preprocessing.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 🔍 Exploration and Data Preprocessing"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this section, we will take a first look at the data and preprocess it to make it more suitable for our analysis. The following steps will be performed:\n",
    "1. Download the data from the OpenTransportData platform for the 10th of March 2023.\n",
    "2. Remove unnecessary columns and translate the column names from German to English."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📂 Download the Data"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data is available as a CSV file. We will use the `pandas` library to read the data into a `DataFrame` and keep only the columns that we are interested in. The data contains the following columns:\n",
    "| Column Name | Type | Description |\n",
    "| --- | --- | --- |\n",
    "| `BETRIEBSTAG` | Date (DD.MM.YYYY) | Date of the journey |\n",
    "| `FAHRT_BEZEICHNER` | String | Identifier of the journey (see https://opentransportdata.swiss/en/cookbook/fahrt-id-journeyref/) |\n",
    "| `BETREIBER_ID` | String | Identifier of the operator |\n",
    "| `BETREIBER_ABK` | String | Abbreviation of the operator |\n",
    "| `BETREIBER_NAME` | String | Name of the operator |\n",
    "| `PRODUKT_ID` | Category | Identifier of the product |\n",
    "| `LINIEN_ID` | String | Identifier of the line |\n",
    "| `LINIEN_TEXT` | String | Name of the line |\n",
    "| `UMLAUF_ID` | String | Identifier of the circuit (see https://opentransportdata.swiss/en/cookbook/umlauf/)|\n",
    "| `VERKEHRSMITTEL_TEXT` | String | Type of transport |\n",
    "| `ZUSATZFAHRT_TF` | Boolean | Whether the journey is an additional one |\n",
    "| `FAELLT_AUS_TF` | Boolean | Whether the journey is cancelled |\n",
    "| `BPUIC` | String | Identifier of the stop (see https://didok.ch/en/glossary-stop-points-location-codes/) |\n",
    "| `HALTESTELLEN_NAME` | String | Name of the stop |\n",
    "| `ANKUNFTSZEIT` | DateTime (DD.MM.YYYY HH:MM:SS) | Arrival time at the stop |\n",
    "| `AN_PROGNOSE` | DateTime (DD.MM.YYYY HH:MM:SS) | Arrival time at the stop (predicted) |\n",
    "| `AN_PROGNOSE_STATUS` | Category | Status of the predicted arrival time |\n",
    "| `ABFAHRTSZEIT` | DateTime (DD.MM.YYYY HH:MM:SS) | Departure time from the stop |\n",
    "| `AB_PROGNOSE` | DateTime (DD.MM.YYYY HH:MM:SS) | Departure time from the stop (predicted) |\n",
    "| `AB_PROGNOSE_STATUS` | Category | Status of the predicted departure time |\n",
    "| `DURCHFAHRT_TF` | Boolean | Whether the stop is a through stop |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# False if you want to download the data from the internet\n",
    "LOCAL_DOWNLOAD = True\n",
    "DATE_FILE = '2023-03-10' # Format YYYY-MM-DD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Download the data\n",
    "DATA_LINK = f\"https://opentransportdata.swiss/dataset/0edc74a3-ad4d-486e-8657-f8f3b34a0979/resource/9de83058-0365-481d-85fa-a8477709b9fa/download/{DATE_FILE}_istdaten.csv\"\n",
    "DATA_FILE = f\"data/{DATE_FILE}_istdaten.csv\"\n",
    "DATA_PATH = DATA_FILE if LOCAL_DOWNLOAD else DATA_LINK\n",
    "transport_data = pd.read_csv(\n",
    "    DATA_PATH, \n",
    "    sep=';',\n",
    "    parse_dates=['BETRIEBSTAG', 'ANKUNFTSZEIT', 'AN_PROGNOSE', 'ABFAHRTSZEIT', 'AB_PROGNOSE'],\n",
    "    dtype={\n",
    "        'FAHRT_BEZEICHNER': 'string',\n",
    "        'BETRIEBER_ID': 'string',\n",
    "        'BETREIBER_ABK': 'string',\n",
    "        'BETREIBER_NAME': 'string',\n",
    "        'PRODUKT_ID': 'category',\n",
    "        'LINIEN_ID': 'string',\n",
    "        'LINIEN_TEXT': 'string',\n",
    "        'UMLAUF_ID': 'string',\n",
    "        'VERKEHRSMITTEL_TEXT': 'string',\n",
    "        'ZUSATZFAHRT_TF': 'boolean',\n",
    "        'FAELLT_AUS_TF': 'boolean',\n",
    "        'BPUIC': 'string',\n",
    "        'HALTESTELLEN_NAME': 'string',\n",
    "        'AN_PROGNOSE_STATUS': 'category',\n",
    "        'AB_PROGNOSE_STATUS': 'category',\n",
    "        'DURCHFAHRT_TF': 'boolean'\n",
    "    },\n",
    "    dayfirst=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>BETRIEBSTAG</th>\n",
       "      <th>FAHRT_BEZEICHNER</th>\n",
       "      <th>BETREIBER_ID</th>\n",
       "      <th>BETREIBER_ABK</th>\n",
       "      <th>BETREIBER_NAME</th>\n",
       "      <th>PRODUKT_ID</th>\n",
       "      <th>LINIEN_ID</th>\n",
       "      <th>LINIEN_TEXT</th>\n",
       "      <th>UMLAUF_ID</th>\n",
       "      <th>VERKEHRSMITTEL_TEXT</th>\n",
       "      <th>...</th>\n",
       "      <th>FAELLT_AUS_TF</th>\n",
       "      <th>BPUIC</th>\n",
       "      <th>HALTESTELLEN_NAME</th>\n",
       "      <th>ANKUNFTSZEIT</th>\n",
       "      <th>AN_PROGNOSE</th>\n",
       "      <th>AN_PROGNOSE_STATUS</th>\n",
       "      <th>ABFAHRTSZEIT</th>\n",
       "      <th>AB_PROGNOSE</th>\n",
       "      <th>AB_PROGNOSE_STATUS</th>\n",
       "      <th>DURCHFAHRT_TF</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17230:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17230</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 04:59:00</td>\n",
       "      <td>2023-03-10 04:59:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17233:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17233</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 06:07:00</td>\n",
       "      <td>2023-03-10 06:07:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17234:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17234</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 05:56:00</td>\n",
       "      <td>NaT</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17235:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17235</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 06:43:00</td>\n",
       "      <td>2023-03-10 06:43:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17236:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17236</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 06:33:00</td>\n",
       "      <td>2023-03-10 06:37:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  BETRIEBSTAG     FAHRT_BEZEICHNER BETREIBER_ID BETREIBER_ABK  \\\n",
       "0  2023-03-10  80:800631:17230:000    80:800631            DB   \n",
       "1  2023-03-10  80:800631:17233:000    80:800631            DB   \n",
       "2  2023-03-10  80:800631:17234:000    80:800631            DB   \n",
       "3  2023-03-10  80:800631:17235:000    80:800631            DB   \n",
       "4  2023-03-10  80:800631:17236:000    80:800631            DB   \n",
       "\n",
       "                  BETREIBER_NAME PRODUKT_ID LINIEN_ID LINIEN_TEXT UMLAUF_ID  \\\n",
       "0  DB Regio AG Baden-Württemberg        Zug     17230          RB      <NA>   \n",
       "1  DB Regio AG Baden-Württemberg        Zug     17233          RB      <NA>   \n",
       "2  DB Regio AG Baden-Württemberg        Zug     17234          RB      <NA>   \n",
       "3  DB Regio AG Baden-Württemberg        Zug     17235          RB      <NA>   \n",
       "4  DB Regio AG Baden-Württemberg        Zug     17236          RB      <NA>   \n",
       "\n",
       "  VERKEHRSMITTEL_TEXT  ...  FAELLT_AUS_TF    BPUIC HALTESTELLEN_NAME  \\\n",
       "0                  RB  ...          False  8500090      Basel Bad Bf   \n",
       "1                  RB  ...          False  8500090      Basel Bad Bf   \n",
       "2                  RB  ...           True  8500090      Basel Bad Bf   \n",
       "3                  RB  ...          False  8500090      Basel Bad Bf   \n",
       "4                  RB  ...          False  8500090      Basel Bad Bf   \n",
       "\n",
       "         ANKUNFTSZEIT         AN_PROGNOSE AN_PROGNOSE_STATUS  \\\n",
       "0 2023-03-10 04:59:00 2023-03-10 04:59:00           PROGNOSE   \n",
       "1                 NaT                 NaT                NaN   \n",
       "2 2023-03-10 05:56:00                 NaT          UNBEKANNT   \n",
       "3                 NaT                 NaT                NaN   \n",
       "4 2023-03-10 06:33:00 2023-03-10 06:37:00           PROGNOSE   \n",
       "\n",
       "         ABFAHRTSZEIT         AB_PROGNOSE AB_PROGNOSE_STATUS DURCHFAHRT_TF  \n",
       "0                 NaT                 NaT                NaN         False  \n",
       "1 2023-03-10 06:07:00 2023-03-10 06:07:00           PROGNOSE         False  \n",
       "2                 NaT                 NaT          UNBEKANNT         False  \n",
       "3 2023-03-10 06:43:00 2023-03-10 06:43:00           PROGNOSE         False  \n",
       "4                 NaT                 NaT                NaN         False  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BETRIEBSTAG            datetime64[ns]\n",
       "FAHRT_BEZEICHNER               string\n",
       "BETREIBER_ID                   object\n",
       "BETREIBER_ABK                  string\n",
       "BETREIBER_NAME                 string\n",
       "PRODUKT_ID                   category\n",
       "LINIEN_ID                      string\n",
       "LINIEN_TEXT                    string\n",
       "UMLAUF_ID                      string\n",
       "VERKEHRSMITTEL_TEXT            string\n",
       "ZUSATZFAHRT_TF                boolean\n",
       "FAELLT_AUS_TF                 boolean\n",
       "BPUIC                          string\n",
       "HALTESTELLEN_NAME              string\n",
       "ANKUNFTSZEIT           datetime64[ns]\n",
       "AN_PROGNOSE            datetime64[ns]\n",
       "AN_PROGNOSE_STATUS           category\n",
       "ABFAHRTSZEIT           datetime64[ns]\n",
       "AB_PROGNOSE            datetime64[ns]\n",
       "AB_PROGNOSE_STATUS           category\n",
       "DURCHFAHRT_TF                 boolean\n",
       "dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data.dtypes"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Translate Column Names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>is_cancelled</th>\n",
       "      <th>stop_id</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17230:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17230</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 04:59:00</td>\n",
       "      <td>2023-03-10 04:59:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17233:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17233</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 06:07:00</td>\n",
       "      <td>2023-03-10 06:07:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17234:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17234</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 05:56:00</td>\n",
       "      <td>NaT</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17235:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17235</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 06:43:00</td>\n",
       "      <td>2023-03-10 06:43:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>80:800631:17236:000</td>\n",
       "      <td>80:800631</td>\n",
       "      <td>DB</td>\n",
       "      <td>DB Regio AG Baden-Württemberg</td>\n",
       "      <td>Zug</td>\n",
       "      <td>17236</td>\n",
       "      <td>RB</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>RB</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8500090</td>\n",
       "      <td>Basel Bad Bf</td>\n",
       "      <td>2023-03-10 06:33:00</td>\n",
       "      <td>2023-03-10 06:37:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        date              trip_id operator_id operator_abbreviation  \\\n",
       "0 2023-03-10  80:800631:17230:000   80:800631                    DB   \n",
       "1 2023-03-10  80:800631:17233:000   80:800631                    DB   \n",
       "2 2023-03-10  80:800631:17234:000   80:800631                    DB   \n",
       "3 2023-03-10  80:800631:17235:000   80:800631                    DB   \n",
       "4 2023-03-10  80:800631:17236:000   80:800631                    DB   \n",
       "\n",
       "                   operator_name product_id line_id line_text circuit_id  \\\n",
       "0  DB Regio AG Baden-Württemberg        Zug   17230        RB       <NA>   \n",
       "1  DB Regio AG Baden-Württemberg        Zug   17233        RB       <NA>   \n",
       "2  DB Regio AG Baden-Württemberg        Zug   17234        RB       <NA>   \n",
       "3  DB Regio AG Baden-Württemberg        Zug   17235        RB       <NA>   \n",
       "4  DB Regio AG Baden-Württemberg        Zug   17236        RB       <NA>   \n",
       "\n",
       "  transport_type  ...  is_cancelled  stop_id     stop_name  \\\n",
       "0             RB  ...         False  8500090  Basel Bad Bf   \n",
       "1             RB  ...         False  8500090  Basel Bad Bf   \n",
       "2             RB  ...          True  8500090  Basel Bad Bf   \n",
       "3             RB  ...         False  8500090  Basel Bad Bf   \n",
       "4             RB  ...         False  8500090  Basel Bad Bf   \n",
       "\n",
       "         arrival_time    arrival_forecast arrival_forecast_status  \\\n",
       "0 2023-03-10 04:59:00 2023-03-10 04:59:00                PROGNOSE   \n",
       "1                 NaT                 NaT                     NaN   \n",
       "2 2023-03-10 05:56:00                 NaT               UNBEKANNT   \n",
       "3                 NaT                 NaT                     NaN   \n",
       "4 2023-03-10 06:33:00 2023-03-10 06:37:00                PROGNOSE   \n",
       "\n",
       "       departure_time  departure_forecast departure_forecast_status  \\\n",
       "0                 NaT                 NaT                       NaN   \n",
       "1 2023-03-10 06:07:00 2023-03-10 06:07:00                  PROGNOSE   \n",
       "2                 NaT                 NaT                 UNBEKANNT   \n",
       "3 2023-03-10 06:43:00 2023-03-10 06:43:00                  PROGNOSE   \n",
       "4                 NaT                 NaT                       NaN   \n",
       "\n",
       "  is_through_trip  \n",
       "0           False  \n",
       "1           False  \n",
       "2           False  \n",
       "3           False  \n",
       "4           False  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "translations = {\n",
    "    'BETRIEBSTAG': 'date',\n",
    "    'FAHRT_BEZEICHNER': 'trip_id',\n",
    "    'BETREIBER_ID': 'operator_id',\n",
    "    'BETREIBER_ABK': 'operator_abbreviation',\n",
    "    'BETREIBER_NAME': 'operator_name',\n",
    "    'PRODUKT_ID': 'product_id',\n",
    "    'LINIEN_ID': 'line_id',\n",
    "    'LINIEN_TEXT': 'line_text',\n",
    "    'UMLAUF_ID': 'circuit_id',\n",
    "    'VERKEHRSMITTEL_TEXT': 'transport_type',\n",
    "    'ZUSATZFAHRT_TF': 'is_additional_trip',\n",
    "    'FAELLT_AUS_TF': 'is_cancelled',\n",
    "    'BPUIC': 'stop_id',\n",
    "    'HALTESTELLEN_NAME': 'stop_name',\n",
    "    'ANKUNFTSZEIT': 'arrival_time',\n",
    "    'AN_PROGNOSE': 'arrival_forecast',\n",
    "    'AN_PROGNOSE_STATUS': 'arrival_forecast_status',\n",
    "    'ABFAHRTSZEIT': 'departure_time',\n",
    "    'AB_PROGNOSE': 'departure_forecast',\n",
    "    'AB_PROGNOSE_STATUS': 'departure_forecast_status',\n",
    "    'DURCHFAHRT_TF': 'is_through_trip'\n",
    "}\n",
    "\n",
    "transport_data = transport_data.rename(columns=translations)\n",
    "transport_data.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Basic Analysis"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Transport Types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<StringArray>\n",
       "[ 'RB', 'IRE',  'RE', 'ICE', 'FLX',   'S', 'Bus',   'R',  'IC',  'EC',  'IR',\n",
       " 'RJX', 'TER', 'EXT',  'NJ', 'TGV',  'CC',   'B',  'BN',   'M', 'BAT',  'PE',\n",
       "   'T',  'TN',  'SN', 'CAR', 'EXB', 'Zug',  <NA>]\n",
       "Length: 29, dtype: string"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_types = transport_data['transport_type'].unique()\n",
    "transport_types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Ozan\\AppData\\Local\\Temp\\ipykernel_26268\\669156617.py:1: FutureWarning: Treating datetime data as categorical rather than numeric in `.describe` is deprecated and will be removed in a future version of pandas. Specify `datetime_is_numeric=True` to silence this warning and adopt the future behavior now.\n",
      "  transport_data.describe()\n",
      "C:\\Users\\Ozan\\AppData\\Local\\Temp\\ipykernel_26268\\669156617.py:1: FutureWarning: Treating datetime data as categorical rather than numeric in `.describe` is deprecated and will be removed in a future version of pandas. Specify `datetime_is_numeric=True` to silence this warning and adopt the future behavior now.\n",
      "  transport_data.describe()\n",
      "C:\\Users\\Ozan\\AppData\\Local\\Temp\\ipykernel_26268\\669156617.py:1: FutureWarning: Treating datetime data as categorical rather than numeric in `.describe` is deprecated and will be removed in a future version of pandas. Specify `datetime_is_numeric=True` to silence this warning and adopt the future behavior now.\n",
      "  transport_data.describe()\n",
      "C:\\Users\\Ozan\\AppData\\Local\\Temp\\ipykernel_26268\\669156617.py:1: FutureWarning: Treating datetime data as categorical rather than numeric in `.describe` is deprecated and will be removed in a future version of pandas. Specify `datetime_is_numeric=True` to silence this warning and adopt the future behavior now.\n",
      "  transport_data.describe()\n",
      "C:\\Users\\Ozan\\AppData\\Local\\Temp\\ipykernel_26268\\669156617.py:1: FutureWarning: Treating datetime data as categorical rather than numeric in `.describe` is deprecated and will be removed in a future version of pandas. Specify `datetime_is_numeric=True` to silence this warning and adopt the future behavior now.\n",
      "  transport_data.describe()\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>is_cancelled</th>\n",
       "      <th>stop_id</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347042</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>1137342</td>\n",
       "      <td>2339682</td>\n",
       "      <td>...</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2347175</td>\n",
       "      <td>2188441</td>\n",
       "      <td>2207604</td>\n",
       "      <td>2179615</td>\n",
       "      <td>2207980</td>\n",
       "      <td>2207949</td>\n",
       "      <td>2179175</td>\n",
       "      <td>2208289</td>\n",
       "      <td>2347175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>1</td>\n",
       "      <td>141527</td>\n",
       "      <td>135</td>\n",
       "      <td>130</td>\n",
       "      <td>126</td>\n",
       "      <td>7</td>\n",
       "      <td>13601</td>\n",
       "      <td>936</td>\n",
       "      <td>2835</td>\n",
       "      <td>28</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>21200</td>\n",
       "      <td>19967</td>\n",
       "      <td>1622</td>\n",
       "      <td>81362</td>\n",
       "      <td>4</td>\n",
       "      <td>1614</td>\n",
       "      <td>81462</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>2023-03-10 00:00:00</td>\n",
       "      <td>85:841:502-2082</td>\n",
       "      <td>85:801</td>\n",
       "      <td>PAG</td>\n",
       "      <td>PostAuto AG</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:827:10</td>\n",
       "      <td>1</td>\n",
       "      <td>601</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>8576646</td>\n",
       "      <td>Bern, Bahnhof</td>\n",
       "      <td>2023-03-10 07:38:00</td>\n",
       "      <td>2023-03-10 16:59:00</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-10 17:05:00</td>\n",
       "      <td>2023-03-10 16:55:00</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>2347175</td>\n",
       "      <td>79</td>\n",
       "      <td>364633</td>\n",
       "      <td>364633</td>\n",
       "      <td>364633</td>\n",
       "      <td>1777815</td>\n",
       "      <td>14112</td>\n",
       "      <td>104034</td>\n",
       "      <td>3801</td>\n",
       "      <td>1905018</td>\n",
       "      <td>...</td>\n",
       "      <td>2323023</td>\n",
       "      <td>3844</td>\n",
       "      <td>3837</td>\n",
       "      <td>2583</td>\n",
       "      <td>141</td>\n",
       "      <td>1656758</td>\n",
       "      <td>2631</td>\n",
       "      <td>142</td>\n",
       "      <td>1700550</td>\n",
       "      <td>2347167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>first</th>\n",
       "      <td>2023-03-10 00:00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 00:00:00</td>\n",
       "      <td>2023-03-10 00:00:53</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-10 00:00:00</td>\n",
       "      <td>2023-03-10 00:00:18</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>last</th>\n",
       "      <td>2023-03-10 00:00:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-11 05:07:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2023-03-11 05:05:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       date          trip_id operator_id  \\\n",
       "count               2347175          2347175     2347175   \n",
       "unique                    1           141527         135   \n",
       "top     2023-03-10 00:00:00  85:841:502-2082      85:801   \n",
       "freq                2347175               79      364633   \n",
       "first   2023-03-10 00:00:00              NaN         NaN   \n",
       "last    2023-03-10 00:00:00              NaN         NaN   \n",
       "\n",
       "       operator_abbreviation operator_name product_id    line_id line_text  \\\n",
       "count                2347175       2347175    2347042    2347175   2347175   \n",
       "unique                   130           126          7      13601       936   \n",
       "top                      PAG   PostAuto AG        Bus  85:827:10         1   \n",
       "freq                  364633        364633    1777815      14112    104034   \n",
       "first                    NaN           NaN        NaN        NaN       NaN   \n",
       "last                     NaN           NaN        NaN        NaN       NaN   \n",
       "\n",
       "       circuit_id transport_type  ... is_cancelled  stop_id      stop_name  \\\n",
       "count     1137342        2339682  ...      2347175  2347175        2188441   \n",
       "unique       2835             28  ...            2    21200          19967   \n",
       "top           601              B  ...        False  8576646  Bern, Bahnhof   \n",
       "freq         3801        1905018  ...      2323023     3844           3837   \n",
       "first         NaN            NaN  ...          NaN      NaN            NaN   \n",
       "last          NaN            NaN  ...          NaN      NaN            NaN   \n",
       "\n",
       "               arrival_time     arrival_forecast arrival_forecast_status  \\\n",
       "count               2207604              2179615                 2207980   \n",
       "unique                 1622                81362                       4   \n",
       "top     2023-03-10 07:38:00  2023-03-10 16:59:00                    REAL   \n",
       "freq                   2583                  141                 1656758   \n",
       "first   2023-03-10 00:00:00  2023-03-10 00:00:53                     NaN   \n",
       "last    2023-03-11 05:07:00  2023-03-11 05:56:41                     NaN   \n",
       "\n",
       "             departure_time   departure_forecast departure_forecast_status  \\\n",
       "count               2207949              2179175                   2208289   \n",
       "unique                 1614                81462                         4   \n",
       "top     2023-03-10 17:05:00  2023-03-10 16:55:00                      REAL   \n",
       "freq                   2631                  142                   1700550   \n",
       "first   2023-03-10 00:00:00  2023-03-10 00:00:18                       NaN   \n",
       "last    2023-03-11 05:05:00  2023-03-11 05:56:41                       NaN   \n",
       "\n",
       "       is_through_trip  \n",
       "count          2347175  \n",
       "unique               2  \n",
       "top              False  \n",
       "freq           2347167  \n",
       "first              NaN  \n",
       "last               NaN  \n",
       "\n",
       "[6 rows x 21 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2               17234\n",
       "39              17275\n",
       "42              17278\n",
       "143             17185\n",
       "148             17192\n",
       "              ...    \n",
       "2305961    85:896:152\n",
       "2305962    85:896:152\n",
       "2305963    85:896:152\n",
       "2305964    85:896:152\n",
       "2305965    85:896:152\n",
       "Name: line_id, Length: 24152, dtype: string"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[(transport_data[\"is_cancelled\"])][\"line_id\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "634"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Here we compute the number of lines that had cancellation\n",
    "num_cancelled = transport_data[(transport_data[\"is_cancelled\"])][\"line_id\"].unique().shape[0]\n",
    "num_cancelled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13601"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The total number of lines that day\n",
    "num_lines = transport_data[\"line_id\"].unique().shape[0]\n",
    "num_lines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 4.66%\n"
     ]
    }
   ],
   "source": [
    "# The percentage of lines that had cancellation\n",
    "print(f\"{num_cancelled / num_lines: .2%}\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🕒 Analysis on Arrival Times"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2179615.00000</td>\n",
       "      <td>2179175.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.41384</td>\n",
       "      <td>1.60938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5.56498</td>\n",
       "      <td>5.54900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1442.01667</td>\n",
       "      <td>-1439.85000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.36667</td>\n",
       "      <td>0.53333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.08333</td>\n",
       "      <td>1.23333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.05000</td>\n",
       "      <td>2.20000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>718.68333</td>\n",
       "      <td>718.68333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       arrival_delay departure_delay\n",
       "count  2179615.00000   2179175.00000\n",
       "mean         1.41384         1.60938\n",
       "std          5.56498         5.54900\n",
       "min      -1442.01667     -1439.85000\n",
       "25%          0.36667         0.53333\n",
       "50%          1.08333         1.23333\n",
       "75%          2.05000         2.20000\n",
       "max        718.68333       718.68333"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We add a new column to the datafram that contains the delay in minutes\n",
    "transport_data[\"arrival_delay\"] = (transport_data[\"arrival_forecast\"] - transport_data[\"arrival_time\"]).dt.total_seconds() / 60\n",
    "transport_data[\"departure_delay\"] = (transport_data[\"departure_forecast\"] - transport_data[\"departure_time\"]).dt.total_seconds() / 60\n",
    "\n",
    "transport_data[['arrival_delay', 'departure_delay']].describe().apply(lambda s: s.apply('{0:.5f}'.format))"
   ]
  },
  {
   "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>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1541021.00000</td>\n",
       "      <td>1541020.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.48480</td>\n",
       "      <td>1.69278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.39073</td>\n",
       "      <td>4.38616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1440.73333</td>\n",
       "      <td>-1439.85000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.51667</td>\n",
       "      <td>0.68333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.18333</td>\n",
       "      <td>1.36667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.11667</td>\n",
       "      <td>2.30000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>75.08333</td>\n",
       "      <td>123.08333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       arrival_delay departure_delay\n",
       "count  1541021.00000   1541020.00000\n",
       "mean         1.48480         1.69278\n",
       "std          4.39073         4.38616\n",
       "min      -1440.73333     -1439.85000\n",
       "25%          0.51667         0.68333\n",
       "50%          1.18333         1.36667\n",
       "75%          2.11667         2.30000\n",
       "max         75.08333       123.08333"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Only take rows where the arrival and delay forecast are \"REAL\"\n",
    "(transport_data[(transport_data[\"arrival_forecast_status\"] == \"REAL\") & (transport_data[\"departure_forecast_status\"] == \"REAL\")]).describe().apply(lambda s: s.apply('{0:.5f}'.format))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='arrival_delay', ylabel='Count'>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This histogram coressponds to the delays in minutes for each arrival of each line (we remove the outliers)\n",
    "sns.histplot(data=transport_data[(transport_data[\"arrival_delay\"] > -10000) & (transport_data[\"arrival_delay\"] < 10000)], x=\"arrival_delay\", bins=100, log_scale=(False, True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='departure_delay', ylabel='Count'>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This histogram coressponds to the delays in minutes for each departure of each line (we remove the outliers)\n",
    "sns.histplot(data=transport_data[(transport_data[\"departure_delay\"] > -200) & (transport_data[\"departure_delay\"] < 200)], x=\"departure_delay\", bins=100, log_scale=(False, True))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the distributions on a logarithmic y are bi-modal. They seem to follow a laplace distribution centered around 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='departure_delay', ylabel='Count'>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# This histogram coressponds to the delays in minutes for each departure of each line (we remove the outliers)\n",
    "sns.histplot(data=transport_data[(transport_data[\"departure_delay\"] > -100) & (transport_data[\"departure_delay\"] < 100)], x=\"departure_delay\", bins=100, log_scale=(False, True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='arrival_delay', ylabel='Count'>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data=transport_data.groupby(\"line_id\").mean(numeric_only=True), x=\"arrival_delay\", bins=10, log_scale=(False, True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='departure_delay', ylabel='Count'>"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data=transport_data.groupby(\"line_id\").mean(numeric_only=True), x=\"departure_delay\", bins=10, log_scale=(False, True))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyse outliers and weird values"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The delay is in minutes."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "10000 minutes is 166 hours, which is 6.9 days. This is a lot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] > 10000].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "      <th>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:11:1093:001</td>\n",
       "      <td>85:11</td>\n",
       "      <td>SBB</td>\n",
       "      <td>Schweizerische Bundesbahnen SBB</td>\n",
       "      <td>Zug</td>\n",
       "      <td>1093</td>\n",
       "      <td>IC6</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>IC</td>\n",
       "      <td>...</td>\n",
       "      <td>Brig</td>\n",
       "      <td>2023-10-03 23:44:00</td>\n",
       "      <td>2023-11-03 00:02:15</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>43218.250000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1974</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:11:14395:001</td>\n",
       "      <td>85:11</td>\n",
       "      <td>SBB</td>\n",
       "      <td>Schweizerische Bundesbahnen SBB</td>\n",
       "      <td>Zug</td>\n",
       "      <td>14395</td>\n",
       "      <td>S30</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Cheyres</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:01:20</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:01:57</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43202.333333</td>\n",
       "      <td>43202.950000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1991</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:11:14397:001</td>\n",
       "      <td>85:11</td>\n",
       "      <td>SBB</td>\n",
       "      <td>Schweizerische Bundesbahnen SBB</td>\n",
       "      <td>Zug</td>\n",
       "      <td>14397</td>\n",
       "      <td>S30</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Corcelles-Sud</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:01:29</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:04:28</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43202.483333</td>\n",
       "      <td>43205.466667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2072</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:11:14492:001</td>\n",
       "      <td>85:11</td>\n",
       "      <td>SBB</td>\n",
       "      <td>Schweizerische Bundesbahnen SBB</td>\n",
       "      <td>Zug</td>\n",
       "      <td>14492</td>\n",
       "      <td>S4</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Vallorbe</td>\n",
       "      <td>2023-10-03 23:49:00</td>\n",
       "      <td>2023-11-03 00:05:52</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>43216.866667</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2461</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:11:1545:001</td>\n",
       "      <td>85:11</td>\n",
       "      <td>SBB</td>\n",
       "      <td>Schweizerische Bundesbahnen SBB</td>\n",
       "      <td>Zug</td>\n",
       "      <td>1545</td>\n",
       "      <td>IC</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>IC</td>\n",
       "      <td>...</td>\n",
       "      <td>Neuchâtel</td>\n",
       "      <td>2023-10-03 23:56:00</td>\n",
       "      <td>2023-11-03 00:01:18</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-10-03 23:58:00</td>\n",
       "      <td>2023-11-03 00:02:38</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43205.300000</td>\n",
       "      <td>43204.633333</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>2338619</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:955:00041010122</td>\n",
       "      <td>85:955</td>\n",
       "      <td>TPL</td>\n",
       "      <td>Trasporti Pubblici Luganesi SA</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:955:410</td>\n",
       "      <td>10</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Albonago, Sentiero di Camà</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:00:10</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:00:18</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43201.166667</td>\n",
       "      <td>43201.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2339922</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:955:00041212092</td>\n",
       "      <td>85:955</td>\n",
       "      <td>TPL</td>\n",
       "      <td>Trasporti Pubblici Luganesi SA</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:955:412</td>\n",
       "      <td>12</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Castagnola, Roncaccio</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:00:02</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:00:12</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43201.033333</td>\n",
       "      <td>43201.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2343121</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:96:173:000</td>\n",
       "      <td>85:96</td>\n",
       "      <td>AVA-wsb</td>\n",
       "      <td>Aargau Verkehr AG</td>\n",
       "      <td>Zug</td>\n",
       "      <td>173</td>\n",
       "      <td>S14</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Unterkulm</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:00:13</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-10-03 23:59:00</td>\n",
       "      <td>2023-11-03 00:03:30</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>43201.216667</td>\n",
       "      <td>43204.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2347126</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:97:78:000</td>\n",
       "      <td>85:97</td>\n",
       "      <td>TRAVYS</td>\n",
       "      <td>Transports Vallée de Joux-Yverdon-Ste-Croix</td>\n",
       "      <td>Zug</td>\n",
       "      <td>78</td>\n",
       "      <td>R</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>R</td>\n",
       "      <td>...</td>\n",
       "      <td>Essert-sous-Champvent</td>\n",
       "      <td>2023-10-03 23:57:00</td>\n",
       "      <td>2023-11-03 00:01:12</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-10-03 23:57:00</td>\n",
       "      <td>2023-11-03 00:01:12</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>43204.200000</td>\n",
       "      <td>43204.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2347145</th>\n",
       "      <td>2023-10-03</td>\n",
       "      <td>85:97:81:000</td>\n",
       "      <td>85:97</td>\n",
       "      <td>TRAVYS</td>\n",
       "      <td>Transports Vallée de Joux-Yverdon-Ste-Croix</td>\n",
       "      <td>Zug</td>\n",
       "      <td>81</td>\n",
       "      <td>R</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>R</td>\n",
       "      <td>...</td>\n",
       "      <td>Baulmes</td>\n",
       "      <td>2023-10-03 23:57:00</td>\n",
       "      <td>2023-11-03 00:01:05</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-10-03 23:57:00</td>\n",
       "      <td>2023-11-03 00:02:18</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>43204.083333</td>\n",
       "      <td>43205.300000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>814 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              date             trip_id operator_id operator_abbreviation  \\\n",
       "902     2023-10-03      85:11:1093:001       85:11                   SBB   \n",
       "1974    2023-10-03     85:11:14395:001       85:11                   SBB   \n",
       "1991    2023-10-03     85:11:14397:001       85:11                   SBB   \n",
       "2072    2023-10-03     85:11:14492:001       85:11                   SBB   \n",
       "2461    2023-10-03      85:11:1545:001       85:11                   SBB   \n",
       "...            ...                 ...         ...                   ...   \n",
       "2338619 2023-10-03  85:955:00041010122      85:955                   TPL   \n",
       "2339922 2023-10-03  85:955:00041212092      85:955                   TPL   \n",
       "2343121 2023-10-03       85:96:173:000       85:96               AVA-wsb   \n",
       "2347126 2023-10-03        85:97:78:000       85:97                TRAVYS   \n",
       "2347145 2023-10-03        85:97:81:000       85:97                TRAVYS   \n",
       "\n",
       "                                       operator_name product_id     line_id  \\\n",
       "902                  Schweizerische Bundesbahnen SBB        Zug        1093   \n",
       "1974                 Schweizerische Bundesbahnen SBB        Zug       14395   \n",
       "1991                 Schweizerische Bundesbahnen SBB        Zug       14397   \n",
       "2072                 Schweizerische Bundesbahnen SBB        Zug       14492   \n",
       "2461                 Schweizerische Bundesbahnen SBB        Zug        1545   \n",
       "...                                              ...        ...         ...   \n",
       "2338619               Trasporti Pubblici Luganesi SA        Bus  85:955:410   \n",
       "2339922               Trasporti Pubblici Luganesi SA        Bus  85:955:412   \n",
       "2343121                            Aargau Verkehr AG        Zug         173   \n",
       "2347126  Transports Vallée de Joux-Yverdon-Ste-Croix        Zug          78   \n",
       "2347145  Transports Vallée de Joux-Yverdon-Ste-Croix        Zug          81   \n",
       "\n",
       "        line_text circuit_id transport_type  ...                   stop_name  \\\n",
       "902           IC6       <NA>             IC  ...                        Brig   \n",
       "1974          S30       <NA>              S  ...                     Cheyres   \n",
       "1991          S30       <NA>              S  ...               Corcelles-Sud   \n",
       "2072           S4       <NA>              S  ...                    Vallorbe   \n",
       "2461           IC       <NA>             IC  ...                   Neuchâtel   \n",
       "...           ...        ...            ...  ...                         ...   \n",
       "2338619        10       <NA>              B  ...  Albonago, Sentiero di Camà   \n",
       "2339922        12       <NA>              B  ...       Castagnola, Roncaccio   \n",
       "2343121       S14       <NA>              S  ...                   Unterkulm   \n",
       "2347126         R       <NA>              R  ...       Essert-sous-Champvent   \n",
       "2347145         R       <NA>              R  ...                     Baulmes   \n",
       "\n",
       "               arrival_time    arrival_forecast arrival_forecast_status  \\\n",
       "902     2023-10-03 23:44:00 2023-11-03 00:02:15                    REAL   \n",
       "1974    2023-10-03 23:59:00 2023-11-03 00:01:20                    REAL   \n",
       "1991    2023-10-03 23:59:00 2023-11-03 00:01:29                PROGNOSE   \n",
       "2072    2023-10-03 23:49:00 2023-11-03 00:05:52                    REAL   \n",
       "2461    2023-10-03 23:56:00 2023-11-03 00:01:18                    REAL   \n",
       "...                     ...                 ...                     ...   \n",
       "2338619 2023-10-03 23:59:00 2023-11-03 00:00:10                    REAL   \n",
       "2339922 2023-10-03 23:59:00 2023-11-03 00:00:02                    REAL   \n",
       "2343121 2023-10-03 23:59:00 2023-11-03 00:00:13                PROGNOSE   \n",
       "2347126 2023-10-03 23:57:00 2023-11-03 00:01:12                PROGNOSE   \n",
       "2347145 2023-10-03 23:57:00 2023-11-03 00:01:05                PROGNOSE   \n",
       "\n",
       "             departure_time  departure_forecast departure_forecast_status  \\\n",
       "902                     NaT                 NaT                       NaN   \n",
       "1974    2023-10-03 23:59:00 2023-11-03 00:01:57                      REAL   \n",
       "1991    2023-10-03 23:59:00 2023-11-03 00:04:28                      REAL   \n",
       "2072                    NaT                 NaT                       NaN   \n",
       "2461    2023-10-03 23:58:00 2023-11-03 00:02:38                      REAL   \n",
       "...                     ...                 ...                       ...   \n",
       "2338619 2023-10-03 23:59:00 2023-11-03 00:00:18                      REAL   \n",
       "2339922 2023-10-03 23:59:00 2023-11-03 00:00:12                      REAL   \n",
       "2343121 2023-10-03 23:59:00 2023-11-03 00:03:30                      REAL   \n",
       "2347126 2023-10-03 23:57:00 2023-11-03 00:01:12                  PROGNOSE   \n",
       "2347145 2023-10-03 23:57:00 2023-11-03 00:02:18                  PROGNOSE   \n",
       "\n",
       "        is_through_trip arrival_delay departure_delay  \n",
       "902               False  43218.250000             NaN  \n",
       "1974              False  43202.333333    43202.950000  \n",
       "1991              False  43202.483333    43205.466667  \n",
       "2072              False  43216.866667             NaN  \n",
       "2461              False  43205.300000    43204.633333  \n",
       "...                 ...           ...             ...  \n",
       "2338619           False  43201.166667    43201.300000  \n",
       "2339922           False  43201.033333    43201.200000  \n",
       "2343121           False  43201.216667    43204.500000  \n",
       "2347126           False  43204.200000    43204.200000  \n",
       "2347145           False  43204.083333    43205.300000  \n",
       "\n",
       "[814 rows x 23 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] > 10000]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that the problem comes from the fact that pandas reads the day in the \"american\" format rather than the ISO format..."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now after specifying day_first = True, we can see that the problem is gone."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] > 1000].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] < -1000].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "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>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "      <th>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>272887</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:22:2005:000</td>\n",
       "      <td>85:22</td>\n",
       "      <td>AB-ab</td>\n",
       "      <td>Appenzeller Bahnen (ab)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>2005</td>\n",
       "      <td>S21</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Gfeld</td>\n",
       "      <td>2023-03-11 00:31:00</td>\n",
       "      <td>2023-03-10 00:30:49</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:31:00</td>\n",
       "      <td>2023-03-10 00:31:22</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1440.183333</td>\n",
       "      <td>-1439.633333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272888</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:22:2005:000</td>\n",
       "      <td>85:22</td>\n",
       "      <td>AB-ab</td>\n",
       "      <td>Appenzeller Bahnen (ab)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>2005</td>\n",
       "      <td>S21</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Bendlehn</td>\n",
       "      <td>2023-03-11 00:32:00</td>\n",
       "      <td>2023-03-10 00:32:07</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:32:00</td>\n",
       "      <td>2023-03-10 00:32:43</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1439.883333</td>\n",
       "      <td>-1439.283333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272889</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:22:2005:000</td>\n",
       "      <td>85:22</td>\n",
       "      <td>AB-ab</td>\n",
       "      <td>Appenzeller Bahnen (ab)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>2005</td>\n",
       "      <td>S21</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Speicher</td>\n",
       "      <td>2023-03-11 00:35:00</td>\n",
       "      <td>2023-03-10 00:34:10</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-1440.833333</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466603</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Frauenfeld Marktplatz</td>\n",
       "      <td>2023-03-11 00:28:00</td>\n",
       "      <td>2023-03-10 00:28:22</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:28:00</td>\n",
       "      <td>2023-03-10 00:28:29</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1439.633333</td>\n",
       "      <td>-1439.516667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466604</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Lüdem</td>\n",
       "      <td>2023-03-11 00:30:00</td>\n",
       "      <td>2023-03-10 00:30:29</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:30:00</td>\n",
       "      <td>2023-03-10 00:30:29</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1439.516667</td>\n",
       "      <td>-1439.516667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466605</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Weberei Matzingen</td>\n",
       "      <td>2023-03-11 00:33:00</td>\n",
       "      <td>2023-03-10 00:34:11</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:33:00</td>\n",
       "      <td>2023-03-10 00:34:11</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1438.816667</td>\n",
       "      <td>-1438.816667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466606</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Matzingen</td>\n",
       "      <td>2023-03-11 00:36:00</td>\n",
       "      <td>2023-03-10 00:35:16</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:36:00</td>\n",
       "      <td>2023-03-10 00:36:13</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1440.733333</td>\n",
       "      <td>-1439.783333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466607</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Jakobstal</td>\n",
       "      <td>2023-03-11 00:38:00</td>\n",
       "      <td>2023-03-10 00:39:13</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:38:00</td>\n",
       "      <td>2023-03-10 00:39:13</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1438.783333</td>\n",
       "      <td>-1438.783333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466608</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Wiesengrund</td>\n",
       "      <td>2023-03-11 00:39:00</td>\n",
       "      <td>2023-03-10 00:40:13</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:39:00</td>\n",
       "      <td>2023-03-10 00:40:13</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1438.783333</td>\n",
       "      <td>-1438.783333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466609</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Wängi</td>\n",
       "      <td>2023-03-11 00:41:00</td>\n",
       "      <td>2023-03-10 00:40:46</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:41:00</td>\n",
       "      <td>2023-03-10 00:41:09</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1440.233333</td>\n",
       "      <td>-1439.850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466610</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Rosental</td>\n",
       "      <td>2023-03-11 00:43:00</td>\n",
       "      <td>2023-03-10 00:43:32</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:43:00</td>\n",
       "      <td>2023-03-10 00:44:01</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1439.466667</td>\n",
       "      <td>-1438.983333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466611</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Münchwilen Pflegeheim</td>\n",
       "      <td>2023-03-11 00:46:00</td>\n",
       "      <td>2023-03-10 00:46:24</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:46:00</td>\n",
       "      <td>2023-03-10 00:46:43</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1439.600000</td>\n",
       "      <td>-1439.283333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466612</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Münchwilen TG</td>\n",
       "      <td>2023-03-11 00:47:00</td>\n",
       "      <td>2023-03-10 00:48:43</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-11 00:47:00</td>\n",
       "      <td>2023-03-10 00:48:43</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-1438.283333</td>\n",
       "      <td>-1438.283333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466613</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:51:7004:000</td>\n",
       "      <td>85:51</td>\n",
       "      <td>AB-fw</td>\n",
       "      <td>Appenzeller Bahnen (fw)</td>\n",
       "      <td>Zug</td>\n",
       "      <td>7004</td>\n",
       "      <td>S15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>S</td>\n",
       "      <td>...</td>\n",
       "      <td>Wil SG</td>\n",
       "      <td>2023-03-11 00:55:00</td>\n",
       "      <td>2023-03-10 00:52:59</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-1442.016667</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             date         trip_id operator_id operator_abbreviation  \\\n",
       "272887 2023-03-10  85:22:2005:000       85:22                 AB-ab   \n",
       "272888 2023-03-10  85:22:2005:000       85:22                 AB-ab   \n",
       "272889 2023-03-10  85:22:2005:000       85:22                 AB-ab   \n",
       "466603 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466604 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466605 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466606 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466607 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466608 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466609 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466610 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466611 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466612 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "466613 2023-03-10  85:51:7004:000       85:51                 AB-fw   \n",
       "\n",
       "                  operator_name product_id line_id line_text circuit_id  \\\n",
       "272887  Appenzeller Bahnen (ab)        Zug    2005       S21       <NA>   \n",
       "272888  Appenzeller Bahnen (ab)        Zug    2005       S21       <NA>   \n",
       "272889  Appenzeller Bahnen (ab)        Zug    2005       S21       <NA>   \n",
       "466603  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466604  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466605  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466606  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466607  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466608  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466609  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466610  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466611  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466612  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "466613  Appenzeller Bahnen (fw)        Zug    7004       S15       <NA>   \n",
       "\n",
       "       transport_type  ...              stop_name        arrival_time  \\\n",
       "272887              S  ...                  Gfeld 2023-03-11 00:31:00   \n",
       "272888              S  ...               Bendlehn 2023-03-11 00:32:00   \n",
       "272889              S  ...               Speicher 2023-03-11 00:35:00   \n",
       "466603              S  ...  Frauenfeld Marktplatz 2023-03-11 00:28:00   \n",
       "466604              S  ...                  Lüdem 2023-03-11 00:30:00   \n",
       "466605              S  ...      Weberei Matzingen 2023-03-11 00:33:00   \n",
       "466606              S  ...              Matzingen 2023-03-11 00:36:00   \n",
       "466607              S  ...              Jakobstal 2023-03-11 00:38:00   \n",
       "466608              S  ...            Wiesengrund 2023-03-11 00:39:00   \n",
       "466609              S  ...                  Wängi 2023-03-11 00:41:00   \n",
       "466610              S  ...               Rosental 2023-03-11 00:43:00   \n",
       "466611              S  ...  Münchwilen Pflegeheim 2023-03-11 00:46:00   \n",
       "466612              S  ...          Münchwilen TG 2023-03-11 00:47:00   \n",
       "466613              S  ...                 Wil SG 2023-03-11 00:55:00   \n",
       "\n",
       "          arrival_forecast arrival_forecast_status      departure_time  \\\n",
       "272887 2023-03-10 00:30:49                    REAL 2023-03-11 00:31:00   \n",
       "272888 2023-03-10 00:32:07                    REAL 2023-03-11 00:32:00   \n",
       "272889 2023-03-10 00:34:10                    REAL                 NaT   \n",
       "466603 2023-03-10 00:28:22                    REAL 2023-03-11 00:28:00   \n",
       "466604 2023-03-10 00:30:29                    REAL 2023-03-11 00:30:00   \n",
       "466605 2023-03-10 00:34:11                    REAL 2023-03-11 00:33:00   \n",
       "466606 2023-03-10 00:35:16                    REAL 2023-03-11 00:36:00   \n",
       "466607 2023-03-10 00:39:13                    REAL 2023-03-11 00:38:00   \n",
       "466608 2023-03-10 00:40:13                    REAL 2023-03-11 00:39:00   \n",
       "466609 2023-03-10 00:40:46                    REAL 2023-03-11 00:41:00   \n",
       "466610 2023-03-10 00:43:32                    REAL 2023-03-11 00:43:00   \n",
       "466611 2023-03-10 00:46:24                    REAL 2023-03-11 00:46:00   \n",
       "466612 2023-03-10 00:48:43                    REAL 2023-03-11 00:47:00   \n",
       "466613 2023-03-10 00:52:59                    REAL                 NaT   \n",
       "\n",
       "        departure_forecast departure_forecast_status is_through_trip  \\\n",
       "272887 2023-03-10 00:31:22                      REAL           False   \n",
       "272888 2023-03-10 00:32:43                      REAL           False   \n",
       "272889                 NaT                       NaN           False   \n",
       "466603 2023-03-10 00:28:29                      REAL           False   \n",
       "466604 2023-03-10 00:30:29                      REAL           False   \n",
       "466605 2023-03-10 00:34:11                      REAL           False   \n",
       "466606 2023-03-10 00:36:13                      REAL           False   \n",
       "466607 2023-03-10 00:39:13                      REAL           False   \n",
       "466608 2023-03-10 00:40:13                      REAL           False   \n",
       "466609 2023-03-10 00:41:09                      REAL           False   \n",
       "466610 2023-03-10 00:44:01                      REAL           False   \n",
       "466611 2023-03-10 00:46:43                      REAL           False   \n",
       "466612 2023-03-10 00:48:43                      REAL           False   \n",
       "466613                 NaT                       NaN           False   \n",
       "\n",
       "       arrival_delay departure_delay  \n",
       "272887  -1440.183333    -1439.633333  \n",
       "272888  -1439.883333    -1439.283333  \n",
       "272889  -1440.833333             NaN  \n",
       "466603  -1439.633333    -1439.516667  \n",
       "466604  -1439.516667    -1439.516667  \n",
       "466605  -1438.816667    -1438.816667  \n",
       "466606  -1440.733333    -1439.783333  \n",
       "466607  -1438.783333    -1438.783333  \n",
       "466608  -1438.783333    -1438.783333  \n",
       "466609  -1440.233333    -1439.850000  \n",
       "466610  -1439.466667    -1438.983333  \n",
       "466611  -1439.600000    -1439.283333  \n",
       "466612  -1438.283333    -1438.283333  \n",
       "466613  -1442.016667             NaN  \n",
       "\n",
       "[14 rows x 23 columns]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] < -1000]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There seem to be a problem with the lines S21 and S15 from the Appenzeller Bahnen. The arrival/departure times and the forecasts are shifted by one day."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "      <th>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1004208</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:801:10659-704</td>\n",
       "      <td>85:801</td>\n",
       "      <td>PAG</td>\n",
       "      <td>PostAuto AG</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:801:704</td>\n",
       "      <td>106</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Bern, Obergericht</td>\n",
       "      <td>2023-03-10 18:39:00</td>\n",
       "      <td>2023-03-10 16:35:37</td>\n",
       "      <td>REAL</td>\n",
       "      <td>2023-03-10 18:39:00</td>\n",
       "      <td>2023-03-10 16:36:04</td>\n",
       "      <td>REAL</td>\n",
       "      <td>False</td>\n",
       "      <td>-123.383333</td>\n",
       "      <td>-122.933333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1004209</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:801:10659-704</td>\n",
       "      <td>85:801</td>\n",
       "      <td>PAG</td>\n",
       "      <td>PostAuto AG</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:801:704</td>\n",
       "      <td>106</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Bern, Mittelstrasse</td>\n",
       "      <td>2023-03-10 18:40:00</td>\n",
       "      <td>2023-03-10 16:37:21</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-122.650000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1008705</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:801:10570-709</td>\n",
       "      <td>85:801</td>\n",
       "      <td>PAG</td>\n",
       "      <td>PostAuto AG</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:801:709</td>\n",
       "      <td>105</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Bern, Hauptbahnhof</td>\n",
       "      <td>2023-03-10 18:28:00</td>\n",
       "      <td>2023-03-10 16:34:53</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-113.116667</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1027259</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:801:18059-903</td>\n",
       "      <td>85:801</td>\n",
       "      <td>PAG</td>\n",
       "      <td>PostAuto AG</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:801:903</td>\n",
       "      <td>180</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Les Diablerets, Les Preises</td>\n",
       "      <td>2023-03-10 14:26:00</td>\n",
       "      <td>2023-03-10 11:55:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 14:26:00</td>\n",
       "      <td>2023-03-10 14:26:00</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-150.316667</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1126838</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:816:166845-46011-1</td>\n",
       "      <td>85:816</td>\n",
       "      <td>AAGU</td>\n",
       "      <td>Auto AG Uri</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:816:407</td>\n",
       "      <td>407</td>\n",
       "      <td>364701</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Amsteg St. Anton</td>\n",
       "      <td>2023-03-10 15:04:00</td>\n",
       "      <td>2023-03-10 13:22:04</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>2023-03-10 15:04:00</td>\n",
       "      <td>2023-03-10 13:22:05</td>\n",
       "      <td>UNBEKANNT</td>\n",
       "      <td>False</td>\n",
       "      <td>-101.933333</td>\n",
       "      <td>-101.916667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1126839</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:816:166845-46011-1</td>\n",
       "      <td>85:816</td>\n",
       "      <td>AAGU</td>\n",
       "      <td>Auto AG Uri</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:816:407</td>\n",
       "      <td>407</td>\n",
       "      <td>364701</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>Amsteg Post</td>\n",
       "      <td>2023-03-10 15:12:00</td>\n",
       "      <td>2023-03-10 13:22:06</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-109.900000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1651761</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:841:506-6906</td>\n",
       "      <td>85:841</td>\n",
       "      <td>AAGS</td>\n",
       "      <td>Auto AG Schwyz</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:841:506</td>\n",
       "      <td>506</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>2023-03-10 15:58:00</td>\n",
       "      <td>2023-03-10 09:58:28</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-359.533333</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1651776</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:841:506-6917</td>\n",
       "      <td>85:841</td>\n",
       "      <td>AAGS</td>\n",
       "      <td>Auto AG Schwyz</td>\n",
       "      <td>Bus</td>\n",
       "      <td>85:841:506</td>\n",
       "      <td>506</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>B</td>\n",
       "      <td>...</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>2023-03-10 16:17:00</td>\n",
       "      <td>2023-03-10 10:18:59</td>\n",
       "      <td>REAL</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-358.016667</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010822</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Collège Sismondi</td>\n",
       "      <td>2023-03-10 17:46:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:46:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-709.316667</td>\n",
       "      <td>-709.316667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010823</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Maison de la Paix</td>\n",
       "      <td>2023-03-10 17:47:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:47:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-710.316667</td>\n",
       "      <td>-710.316667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010824</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Butini</td>\n",
       "      <td>2023-03-10 17:50:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:50:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-713.316667</td>\n",
       "      <td>-713.316667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010825</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Môle</td>\n",
       "      <td>2023-03-10 17:51:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:51:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-714.316667</td>\n",
       "      <td>-714.316667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010826</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, gare Cornavin</td>\n",
       "      <td>2023-03-10 17:53:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:53:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-716.316667</td>\n",
       "      <td>-716.316667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010827</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Mercier</td>\n",
       "      <td>2023-03-10 17:56:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:56:00</td>\n",
       "      <td>2023-03-10 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>-719.316667</td>\n",
       "      <td>-719.316667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>14 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              date                         trip_id operator_id  \\\n",
       "1004208 2023-03-10                85:801:10659-704      85:801   \n",
       "1004209 2023-03-10                85:801:10659-704      85:801   \n",
       "1008705 2023-03-10                85:801:10570-709      85:801   \n",
       "1027259 2023-03-10                85:801:18059-903      85:801   \n",
       "1126838 2023-03-10           85:816:166845-46011-1      85:816   \n",
       "1126839 2023-03-10           85:816:166845-46011-1      85:816   \n",
       "1651761 2023-03-10                 85:841:506-6906      85:841   \n",
       "1651776 2023-03-10                 85:841:506-6917      85:841   \n",
       "2010822 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010823 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010824 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010825 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010826 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010827 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "\n",
       "        operator_abbreviation                operator_name product_id  \\\n",
       "1004208                   PAG                  PostAuto AG        Bus   \n",
       "1004209                   PAG                  PostAuto AG        Bus   \n",
       "1008705                   PAG                  PostAuto AG        Bus   \n",
       "1027259                   PAG                  PostAuto AG        Bus   \n",
       "1126838                  AAGU                  Auto AG Uri        Bus   \n",
       "1126839                  AAGU                  Auto AG Uri        Bus   \n",
       "1651761                  AAGS               Auto AG Schwyz        Bus   \n",
       "1651776                  AAGS               Auto AG Schwyz        Bus   \n",
       "2010822                   TPG  Transports Publics Genevois       Tram   \n",
       "2010823                   TPG  Transports Publics Genevois       Tram   \n",
       "2010824                   TPG  Transports Publics Genevois       Tram   \n",
       "2010825                   TPG  Transports Publics Genevois       Tram   \n",
       "2010826                   TPG  Transports Publics Genevois       Tram   \n",
       "2010827                   TPG  Transports Publics Genevois       Tram   \n",
       "\n",
       "               line_id line_text circuit_id transport_type  ...  \\\n",
       "1004208     85:801:704       106       <NA>              B  ...   \n",
       "1004209     85:801:704       106       <NA>              B  ...   \n",
       "1008705     85:801:709       105       <NA>              B  ...   \n",
       "1027259     85:801:903       180       <NA>              B  ...   \n",
       "1126838     85:816:407       407     364701              B  ...   \n",
       "1126839     85:816:407       407     364701              B  ...   \n",
       "1651761     85:841:506       506       <NA>              B  ...   \n",
       "1651776     85:841:506       506       <NA>              B  ...   \n",
       "2010822  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010823  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010824  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010825  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010826  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010827  85:881:TPG015        15       <NA>              T  ...   \n",
       "\n",
       "                           stop_name        arrival_time    arrival_forecast  \\\n",
       "1004208            Bern, Obergericht 2023-03-10 18:39:00 2023-03-10 16:35:37   \n",
       "1004209          Bern, Mittelstrasse 2023-03-10 18:40:00 2023-03-10 16:37:21   \n",
       "1008705           Bern, Hauptbahnhof 2023-03-10 18:28:00 2023-03-10 16:34:53   \n",
       "1027259  Les Diablerets, Les Preises 2023-03-10 14:26:00 2023-03-10 11:55:41   \n",
       "1126838             Amsteg St. Anton 2023-03-10 15:04:00 2023-03-10 13:22:04   \n",
       "1126839                  Amsteg Post 2023-03-10 15:12:00 2023-03-10 13:22:06   \n",
       "1651761                         <NA> 2023-03-10 15:58:00 2023-03-10 09:58:28   \n",
       "1651776                         <NA> 2023-03-10 16:17:00 2023-03-10 10:18:59   \n",
       "2010822     Genève, Collège Sismondi 2023-03-10 17:46:00 2023-03-10 05:56:41   \n",
       "2010823    Genève, Maison de la Paix 2023-03-10 17:47:00 2023-03-10 05:56:41   \n",
       "2010824               Genève, Butini 2023-03-10 17:50:00 2023-03-10 05:56:41   \n",
       "2010825                 Genève, Môle 2023-03-10 17:51:00 2023-03-10 05:56:41   \n",
       "2010826        Genève, gare Cornavin 2023-03-10 17:53:00 2023-03-10 05:56:41   \n",
       "2010827              Genève, Mercier 2023-03-10 17:56:00 2023-03-10 05:56:41   \n",
       "\n",
       "        arrival_forecast_status      departure_time  departure_forecast  \\\n",
       "1004208                    REAL 2023-03-10 18:39:00 2023-03-10 16:36:04   \n",
       "1004209                    REAL                 NaT                 NaT   \n",
       "1008705                    REAL                 NaT                 NaT   \n",
       "1027259                PROGNOSE 2023-03-10 14:26:00 2023-03-10 14:26:00   \n",
       "1126838               UNBEKANNT 2023-03-10 15:04:00 2023-03-10 13:22:05   \n",
       "1126839                    REAL                 NaT                 NaT   \n",
       "1651761                    REAL                 NaT                 NaT   \n",
       "1651776                    REAL                 NaT                 NaT   \n",
       "2010822                PROGNOSE 2023-03-10 17:46:00 2023-03-10 05:56:41   \n",
       "2010823                PROGNOSE 2023-03-10 17:47:00 2023-03-10 05:56:41   \n",
       "2010824                PROGNOSE 2023-03-10 17:50:00 2023-03-10 05:56:41   \n",
       "2010825                PROGNOSE 2023-03-10 17:51:00 2023-03-10 05:56:41   \n",
       "2010826                PROGNOSE 2023-03-10 17:53:00 2023-03-10 05:56:41   \n",
       "2010827                PROGNOSE 2023-03-10 17:56:00 2023-03-10 05:56:41   \n",
       "\n",
       "        departure_forecast_status is_through_trip arrival_delay  \\\n",
       "1004208                      REAL           False   -123.383333   \n",
       "1004209                       NaN           False   -122.650000   \n",
       "1008705                       NaN           False   -113.116667   \n",
       "1027259                  PROGNOSE           False   -150.316667   \n",
       "1126838                 UNBEKANNT           False   -101.933333   \n",
       "1126839                       NaN           False   -109.900000   \n",
       "1651761                       NaN           False   -359.533333   \n",
       "1651776                       NaN           False   -358.016667   \n",
       "2010822                  PROGNOSE           False   -709.316667   \n",
       "2010823                  PROGNOSE           False   -710.316667   \n",
       "2010824                  PROGNOSE           False   -713.316667   \n",
       "2010825                  PROGNOSE           False   -714.316667   \n",
       "2010826                  PROGNOSE           False   -716.316667   \n",
       "2010827                  PROGNOSE           False   -719.316667   \n",
       "\n",
       "        departure_delay  \n",
       "1004208     -122.933333  \n",
       "1004209             NaN  \n",
       "1008705             NaN  \n",
       "1027259        0.000000  \n",
       "1126838     -101.916667  \n",
       "1126839             NaN  \n",
       "1651761             NaN  \n",
       "1651776             NaN  \n",
       "2010822     -709.316667  \n",
       "2010823     -710.316667  \n",
       "2010824     -713.316667  \n",
       "2010825     -714.316667  \n",
       "2010826     -716.316667  \n",
       "2010827     -719.316667  \n",
       "\n",
       "[14 rows x 23 columns]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[(transport_data[\"arrival_delay\"] < -100) & (transport_data[\"arrival_delay\"] > -1000)]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also see some transports that arrived or departed more than 100 minutes (less than 1000 minutes) before the scheduled time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "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>date</th>\n",
       "      <th>trip_id</th>\n",
       "      <th>operator_id</th>\n",
       "      <th>operator_abbreviation</th>\n",
       "      <th>operator_name</th>\n",
       "      <th>product_id</th>\n",
       "      <th>line_id</th>\n",
       "      <th>line_text</th>\n",
       "      <th>circuit_id</th>\n",
       "      <th>transport_type</th>\n",
       "      <th>...</th>\n",
       "      <th>stop_name</th>\n",
       "      <th>arrival_time</th>\n",
       "      <th>arrival_forecast</th>\n",
       "      <th>arrival_forecast_status</th>\n",
       "      <th>departure_time</th>\n",
       "      <th>departure_forecast</th>\n",
       "      <th>departure_forecast_status</th>\n",
       "      <th>is_through_trip</th>\n",
       "      <th>arrival_delay</th>\n",
       "      <th>departure_delay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010828</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Stand</td>\n",
       "      <td>2023-03-10 17:58:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 17:58:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>718.683333</td>\n",
       "      <td>718.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010829</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Cirque</td>\n",
       "      <td>2023-03-10 18:00:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:00:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>716.683333</td>\n",
       "      <td>716.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010830</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Plainpalais</td>\n",
       "      <td>2023-03-10 18:02:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:02:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>714.683333</td>\n",
       "      <td>714.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010831</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Uni-Mail</td>\n",
       "      <td>2023-03-10 18:03:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:03:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>713.683333</td>\n",
       "      <td>713.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010832</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Acacias</td>\n",
       "      <td>2023-03-10 18:05:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:05:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>711.683333</td>\n",
       "      <td>711.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010833</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Genève, Industrielle</td>\n",
       "      <td>2023-03-10 18:07:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:07:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>709.683333</td>\n",
       "      <td>709.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010834</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Carouge GE, Pictet-Thellusson</td>\n",
       "      <td>2023-03-10 18:08:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:08:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>708.683333</td>\n",
       "      <td>708.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010835</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Lancy-Pont-Rouge, gare/Etoile</td>\n",
       "      <td>2023-03-10 18:10:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:10:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>706.683333</td>\n",
       "      <td>706.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010836</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Grand-Lancy, Mairie de Lancy</td>\n",
       "      <td>2023-03-10 18:12:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:12:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>704.683333</td>\n",
       "      <td>704.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010837</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Grand-Lancy, Place du 1er-Août</td>\n",
       "      <td>2023-03-10 18:13:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:13:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>703.683333</td>\n",
       "      <td>703.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010838</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Grand-Lancy, Lancy Piscine</td>\n",
       "      <td>2023-03-10 18:14:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>2023-03-10 18:14:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>False</td>\n",
       "      <td>702.683333</td>\n",
       "      <td>702.683333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010839</th>\n",
       "      <td>2023-03-10</td>\n",
       "      <td>85:881:TPG015-4503784315733856</td>\n",
       "      <td>85:881</td>\n",
       "      <td>TPG</td>\n",
       "      <td>Transports Publics Genevois</td>\n",
       "      <td>Tram</td>\n",
       "      <td>85:881:TPG015</td>\n",
       "      <td>15</td>\n",
       "      <td>&lt;NA&gt;</td>\n",
       "      <td>T</td>\n",
       "      <td>...</td>\n",
       "      <td>Grand-Lancy, Palettes</td>\n",
       "      <td>2023-03-10 18:15:00</td>\n",
       "      <td>2023-03-11 05:56:41</td>\n",
       "      <td>PROGNOSE</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>701.683333</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              date                         trip_id operator_id  \\\n",
       "2010828 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010829 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010830 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010831 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010832 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010833 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010834 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010835 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010836 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010837 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010838 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "2010839 2023-03-10  85:881:TPG015-4503784315733856      85:881   \n",
       "\n",
       "        operator_abbreviation                operator_name product_id  \\\n",
       "2010828                   TPG  Transports Publics Genevois       Tram   \n",
       "2010829                   TPG  Transports Publics Genevois       Tram   \n",
       "2010830                   TPG  Transports Publics Genevois       Tram   \n",
       "2010831                   TPG  Transports Publics Genevois       Tram   \n",
       "2010832                   TPG  Transports Publics Genevois       Tram   \n",
       "2010833                   TPG  Transports Publics Genevois       Tram   \n",
       "2010834                   TPG  Transports Publics Genevois       Tram   \n",
       "2010835                   TPG  Transports Publics Genevois       Tram   \n",
       "2010836                   TPG  Transports Publics Genevois       Tram   \n",
       "2010837                   TPG  Transports Publics Genevois       Tram   \n",
       "2010838                   TPG  Transports Publics Genevois       Tram   \n",
       "2010839                   TPG  Transports Publics Genevois       Tram   \n",
       "\n",
       "               line_id line_text circuit_id transport_type  ...  \\\n",
       "2010828  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010829  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010830  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010831  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010832  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010833  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010834  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010835  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010836  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010837  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010838  85:881:TPG015        15       <NA>              T  ...   \n",
       "2010839  85:881:TPG015        15       <NA>              T  ...   \n",
       "\n",
       "                              stop_name        arrival_time  \\\n",
       "2010828                   Genève, Stand 2023-03-10 17:58:00   \n",
       "2010829                  Genève, Cirque 2023-03-10 18:00:00   \n",
       "2010830             Genève, Plainpalais 2023-03-10 18:02:00   \n",
       "2010831                Genève, Uni-Mail 2023-03-10 18:03:00   \n",
       "2010832                 Genève, Acacias 2023-03-10 18:05:00   \n",
       "2010833            Genève, Industrielle 2023-03-10 18:07:00   \n",
       "2010834   Carouge GE, Pictet-Thellusson 2023-03-10 18:08:00   \n",
       "2010835   Lancy-Pont-Rouge, gare/Etoile 2023-03-10 18:10:00   \n",
       "2010836    Grand-Lancy, Mairie de Lancy 2023-03-10 18:12:00   \n",
       "2010837  Grand-Lancy, Place du 1er-Août 2023-03-10 18:13:00   \n",
       "2010838      Grand-Lancy, Lancy Piscine 2023-03-10 18:14:00   \n",
       "2010839           Grand-Lancy, Palettes 2023-03-10 18:15:00   \n",
       "\n",
       "           arrival_forecast arrival_forecast_status      departure_time  \\\n",
       "2010828 2023-03-11 05:56:41                PROGNOSE 2023-03-10 17:58:00   \n",
       "2010829 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:00:00   \n",
       "2010830 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:02:00   \n",
       "2010831 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:03:00   \n",
       "2010832 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:05:00   \n",
       "2010833 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:07:00   \n",
       "2010834 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:08:00   \n",
       "2010835 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:10:00   \n",
       "2010836 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:12:00   \n",
       "2010837 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:13:00   \n",
       "2010838 2023-03-11 05:56:41                PROGNOSE 2023-03-10 18:14:00   \n",
       "2010839 2023-03-11 05:56:41                PROGNOSE                 NaT   \n",
       "\n",
       "         departure_forecast departure_forecast_status is_through_trip  \\\n",
       "2010828 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010829 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010830 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010831 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010832 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010833 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010834 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010835 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010836 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010837 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010838 2023-03-11 05:56:41                  PROGNOSE           False   \n",
       "2010839                 NaT                       NaN           False   \n",
       "\n",
       "        arrival_delay departure_delay  \n",
       "2010828    718.683333      718.683333  \n",
       "2010829    716.683333      716.683333  \n",
       "2010830    714.683333      714.683333  \n",
       "2010831    713.683333      713.683333  \n",
       "2010832    711.683333      711.683333  \n",
       "2010833    709.683333      709.683333  \n",
       "2010834    708.683333      708.683333  \n",
       "2010835    706.683333      706.683333  \n",
       "2010836    704.683333      704.683333  \n",
       "2010837    703.683333      703.683333  \n",
       "2010838    702.683333      702.683333  \n",
       "2010839    701.683333             NaN  \n",
       "\n",
       "[12 rows x 23 columns]"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transport_data[transport_data[\"arrival_delay\"] > 700]"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This tram seem to have delayed 11 hours wich seems a bit weird and might be an error or has default values for the arrival/departure times."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABlgAAANBCAYAAACMEbO5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAABZv0lEQVR4nOzdf5xVBZ0//vfwa8AfjCCGDiLSluYNRcVbwYaCJTYmplSLZYIFFcs1M1pN1spiTba1iE950WxNst1W+qFubWw2mj8o1B0QSh0sKViMH4GEjlACDuf7R19mnWBGzjAz587c5/PxmMdj7znnnvO+zJF47evccyqSJEkCAAAAAACAA9Yj6wEAAAAAAAC6GgULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkFKvrAfI2p49e2LDhg1x+OGHR0VFRdbjAABAh0qSJF588cWorq6OHj1cb8Wrk5kAACg3B5qbyr5g2bBhQwwdOjTrMQAAoFM9++yzceyxx2Y9Bl2AzAQAQLl6tdxU9gXL4YcfHhF/+YPq379/xtMAAEDHamhoiKFDhzb9OxhejcwEAEC5OdDcVPYFy96vuPfv319YAACgbLjVEwdKZgIAoFy9Wm5y02UAAAAAAICUyrZgKRaLkcvlIp/PZz0KAAAAAADQxZRtwVIoFKK+vj7q6uqyHgUAAKDkuCgNAABaV7YFCwAAAC1zURoAALROwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASmVbsBSLxcjlcpHP57MeBQAAoOTITAAA0LqKJEmSrIfIUkNDQ1RVVcULL7wQ/fv3z3ocAADoUP79S1rOGQAAys2B/hu4bL/BAgAAAAAA0FYKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQUq+sBwAASsvkKdNiw5Zt+11XfdSAWHTHbZ08EQAAQGlpKTfJTFBeFCwAQDMbtmyLIRddtd916+++sZOnAQAAKD0t5SaZCcqLW4QBAAAAAACkVLYFS7FYjFwuF/l8PutRAAAASo7MBAAArSvbgqVQKER9fX3U1dVlPQoAAEDJkZkAAKB1ZVuwAAAAAAAAtJWCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkFK3KFjWrFkT48ePj1wuFyeffHLs2LEj65EAAABKhswEAADtr1fWA7SHyy67LK6//voYO3Zs/PGPf4zKysqsRwIAACgZMhMAALS/Ll+wPPXUU9G7d+8YO3ZsREQMHDgw44kAAABKh8wEAAAdI/NbhD388MMxceLEqK6ujoqKirjnnnv22WbBggUxfPjw6Nu3b4waNSqWLFnStO6ZZ56Jww47LC644II4/fTT44YbbujE6QEAADqWzAQAAKUp84Jlx44dMXLkyLjpppv2u37RokVx5ZVXxrXXXhsrVqyIsWPHRk1NTaxbty4iInbv3h1LliyJYrEYjzzySNTW1kZtbW1nfgQAAIAOIzMBAEBpyrxgqampieuvvz4mTZq03/Xz5s2LadOmxfTp0+Okk06K+fPnx9ChQ+Pmm2+OiIhjjz028vl8DB06NCorK+O8886LlStXtni8nTt3RkNDQ7MfAACAUiUzAQBAacq8YGnNrl27Yvny5TFhwoRmyydMmBBLly6NiIh8Ph9/+MMfYtu2bbFnz554+OGH46STTmpxn3Pnzo2qqqqmn6FDh3boZwAAAOgoMhMAAGSnpAuW5557LhobG2Pw4MHNlg8ePDg2bdoUERG9evWKG264Ic4888w45ZRT4vWvf32cf/75Le5z9uzZ8cILLzT9PPvssx36GQAAADqKzAQAANnplfUAB6KioqLZ6yRJmi2rqamJmpqaA9pXZWVlVFZWtut8AAAAWZKZAACg85X0N1gGDRoUPXv2bLryaq/Nmzfvc4UWAABAuZGZAAAgOyVdsPTp0ydGjRoVtbW1zZbX1tbGmDFjMpoKAACgNMhMAACQncxvEbZ9+/ZYvXp10+s1a9bEypUrY+DAgXHcccfFrFmz4tJLL40zzjgjRo8eHbfeemusW7cuZsyYcVDHLRaLUSwWo7Gx8WA/AgAAQIeRmQAAoDRlXrAsW7Ysxo8f3/R61qxZERExderUWLhwYUyePDm2bt0ac+bMiY0bN8aIESNi8eLFMWzYsIM6bqFQiEKhEA0NDVFVVXVQ+wIAAOgoMhMAAJSmzAuWcePGRZIkrW4zc+bMmDlzZidNBAAAUDpkJgAAKE0l/QwWAAAAAACAUqRgAQAAAAAASEnBAgAAAAAAkFLZFizFYjFyuVzk8/msRwEAACg5MhMAALSubAuWQqEQ9fX1UVdXl/UoAAAAJUdmAgCA1pVtwQIAAAAAANBWChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSKtuCpVgsRi6Xi3w+n/UoAAAAJUdmAgCA1pVtwVIoFKK+vj7q6uqyHgUAAKDkyEwAANC6si1YAAAAAAAA2krBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJTKtmApFouRy+Uin89nPQoAAAAAANDFlG3BUigUor6+Purq6rIeBQAAoOS4KA0AAFpXtgULAAAALXNRGgAAtE7BAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFIq24KlWCxGLpeLfD6f9SgAAAAlR2YCAIDWlW3BUigUor6+Purq6rIeBQAAoOTITAAA0LqyLVgAAAAAAADaSsECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASKlsC5ZisRi5XC7y+XzWowAAAJQcmQkAAFpXtgVLoVCI+vr6qKury3oUAACAkiMzAQBA68q2YAEAAAAAAGgrBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKZVtwVIsFiOXy0U+n896FAAAgJIjMwEAQOvKtmApFApRX18fdXV1WY8CAABQcmQmAABoXdkWLAAAAAAAAG2lYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApFS2BUuxWIxcLhf5fD7rUQAAAEqOzAQAAK0r24KlUChEfX191NXVZT0KAABAyZGZAACgdWVbsAAAAAAAALSVggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACClblGw9OrVK0499dQ49dRTY/r06VmPAwAAUFJkJgAAaH+9sh6gPRxxxBGxcuXKrMcAAAAoSTITAAC0v27xDRYAAAAAAIDOlHnB8vDDD8fEiROjuro6Kioq4p577tlnmwULFsTw4cOjb9++MWrUqFiyZEmz9Q0NDTFq1Kh461vfGg899FAnTQ4AANDxZCYAAChNmRcsO3bsiJEjR8ZNN9203/WLFi2KK6+8Mq699tpYsWJFjB07NmpqamLdunVN26xduzaWL18et9xyS0yZMiUaGhpaPN7OnTujoaGh2Q8AAECp6uzMBAAAHJjMC5aampq4/vrrY9KkSftdP2/evJg2bVpMnz49TjrppJg/f34MHTo0br755qZtqqurIyJixIgRkcvl4je/+U2Lx5s7d25UVVU1/QwdOrR9PxAAAEA76uzM5KI0AAA4MJkXLK3ZtWtXLF++PCZMmNBs+YQJE2Lp0qUREbFt27bYuXNnRET8/ve/j/r6+njta1/b4j5nz54dL7zwQtPPs88+23EfAAAAoAN1RGZyURoAAByYXlkP0JrnnnsuGhsbY/Dgwc2WDx48ODZt2hQREatWrYqPfvSj0aNHj6ioqIj/9//+XwwcOLDFfVZWVkZlZWWHzg0AANAZOiIzzZ49O2bNmtX0uqGhQckCAAD7UdIFy14VFRXNXidJ0rRszJgx8cQTT2QxFgAAQEloz8zkojQAADgwJX2LsEGDBkXPnj2brrzaa/PmzftcoQUAAFBuZCYAAMhOSRcsffr0iVGjRkVtbW2z5bW1tTFmzJiMpgIAACgNMhMAAGQn81uEbd++PVavXt30es2aNbFy5coYOHBgHHfccTFr1qy49NJL44wzzojRo0fHrbfeGuvWrYsZM2Yc1HGLxWIUi8VobGw82I8AAADQYWQmAAAoTZkXLMuWLYvx48c3vd77MMWpU6fGwoULY/LkybF169aYM2dObNy4MUaMGBGLFy+OYcOGHdRxC4VCFAqFaGhoiKqqqoPaFwAAQEeRmQAAoDRlXrCMGzcukiRpdZuZM2fGzJkzO2kiAACA0iEzAQBAaSrpZ7AAAAAAAACUIgULAAAAAABASmVbsBSLxcjlcpHP57MeBQAAoOTITAAA0LqyLVgKhULU19dHXV1d1qMAAACUHJkJAABaV7YFCwAAAAAAQFspWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJBS2RYsxWIxcrlc5PP5rEcBAAAoOTITAAC0rmwLlkKhEPX19VFXV5f1KAAAACVHZgIAgNaVbcECAAAAAADQVgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApFS2BUuxWIxcLhf5fD7rUQAAAEqOzAQAAK0r24KlUChEfX191NXVZT0KAABAyZGZAACgdWVbsAAAAAAAALSVggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkVLYFS7FYjFwuF/l8PutRAAAASo7MBAAAreuV9QBZKRQKUSgUoqGhIaqqqrIeBwC6hN88vSrG1kzaZ3n1UQNi0R23ZTARAB1FZgIAgNaVbcECAKS3O+kRQy66ap/l6+++MYNpAAAAALJTtrcIAwAAAAAAaCsFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJBS2RYsxWIxcrlc5PP5rEcBAAAoOTITAAC0rmwLlkKhEPX19VFXV5f1KAAAACVHZgIAgNaVbcECAAAAAADQVgoWAAAAAACAlBQsAAAAAAAAKfXKegAAAAAAgO7gN0+virE1k/a7rvqoAbHojts6eSKgIylYAAAAAADawe6kRwy56Kr9rlt/942dPA3Q0dwiDAAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKRUtgVLsViMXC4X+Xw+61EAAAAAAIAupmwLlkKhEPX19VFXV5f1KAAAACXHRWkAANC6si1YAAAAaJmL0gAAoHUKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlMq2YCkWi5HL5SKfz2c9CgAAQMmRmQAAoHVlW7AUCoWor6+Purq6rEcBAAAoOTITAAC0rmwLFgAAAAAAgLZSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFLqNgXLn/70pxg2bFj8wz/8Q9ajAAAAAAAA3Vy3KVi+8IUvxJvf/OasxwAAAChJLkoDAID21S0KlmeeeSaefvrpOO+887IeBQAAoCS5KA0AANpX5gXLww8/HBMnTozq6uqoqKiIe+65Z59tFixYEMOHD4++ffvGqFGjYsmSJc3W/8M//EPMnTu3kyYGAADoWlyUBgAA7S/zgmXHjh0xcuTIuOmmm/a7ftGiRXHllVfGtddeGytWrIixY8dGTU1NrFu3LiIi/vM//zNOOOGEOOGEEzpzbAAAgE7hojQAAChNvbIeoKamJmpqalpcP2/evJg2bVpMnz49IiLmz58f9957b9x8880xd+7cePTRR+POO++M733ve7F9+/bYvXt39O/fPz772c/ud387d+6MnTt3Nr1uaGho3w8EAADQjvZelPbBD34w3v3ud++zfu9FaQsWLIi//du/ja9//etRU1MT9fX1cdxxxzW7KG3p0qWvejyZCQAADkzmBUtrdu3aFcuXL49rrrmm2fIJEyY0BYO5c+c2XYm1cOHCePLJJ1ssV/Zu//nPf77jhgYAAGhHnX1RmswEAAAHJvNbhLXmueeei8bGxhg8eHCz5YMHD45Nmza1aZ+zZ8+OF154oenn2WefbY9RAQAAOt3ei9ImTJjQbPlfX5T27LPPxtq1a+NLX/pSfPjDH271ojSZCQAADkxJf4Nlr4qKimavkyTZZ1lExGWXXfaq+6qsrIzKysr2Gg0AACAzHXFRmswEAAAHpqQLlkGDBkXPnj33CQabN2/eJ0AAAACUq/a8KA0AADgwJX2LsD59+sSoUaOitra22fLa2toYM2ZMRlMBAACUBhelAQBAdjIvWLZv3x4rV66MlStXRkTEmjVrYuXKlbFu3bqIiJg1a1b867/+a3zzm9+MVatWxSc+8YlYt25dzJgx46COWywWI5fLRT6fP9iPAAAAkImOvChNZgIAgNZlfouwZcuWxfjx45tez5o1KyIipk6dGgsXLozJkyfH1q1bY86cObFx48YYMWJELF68OIYNG3ZQxy0UClEoFKKhoSGqqqoOal8AAAAdZfv27bF69eqm13svShs4cGAcd9xxMWvWrLj00kvjjDPOiNGjR8ett97aLhelyUwAANC6zAuWcePGRZIkrW4zc+bMmDlzZidNBAAAUDqyuigNAABoXeYFCwAAAC1zURoAAJSmzJ/BAgAAAAAA0NUoWAAAAAAAAFIq24KlWCxGLpeLfD6f9SgAAAAlR2YCAIDWlW3BUigUor6+Purq6rIeBQAAoOTITAAA0LqyLVgAAAAAAADaSsECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASmVbsBSLxcjlcpHP57MeBQAAoOTITAAA0LqyLVgKhULU19dHXV1d1qMAAACUHJkJAABaV7YFCwAAAAAAQFspWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACClsi1YisVi5HK5yOfzWY8CAABQcmQmAABoXdkWLIVCIerr66Ouri7rUQAAAEqOzAQAAK0r24IFAAAAAACgrRQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkFLZFizFYjFyuVzk8/msRwEAAAAAALqYsi1YCoVC1NfXR11dXdajAAAAlBwXpQEAQOvKtmABAACgZS5KAwCA1ilYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACmVbcFSLBYjl8tFPp/PehQAAICSIzMBAEDryrZgKRQKUV9fH3V1dVmPAgAAUHJkJgAAaF3ZFiwAAAAAAABtpWABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAIKWyLViKxWLkcrnI5/NZjwIAAFByZCYAAGhd2RYshUIh6uvro66uLutRAAAASo7MBAAArSvbggUAAAAAAKCtFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAIKU2FSyvfe1rY+vWrfssf/755+O1r33tQQ8FAADQlclMAADQ/bWpYFm7dm00Njbus3znzp2xfv36gx4KAACgK5OZAACg++uVZuMf/vCHTf/3vffeG1VVVU2vGxsb4/7774/jjz++3YbrSMViMYrF4n5DDwAAQFt0p8wEAAC0LlXBcuGFF0ZEREVFRUydOrXZut69e8fxxx8fX/7yl9ttuI5UKBSiUChEQ0NDs9ADAADQVt0pM7koDQAAWpeqYNmzZ09ERAwfPjzq6upi0KBBHTIUAABAV9SdMpOL0gAAoHWpCpa91qxZ095zAAAAdBsyEwAAdH9tKlgiIu6///64//77Y/PmzU1Xae31zW9+86AHAwAA6MpkJgAA6N7aVLB8/vOfjzlz5sQZZ5wRxxxzTFRUVLT3XAAAAF2WzAQAAN1fmwqWW265JRYuXBiXXnppe88DAADQ5clMAADQ/fVoy5t27doVY8aMae9ZAAAAugWZCQAAur82FSzTp0+P73znO+09CwAAQLcgMwEAQPfXpluEvfTSS3HrrbfGfffdF6ecckr07t272fp58+a1y3AAAABdkcwEAADdX5sKll/96ldx6qmnRkTEk08+2WydhzcCAADlTmYCAIDur00FywMPPNDecwAAAHQbMhMAAHR/bXoGCwAAAAAAQDlr0zdYxo8f3+rX2n/2s5+1eSAAAICuTmYCAIDur00Fy957Ce+1e/fuWLlyZTz55JMxderU9pgLAACgy5KZAACg+2tTwfKVr3xlv8s/97nPxfbt2w9qIAAAgK5OZgIAgO6vXZ/B8oEPfCC++c1vtucuAQAAug2ZCQAAuo92LVgeeeSR6Nu3b3vuEgAAoNuQmQAAoPto0y3CJk2a1Ox1kiSxcePGWLZsWXzmM59pl8EAAAC6KpkJAAC6vzYVLFVVVc1e9+jRI0488cSYM2dOTJgwoV0GAwAA6KpkJgAA6P7aVLDcfvvt7T0HAABAtyEzAQBA99emgmWv5cuXx6pVq6KioiJyuVycdtpp7TUXAABAlyczAQBA99WmgmXz5s1x8cUXx4MPPhhHHHFEJEkSL7zwQowfPz7uvPPOOOqoo9p7TgAAgC5DZgIAgO6vR1ve9LGPfSwaGhriqaeeij/+8Y+xbdu2ePLJJ6OhoSGuuOKK9p4RAChxv3l6VYytmbTfn8lTpmU9HkCnk5kAAKD7a9M3WH7yk5/EfffdFyeddFLTslwuF8Vi0QMbAaAM7U56xJCLrtrvuvV339jJ0wBkT2YCAIDur03fYNmzZ0/07t17n+W9e/eOPXv2HPRQAAAAXZnMBAAA3V+bvsFy9tlnx8c//vH4j//4j6iuro6IiPXr18cnPvGJeNvb3tauAwIAAHQ1MhMA8Nf23lr5r1UfNSAW3XFbBhMBB6tNBctNN90U73rXu+L444+PoUOHRkVFRaxbty5OPvnk+Ld/+7f2nhEAAKBLkZkAgL/W0q2V3VYZuq42FSxDhw6Nxx9/PGpra+Ppp5+OJEkil8vF29/+9vae71W9+OKLcfbZZ8fu3bujsbExrrjiivjwhz/c6XMAAADsJTMBAED3l6pg+dnPfhaXX355PProo9G/f/8455xz4pxzzomIiBdeeCHe+MY3xi233BJjx47tkGH355BDDomHHnooDjnkkPjTn/4UI0aMiEmTJsWRRx7ZaTMAAABEyEwAAFBOUj3kfv78+fHhD384+vfvv8+6qqqq+OhHPxrz5s1rt+EORM+ePeOQQw6JiIiXXnopGhsbI0mSTp0BAAAgQmYCAIBykqpg+eUvfxnveMc7Wlw/YcKEWL58eaoBHn744Zg4cWJUV1dHRUVF3HPPPftss2DBghg+fHj07ds3Ro0aFUuWLGm2/vnnn4+RI0fGscceG1dffXUMGjQo1QwAAADtQWYCAIDykapg+cMf/hC9e/ducX2vXr1iy5YtqQbYsWNHjBw5Mm666ab9rl+0aFFceeWVce2118aKFSti7NixUVNTE+vWrWva5ogjjohf/vKXsWbNmvjOd74Tf/jDH1LNAAAA0B5kJgAAKB+pCpYhQ4bEE0880eL6X/3qV3HMMcekGqCmpiauv/76mDRp0n7Xz5s3L6ZNmxbTp0+Pk046KebPnx9Dhw6Nm2++eZ9tBw8eHKeccko8/PDDLR5v586d0dDQ0OwHAACgPchMAABQPlIVLOedd1589rOfjZdeemmfdX/+85/juuuui/PPP7/dhtu1a1csX748JkyY0Gz5hAkTYunSpRHxlyvE9v6Dv6GhIR5++OE48cQTW9zn3Llzo6qqquln6NCh7TYvAABQ3mQmAAAoH73SbPzpT3867rrrrjjhhBPi8ssvjxNPPDEqKipi1apVUSwWo7GxMa699tp2G+65556LxsbGGDx4cLPlgwcPjk2bNkVExO9///uYNm1aJEkSSZLE5ZdfHqecckqL+5w9e3bMmjWr6XVDQ4PAAAAAtAuZCQAAykeqgmXw4MGxdOnS+Pu///uYPXt2JEkSEREVFRVx7rnnxoIFC/b5h317qKioaPY6SZKmZaNGjYqVK1ce8L4qKyujsrKyPccDAACICJkJAADKSaqCJSJi2LBhsXjx4ti2bVusXr06kiSJ17/+9TFgwIB2H27QoEHRs2fPpiuv9tq8eXOHhBIAAICDJTMBAEB5SPUMllcaMGBA5PP5eNOb3tQhQSEiok+fPjFq1Kiora1ttry2tjbGjBnTIccEAABoDzITAAB0b6m/wdLetm/fHqtXr256vWbNmli5cmUMHDgwjjvuuJg1a1ZceumlccYZZ8To0aPj1ltvjXXr1sWMGTMO6rjFYrHpHsgAAAClSmYCAIDSlHnBsmzZshg/fnzT670PU5w6dWosXLgwJk+eHFu3bo05c+bExo0bY8SIEbF48eIYNmzYQR23UChEoVCIhoaGqKqqOqh9AQAAdBSZCQAASlPmBcu4ceOaHvzYkpkzZ8bMmTM7aSIAAIDSITMBAEBpavMzWAAAAAAAAMqVggUAAAAAACClsi1YisVi5HK5yOfzWY8CAAAAAAB0MWVbsBQKhaivr4+6urqsRwEAACg5LkoDAIDWlW3BAgAAQMtclAYAAK1TsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACClsi1YisVi5HK5yOfzWY8CAABQcmQmAABoXdkWLIVCIerr66Ouri7rUQAAAEqOzAQAAK0r24IFAAAAAACgrRQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASKlsC5ZisRi5XC7y+XzWowAAAJQcmQkAAFpXtgVLoVCI+vr6qKury3oUAACAkiMzAQBA68q2YAEAAAAAAGgrBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIqWwLlmKxGLlcLvL5fNajAAAAlByZCQAAWle2BUuhUIj6+vqoq6vLehQAAICSIzMBAEDryrZgAQAAAAAAaCsFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFLqlfUAAAAAAAClZvKUabFhy7b9rntm9eoY0snzAKVHwQIAAAAA8Fc2bNkWQy66ar/rnrpheidPA5QitwgDAAAAAABIScECAAAAAACQUtkWLMViMXK5XOTz+axHAQAAKDkyEwAAtK5sC5ZCoRD19fVRV1eX9SgAAAAlR2YCAIDWlW3BAgAAAAAA0Fa9sh4AAMjG5CnTYsOWbfssf2b16hiSwTwAAAAAXYmCBQDK1IYt22LIRVfts/ypG6ZnMA0AAABA1+IWYQAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEjJQ+4BgA71m6dXxdiaSfssrz5qQCy647YMJgIAAAA4eAoWAKBD7U56xJCLrtpn+fq7b8xgGgAAAID24RZhAAAAAAAAKSlYAAAAAAAAUnKLMAAAAACAjHhuJXRdZVuwFIvFKBaL0djYmPUoAAAAJUdmAoDO4bmV0HWV7S3CCoVC1NfXR11dXdajAAAAlByZCQAAWle2BQsAAAAAAEBbKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUuqV9QAAQHn6zdOrYmzNpH2WVx81IBbdcVsGEwEAAAAcOAULAJCJ3UmPGHLRVfssX3/3jRlMAwAAAJCOW4QBAAAAAACk5BssAAAAAAAlpqXbKke4tTKUCgULAAAAAECJaem2yhFurQylwi3CAAAAAAAAUlKwAAAAAAAApFS2BUuxWIxcLhf5fD7rUQAAAEqOzAQAAK0r24KlUChEfX191NXVZT0KAABAyZGZAACgdWVbsAAAAAAAALSVggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkFKvrAcAAHil3zy9KsbWTNrvuuqjBsSiO27r5IkAAAAA9qVgAQBKyu6kRwy56Kr9rlt/942dPA0AAEDpaenCNBelQedSsAAAAAAAdCEtXZjmojToXJ7BAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFLq8gXLs88+G+PGjYtcLhennHJKfO9738t6JAAAgJIhMwEAQMfolfUAB6tXr14xf/78OPXUU2Pz5s1x+umnx3nnnReHHnpo1qMBAABkTmYCAICO0eULlmOOOSaOOeaYiIh4zWteEwMHDow//vGPwgIAAEDITAAA0FEyv0XYww8/HBMnTozq6uqoqKiIe+65Z59tFixYEMOHD4++ffvGqFGjYsmSJfvd17Jly2LPnj0xdOjQDp4aAACgc8hMAABQmjIvWHbs2BEjR46Mm266ab/rFy1aFFdeeWVce+21sWLFihg7dmzU1NTEunXrmm23devWmDJlStx6662tHm/nzp3R0NDQ7AcAAKBUyUwAAFCaMi9Yampq4vrrr49Jkybtd/28efNi2rRpMX369DjppJNi/vz5MXTo0Lj55pubttm5c2dcdNFFMXv27BgzZkyrx5s7d25UVVU1/bhyCwAAKGUyEwAAlKbMC5bW7Nq1K5YvXx4TJkxotnzChAmxdOnSiIhIkiQuu+yyOPvss+PSSy991X3Onj07XnjhhaafZ599tkNmBwAA6GgyEwAAZKekH3L/3HPPRWNjYwwePLjZ8sGDB8emTZsiIuIXv/hFLFq0KE455ZSmexF/+9vfjpNPPnm/+6ysrIzKysoOnRsAAKAzyEwAAJCdki5Y9qqoqGj2OkmSpmVvfetbY8+ePVmMBQAAUBJkJgAA6HwlfYuwQYMGRc+ePZuuvNpr8+bN+1yhBQAAUG5kJgAAyE5JFyx9+vSJUaNGRW1tbbPltbW1r/pgxldTLBYjl8tFPp8/qP0AAABkRWYCAIDsZH6LsO3bt8fq1aubXq9ZsyZWrlwZAwcOjOOOOy5mzZoVl156aZxxxhkxevTouPXWW2PdunUxY8aMgzpuoVCIQqEQDQ0NUVVVdbAfAwAAoEPITAAAUJoyL1iWLVsW48ePb3o9a9asiIiYOnVqLFy4MCZPnhxbt26NOXPmxMaNG2PEiBGxePHiGDZsWFYjAwAAdBqZCQAASlPmBcu4ceMiSZJWt5k5c2bMnDmzkyYCAAAoHTITAACUppJ+BgsAAAAAAEApUrAAAAAAAACkVLYFS7FYjFwuF/l8PutRAAAASo7MBAAArSvbgqVQKER9fX3U1dVlPQoAAEDJkZkAAKB1ZVuwAAAAAAAAtJWCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJTKtmApFouRy+Uin89nPQoAAAAAANDFlG3BUigUor6+Purq6rIeBQAAoOS4KA0AAFpXtgULAAAALXNRGgAAtE7BAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACmVbcFSLBYjl8tFPp/PehQAAICSIzMBAEDryrZgKRQKUV9fH3V1dVmPAgAAUHJkJgAAaF3ZFiwAAAAAAABtpWABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKZVtwVIsFiOXy0U+n896FAAAgJIjMwEAQOvKtmApFApRX18fdXV1WY8CAABQcmQmAABoXdkWLAAAAAAAAG2lYAEAAAAAAEhJwQIAAAAAAJBSr6wHAAA4UL95elWMrZm0z/LqowbEojtuy2AiAAAAoFwpWACALmN30iOGXHTVPsvX331jBtMAAAAA5cwtwgAAAAAAAFLyDRYAoMtr6dZhEW4fBgAAEBExecq02LBl2z7LZSZoOwULANDltXTrsAi3DwMAAMpHaxefPbN6dYy76uv7LJeZoO3KtmApFotRLBajsbEx61EAAABKjswEAF1PaxefPXXD9E6eBrq/sn0GS6FQiPr6+qirq8t6FAAAgJIjMwEAQOvKtmABAAAAAABoKwULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUuqV9QAAAAAAAFmZPGVabNiybZ/lz6xeHUMymAfoOhQsAAAAAEDZ2rBlWwy56Kp9lj91w/QMpgG6ErcIAwAAAAAASEnBAgAAAAAAkJJbhAEA3dpvnl4VY2sm7bO8+qgBseiO2zKYCAAAAOgOFCwAQLe2O+mx3/spr7/7xgymAQAAALoLBQsAAAAAQJlq6Vv/Eb75D6+mbAuWYrEYxWIxGhsbsx4FAACg5MhMAHRFk6dMiw1btu13nbJg/1r61n+Eb/7DqynbgqVQKEShUIiGhoaoqqrKehwAAICSIjMB0BVt2LJNWQB0mh5ZDwAAAAAAANDVKFgAAAAAAABSUrAAAAAAAACkVLbPYAGArqilBzZ6WCMAAABA51KwAEAX0tIDGz2sEQAAAKBzuUUYAAAAAABASgoWAAAAAACAlNwiDAC6uZae2/LM6tUxJIN5AAAAALoDBQsAdHMtPbflqRumZzANAAAAQPfgFmEAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKZVtwVIsFiOXy0U+n896FAAAgJIjMwEAQOvKtmApFApRX18fdXV1WY8CAABQcmQmAABoXdkWLAAAAAAAAG2lYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQUq+sBwAADt5vnl4VY2sm7XfdM6tXx5BOngcAAACgu1OwAEA3sDvpEUMuumq/6566YXonTwMAAADQ/blFGAAAAAAAQEoKFgAAAAAAgJTcIqwETJ4yLTZs2bbP8uqjBsSiO27LYCIAAIDS0VJmipCbADhwLT270nMrgbZSsJSADVu27fe++evvvjGDaQAAAEpLS5kpQm4C4MC19OxKz60E2sotwgAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACClXlkPAAAA3dXkKdNiw5Zt+yyvPmpALLrjtgwmAgAAoL0oWAAAoINs2LIthlx01T7L1999YwbTAAAA0J7cIgwAAAAAACClblGwXHTRRTFgwIB4z3vek/UoAAAAJUdmAgCA9tctbhF2xRVXxIc+9KH41re+lfUoAAAAJUdmAgAgSy09n3Ldmt/GccP/Zr/v6QrPruwWBcv48ePjwQcfzHoMAACAkiQzAQCQpZaeT/nUDdP3uzyiazy7MvNbhD388MMxceLEqK6ujoqKirjnnnv22WbBggUxfPjw6Nu3b4waNSqWLFnS+YMCAABkQGYCAIDSlHnBsmPHjhg5cmTcdNNN+12/aNGiuPLKK+Paa6+NFStWxNixY6OmpibWrVvXyZMCAAB0PpkJAABKU+a3CKupqYmampoW18+bNy+mTZsW06dPj4iI+fPnx7333hs333xzzJ07N/Xxdu7cGTt37mx63dDQkH5oAACATiIzAQBAacq8YGnNrl27Yvny5XHNNdc0Wz5hwoRYunRpm/Y5d+7c+PznP98e4wEA0A2lffhiaw9lfGb16hjS7hPC/5GZAADobG15YH13zUYlXbA899xz0djYGIMHD262fPDgwbFp06am1+eee248/vjjsWPHjjj22GPj7rvvjnw+v999zp49O2bNmtX0uqGhIYYOHdoxHwAAgC4n7cMXW3so41M3TG/3+eCVZCYAADpbWx5Y312zUUkXLHtVVFQ0e50kSbNl99577wHvq7KyMiorK9ttNgAAgKzJTAAA0Pkyf8h9awYNGhQ9e/ZsduVVRMTmzZv3uUILAACg3MhMAACQnZIuWPr06ROjRo2K2traZstra2tjzJgxGU0FAABQGmQmAADITua3CNu+fXusXr266fWaNWti5cqVMXDgwDjuuONi1qxZcemll8YZZ5wRo0ePjltvvTXWrVsXM2bMOKjjFovFKBaL0djYeLAfAQAAoMPITAAAUJoyL1iWLVsW48ePb3q992GKU6dOjYULF8bkyZNj69atMWfOnNi4cWOMGDEiFi9eHMOGDTuo4xYKhSgUCtHQ0BBVVVUHtS8AAICOIjMBAEBpyrxgGTduXCRJ0uo2M2fOjJkzZ3bSRAAAAKVDZgIAgNJU0s9gAQAAAAAAKEUKFgAAAAAAgJQULAAAAAAAACmVbcFSLBYjl8tFPp/PehQAAICSIzMBAEDryrZgKRQKUV9fH3V1dVmPAgAAUHJkJgAAaF3ZFiwAAAAAAABtpWABAAAAAABIScECAAAAAACQkoIFAAAAAAAgpV5ZD5CVYrEYxWIxGhsbsx4FAACg5MhM0P1MnjItNmzZtt911UcNiEV33NbJEwFA11a2BUuhUIhCoRANDQ1RVVWV9TgAAAAlRWaC7mfDlm0x5KKr9rtu/d03dvI0AND1uUUYAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASKlX1gNkpVgsRrFYjMbGxqxHAdrJ5CnTYsOWbfssrz5qQCy647YMJgIA6LpkJgAAaF3ZFiyFQiEKhUI0NDREVVVV1uMA7WDDlm0x5KKr9lm+/u4bM5gGAKBrk5kAAKB1bhEGAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUirbgqVYLEYul4t8Pp/1KAAAACVHZgIAgNaVbcFSKBSivr4+6urqsh4FAACg5MhMAADQurItWAAAAAAAANpKwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIqVfWA2SlWCxGsViMxsbGrEcBgGYmT5kWG7Zs2++6Z1avjiGdPA8A5UlmAgCA1pVtwVIoFKJQKERDQ0NUVVVlPQ4ANNmwZVsMueiq/a576obpnTwNAOVKZgIAgNa5RRgAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUuqV9QBZKRaLUSwWo7GxMetRAAAASo7MBEApmzxlWmzYsm2f5c+sXh1DMpgHKE9lW7AUCoUoFArR0NAQVVVVWY8DAABQUmQmAErZhi3bYshFV+2z/KkbpmcwDVCu3CIMAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACClXlkPkJVisRjFYjEaGxuzHgWgbE2eMi02bNm2z/LqowbEojtuy2AiAGAvmQkAAFpXtgVLoVCIQqEQDQ0NUVVVlfU4AGVpw5ZtMeSiq/ZZvv7uGzOYBgB4JZkJAABa5xZhAAAAAAAAKSlYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsAAAAAAAAKSkYAEAAAAAAEhJwQIAAAAAAJCSggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAApKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASgoWAAAAAACAlBQsAAAAAAAAKSlYAAAAAAAAUuoWBct//dd/xYknnhivf/3r41//9V+zHgcAAKCkyEwAAND+emU9wMF6+eWXY9asWfHAAw9E//794/TTT49JkybFwIEDsx4NAAAgczITAAB0jC7/DZb/+Z//iTe+8Y0xZMiQOPzww+O8886Le++9N+uxAAAASoLMBAAAHSPzguXhhx+OiRMnRnV1dVRUVMQ999yzzzYLFiyI4cOHR9++fWPUqFGxZMmSpnUbNmyIIUOGNL0+9thjY/369Z0xOgAAQIeTmQAAoDRlXrDs2LEjRo4cGTfddNN+1y9atCiuvPLKuPbaa2PFihUxduzYqKmpiXXr1kVERJIk+7ynoqKiQ2cGAADoLDITAACUpsyfwVJTUxM1NTUtrp83b15MmzYtpk+fHhER8+fPj3vvvTduvvnmmDt3bgwZMqTZ1Ve///3v481vfnOL+9u5c2fs3Lmz6XVDQ0M7fAoAAICOITMBAEBpyrxgac2uXbti+fLlcc011zRbPmHChFi6dGlERLzpTW+KJ598MtavXx/9+/ePxYsXx2c/+9kW9zl37tz4/Oc/36FzA8ArTZ4yLTZs2bbP8uqjBsSiO27LYCLaW0u/4wi/51LV2u/smdWrY8h+10DpkZkA6A5kpvLQnX7PWX+WrI/P/ynpguW5556LxsbGGDx4cLPlgwcPjk2bNkVERK9eveLLX/5yjB8/Pvbs2RNXX311HHnkkS3uc/bs2TFr1qym1w0NDTF06NCO+QAAEBEbtmyLIRddtc/y9XffmME0dISWfscRfs+lqrXf2VM3TO/kaaDtZCYAugOZqTx0p99z1p8l6+Pzf0q6YNnrr+8PnCRJs2UXXHBBXHDBBQe0r8rKyqisrGzX+QAAALIkMwEAQOfL/CH3rRk0aFD07Nmz6cqrvTZv3rzPFVoAAADlRmYCAIDslHTB0qdPnxg1alTU1tY2W15bWxtjxozJaCoAAIDSIDMBAEB2Mr9F2Pbt22P16tVNr9esWRMrV66MgQMHxnHHHRezZs2KSy+9NM4444wYPXp03HrrrbFu3bqYMWPGQR23WCxGsViMxsbGg/0IAAAAHUZmAgCA0pR5wbJs2bIYP3580+u9D1OcOnVqLFy4MCZPnhxbt26NOXPmxMaNG2PEiBGxePHiGDZs2EEdt1AoRKFQiIaGhqiqqjqofQEAAHQUmQkAAEpT5gXLuHHjIkmSVreZOXNmzJw5s5MmAgAAKB0yEwAAlKaSfgYLAAAAAABAKVKwAAAAAAAApFS2BUuxWIxcLhf5fD7rUQAAAEqOzAQAAK0r24KlUChEfX191NXVZT0KAABAyZGZAACgdWVbsAAAAAAAALSVggUAAAAAACAlBQsAAAAAAEBKChYAAAAAAICUFCwAAAAAAAAplW3BUiwWI5fLRT6fz3oUAACAkiMzAQBA68q2YCkUClFfXx91dXVZjwIAAFByZCYAAGhd2RYsAAAAAAAAbaVgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEplW7AUi8XI5XKRz+ezHgUAAKDkyEwAANC6si1YCoVC1NfXR11dXdajAAAAlByZCQAAWle2BQsAAAAAAEBbKVgAAAAAAABSUrAAAAAAAACkpGABAAAAAABIScECAAAAAACQkoIFAAAAAAAgJQULAAAAAABASmVbsBSLxcjlcpHP57MeBQAAoOTITAAA0LpeWQ+QlUKhEIVCIV544YU44ogjoqGhIbNZXn55d+z+8479Ls9yLuhq/LfU9ZTL7yzt52xp+4iIZE/jfte1tLwt72nPfZXy8dvzPGvtd9bdzufuoj3/O8v6/Etr73GTJMnk+HQdXSEz7V3n71k4MP5bolS1JRu29J6umE1K9fgRnZObuuLfP1l/llI9fimff2kdaG6qSMo8Wf3+97+PoUOHZj0GAAB0qmeffTaOPfbYrMegC5CZAAAoV6+Wm8q+YNmzZ09s2LAhDj/88KioqDjg9zU0NMTQoUPj2Wefjf79+3fghJQL5xTtyflEe3I+0Z6cT9lLkiRefPHFqK6ujh49yvaOwaQgM1EqnFO0J+cT7cn5RHtzTmXvQHNT2d4ibK8ePXoc1JV7/fv3d5LTrpxTtCfnE+3J+UR7cj5lq6qqKusR6EJkJkqNc4r25HyiPTmfaG/OqWwdSG5yyRoAAAAAAEBKChYAAAAAAICUFCxtVFlZGdddd11UVlZmPQrdhHOK9uR8oj05n2hPzicoH/57p705p2hPzifak/OJ9uac6jrK/iH3AAAAAAAAafkGCwAAAAAAQEoKFgAAAAAAgJQULAAAAAAAACkpWAAAAAAAAFJSsPyVL3zhCzFmzJg45JBD4ogjjtjvNhUVFfv83HLLLc22eeKJJ+Kss86Kfv36xZAhQ2LOnDmRJEmzbR566KEYNWpU9O3bN1772tfusw+6hwM5p9atWxcTJ06MQw89NAYNGhRXXHFF7Nq1q9k2zin25/jjj9/n76Nrrrmm2TbtdX5RvhYsWBDDhw+Pvn37xqhRo2LJkiVZj0SJ+dznPrfP30VHH3100/okSeJzn/tcVFdXR79+/WLcuHHx1FNPNdvHzp0742Mf+1gMGjQoDj300Ljgggvi97//fWd/FOAAyEy0N5mJjiQz0RlkJl6NzNR9KVj+yq5du+K9731v/P3f/32r291+++2xcePGpp+pU6c2rWtoaIhzzjknqquro66uLr72ta/Fl770pZg3b17TNmvWrInzzjsvxo4dGytWrIh//Md/jCuuuCJ+8IMfdNhnIxuvdk41NjbGO9/5ztixY0f8/Oc/jzvvvDN+8IMfxCc/+cmmbZxTtGbOnDnN/j769Kc/3bSuvc4vyteiRYviyiuvjGuvvTZWrFgRY8eOjZqamli3bl3Wo1Fi3vjGNzb7u+iJJ55oWvcv//IvMW/evLjpppuirq4ujj766DjnnHPixRdfbNrmyiuvjLvvvjvuvPPO+PnPfx7bt2+P888/PxobG7P4OEArZCbam8xER5OZ6EgyEwdKZuqmEvbr9ttvT6qqqva7LiKSu+++u8X3LliwIKmqqkpeeumlpmVz585Nqqurkz179iRJkiRXX3118oY3vKHZ+z760Y8mb3nLWw56dkpTS+fU4sWLkx49eiTr169vWvYf//EfSWVlZfLCCy8kSeKcomXDhg1LvvKVr7S4vr3OL8rXm970pmTGjBnNlr3hDW9IrrnmmowmohRdd911yciRI/e7bs+ePcnRRx+d/PM//3PTspdeeimpqqpKbrnlliRJkuT5559Pevfundx5551N26xfvz7p0aNH8pOf/KRDZwfaTmaivclMdASZiY4mM3EgZKbuyzdY2ujyyy+PQYMGRT6fj1tuuSX27NnTtO6RRx6Js846KyorK5uWnXvuubFhw4ZYu3Zt0zYTJkxots9zzz03li1bFrt37+6Uz0BpeOSRR2LEiBFRXV3dtOzcc8+NnTt3xvLly5u2cU7Rki9+8Ytx5JFHxqmnnhpf+MIXmn2Vvb3OL8rTrl27Yvny5fv83TJhwoRYunRpRlNRqp555pmorq6O4cOHx8UXXxy/+93vIuIvVwtv2rSp2XlUWVkZZ511VtN5tHz58ti9e3ezbaqrq2PEiBHONejCZCbai8zEwZKZ6CgyE2nITN2TgqUN/umf/im+973vxX333RcXX3xxfPKTn4wbbrihaf2mTZti8ODBzd6z9/WmTZta3ebll1+O5557roM/AaVkf+fCgAEDok+fPq96vuxd19o2zqnu7eMf/3jceeed8cADD8Tll18e8+fPj5kzZzatb6/zi/L03HPPRWNj437PD+cGr/TmN7857rjjjrj33nvjG9/4RmzatCnGjBkTW7dubTpXWjuPNm3aFH369IkBAwa0uA3QtchMtCeZiYMhM9GRZCYOlMzUfZVFwbK/hwj99c+yZcsOeH+f/vSnY/To0XHqqafGJz/5yZgzZ07ceOONzbapqKho9jr5/x989srlB7INpam9z6n9/c6TJEl9vjinuoc059cnPvGJOOuss+KUU06J6dOnxy233BK33XZbbN26tWl/7XV+Ub72d344N3ilmpqaePe73x0nn3xyvP3tb48f//jHERHxrW99q2mbtpxHzjXoPDIT7U1moiPJTJQamYlXIzN1X72yHqAzXH755XHxxRe3us3xxx/f5v2/5S1viYaGhvjDH/4QgwcPjqOPPnqf5nDz5s0R8X9NZEvb9OrVK4488sg2z0LnaM9z6uijj47HHnus2bJt27bF7t27X/V8iXBOdUcHc3695S1viYiI1atXx5FHHtlu5xfladCgQdGzZ8/9nh/ODVpz6KGHxsknnxzPPPNMXHjhhRHxlyuujjnmmKZtXnkeHX300bFr167Ytm1bsyuyNm/eHGPGjOnU2aFcyUy0N5mJjiQzUSpkJtpKZuo+yqJgGTRoUAwaNKjD9r9ixYro27dvHHHEERERMXr06PjHf/zH2LVrV/Tp0yciIn76059GdXV10//Ajx49On70ox81289Pf/rTOOOMM6J3794dNivtoz3PqdGjR8cXvvCF2LhxY9Nfoj/96U+jsrIyRo0a1bSNc6p8HMz5tWLFioiIpnOpvc4vylOfPn1i1KhRUVtbGxdddFHT8tra2njXu96V4WSUup07d8aqVati7NixMXz48Dj66KOjtrY2TjvttIj4y72qH3roofjiF78YERGjRo2K3r17R21tbfzd3/1dRERs3LgxnnzyyfiXf/mXzD4HlBOZifYmM9GRZCZKhcxEW8lM3ch+Hnxf1v73f/83WbFiRfL5z38+Oeyww5IVK1YkK1asSF588cUkSZLkhz/8YXLrrbcmTzzxRLJ69erkG9/4RtK/f//kiiuuaNrH888/nwwePDh53/velzzxxBPJXXfdlfTv3z/50pe+1LTN7373u+SQQw5JPvGJTyT19fXJbbfdlvTu3Tv5/ve/3+mfmY71aufUyy+/nIwYMSJ529veljz++OPJfffdlxx77LHJ5Zdf3rQP5xT7s3Tp0mTevHnJihUrkt/97nfJokWLkurq6uSCCy5o2qa9zi/K15133pn07t07ue2225L6+vrkyiuvTA499NBk7dq1WY9GCfnkJz+ZPPjgg8nvfve75NFHH03OP//85PDDD286T/75n/85qaqqSu66667kiSeeSN73vvclxxxzTNLQ0NC0jxkzZiTHHntsct999yWPP/54cvbZZycjR45MXn755aw+FtACmYn2JjPRUWQmOoPMxIGQmbovBctfmTp1ahIR+/w88MADSZIkyX//938np556anLYYYclhxxySDJixIhk/vz5ye7du5vt51e/+lUyduzYpLKyMjn66KOTz33uc8mePXuabfPggw8mp512WtKnT5/k+OOPT26++ebO+ph0olc7p5LkL4Hine98Z9KvX79k4MCByeWXX5689NJLzfbjnOKvLV++PHnzm9+cVFVVJX379k1OPPHE5Lrrrkt27NjRbLv2Or8oX8ViMRk2bFjSp0+f5PTTT08eeuihrEeixEyePDk55phjkt69eyfV1dXJpEmTkqeeeqpp/Z49e5LrrrsuOfroo5PKysrkzDPPTJ544olm+/jzn/+cXH755cnAgQOTfv36Jeeff36ybt26zv4owAGQmWhvMhMdRWais8hMvBqZqfuqSJL//6lcAAAAAAAAHJAeWQ8AAAAAAADQ1ShYAAAAAAAAUlKwAAAAAAAApKRgAQAAAAAASEnBAgAAAAAAkJKCBQAAAAAAICUFCwAAAAAAQEoKFgBadPzxx8f8+fPbbX9r166NioqKWLlyZZve/+CDD0ZFRUU8//zzB/ye9v4MAABA6Rs3blxceeWVWY/RJaTNTJ/73Ofi1FNP7bB5ALqSXlkPAEDpqquri0MPPTTrMQAAALqktWvXxvDhw2PFihVKCYBuyDdYAMrc7t2791m2a9euiIg46qij4pBDDunskQAAALq8vbmqVPcHwMFTsAB0Mz/5yU/irW99axxxxBFx5JFHxvnnnx+//e1vI+L/btH13e9+N8aNGxd9+/aNf/u3f4vLLrssLrzwwpg7d25UV1fHCSecEBHNvyr+vve9Ly6++OJmx9q9e3cMGjQobr/99lc9dlssXrw4TjjhhOjXr1+MHz8+1q5du882S5cujTPPPDP69esXQ4cOjSuuuCJ27NjR4j7nzZsXJ598chx66KExdOjQmDlzZmzfvj0iInbs2BH9+/eP73//+83e86Mf/SgOPfTQePHFF9v8WQAAgI6xY8eOmDJlShx22GFxzDHHxJe//OVm63ft2hVXX311DBkyJA499NB485vfHA8++GDT+oULF8YRRxwR99xzT5xwwgnRt2/fOOecc+LZZ59t2ua3v/1tvOtd74rBgwfHYYcdFvl8Pu67775mxzn++OPj+uuvj8suuyyqqqriwx/+cAwfPjwiIk477bSoqKiIcePGRcT+b2F24YUXxmWXXdbq/iLSZ6BX2rx5c0ycODH69esXw4cPj3//93/fZ5sXXnghPvKRj8RrXvOa6N+/f5x99tnxy1/+ssV91tXVxTnnnBODBg2KqqqqOOuss+Lxxx9vWv+hD30ozj///Gbvefnll+Poo4+Ob37zmwc0N0CpUrAAdDM7duyIWbNmRV1dXdx///3Ro0ePuOiii2LPnj1N23zqU5+KK664IlatWhXnnntuRETcf//9sWrVqqitrY3/+q//2me/l1xySfzwhz9sKiMiIu69997YsWNHvPvd7z7gYx+oZ599NiZNmhTnnXderFy5MqZPnx7XXHNNs22eeOKJOPfcc2PSpEnxq1/9KhYtWhQ///nP4/LLL29xvz169IivfvWr8eSTT8a3vvWt+NnPfhZXX311REQceuihcfHFFzcVRnvdfvvt8Z73vCcOP/zw1J8DAADoWFdddVU88MADcffdd8dPf/rTePDBB2P58uVN6z/4wQ/GL37xi7jzzjvjV7/6Vbz3ve+Nd7zjHfHMM880bfOnP/0pvvCFL8S3vvWt+MUvfhENDQ3NLjDbvn17nHfeeXHffffFihUr4txzz42JEyfGunXrms1y4403xogRI2L58uXxmc98Jv7nf/4nIiLuu+++2LhxY9x1112pPttf768tGeiVLrvssli7dm387Gc/i+9///uxYMGC2Lx5c9P6JEnine98Z2zatCkWL14cy5cvj9NPPz3e9ra3xR//+Mf97vPFF1+MqVOnxpIlS+LRRx+N17/+9XHeeec1XaA2ffr0+MlPfhIbN25ses/ixYtj+/bt8Xd/93ep/jwASk4CQLe2efPmJCKSJ554IlmzZk0SEcn8+fObbTN16tRk8ODByc6dO5stHzZsWPKVr3wlSZIk2bVrVzJo0KDkjjvuaFr/vve9L3nve997QMdOkqTp+CtWrHjVuWfPnp2cdNJJyZ49e5qWfepTn0oiItm2bVuSJEly6aWXJh/5yEeavW/JkiVJjx49kj//+c/7fIb9+e53v5sceeSRTa8fe+yxpGfPnsn69euTJEmSLVu2JL17904efPDBV50ZAADoXC+++GLSp0+f5M4772xatnXr1qRfv37Jxz/+8WT16tVJRUVF07/v93rb296WzJ49O0mSJLn99tuTiEgeffTRpvWrVq1KIiJ57LHHWjx2LpdLvva1rzW9HjZsWHLhhRc226alDHTWWWclH//4x5ste9e73pVMnTq11f0dSAZqya9//esWP+fezHT//fcn/fv3T1566aVm7/2bv/mb5Otf/3qSJEly3XXXJSNHjmzxOC+//HJy+OGHJz/60Y+aluVyueSLX/xi0+sLL7wwueyyy1qdF6Ar8A0WgG7mt7/9bbz//e+P1772tdG/f/+mr6S/8sqqM844Y5/3nXzyydGnT58W99u7d+9473vf2/QV8h07dsR//ud/xiWXXJLq2Adq1apV8Za3vCUqKiqalo0ePbrZNsuXL4+FCxfGYYcd1vRz7rnnxp49e2LNmjX73e8DDzwQ55xzTgwZMiQOP/zwmDJlSmzdurXpK/VvetOb4o1vfGPccccdERHx7W9/O4477rg488wzU38GAACgY/32t7+NXbt2NcsKAwcOjBNPPDEiIh5//PFIkiROOOGEZrnhoYceanY74169ejXLSW94wxviiCOOiFWrVkXEX/LP1VdfHblcLo444og47LDD4umnn94n6+wvax2Mv95fWzLQXqtWrWrxc75y/9u3b48jjzyy2THWrFnT4u2fN2/eHDNmzIgTTjghqqqqoqqqKrZv397sz2b69OlNdwrYvHlz/PjHP44PfehDaf84AEpOr6wHAKB9TZw4MYYOHRrf+MY3orq6Ovbs2RMjRoxo9kDEQw89dJ/37W/ZX7vkkkvirLPOis2bN0dtbW307ds3ampqUh37QCVJ8qrb7NmzJz760Y/GFVdcsc+64447bp9l//u//xvnnXdezJgxI/7pn/4pBg4cGD//+c9j2rRpsXv37qbtpk+fHjfddFNcc801cfvtt8cHP/jBZkUPAABQGl4tN+zZsyd69uwZy5cvj549ezZbd9hhhzV7vb9/8+9ddtVVV8W9994bX/rSl+J1r3td9OvXL97znvfsk3UOJFdF/OXWxX89+yszSUv7S5uBXmnv8VrLNnv27Iljjjmm2TNq9nplEfNKl112WWzZsiXmz58fw4YNi8rKyhg9enSzP5spU6bENddcE4888kg88sgjcfzxx8fYsWNbnRegK1CwAHQjW7dujVWrVsXXv/71pn+s/vznP2+3/Y8ZMyaGDh0aixYtiv/+7/+O9773vU3femnvY+dyubjnnnuaLXv00UebvT799NPjqaeeite97nUHtM9ly5bFyy+/HF/+8pejR4+/fInzu9/97j7bfeADH4irr746vvrVr8ZTTz0VU6dObduHAAAAOtTrXve66N27dzz66KNNBcO2bdviN7/5TZx11llx2mmnRWNjY2zevLnV/4f+yy+/HMuWLYs3velNERHx61//Op5//vl4wxveEBERS5YsicsuuywuuuiiiPjLM1nWrl37qvPtzUuNjY3Nlh911FHNnknS2NgYTz75ZIwfP77V/aXNQK900kkntfg5X7n/TZs2Ra9eveL4448/oP0uWbIkFixYEOedd15E/OV5ms8991yzbY488si48MIL4/bbb49HHnkkPvjBD6aeH6AUuUUYQDcyYMCAOPLII+PWW2+N1atXx89+9rOYNWtWu+2/oqIi3v/+98ctt9wStbW18YEPfKDDjj1jxoz47W9/G7NmzYpf//rX8Z3vfCcWLlzYbJtPfepT8cgjj0ShUIiVK1fGM888Ez/84Q/jYx/72H73+Td/8zfx8ssvx9e+9rX43e9+F9/+9rfjlltu2We7AQMGxKRJk+Kqq66KCRMmxLHHHtvmzwEAAHScww47LKZNmxZXXXVV3H///fHkk0/GZZdd1nRB1QknnBCXXHJJTJkyJe66665Ys2ZN1NXVxRe/+MVYvHhx03569+4dH/vYx+Kxxx6Lxx9/PD74wQ/GW97ylqYi4nWve13cddddsXLlyvjlL38Z73//+2PPnj2vOt9rXvOa6NevX/zkJz+JP/zhD/HCCy9ERMTZZ58dP/7xj+PHP/5xPP300zFz5sxmRUdL0magVzrxxBPjHe94R3z4wx+Oxx57LJYvXx7Tp0+Pfv36NW3z9re/PUaPHh0XXnhh3HvvvbF27dpYunRpfPrTn45ly5btd7+ve93r4tvf/nasWrUqHnvssbjkkkua7XOv6dOnx7e+9a1YtWqVi9iAbkPBAtCN9OjRI+68885Yvnx5jBgxIj7xiU/EjTfe2K7HuOSSS6K+vj6GDBkSf/u3f9thxz7uuOPiBz/4QfzoRz+KkSNHxi233BI33HBDs21OOeWUeOihh+KZZ56JsWPHxmmnnRaf+cxn4phjjtnvPk899dSYN29efPGLX4wRI0bEv//7v8fcuXP3u+20adNi165d7gsMAAAl7sYbb4wzzzwzLrjggnj7298eb33rW2PUqFFN62+//faYMmVKfPKTn4wTTzwxLrjggnjsscdi6NChTdsccsgh8alPfSre//73x+jRo6Nfv35x5513Nq3/yle+EgMGDIgxY8bExIkT49xzz43TTz/9VWfr1atXfPWrX42vf/3rUV1dHe9617siIuJDH/pQTJ06NaZMmRJnnXVWDB8+/FW/vRKRPgP9tdtvvz2GDh0aZ511VkyaNCk+8pGPxGte85qm9RUVFbF48eI488wz40Mf+lCccMIJcfHFF8fatWtj8ODB+93nN7/5zdj2/7V3xyiKBFEYgN/ChI2GYpt6AL2DYGZgbmxHauoJBMErNCYmXsODCIIiJgMGbeZstMsuOGDvDtvLzPdBRQXNnxZ/1evX1+h2uzEajWIymfz2zR96vV40m83o9/uRpulTeQH+d9/enhlyDwBfzGaziel0GqfT6eezfgAA4PNZr9cxm82eekHCnyuKItI0jTzPYzgcVh0H4EP4BwsA/KIoitjv97FYLGI8HitXAAAA/sL9fo/z+Ryr1Srq9XoMBoOqIwF8GCPCAKhElmWRJMnDlWVZZbmWy2V0Op1oNBoxn88rywEAAFDGbrd794yVJElluQ6HQ7Rardhut5Hneby8uO8NfB5GhAFQicvlEtfr9eFerVZ7OLMXAACAx263WxyPx3f32+32P0wD8DUoWAAAAAAAAEoyIgwAAAAAAKAkBQsAAAAAAEBJChYAAAAAAICSFCwAAAAAAAAlKVgAAAAAAABKUrAAAAAAAACUpGABAAAAAAAoScECAAAAAABQ0ndnL84ub7jJRgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 2000x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot histograms\n",
    "fig, ax = plt.subplots(1, 2, figsize=(20, 10))\n",
    "sns.histplot(data=transport_data[(transport_data[\"arrival_delay\"] > -10000) & (transport_data[\"arrival_delay\"] < 10000)], x=\"arrival_delay\", bins=100, log_scale=(False, True), ax=ax[0])\n",
    "sns.histplot(data=transport_data[(transport_data[\"departure_delay\"] > -10000) & (transport_data[\"departure_delay\"] < 10000)], x=\"departure_delay\", bins=100, log_scale=(False, True), ax=ax[1])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Show statistics delays for each line each day on a website "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "data_viz",
   "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": "3.9.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}