",
      "[52]\tvalid_0's l1: 541.527\tvalid_0's rmse: 824.703\n",
      "[53]\tvalid_0's l1: 540.927\tvalid_0's rmse: 824.08\n",
      "[54]\tvalid_0's l1: 541.104\tvalid_0's rmse: 824.447\n",
      "[55]\tvalid_0's l1: 540.889\tvalid_0's rmse: 824.416\n",
      "[56]\tvalid_0's l1: 540.834\tvalid_0's rmse: 824.453\n",
      "[57]\tvalid_0's l1: 540.802\tvalid_0's rmse: 824.513\n",
      "[58]\tvalid_0's l1: 540.793\tvalid_0's rmse: 824.603\n",
      "[59]\tvalid_0's l1: 540.924\tvalid_0's rmse: 824.722\n",
      "[60]\tvalid_0's l1: 540.753\tvalid_0's rmse: 824.841\n",
      "[61]\tvalid_0's l1: 540.702\tvalid_0's rmse: 824.91\n",
      "[62]\tvalid_0's l1: 540.506\tvalid_0's rmse: 824.788\n",
      "[63]\tvalid_0's l1: 540.518\tvalid_0's rmse: 824.818\n",
      "[64]\tvalid_0's l1: 540.646\tvalid_0's rmse: 824.733\n",
      "[65]\tvalid_0's l1: 540.878\tvalid_0's rmse: 824.741\n",
      "[66]\tvalid_0's l1: 541.03\tvalid_0's rmse: 824.834\n",
      "[67]\tvalid_0's l1: 541.1\tvalid_0's rmse: 824.905\n",
      "[68]\tvalid_0's l1: 541.144\tvalid_0's rmse: 824.965\n",
      "[69]\tvalid_0's l1: 541.352\tvalid_0's rmse: 825.118\n",
      "[70]\tvalid_0's l1: 541.215\tvalid_0's rmse: 824.94\n",
      "[71]\tvalid_0's l1: 541.244\tvalid_0's rmse: 824.929\n",
      "[72]\tvalid_0's l1: 540.887\tvalid_0's rmse: 824.611\n",
      "[73]\tvalid_0's l1: 540.896\tvalid_0's rmse: 824.673\n",
      "[74]\tvalid_0's l1: 540.875\tvalid_0's rmse: 824.752\n",
      "[75]\tvalid_0's l1: 540.802\tvalid_0's rmse: 824.785\n",
      "[76]\tvalid_0's l1: 540.764\tvalid_0's rmse: 824.829\n",
      "[77]\tvalid_0's l1: 540.655\tvalid_0's rmse: 824.772\n",
      "[78]\tvalid_0's l1: 540.621\tvalid_0's rmse: 824.917\n",
      "[79]\tvalid_0's l1: 540.58\tvalid_0's rmse: 824.856\n",
      "[80]\tvalid_0's l1: 540.451\tvalid_0's rmse: 824.772\n",
      "[81]\tvalid_0's l1: 540.253\tvalid_0's rmse: 824.809\n",
      "[82]\tvalid_0's l1: 540.272\tvalid_0's rmse: 824.725\n",
      "[83]\tvalid_0's l1: 540.268\tvalid_0's rmse: 824.76\n",
      "[84]\tvalid_0's l1: 540.232\tvalid_0's rmse: 824.641\n",
      "[85]\tvalid_0's l1: 539.897\tvalid_0's rmse: 824.305\n",
      "[86]\tvalid_0's l1: 539.424\tvalid_0's rmse: 823.975\n",
      "[87]\tvalid_0's l1: 539.151\tvalid_0's rmse: 823.937\n",
      "[88]\tvalid_0's l1: 538.845\tvalid_0's rmse: 823.683\n",
      "[89]\tvalid_0's l1: 539.177\tvalid_0's rmse: 824\n",
      "[90]\tvalid_0's l1: 538.847\tvalid_0's rmse: 823.75\n",
      "[91]\tvalid_0's l1: 538.507\tvalid_0's rmse: 823.464\n",
      "[92]\tvalid_0's l1: 538.658\tvalid_0's rmse: 823.03\n",
      "[93]\tvalid_0's l1: 538.496\tvalid_0's rmse: 822.942\n",
      "[94]\tvalid_0's l1: 538.574\tvalid_0's rmse: 823.104\n",
      "[95]\tvalid_0's l1: 538.658\tvalid_0's rmse: 822.854\n",
      "[96]\tvalid_0's l1: 538.532\tvalid_0's rmse: 822.815\n",
      "[97]\tvalid_0's l1: 538.843\tvalid_0's rmse: 823.068\n",
      "[98]\tvalid_0's l1: 539.159\tvalid_0's rmse: 823.394\n",
      "[99]\tvalid_0's l1: 539.163\tvalid_0's rmse: 823.256\n",
      "[100]\tvalid_0's l1: 539.169\tvalid_0's rmse: 823.278\n",
      "[101]\tvalid_0's l1: 539\tvalid_0's rmse: 823.102\n",
      "[102]\tvalid_0's l1: 538.982\tvalid_0's rmse: 822.836\n",
      "[103]\tvalid_0's l1: 538.887\tvalid_0's rmse: 822.552\n",
      "[104]\tvalid_0's l1: 538.789\tvalid_0's rmse: 822.397\n",
      "[105]\tvalid_0's l1: 538.577\tvalid_0's rmse: 822.463\n",
      "[106]\tvalid_0's l1: 538.533\tvalid_0's rmse: 822.42\n",
      "[107]\tvalid_0's l1: 538.497\tvalid_0's rmse: 822.466\n",
      "[108]\tvalid_0's l1: 538.433\tvalid_0's rmse: 822.561\n",
      "[109]\tvalid_0's l1: 538.348\tvalid_0's rmse: 822.573\n",
      "[110]\tvalid_0's l1: 538.47\tvalid_0's rmse: 822.693\n",
      "[111]\tvalid_0's l1: 538.674\tvalid_0's rmse: 822.784\n",
      "[112]\tvalid_0's l1: 538.963\tvalid_0's rmse: 822.929\n",
      "[113]\tvalid_0's l1: 538.812\tvalid_0's rmse: 822.741\n",
      "[114]\tvalid_0's l1: 538.763\tvalid_0's rmse: 822.754\n",
      "[115]\tvalid_0's l1: 538.919\tvalid_0's rmse: 822.807\n",
      "[116]\tvalid_0's l1: 539.178\tvalid_0's rmse: 822.913\n",
      "[117]\tvalid_0's l1: 539.268\tvalid_0's rmse: 822.874\n",
      "[118]\tvalid_0's l1: 539.257\tvalid_0's rmse: 822.828\n",
      "[119]\tvalid_0's l1: 539.377\tvalid_0's rmse: 822.899\n",
      "[120]\tvalid_0's l1: 539.372\tvalid_0's rmse: 822.936\n",
      "[121]\tvalid_0's l1: 539.578\tvalid_0's rmse: 823.132\n",
      "[122]\tvalid_0's l1: 539.934\tvalid_0's rmse: 823.552\n",
      "[123]\tvalid_0's l1: 539.958\tvalid_0's rmse: 823.707\n",
      "[124]\tvalid_0's l1: 540.155\tvalid_0's rmse: 823.98\n",
      "[125]\tvalid_0's l1: 540.296\tvalid_0's rmse: 823.892\n",
      "[126]\tvalid_0's l1: 540.371\tvalid_0's rmse: 823.724\n",
      "[127]\tvalid_0's l1: 540.609\tvalid_0's rmse: 824\n",
      "[128]\tvalid_0's l1: 540.915\tvalid_0's rmse: 824.219\n",
      "[129]\tvalid_0's l1: 541.187\tvalid_0's rmse: 824.563\n",
      "[130]\tvalid_0's l1: 541.536\tvalid_0's rmse: 824.76\n",
      "[131]\tvalid_0's l1: 541.823\tvalid_0's rmse: 824.963\n",
      "[132]\tvalid_0's l1: 541.56\tvalid_0's rmse: 824.892\n",
      "[133]\tvalid_0's l1: 541.75\tvalid_0's rmse: 825.172\n",
      "[134]\tvalid_0's l1: 541.546\tvalid_0's rmse: 825.192\n",
      "[135]\tvalid_0's l1: 541.656\tvalid_0's rmse: 825.264\n",
      "[136]\tvalid_0's l1: 541.856\tvalid_0's rmse: 825.285\n",
      "[137]\tvalid_0's l1: 541.873\tvalid_0's rmse: 825.307\n",
      "[138]\tvalid_0's l1: 542.147\tvalid_0's rmse: 825.266\n",
      "[139]\tvalid_0's l1: 542.385\tvalid_0's rmse: 825.388\n",
      "[140]\tvalid_0's l1: 542.707\tvalid_0's rmse: 825.734\n",
      "[141]\tvalid_0's l1: 542.237\tvalid_0's rmse: 825.634\n",
      "[142]\tvalid_0's l1: 541.95\tvalid_0's rmse: 825.795\n",
      "[143]\tvalid_0's l1: 541.778\tvalid_0's rmse: 825.846\n",
      "[144]\tvalid_0's l1: 541.667\tvalid_0's rmse: 825.981\n",
      "[145]\tvalid_0's l1: 541.699\tvalid_0's rmse: 826.067\n",
      "[146]\tvalid_0's l1: 541.786\tvalid_0's rmse: 826.313\n",
      "[147]\tvalid_0's l1: 541.772\tvalid_0's rmse: 826.605\n",
      "[148]\tvalid_0's l1: 541.961\tvalid_0's rmse: 826.657\n",
      "[149]\tvalid_0's l1: 541.946\tvalid_0's rmse: 826.707\n",
      "[150]\tvalid_0's l1: 541.758\tvalid_0's rmse: 826.6\n",
      "[151]\tvalid_0's l1: 541.689\tvalid_0's rmse: 826.671\n",
      "[152]\tvalid_0's l1: 541.64\tvalid_0's rmse: 826.743\n",
      "[153]\tvalid_0's l1: 541.672\tvalid_0's rmse: 826.861\n",
      "[154]\tvalid_0's l1: 541.724\tvalid_0's rmse: 826.911\n",
      "[155]\tvalid_0's l1: 541.645\tvalid_0's rmse: 826.84\n",
      "[156]\tvalid_0's l1: 541.455\tvalid_0's rmse: 826.675\n",
      "[157]\tvalid_0's l1: 541.649\tvalid_0's rmse: 826.83\n",
      "[158]\tvalid_0's l1: 541.942\tvalid_0's rmse: 826.992\n",
      "[159]\tvalid_0's l1: 541.787\tvalid_0's rmse: 826.961\n",
      "[160]\tvalid_0's l1: 541.782\tvalid_0's rmse: 826.91\n",
      "[161]\tvalid_0's l1: 541.808\tvalid_0's rmse: 826.845\n",
      "[162]\tvalid_0's l1: 541.985\tvalid_0's rmse: 826.856\n",
      "[163]\tvalid_0's l1: 541.996\tvalid_0's rmse: 826.703\n",
      "[164]\tvalid_0's l1: 542.244\tvalid_0's rmse: 826.751\n",
      "[165]\tvalid_0's l1: 542.216\tvalid_0's rmse: 826.773\n",
      "[166]\tvalid_0's l1: 542.659\tvalid_0's rmse: 826.954\n",
      "[167]\tvalid_0's l1: 542.715\tvalid_0's rmse: 826.956\n",
      "[168]\tvalid_0's l1: 542.712\tvalid_0's rmse: 826.912\n",
      "[169]\tvalid_0's l1: 542.86\tvalid_0's rmse: 826.993\n",
      "[170]\tvalid_0's l1: 542.892\tvalid_0's rmse: 826.934\n",
      "[171]\tvalid_0's l1: 542.707\tvalid_0's rmse: 826.893\n",
      "[172]\tvalid_0's l1: 542.743\tvalid_0's rmse: 827.028\n",
      "[173]\tvalid_0's l1: 542.765\tvalid_0's rmse: 827.102\n",
      "[174]\tvalid_0's l1: 542.623\tvalid_0's rmse: 827.106\n",
      "[175]\tvalid_0's l1: 542.287\tvalid_0's rmse: 827.039\n",
      "[176]\tvalid_0's l1: 542.229\tvalid_0's rmse: 826.921\n",
      "[177]\tvalid_0's l1: 542.202\tvalid_0's rmse: 826.786\n",
      "[178]\tvalid_0's l1: 542.291\tvalid_0's rmse: 826.814\n",
      "[179]\tvalid_0's l1: 542.237\tvalid_0's rmse: 826.804\n",
      "[180]\tvalid_0's l1: 542.215\tvalid_0's rmse: 826.785\n",
      "[181]\tvalid_0's l1: 542.367\tvalid_0's rmse: 826.873\n",
      "[182]\tvalid_0's l1: 542.332\tvalid_0's rmse: 826.808\n",
      "[183]\tvalid_0's l1: 542.428\tvalid_0's rmse: 826.888\n",
      "[184]\tvalid_0's l1: 542.6\tvalid_0's rmse: 827.229\n",
      "[185]\tvalid_0's l1: 542.607\tvalid_0's rmse: 827.173\n",
      "[186]\tvalid_0's l1: 542.569\tvalid_0's rmse: 827.202\n",
      "[187]\tvalid_0's l1: 542.542\tvalid_0's rmse: 827.155\n",
      "[188]\tvalid_0's l1: 542.745\tvalid_0's rmse: 827.396\n",
      "[189]\tvalid_0's l1: 542.645\tvalid_0's rmse: 827.439\n",
      "[190]\tvalid_0's l1: 542.429\tvalid_0's rmse: 827.3\n",
      "[191]\tvalid_0's l1: 542.362\tvalid_0's rmse: 826.924\n",
      "[192]\tvalid_0's l1: 542.328\tvalid_0's rmse: 826.725\n",
      "[193]\tvalid_0's l1: 542.326\tvalid_0's rmse: 826.765\n",
      "[194]\tvalid_0's l1: 542.19\tvalid_0's rmse: 826.63\n",
      "[195]\tvalid_0's l1: 542.48\tvalid_0's rmse: 827.002\n",
      "[196]\tvalid_0's l1: 542.556\tvalid_0's rmse: 826.854\n",
      "[197]\tvalid_0's l1: 542.667\tvalid_0's rmse: 826.791\n",
      "[198]\tvalid_0's l1: 542.547\tvalid_0's rmse: 826.774\n",
      "[199]\tvalid_0's l1: 542.468\tvalid_0's rmse: 826.855\n",
      "[200]\tvalid_0's l1: 542.441\tvalid_0's rmse: 826.892\n",
      "[201]\tvalid_0's l1: 542.186\tvalid_0's rmse: 826.914\n",
      "[202]\tvalid_0's l1: 542.139\tvalid_0's rmse: 827.175\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[203]\tvalid_0's l1: 541.719\tvalid_0's rmse: 826.79\n",
      "[204]\tvalid_0's l1: 541.544\tvalid_0's rmse: 826.879\n",
      "Early stopping, best iteration is:\n",
      "[104]\tvalid_0's l1: 538.789\tvalid_0's rmse: 822.397\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LGBMRegressor(bagging_fraction=0.7, bagging_freq=10, boosting_type='gbdt',\n",
       "              class_weight=None, colsample_bytree=1.0, feature_fraction=0.9,\n",
       "              importance_type='split', learning_rate=0.09, max_bin=512,\n",
       "              max_depth=6, metric='rmse', min_child_samples=20,\n",
       "              min_child_weight=0.001, min_split_gain=0.0, n_estimators=1000,\n",
       "              n_jobs=-1, num_iterations=100000, num_leaves=31,\n",
       "              objective='regression', random_state=None, reg_alpha=0.0,\n",
       "              reg_lambda=0.0, silent=True, subsample=1.0,\n",
       "              subsample_for_bin=200000, subsample_freq=0, task='train',\n",
       "              verbose=1)"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbm.fit(X_train, y_train,\n",
    "        eval_set=[(X_test, y_test)],\n",
    "        eval_metric='l1',\n",
    "        early_stopping_rounds=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[730.1614162105948, 822.3969529757484, 0.45119633764477096, 0.3468156044014733]\n"
     ]
    }
   ],
   "source": [
    "print_score(gbm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = gbm.predict(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = df_sample_submission.copy()\n",
    "submission[\"Time from Pickup to Arrival\"] = y\n",
    "submission = submission.set_index('Order_No')\n",
    "submission.to_csv('gbm.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "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>Order No</th>\n",
       "      <th>Unique number identifying the order</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>User Id</td>\n",
       "      <td>Unique number identifying the customer on a pl...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Vehicle Type</td>\n",
       "      <td>For this competition limited to bikes, however...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Platform Type</td>\n",
       "      <td>Platform used to place the order, there are 4 ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Personal or Business</td>\n",
       "      <td>Customer type</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Placement - Day of Month</td>\n",
       "      <td>Placement - Day of Month i.e 1-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>Placement - Weekday (Mo = 1)</td>\n",
       "      <td>Placement - Weekday (Monday = 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>Placement - Time</td>\n",
       "      <td>Placement - Time - Time of day the order was p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>Confirmation - Day of Month</td>\n",
       "      <td>Confirmation - Day of Month i.e 1-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>Confirmation - Weekday (Mo = 1)</td>\n",
       "      <td>Confirmation - Weekday (Monday = 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>Confirmation - Time</td>\n",
       "      <td>Confirmation - Time - Time of day the order wa...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>Arrival at Pickup - Day of Month</td>\n",
       "      <td>Arrival at Pickup - Day of Month i.e 1-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>Arrival at Pickup - Weekday (Monday = 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>Arrival at Pickup - Time</td>\n",
       "      <td>Time of day the the rider arrived at the locat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>Pickup - Day of Month</td>\n",
       "      <td>Pickup - Day of Month i.e 1-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>Pickup - Weekday (Monday = 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>Pickup - Time</td>\n",
       "      <td>Pickup - Time - Time of day the the rider pick...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>Arrival at Destination - Day of Month</td>\n",
       "      <td>Arrival at Delivery - Day of Month i.e 1-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>Arrival at Destination - Weekday (Mo = 1)</td>\n",
       "      <td>Arrival at Delivery - Weekday (Monday = 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>Arrival at Destination - Time</td>\n",
       "      <td>Arrival at Delivery Time - Time of day the rid...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>Distance (KM)</td>\n",
       "      <td>The distance from Pickup to Destination</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>Temperature</td>\n",
       "      <td>Temperature at the time of order placement in ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>Precipitation in millimeters</td>\n",
       "      <td>Precipitation at the time of order placement ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>Pickup Lat</td>\n",
       "      <td>Latitude of pick up location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>Pickup Long</td>\n",
       "      <td>Longitude of pick up location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>Destination Lat</td>\n",
       "      <td>Latitude of delivery location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>Destination Long</td>\n",
       "      <td>Longitude of delivery location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>Rider Id</td>\n",
       "      <td>ID of the Rider who accepted the order</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>Time from Pickup to Arrival</td>\n",
       "      <td>Time in seconds between 'Pickup' and 'Arrival ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>Rider Metrics</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>Rider Id</td>\n",
       "      <td>Unique number identifying the rider (same as i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>No_Of_Orders</td>\n",
       "      <td>Number of Orders the rider has delivered</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>Age</td>\n",
       "      <td>Number of days since the rider delivered the f...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>Average_Rating</td>\n",
       "      <td>Average rating of the rider</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>No_of_Ratings</td>\n",
       "      <td>Number of ratings the rider has received. Rat...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     Order No  \\\n",
       "0                                     User Id   \n",
       "1                                Vehicle Type   \n",
       "2                               Platform Type   \n",
       "3                        Personal or Business   \n",
       "4                    Placement - Day of Month   \n",
       "5                Placement - Weekday (Mo = 1)   \n",
       "6                            Placement - Time   \n",
       "7                 Confirmation - Day of Month   \n",
       "8             Confirmation - Weekday (Mo = 1)   \n",
       "9                         Confirmation - Time   \n",
       "10           Arrival at Pickup - Day of Month   \n",
       "11       Arrival at Pickup - Weekday (Mo = 1)   \n",
       "12                   Arrival at Pickup - Time   \n",
       "13                      Pickup - Day of Month   \n",
       "14                  Pickup - Weekday (Mo = 1)   \n",
       "15                              Pickup - Time   \n",
       "16      Arrival at Destination - Day of Month   \n",
       "17  Arrival at Destination - Weekday (Mo = 1)   \n",
       "18              Arrival at Destination - Time   \n",
       "19                              Distance (KM)   \n",
       "20                                Temperature   \n",
       "21               Precipitation in millimeters   \n",
       "22                                 Pickup Lat   \n",
       "23                                Pickup Long   \n",
       "24                            Destination Lat   \n",
       "25                           Destination Long   \n",
       "26                                   Rider Id   \n",
       "27                Time from Pickup to Arrival   \n",
       "28                                        NaN   \n",
       "29                              Rider Metrics   \n",
       "30                                   Rider Id   \n",
       "31                               No_Of_Orders   \n",
       "32                                        Age   \n",
       "33                             Average_Rating   \n",
       "34                              No_of_Ratings   \n",
       "\n",
       "                  Unique number identifying the order  \n",
       "0   Unique number identifying the customer on a pl...  \n",
       "1   For this competition limited to bikes, however...  \n",
       "2   Platform used to place the order, there are 4 ...  \n",
       "3                                       Customer type  \n",
       "4                   Placement - Day of Month i.e 1-31  \n",
       "5                    Placement - Weekday (Monday = 1)  \n",
       "6   Placement - Time - Time of day the order was p...  \n",
       "7                Confirmation - Day of Month i.e 1-31  \n",
       "8                 Confirmation - Weekday (Monday = 1)  \n",
       "9   Confirmation - Time - Time of day the order wa...  \n",
       "10          Arrival at Pickup - Day of Month i.e 1-31  \n",
       "11           Arrival at Pickup - Weekday (Monday = 1)  \n",
       "12  Time of day the the rider arrived at the locat...  \n",
       "13                     Pickup - Day of Month i.e 1-31  \n",
       "14                      Pickup - Weekday (Monday = 1)  \n",
       "15  Pickup - Time - Time of day the the rider pick...  \n",
       "16        Arrival at Delivery - Day of Month i.e 1-31  \n",
       "17         Arrival at Delivery - Weekday (Monday = 1)  \n",
       "18  Arrival at Delivery Time - Time of day the rid...  \n",
       "19            The distance from Pickup to Destination  \n",
       "20  Temperature at the time of order placement in ...  \n",
       "21   Precipitation at the time of order placement ...  \n",
       "22                       Latitude of pick up location  \n",
       "23                      Longitude of pick up location  \n",
       "24                      Latitude of delivery location  \n",
       "25                     Longitude of delivery location  \n",
       "26             ID of the Rider who accepted the order  \n",
       "27  Time in seconds between 'Pickup' and 'Arrival ...  \n",
       "28                                                NaN  \n",
       "29                                                NaN  \n",
       "30  Unique number identifying the rider (same as i...  \n",
       "31           Number of Orders the rider has delivered  \n",
       "32  Number of days since the rider delivered the f...  \n",
       "33                        Average rating of the rider  \n",
       "34   Number of ratings the rider has received. Rat...  "
      ]
     },
     "execution_count": 163,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_variable_definitions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "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>Rider Id</th>\n",
       "      <th>No_Of_Orders</th>\n",
       "      <th>Age</th>\n",
       "      <th>Average_Rating</th>\n",
       "      <th>No_of_Ratings</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Rider_Id_396</td>\n",
       "      <td>2946</td>\n",
       "      <td>2298</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1159</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Rider_Id_479</td>\n",
       "      <td>360</td>\n",
       "      <td>951</td>\n",
       "      <td>13.5</td>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Rider_Id_648</td>\n",
       "      <td>1746</td>\n",
       "      <td>821</td>\n",
       "      <td>14.3</td>\n",
       "      <td>466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Rider_Id_753</td>\n",
       "      <td>314</td>\n",
       "      <td>980</td>\n",
       "      <td>12.5</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Rider_Id_335</td>\n",
       "      <td>536</td>\n",
       "      <td>1113</td>\n",
       "      <td>13.7</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>955</td>\n",
       "      <td>Rider_Id_896</td>\n",
       "      <td>152</td>\n",
       "      <td>99</td>\n",
       "      <td>12.4</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>956</td>\n",
       "      <td>Rider_Id_149</td>\n",
       "      <td>69</td>\n",
       "      <td>101</td>\n",
       "      <td>10.2</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>957</td>\n",
       "      <td>Rider_Id_270</td>\n",
       "      <td>338</td>\n",
       "      <td>96</td>\n",
       "      <td>14.4</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>958</td>\n",
       "      <td>Rider_Id_201</td>\n",
       "      <td>159</td>\n",
       "      <td>96</td>\n",
       "      <td>15.0</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>959</td>\n",
       "      <td>Rider_Id_904</td>\n",
       "      <td>172</td>\n",
       "      <td>108</td>\n",
       "      <td>13.7</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>960 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         Rider Id  No_Of_Orders   Age  Average_Rating  No_of_Ratings\n",
       "0    Rider_Id_396          2946  2298            14.0           1159\n",
       "1    Rider_Id_479           360   951            13.5            176\n",
       "2    Rider_Id_648          1746   821            14.3            466\n",
       "3    Rider_Id_753           314   980            12.5             75\n",
       "4    Rider_Id_335           536  1113            13.7            156\n",
       "..            ...           ...   ...             ...            ...\n",
       "955  Rider_Id_896           152    99            12.4             18\n",
       "956  Rider_Id_149            69   101            10.2             10\n",
       "957  Rider_Id_270           338    96            14.4             41\n",
       "958  Rider_Id_201           159    96            15.0              9\n",
       "959  Rider_Id_904           172   108            13.7             18\n",
       "\n",
       "[960 rows x 5 columns]"
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_riders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Order No', 'User Id', 'Vehicle Type', 'Platform Type',\n",
       "       'Personal or Business', 'Placement - Day of Month',\n",
       "       'Placement - Weekday (Mo = 1)', 'Placement - Time',\n",
       "       'Confirmation - Day of Month', 'Confirmation - Weekday (Mo = 1)',\n",
       "       'Confirmation - Time', 'Arrival at Pickup - Day of Month',\n",
       "       'Arrival at Pickup - Weekday (Mo = 1)', 'Arrival at Pickup - Time',\n",
       "       'Pickup - Day of Month', 'Pickup - Weekday (Mo = 1)', 'Pickup - Time',\n",
       "       'Arrival at Destination - Day of Month',\n",
       "       'Arrival at Destination - Weekday (Mo = 1)',\n",
       "       'Arrival at Destination - Time', 'Distance (KM)', 'Temperature',\n",
       "       'Precipitation in millimeters', 'Pickup Lat', 'Pickup Long',\n",
       "       'Destination Lat', 'Destination Long', 'Rider Id',\n",
       "       'Time from Pickup to Arrival'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 166,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>21191</th>\n",
       "      <th>21192</th>\n",
       "      <th>21193</th>\n",
       "      <th>21194</th>\n",
       "      <th>21195</th>\n",
       "      <th>21196</th>\n",
       "      <th>21197</th>\n",
       "      <th>21198</th>\n",
       "      <th>21199</th>\n",
       "      <th>21200</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>Order No</td>\n",
       "      <td>Order_No_4211</td>\n",
       "      <td>Order_No_25375</td>\n",
       "      <td>Order_No_1899</td>\n",
       "      <td>Order_No_9336</td>\n",
       "      <td>Order_No_27883</td>\n",
       "      <td>Order_No_7408</td>\n",
       "      <td>Order_No_22680</td>\n",
       "      <td>Order_No_21578</td>\n",
       "      <td>Order_No_5234</td>\n",
       "      <td>Order_No_1768</td>\n",
       "      <td>...</td>\n",
       "      <td>Order_No_16262</td>\n",
       "      <td>Order_No_11670</td>\n",
       "      <td>Order_No_4988</td>\n",
       "      <td>Order_No_865</td>\n",
       "      <td>Order_No_9932</td>\n",
       "      <td>Order_No_8834</td>\n",
       "      <td>Order_No_22892</td>\n",
       "      <td>Order_No_2831</td>\n",
       "      <td>Order_No_6174</td>\n",
       "      <td>Order_No_9836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>User Id</td>\n",
       "      <td>User_Id_633</td>\n",
       "      <td>User_Id_2285</td>\n",
       "      <td>User_Id_265</td>\n",
       "      <td>User_Id_1402</td>\n",
       "      <td>User_Id_1737</td>\n",
       "      <td>User_Id_1342</td>\n",
       "      <td>User_Id_2803</td>\n",
       "      <td>User_Id_1075</td>\n",
       "      <td>User_Id_733</td>\n",
       "      <td>User_Id_2112</td>\n",
       "      <td>...</td>\n",
       "      <td>User_Id_1231</td>\n",
       "      <td>User_Id_1146</td>\n",
       "      <td>User_Id_875</td>\n",
       "      <td>User_Id_1245</td>\n",
       "      <td>User_Id_3582</td>\n",
       "      <td>User_Id_2001</td>\n",
       "      <td>User_Id_1796</td>\n",
       "      <td>User_Id_2956</td>\n",
       "      <td>User_Id_2524</td>\n",
       "      <td>User_Id_718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Vehicle Type</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>...</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "      <td>Bike</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Platform Type</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Personal or Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>...</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "      <td>Business</td>\n",
       "      <td>Personal</td>\n",
       "      <td>Business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Placement - Time</td>\n",
       "      <td>9:35:46 AM</td>\n",
       "      <td>11:16:16 AM</td>\n",
       "      <td>12:39:25 PM</td>\n",
       "      <td>9:25:34 AM</td>\n",
       "      <td>9:55:18 AM</td>\n",
       "      <td>3:07:35 PM</td>\n",
       "      <td>9:33:45 AM</td>\n",
       "      <td>2:13:01 PM</td>\n",
       "      <td>11:10:44 AM</td>\n",
       "      <td>4:48:54 PM</td>\n",
       "      <td>...</td>\n",
       "      <td>2:09:37 PM</td>\n",
       "      <td>12:41:17 PM</td>\n",
       "      <td>4:48:16 PM</td>\n",
       "      <td>5:02:09 PM</td>\n",
       "      <td>1:08:34 PM</td>\n",
       "      <td>3:54:38 PM</td>\n",
       "      <td>10:13:34 AM</td>\n",
       "      <td>5:06:16 PM</td>\n",
       "      <td>9:31:39 AM</td>\n",
       "      <td>2:19:47 PM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Confirmation - Time</td>\n",
       "      <td>9:40:10 AM</td>\n",
       "      <td>11:23:21 AM</td>\n",
       "      <td>12:42:44 PM</td>\n",
       "      <td>9:26:05 AM</td>\n",
       "      <td>9:56:18 AM</td>\n",
       "      <td>3:08:57 PM</td>\n",
       "      <td>9:49:47 AM</td>\n",
       "      <td>2:14:13 PM</td>\n",
       "      <td>11:15:49 AM</td>\n",
       "      <td>5:17:56 PM</td>\n",
       "      <td>...</td>\n",
       "      <td>2:10:27 PM</td>\n",
       "      <td>12:41:52 PM</td>\n",
       "      <td>4:49:51 PM</td>\n",
       "      <td>5:02:52 PM</td>\n",
       "      <td>1:25:40 PM</td>\n",
       "      <td>3:55:09 PM</td>\n",
       "      <td>10:13:41 AM</td>\n",
       "      <td>5:07:09 PM</td>\n",
       "      <td>9:31:53 AM</td>\n",
       "      <td>2:20:01 PM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Pickup - Time</td>\n",
       "      <td>10:04:47 AM</td>\n",
       "      <td>11:40:22 AM</td>\n",
       "      <td>12:49:34 PM</td>\n",
       "      <td>9:37:56 AM</td>\n",
       "      <td>10:03:53 AM</td>\n",
       "      <td>3:21:36 PM</td>\n",
       "      <td>9:53:12 AM</td>\n",
       "      <td>2:21:33 PM</td>\n",
       "      <td>12:13:18 PM</td>\n",
       "      <td>5:32:41 PM</td>\n",
       "      <td>...</td>\n",
       "      <td>2:51:53 PM</td>\n",
       "      <td>12:44:19 PM</td>\n",
       "      <td>5:06:48 PM</td>\n",
       "      <td>5:08:13 PM</td>\n",
       "      <td>1:52:11 PM</td>\n",
       "      <td>3:58:49 PM</td>\n",
       "      <td>10:20:04 AM</td>\n",
       "      <td>5:30:17 PM</td>\n",
       "      <td>9:38:59 AM</td>\n",
       "      <td>2:24:29 PM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup - Time</td>\n",
       "      <td>10:27:30 AM</td>\n",
       "      <td>11:44:09 AM</td>\n",
       "      <td>12:53:03 PM</td>\n",
       "      <td>9:43:06 AM</td>\n",
       "      <td>10:05:23 AM</td>\n",
       "      <td>3:30:30 PM</td>\n",
       "      <td>9:56:45 AM</td>\n",
       "      <td>2:22:40 PM</td>\n",
       "      <td>12:22:57 PM</td>\n",
       "      <td>5:34:38 PM</td>\n",
       "      <td>...</td>\n",
       "      <td>2:53:08 PM</td>\n",
       "      <td>1:05:11 PM</td>\n",
       "      <td>5:27:22 PM</td>\n",
       "      <td>5:15:15 PM</td>\n",
       "      <td>1:54:36 PM</td>\n",
       "      <td>4:20:08 PM</td>\n",
       "      <td>10:33:27 AM</td>\n",
       "      <td>5:50:52 PM</td>\n",
       "      <td>9:45:15 AM</td>\n",
       "      <td>2:41:55 PM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Destination - Day of Month</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>30</td>\n",
       "      <td>15</td>\n",
       "      <td>13</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>23</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>23</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Destination - Weekday (Mo = 1)</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Arrival at Destination - Time</td>\n",
       "      <td>10:39:55 AM</td>\n",
       "      <td>12:17:22 PM</td>\n",
       "      <td>1:00:38 PM</td>\n",
       "      <td>10:05:27 AM</td>\n",
       "      <td>10:25:37 AM</td>\n",
       "      <td>4:23:41 PM</td>\n",
       "      <td>10:19:45 AM</td>\n",
       "      <td>2:33:26 PM</td>\n",
       "      <td>1:19:35 PM</td>\n",
       "      <td>6:31:57 PM</td>\n",
       "      <td>...</td>\n",
       "      <td>3:39:55 PM</td>\n",
       "      <td>1:21:22 PM</td>\n",
       "      <td>5:56:53 PM</td>\n",
       "      <td>5:53:33 PM</td>\n",
       "      <td>2:33:21 PM</td>\n",
       "      <td>4:20:17 PM</td>\n",
       "      <td>10:46:17 AM</td>\n",
       "      <td>6:40:05 PM</td>\n",
       "      <td>10:08:15 AM</td>\n",
       "      <td>3:17:23 PM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Distance (KM)</td>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "      <td>...</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>10</td>\n",
       "      <td>16</td>\n",
       "      <td>17</td>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>20</td>\n",
       "      <td>13</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Temperature</td>\n",
       "      <td>20.4</td>\n",
       "      <td>26.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19.2</td>\n",
       "      <td>15.4</td>\n",
       "      <td>27.2</td>\n",
       "      <td>20.3</td>\n",
       "      <td>28.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24.6</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>29</td>\n",
       "      <td>28.6</td>\n",
       "      <td>26</td>\n",
       "      <td>29.2</td>\n",
       "      <td>15</td>\n",
       "      <td>30.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Precipitation in millimeters</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup Lat</td>\n",
       "      <td>-1.31775</td>\n",
       "      <td>-1.35145</td>\n",
       "      <td>-1.30828</td>\n",
       "      <td>-1.2813</td>\n",
       "      <td>-1.2666</td>\n",
       "      <td>-1.30258</td>\n",
       "      <td>-1.2794</td>\n",
       "      <td>-1.2528</td>\n",
       "      <td>-1.25519</td>\n",
       "      <td>-1.22532</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.28214</td>\n",
       "      <td>-1.31671</td>\n",
       "      <td>-1.28657</td>\n",
       "      <td>-1.26023</td>\n",
       "      <td>-1.23841</td>\n",
       "      <td>-1.25841</td>\n",
       "      <td>-1.30714</td>\n",
       "      <td>-1.28602</td>\n",
       "      <td>-1.25003</td>\n",
       "      <td>-1.25519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Pickup Long</td>\n",
       "      <td>36.8304</td>\n",
       "      <td>36.8993</td>\n",
       "      <td>36.8434</td>\n",
       "      <td>36.8324</td>\n",
       "      <td>36.7921</td>\n",
       "      <td>36.7671</td>\n",
       "      <td>36.8254</td>\n",
       "      <td>36.8003</td>\n",
       "      <td>36.7822</td>\n",
       "      <td>36.8085</td>\n",
       "      <td>...</td>\n",
       "      <td>36.8162</td>\n",
       "      <td>36.8302</td>\n",
       "      <td>36.8253</td>\n",
       "      <td>36.7991</td>\n",
       "      <td>36.8719</td>\n",
       "      <td>36.8048</td>\n",
       "      <td>36.825</td>\n",
       "      <td>36.8975</td>\n",
       "      <td>36.8742</td>\n",
       "      <td>36.7822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Destination Lat</td>\n",
       "      <td>-1.30041</td>\n",
       "      <td>-1.295</td>\n",
       "      <td>-1.30092</td>\n",
       "      <td>-1.25715</td>\n",
       "      <td>-1.29504</td>\n",
       "      <td>-1.25731</td>\n",
       "      <td>-1.27657</td>\n",
       "      <td>-1.25519</td>\n",
       "      <td>-1.30026</td>\n",
       "      <td>-1.2156</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.2156</td>\n",
       "      <td>-1.30833</td>\n",
       "      <td>-1.30011</td>\n",
       "      <td>-1.31986</td>\n",
       "      <td>-1.27655</td>\n",
       "      <td>-1.27528</td>\n",
       "      <td>-1.33162</td>\n",
       "      <td>-1.25841</td>\n",
       "      <td>-1.27921</td>\n",
       "      <td>-1.32016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Destination Long</td>\n",
       "      <td>36.8297</td>\n",
       "      <td>36.8144</td>\n",
       "      <td>36.8282</td>\n",
       "      <td>36.7951</td>\n",
       "      <td>36.8098</td>\n",
       "      <td>36.806</td>\n",
       "      <td>36.8514</td>\n",
       "      <td>36.7822</td>\n",
       "      <td>36.8257</td>\n",
       "      <td>36.8917</td>\n",
       "      <td>...</td>\n",
       "      <td>36.8917</td>\n",
       "      <td>36.8495</td>\n",
       "      <td>36.7585</td>\n",
       "      <td>36.711</td>\n",
       "      <td>36.767</td>\n",
       "      <td>36.8027</td>\n",
       "      <td>36.848</td>\n",
       "      <td>36.8048</td>\n",
       "      <td>36.7949</td>\n",
       "      <td>36.8309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Rider Id</td>\n",
       "      <td>Rider_Id_432</td>\n",
       "      <td>Rider_Id_856</td>\n",
       "      <td>Rider_Id_155</td>\n",
       "      <td>Rider_Id_855</td>\n",
       "      <td>Rider_Id_770</td>\n",
       "      <td>Rider_Id_124</td>\n",
       "      <td>Rider_Id_114</td>\n",
       "      <td>Rider_Id_913</td>\n",
       "      <td>Rider_Id_394</td>\n",
       "      <td>Rider_Id_660</td>\n",
       "      <td>...</td>\n",
       "      <td>Rider_Id_736</td>\n",
       "      <td>Rider_Id_571</td>\n",
       "      <td>Rider_Id_103</td>\n",
       "      <td>Rider_Id_177</td>\n",
       "      <td>Rider_Id_34</td>\n",
       "      <td>Rider_Id_953</td>\n",
       "      <td>Rider_Id_155</td>\n",
       "      <td>Rider_Id_697</td>\n",
       "      <td>Rider_Id_347</td>\n",
       "      <td>Rider_Id_177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Time from Pickup to Arrival</td>\n",
       "      <td>745</td>\n",
       "      <td>1993</td>\n",
       "      <td>455</td>\n",
       "      <td>1341</td>\n",
       "      <td>1214</td>\n",
       "      <td>3191</td>\n",
       "      <td>1380</td>\n",
       "      <td>646</td>\n",
       "      <td>3398</td>\n",
       "      <td>3439</td>\n",
       "      <td>...</td>\n",
       "      <td>2807</td>\n",
       "      <td>971</td>\n",
       "      <td>1771</td>\n",
       "      <td>2298</td>\n",
       "      <td>2325</td>\n",
       "      <td>9</td>\n",
       "      <td>770</td>\n",
       "      <td>2953</td>\n",
       "      <td>1380</td>\n",
       "      <td>2128</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>29 rows × 21201 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                   0               1      \\\n",
       "Order No                                   Order_No_4211  Order_No_25375   \n",
       "User Id                                      User_Id_633    User_Id_2285   \n",
       "Vehicle Type                                        Bike            Bike   \n",
       "Platform Type                                          3               3   \n",
       "Personal or Business                            Business        Personal   \n",
       "Placement - Day of Month                               9              12   \n",
       "Placement - Weekday (Mo = 1)                           5               5   \n",
       "Placement - Time                              9:35:46 AM     11:16:16 AM   \n",
       "Confirmation - Day of Month                            9              12   \n",
       "Confirmation - Weekday (Mo = 1)                        5               5   \n",
       "Confirmation - Time                           9:40:10 AM     11:23:21 AM   \n",
       "Arrival at Pickup - Day of Month                       9              12   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   5               5   \n",
       "Arrival at Pickup - Time                     10:04:47 AM     11:40:22 AM   \n",
       "Pickup - Day of Month                                  9              12   \n",
       "Pickup - Weekday (Mo = 1)                              5               5   \n",
       "Pickup - Time                                10:27:30 AM     11:44:09 AM   \n",
       "Arrival at Destination - Day of Month                  9              12   \n",
       "Arrival at Destination - Weekday (Mo = 1)              5               5   \n",
       "Arrival at Destination - Time                10:39:55 AM     12:17:22 PM   \n",
       "Distance (KM)                                          4              16   \n",
       "Temperature                                         20.4            26.4   \n",
       "Precipitation in millimeters                         NaN             NaN   \n",
       "Pickup Lat                                      -1.31775        -1.35145   \n",
       "Pickup Long                                      36.8304         36.8993   \n",
       "Destination Lat                                 -1.30041          -1.295   \n",
       "Destination Long                                 36.8297         36.8144   \n",
       "Rider Id                                    Rider_Id_432    Rider_Id_856   \n",
       "Time from Pickup to Arrival                          745            1993   \n",
       "\n",
       "                                                   2              3      \\\n",
       "Order No                                   Order_No_1899  Order_No_9336   \n",
       "User Id                                      User_Id_265   User_Id_1402   \n",
       "Vehicle Type                                        Bike           Bike   \n",
       "Platform Type                                          3              3   \n",
       "Personal or Business                            Business       Business   \n",
       "Placement - Day of Month                              30             15   \n",
       "Placement - Weekday (Mo = 1)                           2              5   \n",
       "Placement - Time                             12:39:25 PM     9:25:34 AM   \n",
       "Confirmation - Day of Month                           30             15   \n",
       "Confirmation - Weekday (Mo = 1)                        2              5   \n",
       "Confirmation - Time                          12:42:44 PM     9:26:05 AM   \n",
       "Arrival at Pickup - Day of Month                      30             15   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   2              5   \n",
       "Arrival at Pickup - Time                     12:49:34 PM     9:37:56 AM   \n",
       "Pickup - Day of Month                                 30             15   \n",
       "Pickup - Weekday (Mo = 1)                              2              5   \n",
       "Pickup - Time                                12:53:03 PM     9:43:06 AM   \n",
       "Arrival at Destination - Day of Month                 30             15   \n",
       "Arrival at Destination - Weekday (Mo = 1)              2              5   \n",
       "Arrival at Destination - Time                 1:00:38 PM    10:05:27 AM   \n",
       "Distance (KM)                                          3              9   \n",
       "Temperature                                          NaN           19.2   \n",
       "Precipitation in millimeters                         NaN            NaN   \n",
       "Pickup Lat                                      -1.30828        -1.2813   \n",
       "Pickup Long                                      36.8434        36.8324   \n",
       "Destination Lat                                 -1.30092       -1.25715   \n",
       "Destination Long                                 36.8282        36.7951   \n",
       "Rider Id                                    Rider_Id_155   Rider_Id_855   \n",
       "Time from Pickup to Arrival                          455           1341   \n",
       "\n",
       "                                                    4              5      \\\n",
       "Order No                                   Order_No_27883  Order_No_7408   \n",
       "User Id                                      User_Id_1737   User_Id_1342   \n",
       "Vehicle Type                                         Bike           Bike   \n",
       "Platform Type                                           1              3   \n",
       "Personal or Business                             Personal       Business   \n",
       "Placement - Day of Month                               13             14   \n",
       "Placement - Weekday (Mo = 1)                            1              5   \n",
       "Placement - Time                               9:55:18 AM     3:07:35 PM   \n",
       "Confirmation - Day of Month                            13             14   \n",
       "Confirmation - Weekday (Mo = 1)                         1              5   \n",
       "Confirmation - Time                            9:56:18 AM     3:08:57 PM   \n",
       "Arrival at Pickup - Day of Month                       13             14   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                    1              5   \n",
       "Arrival at Pickup - Time                      10:03:53 AM     3:21:36 PM   \n",
       "Pickup - Day of Month                                  13             14   \n",
       "Pickup - Weekday (Mo = 1)                               1              5   \n",
       "Pickup - Time                                 10:05:23 AM     3:30:30 PM   \n",
       "Arrival at Destination - Day of Month                  13             14   \n",
       "Arrival at Destination - Weekday (Mo = 1)               1              5   \n",
       "Arrival at Destination - Time                 10:25:37 AM     4:23:41 PM   \n",
       "Distance (KM)                                           9              9   \n",
       "Temperature                                          15.4           27.2   \n",
       "Precipitation in millimeters                          NaN            NaN   \n",
       "Pickup Lat                                        -1.2666       -1.30258   \n",
       "Pickup Long                                       36.7921        36.7671   \n",
       "Destination Lat                                  -1.29504       -1.25731   \n",
       "Destination Long                                  36.8098         36.806   \n",
       "Rider Id                                     Rider_Id_770   Rider_Id_124   \n",
       "Time from Pickup to Arrival                          1214           3191   \n",
       "\n",
       "                                                    6               7      \\\n",
       "Order No                                   Order_No_22680  Order_No_21578   \n",
       "User Id                                      User_Id_2803    User_Id_1075   \n",
       "Vehicle Type                                         Bike            Bike   \n",
       "Platform Type                                           3               3   \n",
       "Personal or Business                             Business        Business   \n",
       "Placement - Day of Month                                9              11   \n",
       "Placement - Weekday (Mo = 1)                            5               1   \n",
       "Placement - Time                               9:33:45 AM      2:13:01 PM   \n",
       "Confirmation - Day of Month                             9              11   \n",
       "Confirmation - Weekday (Mo = 1)                         5               1   \n",
       "Confirmation - Time                            9:49:47 AM      2:14:13 PM   \n",
       "Arrival at Pickup - Day of Month                        9              11   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                    5               1   \n",
       "Arrival at Pickup - Time                       9:53:12 AM      2:21:33 PM   \n",
       "Pickup - Day of Month                                   9              11   \n",
       "Pickup - Weekday (Mo = 1)                               5               1   \n",
       "Pickup - Time                                  9:56:45 AM      2:22:40 PM   \n",
       "Arrival at Destination - Day of Month                   9              11   \n",
       "Arrival at Destination - Weekday (Mo = 1)               5               1   \n",
       "Arrival at Destination - Time                 10:19:45 AM      2:33:26 PM   \n",
       "Distance (KM)                                           5               3   \n",
       "Temperature                                          20.3            28.7   \n",
       "Precipitation in millimeters                          NaN             NaN   \n",
       "Pickup Lat                                        -1.2794         -1.2528   \n",
       "Pickup Long                                       36.8254         36.8003   \n",
       "Destination Lat                                  -1.27657        -1.25519   \n",
       "Destination Long                                  36.8514         36.7822   \n",
       "Rider Id                                     Rider_Id_114    Rider_Id_913   \n",
       "Time from Pickup to Arrival                          1380             646   \n",
       "\n",
       "                                                   8              9      ...  \\\n",
       "Order No                                   Order_No_5234  Order_No_1768  ...   \n",
       "User Id                                      User_Id_733   User_Id_2112  ...   \n",
       "Vehicle Type                                        Bike           Bike  ...   \n",
       "Platform Type                                          3              3  ...   \n",
       "Personal or Business                            Business       Business  ...   \n",
       "Placement - Day of Month                              30             23  ...   \n",
       "Placement - Weekday (Mo = 1)                           2              5  ...   \n",
       "Placement - Time                             11:10:44 AM     4:48:54 PM  ...   \n",
       "Confirmation - Day of Month                           30             23  ...   \n",
       "Confirmation - Weekday (Mo = 1)                        2              5  ...   \n",
       "Confirmation - Time                          11:15:49 AM     5:17:56 PM  ...   \n",
       "Arrival at Pickup - Day of Month                      30             23  ...   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   2              5  ...   \n",
       "Arrival at Pickup - Time                     12:13:18 PM     5:32:41 PM  ...   \n",
       "Pickup - Day of Month                                 30             23  ...   \n",
       "Pickup - Weekday (Mo = 1)                              2              5  ...   \n",
       "Pickup - Time                                12:22:57 PM     5:34:38 PM  ...   \n",
       "Arrival at Destination - Day of Month                 30             23  ...   \n",
       "Arrival at Destination - Weekday (Mo = 1)              2              5  ...   \n",
       "Arrival at Destination - Time                 1:19:35 PM     6:31:57 PM  ...   \n",
       "Distance (KM)                                          9             14  ...   \n",
       "Temperature                                          NaN           24.6  ...   \n",
       "Precipitation in millimeters                         NaN            NaN  ...   \n",
       "Pickup Lat                                      -1.25519       -1.22532  ...   \n",
       "Pickup Long                                      36.7822        36.8085  ...   \n",
       "Destination Lat                                 -1.30026        -1.2156  ...   \n",
       "Destination Long                                 36.8257        36.8917  ...   \n",
       "Rider Id                                    Rider_Id_394   Rider_Id_660  ...   \n",
       "Time from Pickup to Arrival                         3398           3439  ...   \n",
       "\n",
       "                                                    21191           21192  \\\n",
       "Order No                                   Order_No_16262  Order_No_11670   \n",
       "User Id                                      User_Id_1231    User_Id_1146   \n",
       "Vehicle Type                                         Bike            Bike   \n",
       "Platform Type                                           3               3   \n",
       "Personal or Business                             Business        Business   \n",
       "Placement - Day of Month                                1              23   \n",
       "Placement - Weekday (Mo = 1)                            4               2   \n",
       "Placement - Time                               2:09:37 PM     12:41:17 PM   \n",
       "Confirmation - Day of Month                             1              23   \n",
       "Confirmation - Weekday (Mo = 1)                         4               2   \n",
       "Confirmation - Time                            2:10:27 PM     12:41:52 PM   \n",
       "Arrival at Pickup - Day of Month                        1              23   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                    4               2   \n",
       "Arrival at Pickup - Time                       2:51:53 PM     12:44:19 PM   \n",
       "Pickup - Day of Month                                   1              23   \n",
       "Pickup - Weekday (Mo = 1)                               4               2   \n",
       "Pickup - Time                                  2:53:08 PM      1:05:11 PM   \n",
       "Arrival at Destination - Day of Month                   1              23   \n",
       "Arrival at Destination - Weekday (Mo = 1)               4               2   \n",
       "Arrival at Destination - Time                  3:39:55 PM      1:21:22 PM   \n",
       "Distance (KM)                                          13               7   \n",
       "Temperature                                           NaN             NaN   \n",
       "Precipitation in millimeters                          NaN             NaN   \n",
       "Pickup Lat                                       -1.28214        -1.31671   \n",
       "Pickup Long                                       36.8162         36.8302   \n",
       "Destination Lat                                   -1.2156        -1.30833   \n",
       "Destination Long                                  36.8917         36.8495   \n",
       "Rider Id                                     Rider_Id_736    Rider_Id_571   \n",
       "Time from Pickup to Arrival                          2807             971   \n",
       "\n",
       "                                                   21193         21194  \\\n",
       "Order No                                   Order_No_4988  Order_No_865   \n",
       "User Id                                      User_Id_875  User_Id_1245   \n",
       "Vehicle Type                                        Bike          Bike   \n",
       "Platform Type                                          1             3   \n",
       "Personal or Business                            Personal      Business   \n",
       "Placement - Day of Month                              12            15   \n",
       "Placement - Weekday (Mo = 1)                           6             1   \n",
       "Placement - Time                              4:48:16 PM    5:02:09 PM   \n",
       "Confirmation - Day of Month                           12            15   \n",
       "Confirmation - Weekday (Mo = 1)                        6             1   \n",
       "Confirmation - Time                           4:49:51 PM    5:02:52 PM   \n",
       "Arrival at Pickup - Day of Month                      12            15   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   6             1   \n",
       "Arrival at Pickup - Time                      5:06:48 PM    5:08:13 PM   \n",
       "Pickup - Day of Month                                 12            15   \n",
       "Pickup - Weekday (Mo = 1)                              6             1   \n",
       "Pickup - Time                                 5:27:22 PM    5:15:15 PM   \n",
       "Arrival at Destination - Day of Month                 12            15   \n",
       "Arrival at Destination - Weekday (Mo = 1)              6             1   \n",
       "Arrival at Destination - Time                 5:56:53 PM    5:53:33 PM   \n",
       "Distance (KM)                                         10            16   \n",
       "Temperature                                         23.5           NaN   \n",
       "Precipitation in millimeters                         NaN           NaN   \n",
       "Pickup Lat                                      -1.28657      -1.26023   \n",
       "Pickup Long                                      36.8253       36.7991   \n",
       "Destination Lat                                 -1.30011      -1.31986   \n",
       "Destination Long                                 36.7585        36.711   \n",
       "Rider Id                                    Rider_Id_103  Rider_Id_177   \n",
       "Time from Pickup to Arrival                         1771          2298   \n",
       "\n",
       "                                                   21195          21196  \\\n",
       "Order No                                   Order_No_9932  Order_No_8834   \n",
       "User Id                                     User_Id_3582   User_Id_2001   \n",
       "Vehicle Type                                        Bike           Bike   \n",
       "Platform Type                                          3              3   \n",
       "Personal or Business                            Business       Personal   \n",
       "Placement - Day of Month                               2             20   \n",
       "Placement - Weekday (Mo = 1)                           6              3   \n",
       "Placement - Time                              1:08:34 PM     3:54:38 PM   \n",
       "Confirmation - Day of Month                            2             20   \n",
       "Confirmation - Weekday (Mo = 1)                        6              3   \n",
       "Confirmation - Time                           1:25:40 PM     3:55:09 PM   \n",
       "Arrival at Pickup - Day of Month                       2             20   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   6              3   \n",
       "Arrival at Pickup - Time                      1:52:11 PM     3:58:49 PM   \n",
       "Pickup - Day of Month                                  2             20   \n",
       "Pickup - Weekday (Mo = 1)                              6              3   \n",
       "Pickup - Time                                 1:54:36 PM     4:20:08 PM   \n",
       "Arrival at Destination - Day of Month                  2             20   \n",
       "Arrival at Destination - Weekday (Mo = 1)              6              3   \n",
       "Arrival at Destination - Time                 2:33:21 PM     4:20:17 PM   \n",
       "Distance (KM)                                         17              3   \n",
       "Temperature                                           29           28.6   \n",
       "Precipitation in millimeters                         NaN            NaN   \n",
       "Pickup Lat                                      -1.23841       -1.25841   \n",
       "Pickup Long                                      36.8719        36.8048   \n",
       "Destination Lat                                 -1.27655       -1.27528   \n",
       "Destination Long                                  36.767        36.8027   \n",
       "Rider Id                                     Rider_Id_34   Rider_Id_953   \n",
       "Time from Pickup to Arrival                         2325              9   \n",
       "\n",
       "                                                    21197          21198  \\\n",
       "Order No                                   Order_No_22892  Order_No_2831   \n",
       "User Id                                      User_Id_1796   User_Id_2956   \n",
       "Vehicle Type                                         Bike           Bike   \n",
       "Platform Type                                           3              3   \n",
       "Personal or Business                             Business       Business   \n",
       "Placement - Day of Month                               13              7   \n",
       "Placement - Weekday (Mo = 1)                            6              4   \n",
       "Placement - Time                              10:13:34 AM     5:06:16 PM   \n",
       "Confirmation - Day of Month                            13              7   \n",
       "Confirmation - Weekday (Mo = 1)                         6              4   \n",
       "Confirmation - Time                           10:13:41 AM     5:07:09 PM   \n",
       "Arrival at Pickup - Day of Month                       13              7   \n",
       "Arrival at Pickup - Weekday (Mo = 1)                    6              4   \n",
       "Arrival at Pickup - Time                      10:20:04 AM     5:30:17 PM   \n",
       "Pickup - Day of Month                                  13              7   \n",
       "Pickup - Weekday (Mo = 1)                               6              4   \n",
       "Pickup - Time                                 10:33:27 AM     5:50:52 PM   \n",
       "Arrival at Destination - Day of Month                  13              7   \n",
       "Arrival at Destination - Weekday (Mo = 1)               6              4   \n",
       "Arrival at Destination - Time                 10:46:17 AM     6:40:05 PM   \n",
       "Distance (KM)                                           7             20   \n",
       "Temperature                                            26           29.2   \n",
       "Precipitation in millimeters                          NaN            NaN   \n",
       "Pickup Lat                                       -1.30714       -1.28602   \n",
       "Pickup Long                                        36.825        36.8975   \n",
       "Destination Lat                                  -1.33162       -1.25841   \n",
       "Destination Long                                   36.848        36.8048   \n",
       "Rider Id                                     Rider_Id_155   Rider_Id_697   \n",
       "Time from Pickup to Arrival                           770           2953   \n",
       "\n",
       "                                                   21199          21200  \n",
       "Order No                                   Order_No_6174  Order_No_9836  \n",
       "User Id                                     User_Id_2524    User_Id_718  \n",
       "Vehicle Type                                        Bike           Bike  \n",
       "Platform Type                                          1              3  \n",
       "Personal or Business                            Personal       Business  \n",
       "Placement - Day of Month                               4             26  \n",
       "Placement - Weekday (Mo = 1)                           3              2  \n",
       "Placement - Time                              9:31:39 AM     2:19:47 PM  \n",
       "Confirmation - Day of Month                            4             26  \n",
       "Confirmation - Weekday (Mo = 1)                        3              2  \n",
       "Confirmation - Time                           9:31:53 AM     2:20:01 PM  \n",
       "Arrival at Pickup - Day of Month                       4             26  \n",
       "Arrival at Pickup - Weekday (Mo = 1)                   3              2  \n",
       "Arrival at Pickup - Time                      9:38:59 AM     2:24:29 PM  \n",
       "Pickup - Day of Month                                  4             26  \n",
       "Pickup - Weekday (Mo = 1)                              3              2  \n",
       "Pickup - Time                                 9:45:15 AM     2:41:55 PM  \n",
       "Arrival at Destination - Day of Month                  4             26  \n",
       "Arrival at Destination - Weekday (Mo = 1)              3              2  \n",
       "Arrival at Destination - Time                10:08:15 AM     3:17:23 PM  \n",
       "Distance (KM)                                         13             12  \n",
       "Temperature                                           15           30.9  \n",
       "Precipitation in millimeters                         NaN            NaN  \n",
       "Pickup Lat                                      -1.25003       -1.25519  \n",
       "Pickup Long                                      36.8742        36.7822  \n",
       "Destination Lat                                 -1.27921       -1.32016  \n",
       "Destination Long                                 36.7949        36.8309  \n",
       "Rider Id                                    Rider_Id_347   Rider_Id_177  \n",
       "Time from Pickup to Arrival                         1380           2128  \n",
       "\n",
       "[29 rows x 21201 columns]"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train.T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### New features "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}