Travel times GTFS .ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Estimating travel time "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebooks estimates travel time within the transmilenio system. For this purpose, we use the lastest GTFSs information from Bogota, available here: http://transitfeeds.com/p/simur-bogota/973"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### General process \n",
    "\n",
    "As many BRT system, Transmilenio services are not requeired to stop at all stations, in fact, the infraestrucute and the system allows for certain services to stop at predifined stops by design. To estimate a travel time, we will transform the the gtfs data into a transit netwrok using UrbanAccess python package. The transit netwrok will have stations as nodes, and the links will form when two stations are connected by one route. It is possible that two pair of nodes have multiple links. The weight of each link is the schedule travel time. \n",
    "\n",
    "Finally, we will use networkX, another python package, to calculate the shortest path lenght between two stations. The lenght of the path will be the estimates/shcedule travel time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np\n",
    "import urbanaccess as ua\n",
    "from urbanaccess.config import settings\n",
    "from urbanaccess.gtfsfeeds import feeds\n",
    "from urbanaccess import gtfsfeeds\n",
    "from urbanaccess.gtfs.gtfsfeeds_dataframe import gtfsfeeds_dfs\n",
    "from urbanaccess.network import ua_network, load_network\n",
    "import matplotlib.pyplot as plt\n",
    "import re\n",
    "import os; os.chdir('../..')\n",
    "import difflib\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gpd\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "stations = pd.read_csv( os.getcwd() +'/08_Results/helper_tables/stations_w_strata')\n",
    "path = '/Users/juandavidcaicedocastro/TM_datos/gtfs-2'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading GTFS"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We load GFTS data using UrbanAcess library. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GTFS text file header whitespace check completed. Took 1.37 seconds\n",
      "--------------------------------\n",
      "Processing GTFS feed: gtfs-2\n",
      "GTFS feed: gtfs-2, calendar_dates uses the same service_id across multiple agency_ids. This feed calendar_dates table will be modified from its original format to provide service_ids for each agency using a one to many join\n",
      "GTFS feed: gtfs-2, calendar uses the same service_id across multiple agency_ids. This feed calendar table will be modified from its original format to provide service_ids for each agency using a one to many join\n",
      "GTFS feed: gtfs-2, stops uses the same stop_id across multiple agency_ids. This feed stops table will be modified from its original format to provide stop_ids for each agency using a one to many join\n",
      "agency.txt agency_name column has more than one agency name listed. Unique agency id was assigned using the agency id and associated agency name.\n",
      "Unique agency id operation complete. Took 31.51 seconds\n",
      "Unique GTFS feed id operation complete. Took 0.03 seconds\n",
      "Appended route type to stops\n",
      "Appended route type to stop_times\n",
      "--------------------------------\n",
      "Successfully converted ['departure_time'] to seconds past midnight and appended new columns to stop_times. Took 12.36 seconds\n",
      "1 GTFS feed file(s) successfully read as dataframes:\n",
      "     gtfs-2\n",
      "     Took 54.48 seconds\n"
     ]
    }
   ],
   "source": [
    "loaded_feeds = ua.gtfs.load.gtfsfeed_to_df(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['transmilenio-urbana', 'transmilenio-alimentadores',\n",
       "       'transmilenio-complementaria', 'transmilenio-especial',\n",
       "       'transmilenio-dual', 'transmilenio-troncal'], dtype=object)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loaded_feeds.stops.unique_agency_id.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The GTFSs data contains the entire public transit system in Bogota. BRT is only \"transmilenio-troncal\". So, let's filter out the other agencies. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "loaded_feeds.stops = loaded_feeds.stops[loaded_feeds.stops.unique_agency_id == 'transmilenio-troncal']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>stop_id</th>\n",
       "      <th>unique_agency_id</th>\n",
       "      <th>stop_code</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>stop_desc</th>\n",
       "      <th>stop_lat</th>\n",
       "      <th>stop_lon</th>\n",
       "      <th>stop_url</th>\n",
       "      <th>wheelchair_boarding</th>\n",
       "      <th>unique_feed_id</th>\n",
       "      <th>route_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11201</th>\n",
       "      <td>991141</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B002</td>\n",
       "      <td>Alcala C - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721598</td>\n",
       "      <td>-74.051421</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11212</th>\n",
       "      <td>991142</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B003</td>\n",
       "      <td>Alcala B - 1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721128</td>\n",
       "      <td>-74.051512</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11214</th>\n",
       "      <td>991143</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B004</td>\n",
       "      <td>Alcala B - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720956</td>\n",
       "      <td>-74.051521</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11216</th>\n",
       "      <td>991144</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B005</td>\n",
       "      <td>Alcala A - 1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720450</td>\n",
       "      <td>-74.051603</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11217</th>\n",
       "      <td>991145</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B006</td>\n",
       "      <td>Alcala A - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720278</td>\n",
       "      <td>-74.051640</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11218</th>\n",
       "      <td>991147</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B008</td>\n",
       "      <td>Alcala A - 4 o 6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720594</td>\n",
       "      <td>-74.051504</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11224</th>\n",
       "      <td>991149</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B010</td>\n",
       "      <td>Alcala B - 4 o 6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721263</td>\n",
       "      <td>-74.051413</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11232</th>\n",
       "      <td>991151</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B012</td>\n",
       "      <td>Alcala C - 4 o 6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721906</td>\n",
       "      <td>-74.051304</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11235</th>\n",
       "      <td>991152</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B013</td>\n",
       "      <td>C.C. Santafe  A - 3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.763170</td>\n",
       "      <td>-74.044296</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11236</th>\n",
       "      <td>991153</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B014</td>\n",
       "      <td>C.C. Santafe  A - 4 o 6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.763342</td>\n",
       "      <td>-74.044278</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      stop_id      unique_agency_id stop_code                stop_name  \\\n",
       "11201  991141  transmilenio-troncal      B002         Alcala C - 2 o 5   \n",
       "11212  991142  transmilenio-troncal      B003             Alcala B - 1   \n",
       "11214  991143  transmilenio-troncal      B004         Alcala B - 2 o 5   \n",
       "11216  991144  transmilenio-troncal      B005             Alcala A - 1   \n",
       "11217  991145  transmilenio-troncal      B006         Alcala A - 2 o 5   \n",
       "11218  991147  transmilenio-troncal      B008         Alcala A - 4 o 6   \n",
       "11224  991149  transmilenio-troncal      B010         Alcala B - 4 o 6   \n",
       "11232  991151  transmilenio-troncal      B012         Alcala C - 4 o 6   \n",
       "11235  991152  transmilenio-troncal      B013      C.C. Santafe  A - 3   \n",
       "11236  991153  transmilenio-troncal      B014  C.C. Santafe  A - 4 o 6   \n",
       "\n",
       "      stop_desc  stop_lat   stop_lon  stop_url  wheelchair_boarding  \\\n",
       "11201       NaN  4.721598 -74.051421       NaN                  1.0   \n",
       "11212       NaN  4.721128 -74.051512       NaN                  1.0   \n",
       "11214       NaN  4.720956 -74.051521       NaN                  1.0   \n",
       "11216       NaN  4.720450 -74.051603       NaN                  1.0   \n",
       "11217       NaN  4.720278 -74.051640       NaN                  1.0   \n",
       "11218       NaN  4.720594 -74.051504       NaN                  1.0   \n",
       "11224       NaN  4.721263 -74.051413       NaN                  1.0   \n",
       "11232       NaN  4.721906 -74.051304       NaN                  1.0   \n",
       "11235       NaN  4.763170 -74.044296       NaN                  1.0   \n",
       "11236       NaN  4.763342 -74.044278       NaN                  1.0   \n",
       "\n",
       "      unique_feed_id  route_type  \n",
       "11201       gtfs-2_1           3  \n",
       "11212       gtfs-2_1           3  \n",
       "11214       gtfs-2_1           3  \n",
       "11216       gtfs-2_1           3  \n",
       "11217       gtfs-2_1           3  \n",
       "11218       gtfs-2_1           3  \n",
       "11224       gtfs-2_1           3  \n",
       "11232       gtfs-2_1           3  \n",
       "11235       gtfs-2_1           3  \n",
       "11236       gtfs-2_1           3  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loaded_feeds.stops.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunatelly, the stop ID correspond to a door in a station, which means that one station can have multiple IDs. To solve this problem, we will group the stop IDs that belong to the same BRT stop and assign a new ID. Additionally, the new ID should be replace in the \"stops_times\" table. \n",
    "\n",
    "As the table seems to have no easy way to group stops, I will use the location (lon, lat) of the station in my stations data, and create a buffer of 200 meters. The stop ids within that buffer will be assinged to the station of the buffer. "
   ]
  },
  {
   "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>stop_id</th>\n",
       "      <th>unique_agency_id</th>\n",
       "      <th>stop_code</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>stop_desc</th>\n",
       "      <th>stop_lat</th>\n",
       "      <th>stop_lon</th>\n",
       "      <th>stop_url</th>\n",
       "      <th>wheelchair_boarding</th>\n",
       "      <th>unique_feed_id</th>\n",
       "      <th>route_type</th>\n",
       "      <th>geometry_left</th>\n",
       "      <th>index_right</th>\n",
       "      <th>idestacion</th>\n",
       "      <th>recaudoestacion</th>\n",
       "      <th>nombreestacion</th>\n",
       "      <th>ESTRATO</th>\n",
       "      <th>latitud</th>\n",
       "      <th>longitud</th>\n",
       "      <th>geometry_right</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11201</th>\n",
       "      <td>991141</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B002</td>\n",
       "      <td>Alcala C - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721598</td>\n",
       "      <td>-74.051421</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524236.778 -2179071.206)</td>\n",
       "      <td>133.0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.72085</td>\n",
       "      <td>-74.0515</td>\n",
       "      <td>POINT (5524262.279 -2179140.724)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11212</th>\n",
       "      <td>991142</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B003</td>\n",
       "      <td>Alcala B - 1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721128</td>\n",
       "      <td>-74.051512</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524248.221 -2179117.313)</td>\n",
       "      <td>133.0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.72085</td>\n",
       "      <td>-74.0515</td>\n",
       "      <td>POINT (5524262.279 -2179140.724)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11214</th>\n",
       "      <td>991143</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B004</td>\n",
       "      <td>Alcala B - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720956</td>\n",
       "      <td>-74.051521</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524255.052 -2179132.768)</td>\n",
       "      <td>133.0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.72085</td>\n",
       "      <td>-74.0515</td>\n",
       "      <td>POINT (5524262.279 -2179140.724)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11216</th>\n",
       "      <td>991144</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B005</td>\n",
       "      <td>Alcala A - 1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720450</td>\n",
       "      <td>-74.051603</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524269.141 -2179181.494)</td>\n",
       "      <td>133.0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.72085</td>\n",
       "      <td>-74.0515</td>\n",
       "      <td>POINT (5524262.279 -2179140.724)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11217</th>\n",
       "      <td>991145</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B006</td>\n",
       "      <td>Alcala A - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.720278</td>\n",
       "      <td>-74.051640</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524272.976 -2179198.523)</td>\n",
       "      <td>133.0</td>\n",
       "      <td>136.0</td>\n",
       "      <td>2200.0</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.72085</td>\n",
       "      <td>-74.0515</td>\n",
       "      <td>POINT (5524262.279 -2179140.724)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      stop_id      unique_agency_id stop_code         stop_name stop_desc  \\\n",
       "11201  991141  transmilenio-troncal      B002  Alcala C - 2 o 5       NaN   \n",
       "11212  991142  transmilenio-troncal      B003      Alcala B - 1       NaN   \n",
       "11214  991143  transmilenio-troncal      B004  Alcala B - 2 o 5       NaN   \n",
       "11216  991144  transmilenio-troncal      B005      Alcala A - 1       NaN   \n",
       "11217  991145  transmilenio-troncal      B006  Alcala A - 2 o 5       NaN   \n",
       "\n",
       "       stop_lat   stop_lon  stop_url  wheelchair_boarding unique_feed_id  \\\n",
       "11201  4.721598 -74.051421       NaN                  1.0       gtfs-2_1   \n",
       "11212  4.721128 -74.051512       NaN                  1.0       gtfs-2_1   \n",
       "11214  4.720956 -74.051521       NaN                  1.0       gtfs-2_1   \n",
       "11216  4.720450 -74.051603       NaN                  1.0       gtfs-2_1   \n",
       "11217  4.720278 -74.051640       NaN                  1.0       gtfs-2_1   \n",
       "\n",
       "       route_type                     geometry_left  index_right  idestacion  \\\n",
       "11201           3  POINT (5524236.778 -2179071.206)        133.0       136.0   \n",
       "11212           3  POINT (5524248.221 -2179117.313)        133.0       136.0   \n",
       "11214           3  POINT (5524255.052 -2179132.768)        133.0       136.0   \n",
       "11216           3  POINT (5524269.141 -2179181.494)        133.0       136.0   \n",
       "11217           3  POINT (5524272.976 -2179198.523)        133.0       136.0   \n",
       "\n",
       "       recaudoestacion nombreestacion  ESTRATO  latitud  longitud  \\\n",
       "11201           2200.0  Alcal?(02200)      3.0  4.72085  -74.0515   \n",
       "11212           2200.0  Alcal?(02200)      3.0  4.72085  -74.0515   \n",
       "11214           2200.0  Alcal?(02200)      3.0  4.72085  -74.0515   \n",
       "11216           2200.0  Alcal?(02200)      3.0  4.72085  -74.0515   \n",
       "11217           2200.0  Alcal?(02200)      3.0  4.72085  -74.0515   \n",
       "\n",
       "                         geometry_right  \n",
       "11201  POINT (5524262.279 -2179140.724)  \n",
       "11212  POINT (5524262.279 -2179140.724)  \n",
       "11214  POINT (5524262.279 -2179140.724)  \n",
       "11216  POINT (5524262.279 -2179140.724)  \n",
       "11217  POINT (5524262.279 -2179140.724)  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#loading stations in geopandas (TM database)\n",
    "stations_gpd = gpd.GeoDataFrame(stations, \n",
    "                                geometry = gpd.points_from_xy(stations.longitud, stations.latitud))\n",
    "\n",
    "#assign projection (regular coordinate system)\n",
    "stations_gpd.crs = {'init':'EPSG:4326'}\n",
    "stations_gpd.to_crs(epsg=3310, inplace = True)\n",
    "\n",
    "#Loading stations stops (GTFS files)\n",
    "gtfs_gpd = gpd.GeoDataFrame(loaded_feeds.stops, \n",
    "                         geometry = gpd.points_from_xy(loaded_feeds.stops.stop_lon, \n",
    "                                                       loaded_feeds.stops.stop_lat))\n",
    "gtfs_gpd.crs = {'init':'EPSG:4326'}\n",
    "gtfs_gpd.to_crs(epsg=3310, inplace = True)\n",
    "\n",
    "#creating buffer around stations \n",
    "stations_gpd['buffer'] = stations_gpd.buffer(200)\n",
    "stations_gpd = gpd.GeoDataFrame(stations_gpd, geometry = 'buffer')\n",
    "\n",
    "#spatial join \n",
    "merged = gpd.sjoin(gtfs_gpd, stations_gpd, how = \"left\", op='intersects', )\n",
    "merged[['stop_name', 'nombreestacion']].to_csv('check_stations.csv')\n",
    "merged.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems that the spatial join has worked. The 'stop_name' and 'nombreestacion' seem to be similar. While most of the stops have been assinged to a station, there are some missing stations. To make sure there is a perfect match, I have manually verified each match, and change or match a missing station to a stop. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Getting final tabel with the corresponding corresponding station name \n",
    "#both in TM and GTFS files. \n",
    "GTFS_TM_stations = pd.read_csv(os.getcwd() +'/08_Results/helper_tables/GTFS_tm_stations.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_relation = GTFS_TM_stations.merge(loaded_feeds.stops[['stop_id','stop_name']], how = 'left', on = 'stop_name').\\\n",
    "merge(stations[['nombreestacion', 'recaudoestacion']],\n",
    "      how = 'left', on = 'nombreestacion')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Stop ID</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>nombreestacion</th>\n",
       "      <th>stop_id</th>\n",
       "      <th>recaudoestacion</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11201</td>\n",
       "      <td>Alcala C - 2 o 5</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>991141</td>\n",
       "      <td>2200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11212</td>\n",
       "      <td>Alcala B - 1</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>991142</td>\n",
       "      <td>2200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11214</td>\n",
       "      <td>Alcala B - 2 o 5</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>991143</td>\n",
       "      <td>2200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11216</td>\n",
       "      <td>Alcala A - 1</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>991144</td>\n",
       "      <td>2200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11217</td>\n",
       "      <td>Alcala A - 2 o 5</td>\n",
       "      <td>Alcal?(02200)</td>\n",
       "      <td>991145</td>\n",
       "      <td>2200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Stop ID          stop_name nombreestacion stop_id  recaudoestacion\n",
       "0     11201  Alcala C - 2 o 5  Alcal?(02200)  991141             2200\n",
       "1     11212      Alcala B - 1  Alcal?(02200)  991142             2200\n",
       "2     11214  Alcala B - 2 o 5  Alcal?(02200)  991143             2200\n",
       "3     11216      Alcala A - 1  Alcal?(02200)  991144             2200\n",
       "4     11217  Alcala A - 2 o 5  Alcal?(02200)  991145             2200"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_relation.head()\n",
    "#The corresponding stop_id to a \"recaudoestacion\" id. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "stations_replace_dict = final_relation.groupby('stop_id')\\\n",
    "                        ['recaudoestacion'].apply(lambda g: g.values[0]).to_dict()\n",
    "\n",
    "loaded_feeds.stops.stop_id.replace(stations_replace_dict, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>stop_id</th>\n",
       "      <th>unique_agency_id</th>\n",
       "      <th>stop_code</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>stop_desc</th>\n",
       "      <th>stop_lat</th>\n",
       "      <th>stop_lon</th>\n",
       "      <th>stop_url</th>\n",
       "      <th>wheelchair_boarding</th>\n",
       "      <th>unique_feed_id</th>\n",
       "      <th>route_type</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11201</th>\n",
       "      <td>2200</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B002</td>\n",
       "      <td>Alcala C - 2 o 5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.721598</td>\n",
       "      <td>-74.051421</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5524236.778 -2179071.206)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11235</th>\n",
       "      <td>2001</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B013</td>\n",
       "      <td>C.C. Santafe  A - 3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.763170</td>\n",
       "      <td>-74.044296</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5523124.810 -2175047.180)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11243</th>\n",
       "      <td>2303</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>B017</td>\n",
       "      <td>Calle 85 C - 1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.673767</td>\n",
       "      <td>-74.059438</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>gtfs-2_1</td>\n",
       "      <td>3</td>\n",
       "      <td>POINT (5525534.942 -2183689.424)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       stop_id      unique_agency_id stop_code            stop_name stop_desc  \\\n",
       "11201     2200  transmilenio-troncal      B002     Alcala C - 2 o 5       NaN   \n",
       "11235     2001  transmilenio-troncal      B013  C.C. Santafe  A - 3       NaN   \n",
       "11243     2303  transmilenio-troncal      B017       Calle 85 C - 1       NaN   \n",
       "\n",
       "       stop_lat   stop_lon  stop_url  wheelchair_boarding unique_feed_id  \\\n",
       "11201  4.721598 -74.051421       NaN                  1.0       gtfs-2_1   \n",
       "11235  4.763170 -74.044296       NaN                  1.0       gtfs-2_1   \n",
       "11243  4.673767 -74.059438       NaN                  1.0       gtfs-2_1   \n",
       "\n",
       "       route_type                          geometry  \n",
       "11201           3  POINT (5524236.778 -2179071.206)  \n",
       "11235           3  POINT (5523124.810 -2175047.180)  \n",
       "11243           3  POINT (5525534.942 -2183689.424)  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking replace worked ok >> YES. \n",
    "loaded_feeds.stops.drop_duplicates(subset = 'stop_id', inplace = True)\n",
    "loaded_feeds.stops.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Replace stop ID in the \"stops_time\" table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "loaded_feeds.stop_times = loaded_feeds.stop_times[loaded_feeds.stop_times.unique_agency_id == 'transmilenio-troncal']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "loaded_feeds.stop_times.stop_id.replace(stations_replace_dict, inplace = True)\n",
    "loaded_feeds.stop_times.stop_id = loaded_feeds.stop_times.stop_id.astype(str)\n",
    "loaded_feeds.stops.stop_id = loaded_feeds.stops.stop_id.astype(str)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Weighted public transit Network \n",
    "### Nodes and link tables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For now, we will estimate the travel time for the morning peak. The results will be usefull for trips starting on working days, from 7a.m to 10.am. The result of this step is to get two tables of the netwrok 1) Nodes table, and 2) Links table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.21 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.62 seconds\n",
      "Stop times from 07:00:00 to 10:00:00 successfully selected 64,783 records out of 317,952 total records (20.38 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 4,922 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 7.23 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.57 seconds\n",
      "route id successfully joined to transit edges. Took 0.15 seconds\n",
      "Successfully created transit network. Took 11.56 seconds\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<urbanaccess.network.urbanaccess_network at 0x11caac9a0>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ua.gtfs.network.create_transit_net(gtfsfeeds_dfs=loaded_feeds,\n",
    "                                   day='monday',\n",
    "                                   timerange=['07:00:00', '10:00:00'],\n",
    "                                   calendar_dates_lookup=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "urbanaccess_net = ua.network.ua_network"
   ]
  },
  {
   "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>node_id_from</th>\n",
       "      <th>node_id_to</th>\n",
       "      <th>weight</th>\n",
       "      <th>unique_agency_id</th>\n",
       "      <th>unique_trip_id</th>\n",
       "      <th>sequence</th>\n",
       "      <th>id</th>\n",
       "      <th>route_type</th>\n",
       "      <th>unique_route_id</th>\n",
       "      <th>net_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8000_transmilenio-troncal</td>\n",
       "      <td>8001_transmilenio-troncal</td>\n",
       "      <td>1.600000</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal</td>\n",
       "      <td>1</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal_1</td>\n",
       "      <td>3</td>\n",
       "      <td>99102_transmilenio-troncal</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8001_transmilenio-troncal</td>\n",
       "      <td>9004_transmilenio-troncal</td>\n",
       "      <td>2.666667</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal</td>\n",
       "      <td>2</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal_2</td>\n",
       "      <td>3</td>\n",
       "      <td>99102_transmilenio-troncal</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9004_transmilenio-troncal</td>\n",
       "      <td>9100_transmilenio-troncal</td>\n",
       "      <td>1.733333</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal_3</td>\n",
       "      <td>3</td>\n",
       "      <td>99102_transmilenio-troncal</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9100_transmilenio-troncal</td>\n",
       "      <td>9103_transmilenio-troncal</td>\n",
       "      <td>2.966667</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal</td>\n",
       "      <td>4</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal_4</td>\n",
       "      <td>3</td>\n",
       "      <td>99102_transmilenio-troncal</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9103_transmilenio-troncal</td>\n",
       "      <td>9105_transmilenio-troncal</td>\n",
       "      <td>2.100000</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal</td>\n",
       "      <td>5</td>\n",
       "      <td>TR_102_100_TM_transmilenio-troncal_5</td>\n",
       "      <td>3</td>\n",
       "      <td>99102_transmilenio-troncal</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                node_id_from                 node_id_to    weight  \\\n",
       "0  8000_transmilenio-troncal  8001_transmilenio-troncal  1.600000   \n",
       "1  8001_transmilenio-troncal  9004_transmilenio-troncal  2.666667   \n",
       "2  9004_transmilenio-troncal  9100_transmilenio-troncal  1.733333   \n",
       "3  9100_transmilenio-troncal  9103_transmilenio-troncal  2.966667   \n",
       "4  9103_transmilenio-troncal  9105_transmilenio-troncal  2.100000   \n",
       "\n",
       "       unique_agency_id                      unique_trip_id  sequence  \\\n",
       "0  transmilenio-troncal  TR_102_100_TM_transmilenio-troncal         1   \n",
       "1  transmilenio-troncal  TR_102_100_TM_transmilenio-troncal         2   \n",
       "2  transmilenio-troncal  TR_102_100_TM_transmilenio-troncal         3   \n",
       "3  transmilenio-troncal  TR_102_100_TM_transmilenio-troncal         4   \n",
       "4  transmilenio-troncal  TR_102_100_TM_transmilenio-troncal         5   \n",
       "\n",
       "                                     id  route_type  \\\n",
       "0  TR_102_100_TM_transmilenio-troncal_1           3   \n",
       "1  TR_102_100_TM_transmilenio-troncal_2           3   \n",
       "2  TR_102_100_TM_transmilenio-troncal_3           3   \n",
       "3  TR_102_100_TM_transmilenio-troncal_4           3   \n",
       "4  TR_102_100_TM_transmilenio-troncal_5           3   \n",
       "\n",
       "              unique_route_id net_type  \n",
       "0  99102_transmilenio-troncal  transit  \n",
       "1  99102_transmilenio-troncal  transit  \n",
       "2  99102_transmilenio-troncal  transit  \n",
       "3  99102_transmilenio-troncal  transit  \n",
       "4  99102_transmilenio-troncal  transit  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urbanaccess_net.transit_edges.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>x</th>\n",
       "      <th>y</th>\n",
       "      <th>unique_agency_id</th>\n",
       "      <th>route_type</th>\n",
       "      <th>stop_id</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>stop_code</th>\n",
       "      <th>wheelchair_boarding</th>\n",
       "      <th>net_type</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>node_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2200_transmilenio-troncal</th>\n",
       "      <td>-74.051421</td>\n",
       "      <td>4.721598</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>2200</td>\n",
       "      <td>Alcala C - 2 o 5</td>\n",
       "      <td>B002</td>\n",
       "      <td>1.0</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001_transmilenio-troncal</th>\n",
       "      <td>-74.044296</td>\n",
       "      <td>4.763170</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>2001</td>\n",
       "      <td>C.C. Santafe  A - 3</td>\n",
       "      <td>B013</td>\n",
       "      <td>1.0</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2303_transmilenio-troncal</th>\n",
       "      <td>-74.059438</td>\n",
       "      <td>4.673767</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>2303</td>\n",
       "      <td>Calle 85 C - 1</td>\n",
       "      <td>B017</td>\n",
       "      <td>1.0</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2300_transmilenio-troncal</th>\n",
       "      <td>-74.057638</td>\n",
       "      <td>4.684538</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>2300</td>\n",
       "      <td>Calle 100 C - 1</td>\n",
       "      <td>B029</td>\n",
       "      <td>1.0</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2205_transmilenio-troncal</th>\n",
       "      <td>-74.055948</td>\n",
       "      <td>4.694414</td>\n",
       "      <td>transmilenio-troncal</td>\n",
       "      <td>3</td>\n",
       "      <td>2205</td>\n",
       "      <td>Calle 106 B - 1</td>\n",
       "      <td>B041</td>\n",
       "      <td>1.0</td>\n",
       "      <td>transit</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   x         y      unique_agency_id  \\\n",
       "node_id                                                                \n",
       "2200_transmilenio-troncal -74.051421  4.721598  transmilenio-troncal   \n",
       "2001_transmilenio-troncal -74.044296  4.763170  transmilenio-troncal   \n",
       "2303_transmilenio-troncal -74.059438  4.673767  transmilenio-troncal   \n",
       "2300_transmilenio-troncal -74.057638  4.684538  transmilenio-troncal   \n",
       "2205_transmilenio-troncal -74.055948  4.694414  transmilenio-troncal   \n",
       "\n",
       "                           route_type stop_id            stop_name stop_code  \\\n",
       "node_id                                                                        \n",
       "2200_transmilenio-troncal           3    2200     Alcala C - 2 o 5      B002   \n",
       "2001_transmilenio-troncal           3    2001  C.C. Santafe  A - 3      B013   \n",
       "2303_transmilenio-troncal           3    2303       Calle 85 C - 1      B017   \n",
       "2300_transmilenio-troncal           3    2300      Calle 100 C - 1      B029   \n",
       "2205_transmilenio-troncal           3    2205      Calle 106 B - 1      B041   \n",
       "\n",
       "                           wheelchair_boarding net_type  \n",
       "node_id                                                  \n",
       "2200_transmilenio-troncal                  1.0  transit  \n",
       "2001_transmilenio-troncal                  1.0  transit  \n",
       "2303_transmilenio-troncal                  1.0  transit  \n",
       "2300_transmilenio-troncal                  1.0  transit  \n",
       "2205_transmilenio-troncal                  1.0  transit  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urbanaccess_net.transit_nodes.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Figure created. Took 5.50 seconds\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 296.85x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ua.plot.plot_net(nodes=urbanaccess_net.transit_nodes,\n",
    "                 edges=urbanaccess_net.transit_edges);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building the Network and estimating trave time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import networkx as nx\n",
    "import pandana as pdna"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Creating the graph\n",
    "G=nx.Graph()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Adding nodes \n",
    "G.add_nodes_from(urbanaccess_net.transit_nodes.index)\n",
    "\n",
    "#Adding edges and weights\n",
    "for x in range(len(urbanaccess_net.transit_edges)):\n",
    "    G.add_edge(urbanaccess_net.transit_edges['node_id_from'][x],\n",
    "           urbanaccess_net.transit_edges['node_id_to'][x],\n",
    "           weight = urbanaccess_net.transit_edges['weight'][x])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9.183333333333334"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Experimenting\n",
    "nx.shortest_path_length(G, source = '8000_transmilenio-troncal', target = '9105_transmilenio-troncal',\n",
    "                       weight = 'weight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The travel time between station 8000 and 9105 is 9.183 minutes, wich is a plausible travel time for the two stations."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Building the travel time for the O-D pairs matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import permutations "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_tt_network(day = 'monday', time = ['07:00:00', '10:00:00']):\n",
    "    ''' Builds a travel time network for the time and day specified \n",
    "    input: \n",
    "    - day:\n",
    "    - time: \n",
    "    \n",
    "    Output: NetworkX object\n",
    "    '''\n",
    "\n",
    "    ua.gtfs.network.create_transit_net(gtfsfeeds_dfs=loaded_feeds,\n",
    "                                   day=day,\n",
    "                                   timerange=time,\n",
    "                                   calendar_dates_lookup=None)\n",
    "    \n",
    "    urbanaccess_net = ua.network.ua_network\n",
    "    G=nx.Graph()\n",
    "    #Adding nodes \n",
    "    G.add_nodes_from(urbanaccess_net.transit_nodes.index)\n",
    "    \n",
    "    #Adding edges and weights\n",
    "    for x in range(len(urbanaccess_net.transit_edges)):\n",
    "        G.add_edge(urbanaccess_net.transit_edges['node_id_from'][x],\n",
    "               urbanaccess_net.transit_edges['node_id_to'][x],\n",
    "               weight = urbanaccess_net.transit_edges['weight'][x])\n",
    "        \n",
    "    return G"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [],
   "source": [
    "def travel_time(G,ODs):\n",
    "    '''\n",
    "    Calculates the travel time given between origin and destination\n",
    "    Input: \n",
    "    - G: Networkx weighted travel time network \n",
    "    - ODs: Pandas dataframe with columns \"origin\" and \"destination\"\n",
    "    \n",
    "    Output: \n",
    "    - list of travel times for all OD pairs\n",
    "    '''\n",
    "    tt = []\n",
    "    for x in range(len(ODs)):\n",
    "        value = nx.shortest_path_length(G, \n",
    "                                source = ODs.origin.iloc[x], \n",
    "                                target = ODs.destination.iloc[x],\n",
    "                                weight = 'weight')\n",
    "        tt.append(value)\n",
    "    return tt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Building weighted travel time newwork for different times of the day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.27 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.67 seconds\n",
      "Stop times from 04:00:00 to 07:00:00 successfully selected 39,443 records out of 317,952 total records (12.41 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 2,959 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 4.47 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.00 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.44 seconds\n",
      "route id successfully joined to transit edges. Took 0.15 seconds\n",
      "Successfully created transit network. Took 8.73 seconds\n",
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.24 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.70 seconds\n",
      "Stop times from 07:00:00 to 10:00:00 successfully selected 64,783 records out of 317,952 total records (20.38 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 4,922 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 7.04 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.56 seconds\n",
      "route id successfully joined to transit edges. Took 0.15 seconds\n",
      "Successfully created transit network. Took 11.46 seconds\n",
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.24 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.67 seconds\n",
      "Stop times from 10:00:00 to 13:00:00 successfully selected 47,236 records out of 317,952 total records (14.86 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 3,219 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 4.99 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.50 seconds\n",
      "route id successfully joined to transit edges. Took 0.17 seconds\n",
      "Successfully created transit network. Took 9.29 seconds\n",
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.22 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.92 seconds\n",
      "Stop times from 13:00:00 to 16:00:00 successfully selected 47,052 records out of 317,952 total records (14.80 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 3,190 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 4.89 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.01 seconds\n",
      "route type successfully joined to transit edges. Took 0.51 seconds\n",
      "route id successfully joined to transit edges. Took 0.16 seconds\n",
      "Successfully created transit network. Took 9.43 seconds\n",
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.25 seconds\n",
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.84 seconds\n",
      "Stop times from 16:00:00 to 19:00:00 successfully selected 52,464 records out of 317,952 total records (16.50 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 3,725 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 5.28 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.51 seconds\n",
      "route id successfully joined to transit edges. Took 0.16 seconds\n",
      "Successfully created transit network. Took 9.78 seconds\n",
      "Using calendar to extract service_ids to select trips.\n",
      "15 service_ids were extracted from calendar\n",
      "113,712 trip(s) 95.99 percent of 118,462 total trip records were found in calendar for GTFS feed(s): ['gtfs-2']\n",
      "NOTE: If you expected more trips to have been extracted and your GTFS feed(s) have a calendar_dates file, consider utilizing the calendar_dates_lookup parameter in order to add additional trips based on information inside of calendar_dates. This should only be done if you know the corresponding GTFS feed is using calendar_dates instead of calendar to specify service_ids. When in doubt do not use the calendar_dates_lookup parameter.\n",
      "113,712 of 118,462 total trips were extracted representing calendar day: monday. Took 0.24 seconds\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are no departure time records missing from trips following monday schedule. There are no records to interpolate.\n",
      "Difference between stop times has been successfully calculated. Took 2.54 seconds\n",
      "Stop times from 19:00:00 to 22:00:00 successfully selected 51,302 records out of 317,952 total records (16.14 percent of total). Took 0.01 seconds\n",
      "Starting transformation process for 3,579 total trips...\n",
      "stop time table transformation to Pandana format edge table completed. Took 4.95 seconds\n",
      "Time conversion completed: seconds converted to minutes.\n",
      "146 of 146 records selected from stops. Took 0.01 seconds\n",
      "stop time table transformation to Pandana format node table completed. Took 0.00 seconds\n",
      "route type successfully joined to transit edges. Took 0.51 seconds\n",
      "route id successfully joined to transit edges. Took 0.16 seconds\n",
      "Successfully created transit network. Took 9.14 seconds\n"
     ]
    }
   ],
   "source": [
    "times_of_day = [['04:00:00', '07:00:00'], ['07:00:00', '10:00:00'], ['10:00:00', '13:00:00'],\n",
    "                 ['13:00:00', '16:00:00'], ['16:00:00', '19:00:00'], ['19:00:00', '22:00:00']]\n",
    "\n",
    "tt_networks = []\n",
    "for time in times_of_day:\n",
    "    G = build_tt_network(day = 'monday', time = time)\n",
    "    tt_networks.append(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def travel_time(G,ODs):\n",
    "    tt = []\n",
    "    for x in range(len(ODs)):\n",
    "        try: \n",
    "            value = nx.shortest_path_length(G, \n",
    "                                source = ODs.origin.iloc[x], \n",
    "                                target = ODs.destination.iloc[x],\n",
    "                                weight = 'weight')\n",
    "            \n",
    "        except 'NetworkXNoPath':\n",
    "            value = np.nan\n",
    "        tt.append(value)\n",
    "    return tt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compile all posible ODs in a dataframe \n",
    "ODs = urbanaccess_net.transit_nodes.index\n",
    "ODs = list(permutations(ODs, 2))\n",
    "ODs = pd.DataFrame(ODs, columns = ['origin', 'destination'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculates travel time for all OD pairs for all possible times\n",
    "start_time = 4\n",
    "for x in range(len(tt_networks)):\n",
    "    \n",
    "    name = str(start_time) + '_' + str(start_time + 3)\n",
    "    ODs[name] = travel_time(tt_networks[x],ODs)\n",
    "    start_time = start_time + 3\n",
    "\n",
    "# spliting the nodes name to extract the station ID that is consistent\n",
    "# with the station id of the other notebooks\n",
    "ODs['origin_id'] = ODs.origin.str.split(\"_\").str[0]\n",
    "ODs['destination_id'] = ODs.destination.str.split(\"_\").str[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [],
   "source": [
    "ODs['origin_id'] = ODs.origin.str.split(\"_\").str[0]\n",
    "ODs['destination_id'] = ODs.destination.str.split(\"_\").str[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "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>origin</th>\n",
       "      <th>destination</th>\n",
       "      <th>4_7</th>\n",
       "      <th>7_10</th>\n",
       "      <th>10_13</th>\n",
       "      <th>13_16</th>\n",
       "      <th>16_19</th>\n",
       "      <th>19_22</th>\n",
       "      <th>origin_id</th>\n",
       "      <th>destination_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2200_transmilenio-troncal</td>\n",
       "      <td>2001_transmilenio-troncal</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>2200</td>\n",
       "      <td>2001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2200_transmilenio-troncal</td>\n",
       "      <td>2303_transmilenio-troncal</td>\n",
       "      <td>9.083333</td>\n",
       "      <td>7.900000</td>\n",
       "      <td>7.900000</td>\n",
       "      <td>7.900000</td>\n",
       "      <td>7.850000</td>\n",
       "      <td>7.900000</td>\n",
       "      <td>2200</td>\n",
       "      <td>2303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2200_transmilenio-troncal</td>\n",
       "      <td>2300_transmilenio-troncal</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>4.383333</td>\n",
       "      <td>4.433333</td>\n",
       "      <td>2200</td>\n",
       "      <td>2300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2200_transmilenio-troncal</td>\n",
       "      <td>2205_transmilenio-troncal</td>\n",
       "      <td>6.733333</td>\n",
       "      <td>6.516667</td>\n",
       "      <td>6.516667</td>\n",
       "      <td>6.516667</td>\n",
       "      <td>6.466667</td>\n",
       "      <td>6.516667</td>\n",
       "      <td>2200</td>\n",
       "      <td>2205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2200_transmilenio-troncal</td>\n",
       "      <td>2202_transmilenio-troncal</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2200</td>\n",
       "      <td>2202</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      origin                destination       4_7      7_10  \\\n",
       "0  2200_transmilenio-troncal  2001_transmilenio-troncal  4.250000  4.250000   \n",
       "1  2200_transmilenio-troncal  2303_transmilenio-troncal  9.083333  7.900000   \n",
       "2  2200_transmilenio-troncal  2300_transmilenio-troncal  7.000000  4.433333   \n",
       "3  2200_transmilenio-troncal  2205_transmilenio-troncal  6.733333  6.516667   \n",
       "4  2200_transmilenio-troncal  2202_transmilenio-troncal  3.000000  3.000000   \n",
       "\n",
       "      10_13     13_16     16_19     19_22 origin_id destination_id  \n",
       "0  4.250000  4.250000  4.250000  4.250000      2200           2001  \n",
       "1  7.900000  7.900000  7.850000  7.900000      2200           2303  \n",
       "2  4.433333  4.433333  4.383333  4.433333      2200           2300  \n",
       "3  6.516667  6.516667  6.466667  6.516667      2200           2205  \n",
       "4  3.000000  3.000000  3.000000  3.000000      2200           2202  "
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ODs.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Since GTFS are public data, I can share the results of this table :) \n",
    "ODs.to_csv(os.getcwd() + '/07_Notebooks/cp255/data/tm_skims.csv', index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "#This could be a faster way to calculate travel time (I will need to explore this further)\n",
    "# nx.shortest_path_length(G, list_a, list_a, weight='weight')"
   ]
  }
 ],
 "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
}