sendy-logistics-regtree.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import missingno as msno\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "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>order_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>platform</th>\n",
       "      <th>personal_or_business</th>\n",
       "      <th>placement_date</th>\n",
       "      <th>placement_weekday</th>\n",
       "      <th>placement_time</th>\n",
       "      <th>confirm_date</th>\n",
       "      <th>confirm_weekday</th>\n",
       "      <th>confirm_time</th>\n",
       "      <th>...</th>\n",
       "      <th>distance</th>\n",
       "      <th>temperature</th>\n",
       "      <th>pickup_lat</th>\n",
       "      <th>pickup_long</th>\n",
       "      <th>dest_lat</th>\n",
       "      <th>dest_long</th>\n",
       "      <th>rider_id</th>\n",
       "      <th>pickup_to_arrival_time</th>\n",
       "      <th>placement_to_confirm_time</th>\n",
       "      <th>confirm_to_pickup_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4211</td>\n",
       "      <td>633</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>34546</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>34810</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>20.400000</td>\n",
       "      <td>-1.317755</td>\n",
       "      <td>36.830370</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>432</td>\n",
       "      <td>745</td>\n",
       "      <td>264</td>\n",
       "      <td>2840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25375</td>\n",
       "      <td>2285</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>40576</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>41001</td>\n",
       "      <td>...</td>\n",
       "      <td>16</td>\n",
       "      <td>26.400000</td>\n",
       "      <td>-1.351453</td>\n",
       "      <td>36.899315</td>\n",
       "      <td>-1.295004</td>\n",
       "      <td>36.814358</td>\n",
       "      <td>856</td>\n",
       "      <td>1993</td>\n",
       "      <td>425</td>\n",
       "      <td>1248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1899</td>\n",
       "      <td>265</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>30</td>\n",
       "      <td>2</td>\n",
       "      <td>45565</td>\n",
       "      <td>30</td>\n",
       "      <td>2</td>\n",
       "      <td>45764</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>23.258889</td>\n",
       "      <td>-1.308284</td>\n",
       "      <td>36.843419</td>\n",
       "      <td>-1.300921</td>\n",
       "      <td>36.828195</td>\n",
       "      <td>155</td>\n",
       "      <td>455</td>\n",
       "      <td>199</td>\n",
       "      <td>619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9336</td>\n",
       "      <td>1402</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>33934</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>33965</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>19.200000</td>\n",
       "      <td>-1.281301</td>\n",
       "      <td>36.832396</td>\n",
       "      <td>-1.257147</td>\n",
       "      <td>36.795063</td>\n",
       "      <td>855</td>\n",
       "      <td>1341</td>\n",
       "      <td>31</td>\n",
       "      <td>1021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>27883</td>\n",
       "      <td>1737</td>\n",
       "      <td>1</td>\n",
       "      <td>Personal</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>35718</td>\n",
       "      <td>13</td>\n",
       "      <td>1</td>\n",
       "      <td>35778</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>15.400000</td>\n",
       "      <td>-1.266597</td>\n",
       "      <td>36.792118</td>\n",
       "      <td>-1.295041</td>\n",
       "      <td>36.809817</td>\n",
       "      <td>770</td>\n",
       "      <td>1214</td>\n",
       "      <td>60</td>\n",
       "      <td>545</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id platform personal_or_business  placement_date  \\\n",
       "0      4211      633        3             Business               9   \n",
       "1     25375     2285        3             Personal              12   \n",
       "2      1899      265        3             Business              30   \n",
       "3      9336     1402        3             Business              15   \n",
       "4     27883     1737        1             Personal              13   \n",
       "\n",
       "   placement_weekday  placement_time  confirm_date  confirm_weekday  \\\n",
       "0                  5           34546             9                5   \n",
       "1                  5           40576            12                5   \n",
       "2                  2           45565            30                2   \n",
       "3                  5           33934            15                5   \n",
       "4                  1           35718            13                1   \n",
       "\n",
       "   confirm_time  ...  distance  temperature  pickup_lat  pickup_long  \\\n",
       "0         34810  ...         4    20.400000   -1.317755    36.830370   \n",
       "1         41001  ...        16    26.400000   -1.351453    36.899315   \n",
       "2         45764  ...         3    23.258889   -1.308284    36.843419   \n",
       "3         33965  ...         9    19.200000   -1.281301    36.832396   \n",
       "4         35778  ...         9    15.400000   -1.266597    36.792118   \n",
       "\n",
       "   dest_lat  dest_long  rider_id  pickup_to_arrival_time  \\\n",
       "0 -1.300406  36.829741       432                     745   \n",
       "1 -1.295004  36.814358       856                    1993   \n",
       "2 -1.300921  36.828195       155                     455   \n",
       "3 -1.257147  36.795063       855                    1341   \n",
       "4 -1.295041  36.809817       770                    1214   \n",
       "\n",
       "   placement_to_confirm_time  confirm_to_pickup_time  \n",
       "0                        264                    2840  \n",
       "1                        425                    1248  \n",
       "2                        199                     619  \n",
       "3                         31                    1021  \n",
       "4                         60                     545  \n",
       "\n",
       "[5 rows x 29 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = pd.read_pickle(\"clean-data/train.pkl\")\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>PC1</th>\n",
       "      <th>PC2</th>\n",
       "      <th>PC3</th>\n",
       "      <th>PC4</th>\n",
       "      <th>PC5</th>\n",
       "      <th>PC6</th>\n",
       "      <th>pickup_to_arrival_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.471082</td>\n",
       "      <td>0.270344</td>\n",
       "      <td>-0.486887</td>\n",
       "      <td>0.600463</td>\n",
       "      <td>-0.540924</td>\n",
       "      <td>-1.615606</td>\n",
       "      <td>745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.624127</td>\n",
       "      <td>1.498240</td>\n",
       "      <td>-0.350465</td>\n",
       "      <td>-0.674034</td>\n",
       "      <td>0.329674</td>\n",
       "      <td>0.866783</td>\n",
       "      <td>2886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.528462</td>\n",
       "      <td>-0.477729</td>\n",
       "      <td>0.776858</td>\n",
       "      <td>0.343377</td>\n",
       "      <td>0.045975</td>\n",
       "      <td>-0.084287</td>\n",
       "      <td>2615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.635800</td>\n",
       "      <td>0.729681</td>\n",
       "      <td>-2.567904</td>\n",
       "      <td>0.609702</td>\n",
       "      <td>-0.724023</td>\n",
       "      <td>0.841407</td>\n",
       "      <td>2986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.562681</td>\n",
       "      <td>-0.769030</td>\n",
       "      <td>-1.075503</td>\n",
       "      <td>0.506759</td>\n",
       "      <td>-0.834757</td>\n",
       "      <td>-0.026859</td>\n",
       "      <td>1602</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        PC1       PC2       PC3       PC4       PC5       PC6  \\\n",
       "0  0.471082  0.270344 -0.486887  0.600463 -0.540924 -1.615606   \n",
       "1  0.624127  1.498240 -0.350465 -0.674034  0.329674  0.866783   \n",
       "2  0.528462 -0.477729  0.776858  0.343377  0.045975 -0.084287   \n",
       "3  0.635800  0.729681 -2.567904  0.609702 -0.724023  0.841407   \n",
       "4  0.562681 -0.769030 -1.075503  0.506759 -0.834757 -0.026859   \n",
       "\n",
       "   pickup_to_arrival_time  \n",
       "0                     745  \n",
       "1                    2886  \n",
       "2                    2615  \n",
       "3                    2986  \n",
       "4                    1602  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_pca = pd.read_pickle(\"clean-data/train-pca.pkl\")\n",
    "train_pca.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Combining riders and train into one df\n",
    "riders = pd.read_pickle(\"clean-data/riders.pkl\")\n",
    "train = train.merge(riders, on=\"rider_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>platform</th>\n",
       "      <th>personal_or_business</th>\n",
       "      <th>placement_date</th>\n",
       "      <th>placement_weekday</th>\n",
       "      <th>placement_time</th>\n",
       "      <th>confirm_date</th>\n",
       "      <th>confirm_weekday</th>\n",
       "      <th>confirm_time</th>\n",
       "      <th>...</th>\n",
       "      <th>dest_lat</th>\n",
       "      <th>dest_long</th>\n",
       "      <th>rider_id</th>\n",
       "      <th>pickup_to_arrival_time</th>\n",
       "      <th>placement_to_confirm_time</th>\n",
       "      <th>confirm_to_pickup_time</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",
       "      <th>0</th>\n",
       "      <td>4211</td>\n",
       "      <td>633</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>34546</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>34810</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>432</td>\n",
       "      <td>745</td>\n",
       "      <td>264</td>\n",
       "      <td>2840</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>27440</td>\n",
       "      <td>2642</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>56477</td>\n",
       "      <td>18</td>\n",
       "      <td>5</td>\n",
       "      <td>56490</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.356237</td>\n",
       "      <td>36.904295</td>\n",
       "      <td>432</td>\n",
       "      <td>2886</td>\n",
       "      <td>13</td>\n",
       "      <td>1927</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14170</td>\n",
       "      <td>3339</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>46301</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>47569</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.273412</td>\n",
       "      <td>36.818206</td>\n",
       "      <td>432</td>\n",
       "      <td>2615</td>\n",
       "      <td>1268</td>\n",
       "      <td>1219</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21727</td>\n",
       "      <td>3523</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>25930</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>25949</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.223520</td>\n",
       "      <td>36.802061</td>\n",
       "      <td>432</td>\n",
       "      <td>2986</td>\n",
       "      <td>19</td>\n",
       "      <td>1577</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8301</td>\n",
       "      <td>87</td>\n",
       "      <td>2</td>\n",
       "      <td>Personal</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>38458</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>38544</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.300431</td>\n",
       "      <td>36.752427</td>\n",
       "      <td>432</td>\n",
       "      <td>1602</td>\n",
       "      <td>86</td>\n",
       "      <td>1036</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21191</th>\n",
       "      <td>27103</td>\n",
       "      <td>3718</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>40177</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>41538</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.265432</td>\n",
       "      <td>36.795034</td>\n",
       "      <td>528</td>\n",
       "      <td>919</td>\n",
       "      <td>1361</td>\n",
       "      <td>1230</td>\n",
       "      <td>5770</td>\n",
       "      <td>1793</td>\n",
       "      <td>14.2</td>\n",
       "      <td>2205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21192</th>\n",
       "      <td>3786</td>\n",
       "      <td>1558</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>21</td>\n",
       "      <td>4</td>\n",
       "      <td>59597</td>\n",
       "      <td>21</td>\n",
       "      <td>4</td>\n",
       "      <td>60447</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.278067</td>\n",
       "      <td>36.783487</td>\n",
       "      <td>638</td>\n",
       "      <td>2331</td>\n",
       "      <td>850</td>\n",
       "      <td>1406</td>\n",
       "      <td>102</td>\n",
       "      <td>873</td>\n",
       "      <td>13.7</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21193</th>\n",
       "      <td>7281</td>\n",
       "      <td>2008</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>61240</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>61881</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.285850</td>\n",
       "      <td>36.830629</td>\n",
       "      <td>773</td>\n",
       "      <td>2418</td>\n",
       "      <td>641</td>\n",
       "      <td>3356</td>\n",
       "      <td>5</td>\n",
       "      <td>105</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21194</th>\n",
       "      <td>15575</td>\n",
       "      <td>1346</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>29</td>\n",
       "      <td>2</td>\n",
       "      <td>52315</td>\n",
       "      <td>29</td>\n",
       "      <td>2</td>\n",
       "      <td>52363</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.298575</td>\n",
       "      <td>36.808800</td>\n",
       "      <td>860</td>\n",
       "      <td>717</td>\n",
       "      <td>48</td>\n",
       "      <td>940</td>\n",
       "      <td>5</td>\n",
       "      <td>448</td>\n",
       "      <td>15.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21195</th>\n",
       "      <td>15578</td>\n",
       "      <td>2174</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>35483</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>35492</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.286808</td>\n",
       "      <td>36.821940</td>\n",
       "      <td>53</td>\n",
       "      <td>1647</td>\n",
       "      <td>9</td>\n",
       "      <td>479</td>\n",
       "      <td>241</td>\n",
       "      <td>1109</td>\n",
       "      <td>13.7</td>\n",
       "      <td>73</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>21196 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       order_id  user_id platform personal_or_business  placement_date  \\\n",
       "0          4211      633        3             Business               9   \n",
       "1         27440     2642        3             Personal              18   \n",
       "2         14170     3339        3             Business              31   \n",
       "3         21727     3523        3             Personal               2   \n",
       "4          8301       87        2             Personal              22   \n",
       "...         ...      ...      ...                  ...             ...   \n",
       "21191     27103     3718        3             Business              13   \n",
       "21192      3786     1558        3             Personal              21   \n",
       "21193      7281     2008        3             Business              10   \n",
       "21194     15575     1346        3             Business              29   \n",
       "21195     15578     2174        3             Business              30   \n",
       "\n",
       "       placement_weekday  placement_time  confirm_date  confirm_weekday  \\\n",
       "0                      5           34546             9                5   \n",
       "1                      5           56477            18                5   \n",
       "2                      5           46301            31                5   \n",
       "3                      2           25930             2                2   \n",
       "4                      2           38458            22                2   \n",
       "...                  ...             ...           ...              ...   \n",
       "21191                  2           40177            13                2   \n",
       "21192                  4           59597            21                4   \n",
       "21193                  1           61240            10                1   \n",
       "21194                  2           52315            29                2   \n",
       "21195                  5           35483            30                5   \n",
       "\n",
       "       confirm_time  ...  dest_lat  dest_long  rider_id  \\\n",
       "0             34810  ... -1.300406  36.829741       432   \n",
       "1             56490  ... -1.356237  36.904295       432   \n",
       "2             47569  ... -1.273412  36.818206       432   \n",
       "3             25949  ... -1.223520  36.802061       432   \n",
       "4             38544  ... -1.300431  36.752427       432   \n",
       "...             ...  ...       ...        ...       ...   \n",
       "21191         41538  ... -1.265432  36.795034       528   \n",
       "21192         60447  ... -1.278067  36.783487       638   \n",
       "21193         61881  ... -1.285850  36.830629       773   \n",
       "21194         52363  ... -1.298575  36.808800       860   \n",
       "21195         35492  ... -1.286808  36.821940        53   \n",
       "\n",
       "       pickup_to_arrival_time  placement_to_confirm_time  \\\n",
       "0                         745                        264   \n",
       "1                        2886                         13   \n",
       "2                        2615                       1268   \n",
       "3                        2986                         19   \n",
       "4                        1602                         86   \n",
       "...                       ...                        ...   \n",
       "21191                     919                       1361   \n",
       "21192                    2331                        850   \n",
       "21193                    2418                        641   \n",
       "21194                     717                         48   \n",
       "21195                    1647                          9   \n",
       "\n",
       "       confirm_to_pickup_time  no_of_orders   age  average_rating  \\\n",
       "0                        2840          1637  1309            13.8   \n",
       "1                        1927          1637  1309            13.8   \n",
       "2                        1219          1637  1309            13.8   \n",
       "3                        1577          1637  1309            13.8   \n",
       "4                        1036          1637  1309            13.8   \n",
       "...                       ...           ...   ...             ...   \n",
       "21191                    1230          5770  1793            14.2   \n",
       "21192                    1406           102   873            13.7   \n",
       "21193                    3356             5   105             0.0   \n",
       "21194                     940             5   448            15.0   \n",
       "21195                     479           241  1109            13.7   \n",
       "\n",
       "       no_of_ratings  \n",
       "0                549  \n",
       "1                549  \n",
       "2                549  \n",
       "3                549  \n",
       "4                549  \n",
       "...              ...  \n",
       "21191           2205  \n",
       "21192             32  \n",
       "21193              0  \n",
       "21194              2  \n",
       "21195             73  \n",
       "\n",
       "[21196 rows x 33 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping date/time columns\n",
    "dropped = [\"placement_date\", \"placement_weekday\", \"placement_time\",\n",
    "           \"confirm_date\", \"confirm_weekday\", \"confirm_time\",\n",
    "           \"arrival_pickup_date\", \"arrive_pickup_weekday\", \"arrive_time\",\n",
    "           \"pickup_date\", \"pickup_weekday\", \"pickup_time\",\n",
    "           \"arrival_dest_date\", \"arrival_dest_weekday\", \"arrival_dest_time\"]\n",
    "\n",
    "train.drop(columns=dropped, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>platform</th>\n",
       "      <th>personal_or_business</th>\n",
       "      <th>distance</th>\n",
       "      <th>temperature</th>\n",
       "      <th>pickup_lat</th>\n",
       "      <th>pickup_long</th>\n",
       "      <th>dest_lat</th>\n",
       "      <th>dest_long</th>\n",
       "      <th>rider_id</th>\n",
       "      <th>pickup_to_arrival_time</th>\n",
       "      <th>placement_to_confirm_time</th>\n",
       "      <th>confirm_to_pickup_time</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",
       "      <th>0</th>\n",
       "      <td>4211</td>\n",
       "      <td>633</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>4</td>\n",
       "      <td>20.4</td>\n",
       "      <td>-1.317755</td>\n",
       "      <td>36.830370</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>432</td>\n",
       "      <td>745</td>\n",
       "      <td>264</td>\n",
       "      <td>2840</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>27440</td>\n",
       "      <td>2642</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>20</td>\n",
       "      <td>24.5</td>\n",
       "      <td>-1.326774</td>\n",
       "      <td>36.787807</td>\n",
       "      <td>-1.356237</td>\n",
       "      <td>36.904295</td>\n",
       "      <td>432</td>\n",
       "      <td>2886</td>\n",
       "      <td>13</td>\n",
       "      <td>1927</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14170</td>\n",
       "      <td>3339</td>\n",
       "      <td>3</td>\n",
       "      <td>Business</td>\n",
       "      <td>6</td>\n",
       "      <td>24.7</td>\n",
       "      <td>-1.255189</td>\n",
       "      <td>36.782203</td>\n",
       "      <td>-1.273412</td>\n",
       "      <td>36.818206</td>\n",
       "      <td>432</td>\n",
       "      <td>2615</td>\n",
       "      <td>1268</td>\n",
       "      <td>1219</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21727</td>\n",
       "      <td>3523</td>\n",
       "      <td>3</td>\n",
       "      <td>Personal</td>\n",
       "      <td>18</td>\n",
       "      <td>15.2</td>\n",
       "      <td>-1.290315</td>\n",
       "      <td>36.757377</td>\n",
       "      <td>-1.223520</td>\n",
       "      <td>36.802061</td>\n",
       "      <td>432</td>\n",
       "      <td>2986</td>\n",
       "      <td>19</td>\n",
       "      <td>1577</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8301</td>\n",
       "      <td>87</td>\n",
       "      <td>2</td>\n",
       "      <td>Personal</td>\n",
       "      <td>7</td>\n",
       "      <td>19.2</td>\n",
       "      <td>-1.273524</td>\n",
       "      <td>36.799220</td>\n",
       "      <td>-1.300431</td>\n",
       "      <td>36.752427</td>\n",
       "      <td>432</td>\n",
       "      <td>1602</td>\n",
       "      <td>86</td>\n",
       "      <td>1036</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   order_id  user_id platform personal_or_business  distance  temperature  \\\n",
       "0      4211      633        3             Business         4         20.4   \n",
       "1     27440     2642        3             Personal        20         24.5   \n",
       "2     14170     3339        3             Business         6         24.7   \n",
       "3     21727     3523        3             Personal        18         15.2   \n",
       "4      8301       87        2             Personal         7         19.2   \n",
       "\n",
       "   pickup_lat  pickup_long  dest_lat  dest_long  rider_id  \\\n",
       "0   -1.317755    36.830370 -1.300406  36.829741       432   \n",
       "1   -1.326774    36.787807 -1.356237  36.904295       432   \n",
       "2   -1.255189    36.782203 -1.273412  36.818206       432   \n",
       "3   -1.290315    36.757377 -1.223520  36.802061       432   \n",
       "4   -1.273524    36.799220 -1.300431  36.752427       432   \n",
       "\n",
       "   pickup_to_arrival_time  placement_to_confirm_time  confirm_to_pickup_time  \\\n",
       "0                     745                        264                    2840   \n",
       "1                    2886                         13                    1927   \n",
       "2                    2615                       1268                    1219   \n",
       "3                    2986                         19                    1577   \n",
       "4                    1602                         86                    1036   \n",
       "\n",
       "   no_of_orders   age  average_rating  no_of_ratings  \n",
       "0          1637  1309            13.8            549  \n",
       "1          1637  1309            13.8            549  \n",
       "2          1637  1309            13.8            549  \n",
       "3          1637  1309            13.8            549  \n",
       "4          1637  1309            13.8            549  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_dummies(df):\n",
    "    # Creating dummy variables for platforms 1, 2, 3. They will all be 0 for\n",
    "    # platform 4.\n",
    "    for i in range(1, 4):\n",
    "        plat_str = \"platform\" + str(i)\n",
    "        df[plat_str] = 0\n",
    "        df.loc[df.platform == i, plat_str] = 1\n",
    "        \n",
    "    # Creating dummy variable for personal_or_business (1 if personal 0 if business)\n",
    "    df[\"personal\"] = 0\n",
    "    df.loc[df.personal_or_business == \"Personal\", \"personal\"] = 1\n",
    "\n",
    "create_dummies(train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>distance</th>\n",
       "      <th>temperature</th>\n",
       "      <th>pickup_lat</th>\n",
       "      <th>pickup_long</th>\n",
       "      <th>dest_lat</th>\n",
       "      <th>dest_long</th>\n",
       "      <th>pickup_to_arrival_time</th>\n",
       "      <th>placement_to_confirm_time</th>\n",
       "      <th>confirm_to_pickup_time</th>\n",
       "      <th>no_of_orders</th>\n",
       "      <th>age</th>\n",
       "      <th>average_rating</th>\n",
       "      <th>no_of_ratings</th>\n",
       "      <th>platform1</th>\n",
       "      <th>platform2</th>\n",
       "      <th>platform3</th>\n",
       "      <th>personal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>20.4</td>\n",
       "      <td>-1.317755</td>\n",
       "      <td>36.830370</td>\n",
       "      <td>-1.300406</td>\n",
       "      <td>36.829741</td>\n",
       "      <td>745</td>\n",
       "      <td>264</td>\n",
       "      <td>2840</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20</td>\n",
       "      <td>24.5</td>\n",
       "      <td>-1.326774</td>\n",
       "      <td>36.787807</td>\n",
       "      <td>-1.356237</td>\n",
       "      <td>36.904295</td>\n",
       "      <td>2886</td>\n",
       "      <td>13</td>\n",
       "      <td>1927</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>24.7</td>\n",
       "      <td>-1.255189</td>\n",
       "      <td>36.782203</td>\n",
       "      <td>-1.273412</td>\n",
       "      <td>36.818206</td>\n",
       "      <td>2615</td>\n",
       "      <td>1268</td>\n",
       "      <td>1219</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18</td>\n",
       "      <td>15.2</td>\n",
       "      <td>-1.290315</td>\n",
       "      <td>36.757377</td>\n",
       "      <td>-1.223520</td>\n",
       "      <td>36.802061</td>\n",
       "      <td>2986</td>\n",
       "      <td>19</td>\n",
       "      <td>1577</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>19.2</td>\n",
       "      <td>-1.273524</td>\n",
       "      <td>36.799220</td>\n",
       "      <td>-1.300431</td>\n",
       "      <td>36.752427</td>\n",
       "      <td>1602</td>\n",
       "      <td>86</td>\n",
       "      <td>1036</td>\n",
       "      <td>1637</td>\n",
       "      <td>1309</td>\n",
       "      <td>13.8</td>\n",
       "      <td>549</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21191</th>\n",
       "      <td>3</td>\n",
       "      <td>22.2</td>\n",
       "      <td>-1.265003</td>\n",
       "      <td>36.812624</td>\n",
       "      <td>-1.265432</td>\n",
       "      <td>36.795034</td>\n",
       "      <td>919</td>\n",
       "      <td>1361</td>\n",
       "      <td>1230</td>\n",
       "      <td>5770</td>\n",
       "      <td>1793</td>\n",
       "      <td>14.2</td>\n",
       "      <td>2205</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21192</th>\n",
       "      <td>6</td>\n",
       "      <td>19.9</td>\n",
       "      <td>-1.269609</td>\n",
       "      <td>36.825741</td>\n",
       "      <td>-1.278067</td>\n",
       "      <td>36.783487</td>\n",
       "      <td>2331</td>\n",
       "      <td>850</td>\n",
       "      <td>1406</td>\n",
       "      <td>102</td>\n",
       "      <td>873</td>\n",
       "      <td>13.7</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21193</th>\n",
       "      <td>7</td>\n",
       "      <td>19.0</td>\n",
       "      <td>-1.250823</td>\n",
       "      <td>36.789526</td>\n",
       "      <td>-1.285850</td>\n",
       "      <td>36.830629</td>\n",
       "      <td>2418</td>\n",
       "      <td>641</td>\n",
       "      <td>3356</td>\n",
       "      <td>5</td>\n",
       "      <td>105</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21194</th>\n",
       "      <td>2</td>\n",
       "      <td>27.0</td>\n",
       "      <td>-1.291787</td>\n",
       "      <td>36.787267</td>\n",
       "      <td>-1.298575</td>\n",
       "      <td>36.808800</td>\n",
       "      <td>717</td>\n",
       "      <td>48</td>\n",
       "      <td>940</td>\n",
       "      <td>5</td>\n",
       "      <td>448</td>\n",
       "      <td>15.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21195</th>\n",
       "      <td>10</td>\n",
       "      <td>19.1</td>\n",
       "      <td>-1.248278</td>\n",
       "      <td>36.884092</td>\n",
       "      <td>-1.286808</td>\n",
       "      <td>36.821940</td>\n",
       "      <td>1647</td>\n",
       "      <td>9</td>\n",
       "      <td>479</td>\n",
       "      <td>241</td>\n",
       "      <td>1109</td>\n",
       "      <td>13.7</td>\n",
       "      <td>73</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>21196 rows × 17 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       distance  temperature  pickup_lat  pickup_long  dest_lat  dest_long  \\\n",
       "0             4         20.4   -1.317755    36.830370 -1.300406  36.829741   \n",
       "1            20         24.5   -1.326774    36.787807 -1.356237  36.904295   \n",
       "2             6         24.7   -1.255189    36.782203 -1.273412  36.818206   \n",
       "3            18         15.2   -1.290315    36.757377 -1.223520  36.802061   \n",
       "4             7         19.2   -1.273524    36.799220 -1.300431  36.752427   \n",
       "...         ...          ...         ...          ...       ...        ...   \n",
       "21191         3         22.2   -1.265003    36.812624 -1.265432  36.795034   \n",
       "21192         6         19.9   -1.269609    36.825741 -1.278067  36.783487   \n",
       "21193         7         19.0   -1.250823    36.789526 -1.285850  36.830629   \n",
       "21194         2         27.0   -1.291787    36.787267 -1.298575  36.808800   \n",
       "21195        10         19.1   -1.248278    36.884092 -1.286808  36.821940   \n",
       "\n",
       "       pickup_to_arrival_time  placement_to_confirm_time  \\\n",
       "0                         745                        264   \n",
       "1                        2886                         13   \n",
       "2                        2615                       1268   \n",
       "3                        2986                         19   \n",
       "4                        1602                         86   \n",
       "...                       ...                        ...   \n",
       "21191                     919                       1361   \n",
       "21192                    2331                        850   \n",
       "21193                    2418                        641   \n",
       "21194                     717                         48   \n",
       "21195                    1647                          9   \n",
       "\n",
       "       confirm_to_pickup_time  no_of_orders   age  average_rating  \\\n",
       "0                        2840          1637  1309            13.8   \n",
       "1                        1927          1637  1309            13.8   \n",
       "2                        1219          1637  1309            13.8   \n",
       "3                        1577          1637  1309            13.8   \n",
       "4                        1036          1637  1309            13.8   \n",
       "...                       ...           ...   ...             ...   \n",
       "21191                    1230          5770  1793            14.2   \n",
       "21192                    1406           102   873            13.7   \n",
       "21193                    3356             5   105             0.0   \n",
       "21194                     940             5   448            15.0   \n",
       "21195                     479           241  1109            13.7   \n",
       "\n",
       "       no_of_ratings  platform1  platform2  platform3  personal  \n",
       "0                549          0          0          1         0  \n",
       "1                549          0          0          1         1  \n",
       "2                549          0          0          1         0  \n",
       "3                549          0          0          1         1  \n",
       "4                549          0          1          0         1  \n",
       "...              ...        ...        ...        ...       ...  \n",
       "21191           2205          0          0          1         0  \n",
       "21192             32          0          0          1         1  \n",
       "21193              0          0          0          1         0  \n",
       "21194              2          0          0          1         0  \n",
       "21195             73          0          0          1         0  \n",
       "\n",
       "[21196 rows x 17 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dropped = [\"order_id\", \"user_id\", \"platform\", \"personal_or_business\", \"rider_id\"]\n",
    "train.drop(columns=dropped, inplace=True)\n",
    "train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating standard linear model\n",
    "from sklearn.linear_model import LinearRegression\n",
    "linear = LinearRegression()\n",
    "X = train.loc[:, train.columns != \"pickup_to_arrival_time\"]\n",
    "y = train.pickup_to_arrival_time\n",
    "linear.fit(X=X, y=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.34104294576737526"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Running cross-validation\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "cross_val_score(linear, X=X, y=y, cv=10).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor()"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Trying regression tree\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "\n",
    "tree = DecisionTreeRegressor()\n",
    "tree.fit(X=X, y=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "import graphviz\n",
    "from sklearn.tree import export_graphviz\n",
    "\n",
    "def draw_regression_tree(tree, feature_names):\n",
    "    dot_data = export_graphviz(tree, out_file=None, \n",
    "                         feature_names=feature_names,  \n",
    "                         filled=True, \n",
    "                         impurity=False,\n",
    "                         rounded=True,  \n",
    "                         special_characters=True)  \n",
    "    \n",
    "    graph = graphviz.Source(dot_data)\n",
    "    graph.format = 'png'\n",
    "    graph.render('tree',view=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = tree.cost_complexity_pruning_path(X=X, y=y)\n",
    "ccp_alphas, impurities = path.ccp_alphas, path.impurities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>alpha</th>\n",
       "      <th>MSE</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.000018</td>\n",
       "      <td>0.000035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.000024</td>\n",
       "      <td>0.000106</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      alpha       MSE\n",
       "0  0.000000  0.000000\n",
       "1  0.000018  0.000035\n",
       "2  0.000024  0.000059\n",
       "3  0.000024  0.000083\n",
       "4  0.000024  0.000106"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame({'alpha':ccp_alphas, 'MSE':impurities}).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 554.93836772,  711.4507028 , 1045.15830372])"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ccp_alphas[19000::20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Finding the optimal alpha\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "#for a in ccp_alphas[19000::20]:\n",
    "    #model = DecisionTreeRegressor(ccp_alpha=a)\n",
    "    #cv_score = -cross_val_score(model, X=X, y=y, scoring=\"neg_mean_squared_error\", cv=10).mean()\n",
    "    #print(str(a) + \": \" + str(cv_score))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(ccp_alpha=1037.677)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_pruned = DecisionTreeRegressor(ccp_alpha=1037.677)\n",
    "tree_pruned.fit(X=X, y=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fd37d6b52e0>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = tree_pruned.predict(X)\n",
    "sns.relplot(x=y, y=y_pred, kind=\"scatter\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The unpruned tree had 21070 leaves.\n",
      "The pruned tree has 24 leaves.\n"
     ]
    }
   ],
   "source": [
    "print('The unpruned tree had', tree.get_n_leaves(), 'leaves.')\n",
    "print('The pruned tree has', tree_pruned.get_n_leaves(), 'leaves.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "draw_regression_tree(tree_pruned, X.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = sns.barplot(x=X.columns, y=tree_pruned.feature_importances_)\n",
    "p = p.set_xticklabels(labels=X.columns, rotation=90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3426585939047076"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(tree_pruned, X=X, y=y, cv=10).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "545.207666120904\n",
      "617449.4309227142\n",
      "0.3665565106521912\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score\n",
    "\n",
    "print(mean_absolute_error(y_true=y, y_pred=y_pred))\n",
    "print(mean_squared_error(y_true=y, y_pred=y_pred))\n",
    "print(r2_score(y_true=y, y_pred=y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    21196.000000\n",
       "mean      1557.013163\n",
       "std        987.317884\n",
       "min          1.000000\n",
       "25%        882.000000\n",
       "50%       1369.000000\n",
       "75%       2040.000000\n",
       "max       7883.000000\n",
       "Name: pickup_to_arrival_time, dtype: float64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.describe()"
   ]
  },
  {
   "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}