{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\n", "\"GMV\n", "\"UPM\n", "

QA: EMT dataset 🚌

\n", "
INESDATA-MOV
\n", "
\n", "\n", "# Análisis de calidad\n", "Este cuaderno analiza la calidad del dataset proveniente de la fuente de datos de la Empresa Municipal de Transportes de Madrid ([EMT](https://www.emtmadrid.es/Home)). La calidad del mismo se validará teniendo en cuenta los siguientes aspectos:\n", "* Análisis de las variables\n", "* Conversiones de tipos de datos\n", "* Checks de calidad del dato\n", "* Análisis Exploratorio de los datos (EDA)\n", "\n", "La **calidad del dato** se refiere a la medida en que los datos son adecuados para su uso, por lo que es esencial para garantizar la confiabilidad y utilidad de los datos en diversas aplicaciones y contextos. Así, en este notebook se evaluarán también las cinco dimensiones de la calidad del dato:\n", "1. **Unicidad**: Ausencia de duplicados o registros repetidos en un conjunto de datos. Los datos son únicos cuando cada registro o entidad en el conjunto de datos es único y no hay duplicados presentes.\n", "2. **Exactitud**: Los datos exactos son libres de errores y representan con precisión la realidad que están destinados a describir. Esto implica que los datos deben ser correctos y confiables para su uso en análisis y toma de decisiones.\n", "3. **Completitud**: Los datos completos contienen toda la información necesaria para el análisis y no tienen valores faltantes o nulos que puedan afectar la interpretación o validez de los resultados.\n", "4. **Consistencia**: Los datos consistentes mantienen el mismo formato, estructura y significado en todas las instancias, lo que facilita su comparación y análisis sin ambigüedad.\n", "5. **Validez**: Medida en que los datos son precisos y representan con exactitud la realidad que están destinados a describir. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "

Nota

\n", "

\n", "Este dataset ha sido creado ejecutando el comando create del paquete de Python inesdata_mov_datasets.
\n", "Para poder ejecutar este comando es necesario haber ejecutado antes el comando extract, que realiza la extracción de datos de la API de la EMT y los almacena en Minio. El comando create se encargaría de descargar dichos datos y unirlos todos en un único dataset.\n", "

\n", "
" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import os\n", "from datetime import datetime\n", "\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "import seaborn as sns\n", "from ydata_profiling import ProfileReport\n", "\n", "sns.set_palette(\"deep\")\n", "import warnings\n", "\n", "warnings.filterwarnings(\"ignore\")" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'/home/code/inesdata-mov/data-generation/data/processed/emt'" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "ROOT_PATH = os.path.dirname(os.path.dirname(os.path.abspath(os.getcwd())))\n", "DATA_PATH = os.path.join(ROOT_PATH, \"data\", \"processed\")\n", "EMT_DATA_PATH = os.path.join(DATA_PATH, \"emt\")\n", "EMT_DATA_PATH" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "('/home/code/inesdata-mov/data-generation/data/processed/emt', ['2024'], [])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024', ['03'], [])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024/03', ['12', '02', '11', '13'], [])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024/03/12', [], ['emt_20240312.csv'])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024/03/02', [], ['emt_20240302.csv'])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024/03/11', [], ['emt_20240311.csv'])\n", "('/home/code/inesdata-mov/data-generation/data/processed/emt/2024/03/13', [], ['emt_20240313.csv:Zone.Identifier', 'emt_20240313.csv'])\n" ] } ], "source": [ "for w in os.walk(EMT_DATA_PATH):\n", " print(w)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Cada fila de este dataset representa el tiempo de espera de un autobus desde una parada A, a una determinada parada B de una cierta línea L, en una fecha y hora concretos.**" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "

-

\n", "

\n", "Vamos a analizar la calidad del dataset generado solamente para el día 13 de marzo, en el futuro dispondremos de más días.\n", "

\n", "
" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
linestopisHeaddestinationdeviationbusestimateArriveDistanceBuspositionTypeBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
02756FalsePLAZA CASTILLA0513473184102024-03-13 08:00:01.7653172024-03-13-3.69054240.423739LA05:5523:303.012.0N
12756FalsePLAZA CASTILLA0521313122102024-03-13 08:00:01.7653172024-03-13-3.68901940.429011LA05:5523:303.012.0N
215056FalseVIRGEN CORTIJO02549547208102024-03-13 08:00:01.7653172024-03-13-3.69160840.421366NaNNaNNaNNaNNaNNaN
315056FalseVIRGEN CORTIJO025611080277902024-03-13 08:00:01.7653172024-03-13-3.69841140.418013NaNNaNNaNNaNNaNNaN
41456FalsePIO XII0557119767902024-03-13 08:00:01.7653172024-03-13-3.68791840.432850NaNNaNNaNNaNNaNNaN
............................................................
113445717451023FalseVALDEBEBAS021411225811402024-03-13 22:59:07.7700612024-03-13-3.66704840.489136LA06:0023:457.022.0N
113445817151023FalseVALDEBEBAS0881021631329002024-03-13 22:59:07.7700612024-03-130.0000000.000000NaNNaNNaNNaNNaNNaN
113445917151023FalseVALDEBEBAS08829396208802024-03-13 22:59:07.7700612024-03-130.0000000.000000NaNNaNNaNNaNNaNNaN
11344601743256FalseVALDEBEBAS0229019601384702024-03-13 22:59:36.3784322024-03-13-3.62309240.482330LA06:0023:457.022.0N
11344611743256FalseVALDEBEBAS08879746489102024-03-13 22:59:36.3784322024-03-13-3.67613740.469620LA06:0023:457.022.0N
\n", "

1134462 rows × 19 columns

\n", "
" ], "text/plain": [ " line stop isHead destination deviation bus estimateArrive \\\n", "0 27 56 False PLAZA CASTILLA 0 513 473 \n", "1 27 56 False PLAZA CASTILLA 0 521 313 \n", "2 150 56 False VIRGEN CORTIJO 0 2549 547 \n", "3 150 56 False VIRGEN CORTIJO 0 2561 1080 \n", "4 14 56 False PIO XII 0 5571 197 \n", "... ... ... ... ... ... ... ... \n", "1134457 174 51023 False VALDEBEBAS 0 2141 1225 \n", "1134458 171 51023 False VALDEBEBAS 0 8810 2163 \n", "1134459 171 51023 False VALDEBEBAS 0 8829 396 \n", "1134460 174 3256 False VALDEBEBAS 0 2290 1960 \n", "1134461 174 3256 False VALDEBEBAS 0 8879 746 \n", "\n", " DistanceBus positionTypeBus datetime date \\\n", "0 1841 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "1 1221 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "2 2081 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "3 2779 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "4 679 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "... ... ... ... ... \n", "1134457 8114 0 2024-03-13 22:59:07.770061 2024-03-13 \n", "1134458 13290 0 2024-03-13 22:59:07.770061 2024-03-13 \n", "1134459 2088 0 2024-03-13 22:59:07.770061 2024-03-13 \n", "1134460 13847 0 2024-03-13 22:59:36.378432 2024-03-13 \n", "1134461 4891 0 2024-03-13 22:59:36.378432 2024-03-13 \n", "\n", " positionBusLon positionBusLat dayType StartTime StopTime \\\n", "0 -3.690542 40.423739 LA 05:55 23:30 \n", "1 -3.689019 40.429011 LA 05:55 23:30 \n", "2 -3.691608 40.421366 NaN NaN NaN \n", "3 -3.698411 40.418013 NaN NaN NaN \n", "4 -3.687918 40.432850 NaN NaN NaN \n", "... ... ... ... ... ... \n", "1134457 -3.667048 40.489136 LA 06:00 23:45 \n", "1134458 0.000000 0.000000 NaN NaN NaN \n", "1134459 0.000000 0.000000 NaN NaN NaN \n", "1134460 -3.623092 40.482330 LA 06:00 23:45 \n", "1134461 -3.676137 40.469620 LA 06:00 23:45 \n", "\n", " MinimunFrequency MaximumFrequency strike \n", "0 3.0 12.0 N \n", "1 3.0 12.0 N \n", "2 NaN NaN NaN \n", "3 NaN NaN NaN \n", "4 NaN NaN NaN \n", "... ... ... ... \n", "1134457 7.0 22.0 N \n", "1134458 NaN NaN NaN \n", "1134459 NaN NaN NaN \n", "1134460 7.0 22.0 N \n", "1134461 7.0 22.0 N \n", "\n", "[1134462 rows x 19 columns]" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_csv(\n", " os.path.join(EMT_DATA_PATH, \"2024\", \"03\", \"13\", \"emt_20240313.csv\"),\n", " parse_dates=[\"date\", \"datetime\"],\n", ")\n", "df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "

Nota

\n", "

\n", "La API de la EMT nos devuelve información sobre los tiempos de llegada de cada parada, con todas sus líneas conectadas.
\n", "En esta solución se ha decidido estudiar solo las líneas de Plaza Castilla, por lo que es necesario realizar un filtrado del dataset.\n", "

\n", "
" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
linestopisHeaddestinationdeviationbusestimateArriveDistanceBuspositionTypeBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
02756FalsePLAZA CASTILLA0513473184102024-03-13 08:00:01.7653172024-03-13-3.69054240.423739LA05:5523:303.012.0N
12756FalsePLAZA CASTILLA0521313122102024-03-13 08:00:01.7653172024-03-13-3.68901940.429011LA05:5523:303.012.0N
122760FalsePLAZA CASTILLA0513282115902024-03-13 08:00:01.8101822024-03-13-3.69054240.423739LA05:5523:303.012.0N
132760FalsePLAZA CASTILLA052111664602024-03-13 08:00:01.8101822024-03-13-3.68901940.429011LA05:5523:303.012.0N
222766FalsePLAZA CASTILLA051312339102024-03-13 08:00:01.8107592024-03-13-3.69054240.423739LA05:5523:303.012.0N
............................................................
113445317451022TruePLAZA CASTILLA02067651370702024-03-13 22:59:07.7486102024-03-13-3.62195440.482080LA06:0023:457.022.0N
113445617451023FalseVALDEBEBAS02067409335202024-03-13 22:59:07.7700612024-03-13-3.62195440.482080LA06:0023:457.022.0N
113445717451023FalseVALDEBEBAS021411225811402024-03-13 22:59:07.7700612024-03-13-3.66704840.489136LA06:0023:457.022.0N
11344601743256FalseVALDEBEBAS0229019601384702024-03-13 22:59:36.3784322024-03-13-3.62309240.482330LA06:0023:457.022.0N
11344611743256FalseVALDEBEBAS08879746489102024-03-13 22:59:36.3784322024-03-13-3.67613740.469620LA06:0023:457.022.0N
\n", "

598334 rows × 19 columns

\n", "
" ], "text/plain": [ " line stop isHead destination deviation bus estimateArrive \\\n", "0 27 56 False PLAZA CASTILLA 0 513 473 \n", "1 27 56 False PLAZA CASTILLA 0 521 313 \n", "12 27 60 False PLAZA CASTILLA 0 513 282 \n", "13 27 60 False PLAZA CASTILLA 0 521 116 \n", "22 27 66 False PLAZA CASTILLA 0 513 123 \n", "... ... ... ... ... ... ... ... \n", "1134453 174 51022 True PLAZA CASTILLA 0 2067 651 \n", "1134456 174 51023 False VALDEBEBAS 0 2067 409 \n", "1134457 174 51023 False VALDEBEBAS 0 2141 1225 \n", "1134460 174 3256 False VALDEBEBAS 0 2290 1960 \n", "1134461 174 3256 False VALDEBEBAS 0 8879 746 \n", "\n", " DistanceBus positionTypeBus datetime date \\\n", "0 1841 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "1 1221 0 2024-03-13 08:00:01.765317 2024-03-13 \n", "12 1159 0 2024-03-13 08:00:01.810182 2024-03-13 \n", "13 646 0 2024-03-13 08:00:01.810182 2024-03-13 \n", "22 391 0 2024-03-13 08:00:01.810759 2024-03-13 \n", "... ... ... ... ... \n", "1134453 3707 0 2024-03-13 22:59:07.748610 2024-03-13 \n", "1134456 3352 0 2024-03-13 22:59:07.770061 2024-03-13 \n", "1134457 8114 0 2024-03-13 22:59:07.770061 2024-03-13 \n", "1134460 13847 0 2024-03-13 22:59:36.378432 2024-03-13 \n", "1134461 4891 0 2024-03-13 22:59:36.378432 2024-03-13 \n", "\n", " positionBusLon positionBusLat dayType StartTime StopTime \\\n", "0 -3.690542 40.423739 LA 05:55 23:30 \n", "1 -3.689019 40.429011 LA 05:55 23:30 \n", "12 -3.690542 40.423739 LA 05:55 23:30 \n", "13 -3.689019 40.429011 LA 05:55 23:30 \n", "22 -3.690542 40.423739 LA 05:55 23:30 \n", "... ... ... ... ... ... \n", "1134453 -3.621954 40.482080 LA 06:00 23:45 \n", "1134456 -3.621954 40.482080 LA 06:00 23:45 \n", "1134457 -3.667048 40.489136 LA 06:00 23:45 \n", "1134460 -3.623092 40.482330 LA 06:00 23:45 \n", "1134461 -3.676137 40.469620 LA 06:00 23:45 \n", "\n", " MinimunFrequency MaximumFrequency strike \n", "0 3.0 12.0 N \n", "1 3.0 12.0 N \n", "12 3.0 12.0 N \n", "13 3.0 12.0 N \n", "22 3.0 12.0 N \n", "... ... ... ... \n", "1134453 7.0 22.0 N \n", "1134456 7.0 22.0 N \n", "1134457 7.0 22.0 N \n", "1134460 7.0 22.0 N \n", "1134461 7.0 22.0 N \n", "\n", "[598334 rows x 19 columns]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "lines = ['27', '42', '49', '67', '70', '107', '129', '134', '135', '173', '174', '175', '176', '177', '178']\n", "\n", "df = df[df[\"line\"].isin(lines)]\n", "df" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "Index: 598334 entries, 0 to 1134461\n", "Data columns (total 19 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 line 598334 non-null object \n", " 1 stop 598334 non-null int64 \n", " 2 isHead 598334 non-null bool \n", " 3 destination 598334 non-null object \n", " 4 deviation 598334 non-null int64 \n", " 5 bus 598334 non-null int64 \n", " 6 estimateArrive 598334 non-null int64 \n", " 7 DistanceBus 598334 non-null int64 \n", " 8 positionTypeBus 598334 non-null int64 \n", " 9 datetime 598334 non-null datetime64[ns]\n", " 10 date 598334 non-null datetime64[ns]\n", " 11 positionBusLon 598334 non-null float64 \n", " 12 positionBusLat 598334 non-null float64 \n", " 13 dayType 598334 non-null object \n", " 14 StartTime 598334 non-null object \n", " 15 StopTime 598334 non-null object \n", " 16 MinimunFrequency 598334 non-null float64 \n", " 17 MaximumFrequency 598334 non-null float64 \n", " 18 strike 598334 non-null object \n", "dtypes: bool(1), datetime64[ns](2), float64(4), int64(6), object(6)\n", "memory usage: 87.3+ MB\n" ] } ], "source": [ "df.info()" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index(['line', 'stop', 'isHead', 'destination', 'deviation', 'bus',\n", " 'estimateArrive', 'DistanceBus', 'positionTypeBus', 'datetime', 'date',\n", " 'positionBusLon', 'positionBusLat', 'dayType', 'StartTime', 'StopTime',\n", " 'MinimunFrequency', 'MaximumFrequency', 'strike'],\n", " dtype='object')" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.columns" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "283" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"stop\"].nunique()" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "15" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"line\"].nunique()" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "182" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[\"bus\"].nunique()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "La variable `positionTypeBus` parece no dar mucha información, y en la documentación de la [API](https://apidocs.emtmadrid.es/#api-Block_3_TRANSPORT_BUSEMTMAD-arrives) no aparece el significado de sus valores. Por tanto, decidimos eliminarla." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "positionTypeBus\n", "0 597488\n", "5 846\n", "Name: count, dtype: int64\n" ] } ], "source": [ "print(df[\"positionTypeBus\"].value_counts())\n", "df.drop(columns=\"positionTypeBus\", inplace=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Conversiones de tipos" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Numeric cols: ['stop', 'deviation', 'bus', 'estimateArrive', 'DistanceBus', 'positionBusLon', 'positionBusLat', 'MinimunFrequency', 'MaximumFrequency']\n", "Categoric cols: ['line', 'isHead', 'destination', 'dayType', 'StartTime', 'StopTime', 'strike']\n", "Date cols: ['datetime', 'date']\n" ] } ], "source": [ "num_cols = list(df.select_dtypes(include=np.number).columns)\n", "cat_cols = list(df.select_dtypes(include=[\"object\", bool]).columns)\n", "date_cols = list(df.select_dtypes(exclude=[np.number, \"object\", bool]).columns)\n", "\n", "print(f\"Numeric cols: {num_cols}\")\n", "print(f\"Categoric cols: {cat_cols}\")\n", "print(f\"Date cols: {date_cols}\")" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "# Convert line, stop and bus to categoric\n", "df[\"stop\"] = df[\"stop\"].astype(\"category\")\n", "df[\"line\"] = df[\"line\"].astype(\"category\")\n", "df[\"bus\"] = df[\"bus\"].astype(\"category\")" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Numeric cols: ['deviation', 'estimateArrive', 'DistanceBus', 'positionBusLon', 'positionBusLat', 'MinimunFrequency', 'MaximumFrequency']\n", "Categoric cols: ['line', 'stop', 'isHead', 'destination', 'bus', 'dayType', 'StartTime', 'StopTime', 'strike']\n", "Date cols: ['datetime', 'date']\n" ] } ], "source": [ "# Update dytpes cols\n", "num_cols = list(df.select_dtypes(include=np.number).columns)\n", "cat_cols = list(df.select_dtypes(include=[\"object\", bool, \"category\"]).columns)\n", "date_cols = list(df.select_dtypes(exclude=[np.number, \"object\", bool, \"category\"]).columns)\n", "\n", "print(f\"Numeric cols: {num_cols}\")\n", "print(f\"Categoric cols: {cat_cols}\")\n", "print(f\"Date cols: {date_cols}\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## QA checks ✅" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Unicidad\n", "Como hemos comentado anteriormente, **cada fila de este dataset representa el tiempo de espera de un autobus desde una parada A, a una determinada parada B de una cierta línea L, en una fecha y hora concretos.** Por tanto, las claves primarias de este dataset se conformarán teniendo en cuenta dichos atributos:" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PKlinestopisHeaddestinationdeviationbusestimateArriveDistanceBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
02024-03-13 08:00:01.765317_B513_L27_S562756FalsePLAZA CASTILLA051347318412024-03-13 08:00:01.7653172024-03-13-3.69054240.423739LA05:5523:303.012.0N
12024-03-13 08:00:01.765317_B521_L27_S562756FalsePLAZA CASTILLA052131312212024-03-13 08:00:01.7653172024-03-13-3.68901940.429011LA05:5523:303.012.0N
122024-03-13 08:00:01.810182_B513_L27_S602760FalsePLAZA CASTILLA051328211592024-03-13 08:00:01.8101822024-03-13-3.69054240.423739LA05:5523:303.012.0N
132024-03-13 08:00:01.810182_B521_L27_S602760FalsePLAZA CASTILLA05211166462024-03-13 08:00:01.8101822024-03-13-3.68901940.429011LA05:5523:303.012.0N
222024-03-13 08:00:01.810759_B513_L27_S662766FalsePLAZA CASTILLA05131233912024-03-13 08:00:01.8107592024-03-13-3.69054240.423739LA05:5523:303.012.0N
............................................................
11344532024-03-13 22:59:07.748610_B2067_L174_S5102217451022TruePLAZA CASTILLA0206765137072024-03-13 22:59:07.7486102024-03-13-3.62195440.482080LA06:0023:457.022.0N
11344562024-03-13 22:59:07.770061_B2067_L174_S5102317451023FalseVALDEBEBAS0206740933522024-03-13 22:59:07.7700612024-03-13-3.62195440.482080LA06:0023:457.022.0N
11344572024-03-13 22:59:07.770061_B2141_L174_S5102317451023FalseVALDEBEBAS02141122581142024-03-13 22:59:07.7700612024-03-13-3.66704840.489136LA06:0023:457.022.0N
11344602024-03-13 22:59:36.378432_B2290_L174_S32561743256FalseVALDEBEBAS022901960138472024-03-13 22:59:36.3784322024-03-13-3.62309240.482330LA06:0023:457.022.0N
11344612024-03-13 22:59:36.378432_B8879_L174_S32561743256FalseVALDEBEBAS0887974648912024-03-13 22:59:36.3784322024-03-13-3.67613740.469620LA06:0023:457.022.0N
\n", "

598334 rows × 19 columns

\n", "
" ], "text/plain": [ " PK line stop isHead \\\n", "0 2024-03-13 08:00:01.765317_B513_L27_S56 27 56 False \n", "1 2024-03-13 08:00:01.765317_B521_L27_S56 27 56 False \n", "12 2024-03-13 08:00:01.810182_B513_L27_S60 27 60 False \n", "13 2024-03-13 08:00:01.810182_B521_L27_S60 27 60 False \n", "22 2024-03-13 08:00:01.810759_B513_L27_S66 27 66 False \n", "... ... ... ... ... \n", "1134453 2024-03-13 22:59:07.748610_B2067_L174_S51022 174 51022 True \n", "1134456 2024-03-13 22:59:07.770061_B2067_L174_S51023 174 51023 False \n", "1134457 2024-03-13 22:59:07.770061_B2141_L174_S51023 174 51023 False \n", "1134460 2024-03-13 22:59:36.378432_B2290_L174_S3256 174 3256 False \n", "1134461 2024-03-13 22:59:36.378432_B8879_L174_S3256 174 3256 False \n", "\n", " destination deviation bus estimateArrive DistanceBus \\\n", "0 PLAZA CASTILLA 0 513 473 1841 \n", "1 PLAZA CASTILLA 0 521 313 1221 \n", "12 PLAZA CASTILLA 0 513 282 1159 \n", "13 PLAZA CASTILLA 0 521 116 646 \n", "22 PLAZA CASTILLA 0 513 123 391 \n", "... ... ... ... ... ... \n", "1134453 PLAZA CASTILLA 0 2067 651 3707 \n", "1134456 VALDEBEBAS 0 2067 409 3352 \n", "1134457 VALDEBEBAS 0 2141 1225 8114 \n", "1134460 VALDEBEBAS 0 2290 1960 13847 \n", "1134461 VALDEBEBAS 0 8879 746 4891 \n", "\n", " datetime date positionBusLon positionBusLat \\\n", "0 2024-03-13 08:00:01.765317 2024-03-13 -3.690542 40.423739 \n", "1 2024-03-13 08:00:01.765317 2024-03-13 -3.689019 40.429011 \n", "12 2024-03-13 08:00:01.810182 2024-03-13 -3.690542 40.423739 \n", "13 2024-03-13 08:00:01.810182 2024-03-13 -3.689019 40.429011 \n", "22 2024-03-13 08:00:01.810759 2024-03-13 -3.690542 40.423739 \n", "... ... ... ... ... \n", "1134453 2024-03-13 22:59:07.748610 2024-03-13 -3.621954 40.482080 \n", "1134456 2024-03-13 22:59:07.770061 2024-03-13 -3.621954 40.482080 \n", "1134457 2024-03-13 22:59:07.770061 2024-03-13 -3.667048 40.489136 \n", "1134460 2024-03-13 22:59:36.378432 2024-03-13 -3.623092 40.482330 \n", "1134461 2024-03-13 22:59:36.378432 2024-03-13 -3.676137 40.469620 \n", "\n", " dayType StartTime StopTime MinimunFrequency MaximumFrequency strike \n", "0 LA 05:55 23:30 3.0 12.0 N \n", "1 LA 05:55 23:30 3.0 12.0 N \n", "12 LA 05:55 23:30 3.0 12.0 N \n", "13 LA 05:55 23:30 3.0 12.0 N \n", "22 LA 05:55 23:30 3.0 12.0 N \n", "... ... ... ... ... ... ... \n", "1134453 LA 06:00 23:45 7.0 22.0 N \n", "1134456 LA 06:00 23:45 7.0 22.0 N \n", "1134457 LA 06:00 23:45 7.0 22.0 N \n", "1134460 LA 06:00 23:45 7.0 22.0 N \n", "1134461 LA 06:00 23:45 7.0 22.0 N \n", "\n", "[598334 rows x 19 columns]" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Create dataset primary key\n", "df.insert(0, \"PK\", \"\")\n", "df[\"PK\"] = (\n", " df[\"datetime\"].astype(str)\n", " + \"_B\"\n", " + df[\"bus\"].astype(str)\n", " + \"_L\"\n", " + df[\"line\"].astype(str)\n", " + \"_S\"\n", " + df[\"stop\"].astype(str)\n", ")\n", "df" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "PK/Unique identifier check\n", "❌ PK is not unique\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PKdatetimebuslinestopestimateArrive
9301382024-03-13 20:16:53.420475_B581_L70_S2502024-03-13 20:16:53.42047558170250999999
11284522024-03-13 22:54:53.562759_B2260_L129_S1362024-03-13 22:54:53.5627592260129136999999
11284972024-03-13 22:54:53.609227_B2260_L129_S2032024-03-13 22:54:53.6092272260129203999999
11284992024-03-13 22:54:53.613518_B2260_L129_S1302024-03-13 22:54:53.6135182260129130999999
11286822024-03-13 22:54:53.740388_B2260_L129_S1342024-03-13 22:54:53.7403882260129134999999
.....................
11339762024-03-13 22:59:02.690123_B2260_L129_S1382024-03-13 22:59:02.6901232260129138999999
11341692024-03-13 22:59:04.626340_B2260_L129_S21502024-03-13 22:59:04.62634022601292150999999
11341932024-03-13 22:59:04.689090_B2260_L129_S32452024-03-13 22:59:04.68909022601293245999999
11342052024-03-13 22:59:04.703750_B2260_L129_S32472024-03-13 22:59:04.70375022601293247999999
11343762024-03-13 22:59:06.845125_B2263_L135_S56042024-03-13 22:59:06.84512522631355604999999
\n", "

98 rows × 6 columns

\n", "
" ], "text/plain": [ " PK \\\n", "930138 2024-03-13 20:16:53.420475_B581_L70_S250 \n", "1128452 2024-03-13 22:54:53.562759_B2260_L129_S136 \n", "1128497 2024-03-13 22:54:53.609227_B2260_L129_S203 \n", "1128499 2024-03-13 22:54:53.613518_B2260_L129_S130 \n", "1128682 2024-03-13 22:54:53.740388_B2260_L129_S134 \n", "... ... \n", "1133976 2024-03-13 22:59:02.690123_B2260_L129_S138 \n", "1134169 2024-03-13 22:59:04.626340_B2260_L129_S2150 \n", "1134193 2024-03-13 22:59:04.689090_B2260_L129_S3245 \n", "1134205 2024-03-13 22:59:04.703750_B2260_L129_S3247 \n", "1134376 2024-03-13 22:59:06.845125_B2263_L135_S5604 \n", "\n", " datetime bus line stop estimateArrive \n", "930138 2024-03-13 20:16:53.420475 581 70 250 999999 \n", "1128452 2024-03-13 22:54:53.562759 2260 129 136 999999 \n", "1128497 2024-03-13 22:54:53.609227 2260 129 203 999999 \n", "1128499 2024-03-13 22:54:53.613518 2260 129 130 999999 \n", "1128682 2024-03-13 22:54:53.740388 2260 129 134 999999 \n", "... ... ... ... ... ... \n", "1133976 2024-03-13 22:59:02.690123 2260 129 138 999999 \n", "1134169 2024-03-13 22:59:04.626340 2260 129 2150 999999 \n", "1134193 2024-03-13 22:59:04.689090 2260 129 3245 999999 \n", "1134205 2024-03-13 22:59:04.703750 2260 129 3247 999999 \n", "1134376 2024-03-13 22:59:06.845125 2263 135 5604 999999 \n", "\n", "[98 rows x 6 columns]" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "print(\"PK/Unique identifier check\")\n", "if df[\"PK\"].nunique() == df.shape[0]:\n", " print(\"✅ PK is unique\")\n", " # As we passed the PK quality check, we can set this PK as dataframe index\n", " df.set_index(\"PK\", inplace=True)\n", "else:\n", " print(\"❌ PK is not unique\")\n", " display(df[df[\"PK\"].duplicated()][[\"PK\", \"datetime\", \"bus\", \"line\", \"stop\", \"estimateArrive\"]])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "

-

\n", "

\n", "Con este check de unicidad nos damos cuenta de que la clave primaria del dataset no es única. En algunos casos, para el mismo datetime, bus, línea y parada, tenemos dos estimaciones de tiempos de llegada diferentes, donde una de ellas siempre es 9999999, que es el valor por defecto que indica que el ETA sería mayor de 45 minutos.\n", "

\n", "
" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "(598334, 19)\n", "(598236, 19)\n", "(598334, 19)\n", "(598334, 19)\n" ] } ], "source": [ "print(df.shape)\n", "print(df.drop_duplicates(subset=[\"datetime\", \"bus\", \"line\", \"stop\"]).shape)\n", "print(df.drop_duplicates(subset=[\"datetime\", \"bus\", \"line\", \"stop\", \"estimateArrive\"]).shape)\n", "print(df.drop_duplicates(subset=[\"datetime\", \"bus\", \"line\", \"stop\", \"DistanceBus\"]).shape)" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PKlinestopisHeaddestinationdeviationbusestimateArriveDistanceBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
9301382024-03-13 20:16:53.420475_B581_L70_S25070250FalseALSACIA0581999999208102024-03-13 20:16:53.4204752024-03-13-3.63569240.434676LA05:3023:454.017.0N
11284522024-03-13 22:54:53.562759_B2260_L129_S136129136FalseMANOTERAS02260999999166482024-03-13 22:54:53.5627592024-03-13-3.70395440.462515LA06:0023:4512.024.0N
11284972024-03-13 22:54:53.609227_B2260_L129_S203129203FalseMANOTERAS02260999999132752024-03-13 22:54:53.6092272024-03-13-3.70395440.462515LA06:0023:4512.024.0N
11284992024-03-13 22:54:53.613518_B2260_L129_S130129130FalseMANOTERAS02260999999157172024-03-13 22:54:53.6135182024-03-13-3.70395440.462515LA06:0023:4512.024.0N
11286822024-03-13 22:54:53.740388_B2260_L129_S134129134FalseMANOTERAS02260999999164442024-03-13 22:54:53.7403882024-03-13-3.70395440.462515LA06:0023:4512.024.0N
............................................................
11339762024-03-13 22:59:02.690123_B2260_L129_S138129138FalseMANOTERAS02260999999157202024-03-13 22:59:02.6901232024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11341692024-03-13 22:59:04.626340_B2260_L129_S21501292150FalseMANOTERAS02260999999131932024-03-13 22:59:04.6263402024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11341932024-03-13 22:59:04.689090_B2260_L129_S32451293245FalseMANOTERAS02260999999162792024-03-13 22:59:04.6890902024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11342052024-03-13 22:59:04.703750_B2260_L129_S32471293247FalseMANOTERAS02260999999165522024-03-13 22:59:04.7037502024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11343762024-03-13 22:59:06.845125_B2263_L135_S56041355604TrueHOSPITAL RAMON Y CAJAL0226399999989552024-03-13 22:59:06.8451252024-03-13-3.69234440.487187LA06:3023:459.029.0N
\n", "

98 rows × 19 columns

\n", "
" ], "text/plain": [ " PK line stop isHead \\\n", "930138 2024-03-13 20:16:53.420475_B581_L70_S250 70 250 False \n", "1128452 2024-03-13 22:54:53.562759_B2260_L129_S136 129 136 False \n", "1128497 2024-03-13 22:54:53.609227_B2260_L129_S203 129 203 False \n", "1128499 2024-03-13 22:54:53.613518_B2260_L129_S130 129 130 False \n", "1128682 2024-03-13 22:54:53.740388_B2260_L129_S134 129 134 False \n", "... ... ... ... ... \n", "1133976 2024-03-13 22:59:02.690123_B2260_L129_S138 129 138 False \n", "1134169 2024-03-13 22:59:04.626340_B2260_L129_S2150 129 2150 False \n", "1134193 2024-03-13 22:59:04.689090_B2260_L129_S3245 129 3245 False \n", "1134205 2024-03-13 22:59:04.703750_B2260_L129_S3247 129 3247 False \n", "1134376 2024-03-13 22:59:06.845125_B2263_L135_S5604 135 5604 True \n", "\n", " destination deviation bus estimateArrive DistanceBus \\\n", "930138 ALSACIA 0 581 999999 20810 \n", "1128452 MANOTERAS 0 2260 999999 16648 \n", "1128497 MANOTERAS 0 2260 999999 13275 \n", "1128499 MANOTERAS 0 2260 999999 15717 \n", "1128682 MANOTERAS 0 2260 999999 16444 \n", "... ... ... ... ... ... \n", "1133976 MANOTERAS 0 2260 999999 15720 \n", "1134169 MANOTERAS 0 2260 999999 13193 \n", "1134193 MANOTERAS 0 2260 999999 16279 \n", "1134205 MANOTERAS 0 2260 999999 16552 \n", "1134376 HOSPITAL RAMON Y CAJAL 0 2263 999999 8955 \n", "\n", " datetime date positionBusLon positionBusLat \\\n", "930138 2024-03-13 20:16:53.420475 2024-03-13 -3.635692 40.434676 \n", "1128452 2024-03-13 22:54:53.562759 2024-03-13 -3.703954 40.462515 \n", "1128497 2024-03-13 22:54:53.609227 2024-03-13 -3.703954 40.462515 \n", "1128499 2024-03-13 22:54:53.613518 2024-03-13 -3.703954 40.462515 \n", "1128682 2024-03-13 22:54:53.740388 2024-03-13 -3.703954 40.462515 \n", "... ... ... ... ... \n", "1133976 2024-03-13 22:59:02.690123 2024-03-13 -3.688213 40.467646 \n", "1134169 2024-03-13 22:59:04.626340 2024-03-13 -3.688213 40.467646 \n", "1134193 2024-03-13 22:59:04.689090 2024-03-13 -3.688213 40.467646 \n", "1134205 2024-03-13 22:59:04.703750 2024-03-13 -3.688213 40.467646 \n", "1134376 2024-03-13 22:59:06.845125 2024-03-13 -3.692344 40.487187 \n", "\n", " dayType StartTime StopTime MinimunFrequency MaximumFrequency strike \n", "930138 LA 05:30 23:45 4.0 17.0 N \n", "1128452 LA 06:00 23:45 12.0 24.0 N \n", "1128497 LA 06:00 23:45 12.0 24.0 N \n", "1128499 LA 06:00 23:45 12.0 24.0 N \n", "1128682 LA 06:00 23:45 12.0 24.0 N \n", "... ... ... ... ... ... ... \n", "1133976 LA 06:00 23:45 12.0 24.0 N \n", "1134169 LA 06:00 23:45 12.0 24.0 N \n", "1134193 LA 06:00 23:45 12.0 24.0 N \n", "1134205 LA 06:00 23:45 12.0 24.0 N \n", "1134376 LA 06:30 23:45 9.0 29.0 N \n", "\n", "[98 rows x 19 columns]" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df = df[df[\"PK\"].duplicated()]\n", "duplicated_df" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[581, 2260, 2263]\n", "Categories (182, int64): [501, 504, 505, 506, ..., 9120, 9124, 9125, 9130]" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df[[\"datetime\", \"line\", \"stop\", \"bus\"]][\"bus\"].unique()" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['70', '129', '135']\n", "Categories (15, object): ['107', '129', '134', '135', ..., '42', '49', '67', '70']" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df[[\"datetime\", \"line\", \"stop\", \"bus\"]][\"line\"].unique()" ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[250, 136, 203, 130, 134, ..., 140, 509, 2148, 3247, 5606]\n", "Length: 26\n", "Categories (283, int64): [29, 30, 33, 35, ..., 5984, 5996, 51022, 51023]" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df[[\"datetime\", \"line\", \"stop\", \"bus\"]][\"stop\"].unique()" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([999999, 2042, 1981])" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df[[\"datetime\", \"line\", \"stop\", \"bus\", \"estimateArrive\"]][\"estimateArrive\"].unique()" ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "estimateArrive\n", "1981 1\n", "2042 1\n", "999999 96\n", "Name: PK, dtype: int64" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "duplicated_df[[\"PK\", \"datetime\", \"line\", \"stop\", \"bus\", \"estimateArrive\"]].groupby(\n", " \"estimateArrive\"\n", ")[\"PK\"].nunique()" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
estimateArrive
PK
2024-03-13 22:55:53.067079_B2260_L129_S1322
2024-03-13 22:54:57.042314_B2260_L129_S35942
2024-03-13 22:56:53.304306_B2260_L129_S5102
2024-03-13 22:56:53.308814_B2260_L129_S1262
2024-03-13 22:57:02.752661_B2260_L129_S2032
......
2024-03-13 13:04:08.893280_B2072_L174_S510231
2024-03-13 13:04:08.893280_B8879_L174_S510231
2024-03-13 13:04:26.438469_B2079_L134_S56101
2024-03-13 13:04:26.438469_B2468_L134_S56101
2024-03-13 22:59:36.378432_B8879_L174_S32561
\n", "

598236 rows × 1 columns

\n", "
" ], "text/plain": [ " estimateArrive\n", "PK \n", "2024-03-13 22:55:53.067079_B2260_L129_S132 2\n", "2024-03-13 22:54:57.042314_B2260_L129_S3594 2\n", "2024-03-13 22:56:53.304306_B2260_L129_S510 2\n", "2024-03-13 22:56:53.308814_B2260_L129_S126 2\n", "2024-03-13 22:57:02.752661_B2260_L129_S203 2\n", "... ...\n", "2024-03-13 13:04:08.893280_B2072_L174_S51023 1\n", "2024-03-13 13:04:08.893280_B8879_L174_S51023 1\n", "2024-03-13 13:04:26.438469_B2079_L134_S5610 1\n", "2024-03-13 13:04:26.438469_B2468_L134_S5610 1\n", "2024-03-13 22:59:36.378432_B8879_L174_S3256 1\n", "\n", "[598236 rows x 1 columns]" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gr = (\n", " df[[\"PK\", \"datetime\", \"line\", \"stop\", \"bus\", \"estimateArrive\"]]\n", " .groupby(\"PK\")[[\"estimateArrive\"]]\n", " .nunique()\n", " .sort_values(by=\"estimateArrive\", ascending=False)\n", ")\n", "gr" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PKlinestopisHeaddestinationdeviationbusestimateArriveDistanceBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
9301372024-03-13 20:16:53.420475_B581_L70_S25070250FalseALSACIA058136213132024-03-13 20:16:53.4204752024-03-13-3.63569240.434676LA05:3023:454.017.0N
9301382024-03-13 20:16:53.420475_B581_L70_S25070250FalseALSACIA0581999999208102024-03-13 20:16:53.4204752024-03-13-3.63569240.434676LA05:3023:454.017.0N
11284512024-03-13 22:54:53.562759_B2260_L129_S136129136FalseMANOTERAS0226095742102024-03-13 22:54:53.5627592024-03-13-3.70395440.462515LA06:0023:4512.024.0N
11284522024-03-13 22:54:53.562759_B2260_L129_S136129136FalseMANOTERAS02260999999166482024-03-13 22:54:53.5627592024-03-13-3.70395440.462515LA06:0023:4512.024.0N
11284962024-03-13 22:54:53.609227_B2260_L129_S203129203FalseMANOTERAS022604407692024-03-13 22:54:53.6092272024-03-13-3.70395440.462515LA06:0023:4512.024.0N
............................................................
11341932024-03-13 22:59:04.689090_B2260_L129_S32451293245FalseMANOTERAS02260999999162792024-03-13 22:59:04.6890902024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11342042024-03-13 22:59:04.703750_B2260_L129_S32471293247FalseMANOTERAS0226091550192024-03-13 22:59:04.7037502024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11342052024-03-13 22:59:04.703750_B2260_L129_S32471293247FalseMANOTERAS02260999999165522024-03-13 22:59:04.7037502024-03-13-3.68821340.467646LA06:0023:4512.024.0N
11343752024-03-13 22:59:06.845125_B2263_L135_S56041355604TrueHOSPITAL RAMON Y CAJAL02263101128722024-03-13 22:59:06.8451252024-03-13-3.69234440.487187LA06:3023:459.029.0N
11343762024-03-13 22:59:06.845125_B2263_L135_S56041355604TrueHOSPITAL RAMON Y CAJAL0226399999989552024-03-13 22:59:06.8451252024-03-13-3.69234440.487187LA06:3023:459.029.0N
\n", "

196 rows × 19 columns

\n", "
" ], "text/plain": [ " PK line stop isHead \\\n", "930137 2024-03-13 20:16:53.420475_B581_L70_S250 70 250 False \n", "930138 2024-03-13 20:16:53.420475_B581_L70_S250 70 250 False \n", "1128451 2024-03-13 22:54:53.562759_B2260_L129_S136 129 136 False \n", "1128452 2024-03-13 22:54:53.562759_B2260_L129_S136 129 136 False \n", "1128496 2024-03-13 22:54:53.609227_B2260_L129_S203 129 203 False \n", "... ... ... ... ... \n", "1134193 2024-03-13 22:59:04.689090_B2260_L129_S3245 129 3245 False \n", "1134204 2024-03-13 22:59:04.703750_B2260_L129_S3247 129 3247 False \n", "1134205 2024-03-13 22:59:04.703750_B2260_L129_S3247 129 3247 False \n", "1134375 2024-03-13 22:59:06.845125_B2263_L135_S5604 135 5604 True \n", "1134376 2024-03-13 22:59:06.845125_B2263_L135_S5604 135 5604 True \n", "\n", " destination deviation bus estimateArrive DistanceBus \\\n", "930137 ALSACIA 0 581 362 1313 \n", "930138 ALSACIA 0 581 999999 20810 \n", "1128451 MANOTERAS 0 2260 957 4210 \n", "1128452 MANOTERAS 0 2260 999999 16648 \n", "1128496 MANOTERAS 0 2260 440 769 \n", "... ... ... ... ... ... \n", "1134193 MANOTERAS 0 2260 999999 16279 \n", "1134204 MANOTERAS 0 2260 915 5019 \n", "1134205 MANOTERAS 0 2260 999999 16552 \n", "1134375 HOSPITAL RAMON Y CAJAL 0 2263 1011 2872 \n", "1134376 HOSPITAL RAMON Y CAJAL 0 2263 999999 8955 \n", "\n", " datetime date positionBusLon positionBusLat \\\n", "930137 2024-03-13 20:16:53.420475 2024-03-13 -3.635692 40.434676 \n", "930138 2024-03-13 20:16:53.420475 2024-03-13 -3.635692 40.434676 \n", "1128451 2024-03-13 22:54:53.562759 2024-03-13 -3.703954 40.462515 \n", "1128452 2024-03-13 22:54:53.562759 2024-03-13 -3.703954 40.462515 \n", "1128496 2024-03-13 22:54:53.609227 2024-03-13 -3.703954 40.462515 \n", "... ... ... ... ... \n", "1134193 2024-03-13 22:59:04.689090 2024-03-13 -3.688213 40.467646 \n", "1134204 2024-03-13 22:59:04.703750 2024-03-13 -3.688213 40.467646 \n", "1134205 2024-03-13 22:59:04.703750 2024-03-13 -3.688213 40.467646 \n", "1134375 2024-03-13 22:59:06.845125 2024-03-13 -3.692344 40.487187 \n", "1134376 2024-03-13 22:59:06.845125 2024-03-13 -3.692344 40.487187 \n", "\n", " dayType StartTime StopTime MinimunFrequency MaximumFrequency strike \n", "930137 LA 05:30 23:45 4.0 17.0 N \n", "930138 LA 05:30 23:45 4.0 17.0 N \n", "1128451 LA 06:00 23:45 12.0 24.0 N \n", "1128452 LA 06:00 23:45 12.0 24.0 N \n", "1128496 LA 06:00 23:45 12.0 24.0 N \n", "... ... ... ... ... ... ... \n", "1134193 LA 06:00 23:45 12.0 24.0 N \n", "1134204 LA 06:00 23:45 12.0 24.0 N \n", "1134205 LA 06:00 23:45 12.0 24.0 N \n", "1134375 LA 06:30 23:45 9.0 29.0 N \n", "1134376 LA 06:30 23:45 9.0 29.0 N \n", "\n", "[196 rows x 19 columns]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df[(df[\"PK\"].isin(gr[gr[\"estimateArrive\"] > 1].index))]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exactitud y Completitud" ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuray Checks for stop\n", " Expected stops are : [29, 30, 33, 35, 37, 39, 42, 44, 47, 50, 52, 54, 56, 60, 62, 65, 66, 72, 78, 82, 85, 86, 87, 88, 89, 126, 128, 130, 132, 134, 136, 138, 140, 145, 146, 203, 205, 207, 208, 210, 212, 214, 215, 218, 220, 222, 224, 226, 228, 230, 232, 234, 236, 240, 242, 244, 246, 250, 253, 269, 271, 399, 402, 497, 499, 501, 503, 509, 510, 1021, 1022, 1025, 1357, 1360, 1364, 1366, 1368, 1374, 1376, 1377, 1487, 1488, 1489, 1494, 1496, 1498, 1500, 1501, 1503, 1505, 1507, 1509, 1511, 1515, 1517, 1530, 1532, 1537, 1542, 1544, 1546, 1548, 1550, 1552, 1554, 1556, 1558, 1602, 1604, 1606, 1608, 1610, 1612, 1615, 1617, 1619, 1621, 1623, 1625, 1627, 1629, 1631, 1633, 1732, 1734, 1736, 1739, 1741, 1743, 1745, 1749, 1751, 1753, 1760, 1762, 1837, 1839, 1841, 1843, 1845, 1846, 1848, 1850, 1851, 1852, 1866, 2148, 2150, 2653, 2864, 2968, 2970, 3245, 3247, 3249, 3252, 3253, 3254, 3256, 3258, 3260, 3261, 3267, 3269, 3270, 3272, 3279, 3544, 3566, 3568, 3578, 3580, 3594, 3603, 3605, 3620, 3621, 3623, 3634, 3664, 3670, 3671, 3683, 3733, 3769, 3771, 3778, 3790, 3794, 3796, 3821, 3822, 3823, 3826, 3869, 3870, 3872, 3874, 3918, 4033, 4230, 4264, 4266, 4268, 4273, 4354, 4366, 4382, 4493, 4495, 4501, 4502, 4508, 4651, 4708, 4731, 4732, 4734, 4736, 4741, 4752, 4777, 4968, 5016, 5018, 5020, 5133, 5267, 5329, 5333, 5337, 5395, 5397, 5399, 5443, 5458, 5511, 5517, 5518, 5602, 5603, 5604, 5605, 5606, 5607, 5608, 5609, 5610, 5611, 5612, 5632, 5633, 5634, 5635, 5636, 5722, 5747, 5798, 5799, 5800, 5801, 5802, 5803, 5804, 5805, 5806, 5807, 5887, 5892, 5894, 5899, 5911, 5915, 5917, 5919, 5921, 5926, 5927, 5982, 5984, 5996, 51022, 51023]\n", "✅ - No stops apart from Plaza Castilla stops\n" ] } ], "source": [ "print(\"Accuray Checks for stop\")\n", "stops = ['29', '30', '33', '35', '37', '39', '42', '44', '47', '50', '52', '54', '56', '60', '62', '65', '66', '72', '78', '82', '85', '86', '87', '88', '89', '126', '128', '130', '132', '134', '136', '138', '140', '145', '146', '203', '205', '207', '208', '210', '212', '214', '215', '218', '220', '222', '224', '226', '228', '230', '232', '234', '236', '240', '242', '244', '246', '250', '253', '269', '271', '399', '402', '497', '499', '501', '503', '509', '510', '1021', '1022', '1025', '1357', '1360', '1364', '1366', '1368', '1374', '1376', '1377', '1487', '1488', '1489', '1494', '1496', '1498', '1500', '1501', '1503', '1505', '1507', '1509', '1511', '1515', '1517', '1530', '1532', '1537', '1542', '1544', '1546', '1548', '1550', '1552', '1554', '1556', '1558', '1602', '1604', '1606', '1608', '1610', '1612', '1615', '1617', '1619', '1621', '1623', '1625', '1627', '1629', '1631', '1633', '1732', '1734', '1736', '1739', '1741', '1743', '1745', '1749', '1751', '1753', '1760', '1762', '1837', '1839', '1841', '1843', '1845', '1846', '1848', '1850', '1851', '1852', '1866', '2148', '2150', '2653', '2864', '2968', '2970', '3245', '3247', '3249', '3252', '3253', '3254', '3256', '3258', '3260', '3261', '3267', '3269', '3270', '3272', '3279', '3544', '3566', '3568', '3578', '3580', '3594', '3603', '3605', '3620', '3621', '3623', '3634', '3664', '3670', '3671', '3683', '3733', '3769', '3771', '3778', '3790', '3794', '3796', '3821', '3822', '3823', '3826', '3869', '3870', '3872', '3874', '3918', '4033', '4230', '4264', '4266', '4268', '4273', '4354', '4366', '4382', '4493', '4495', '4501', '4502', '4508', '4651', '4708', '4731', '4732', '4734', '4736', '4741', '4752', '4777', '4968', '5016', '5018', '5020', '5133', '5267', '5329', '5333', '5337', '5395', '5397', '5399', '5443', '5458', '5511', '5517', '5518', '5602', '5603', '5604', '5605', '5606', '5607', '5608', '5609', '5610', '5611', '5612', '5632', '5633', '5634', '5635', '5636', '5722', '5747', '5798', '5799', '5800', '5801', '5802', '5803', '5804', '5805', '5806', '5807', '5887', '5892', '5894', '5899', '5911', '5915', '5917', '5919', '5921', '5926', '5927', '5982', '5984', '5996', '51022', '51023']\n", "stops = [eval(stop) for stop in stops]\n", "print(\" Expected stops are : \", stops)\n", "if sum(~df[\"stop\"].isin(stops)) == 0:\n", " print(\"✅ - No stops apart from Plaza Castilla stops\")\n", "else:\n", " print(\"❌ - Below records does belong to expected stops\")\n", " display(df[~df[\"stop\"].isin(stops)][[\"stop\"]])" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Completeness Check for stop\n", "✅ - Data Set have all stops\n" ] } ], "source": [ "print(\"Completeness Check for stop\")\n", "l1 = df[\"stop\"].unique().tolist()\n", "l1.sort()\n", "l2 = stops\n", "if l1 == l2:\n", " print(\"✅ - Data Set have all stops\")\n", "else:\n", " print(\"❌ - Data Set does not have all stops\")\n", " print(\n", " f\"Expected stops are {stops.split('|')}, but we have only stops {df['stop'].unique().tolist()}\"\n", " )" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Consistencia y Validez" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Buses de cada línea" ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['Buses de la línea 27: [513, 521, 526, 537, 511, 524, 510, 528, 522, 538, 505, 520, 527, 504, 525, 514, 508, 530, 507, 506, 501, 531, 516, 536, 532]',\n", " 'Buses de la línea 42: [5546, 5551, 5552, 5547, 5548, 5541, 5553, 5550, 4965]',\n", " 'Buses de la línea 49: [4724, 4737, 4728, 4739, 4725, 4736, 4735, 4733, 4734, 4926, 4833, 4859, 4726, 4834, 4727, 4740, 4860, 4835, 4730, 4731, 4741, 4723, 4689, 4710]',\n", " 'Buses de la línea 67: [5565, 5633, 5562, 5566, 5532, 5564, 5568, 5567, 5563, 5559, 5720]',\n", " 'Buses de la línea 70: [542, 543, 544, 584, 576, 577, 579, 517, 545, 586, 589, 541, 578, 588, 581, 518, 585, 540, 587, 582]',\n", " 'Buses de la línea 107: [2135, 2157, 2133, 2132, 2134]',\n", " 'Buses de la línea 129: [2136, 2138, 2140, 2131, 2137, 2260, 2561]',\n", " 'Buses de la línea 134: [2469, 2472, 2465, 2467, 2589, 2470, 2471, 2474, 2464, 2473, 2468, 2079, 2450]',\n", " 'Buses de la línea 135: [2263, 2341, 2340, 2338]',\n", " 'Buses de la línea 173: [2264, 2496, 2260, 2497, 2495, 2266, 2490, 2493, 2068, 2154, 2545]',\n", " 'Buses de la línea 174: [2071, 2163, 2066, 2141, 2067, 2068, 2063, 2299, 2072, 2065, 2290, 2077, 8879]',\n", " 'Buses de la línea 175: [2512, 2513, 2514, 2569, 2147, 2517, 2516, 2519, 2543]',\n", " 'Buses de la línea 176: [8847, 8860, 8841, 8861, 8845, 8854, 8850, 8853, 8846, 8857, 8852, 2052, 8864, 8856, 8851]',\n", " 'Buses de la línea 177: [9109, 9120, 9125, 9124, 9117, 9130]',\n", " 'Buses de la línea 178: [2501, 2558, 2502, 2259, 2506, 2505, 2584, 2507, 2503, 2571, 2305, 2310]']" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "[f\"Buses de la línea {line}: {list(df[df['line'] == line]['bus'].unique())}\" for line in lines]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Paradas de cada línea" ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['Paradas de la línea 27: [56, 60, 66, 52, 85, 50, 72, 87, 88, 5333, 5511, 39, 65, 62, 42, 29, 35, 44, 47, 82, 54, 89, 33, 86, 5443, 5602, 78, 37]',\n", " 'Paradas de la línea 42: [1509, 1511, 1503, 3568, 5020, 5016, 1507, 1517, 1515, 1500, 1501, 1494, 1498, 1496, 3580, 3733, 3821, 1505, 3578, 5018, 5611, 5996]',\n", " 'Paradas de la línea 49: [1530, 1537, 1366, 1368, 1532, 1503, 1364, 1548, 1554, 1558, 4264, 4273, 4354, 4734, 4736, 4777, 5632, 5635, 1357, 1544, 1546, 1550, 1552, 1556, 1505, 4732, 5329, 3822, 3823, 5634, 4731, 5636, 5611, 5633, 5982, 1542]',\n", " 'Paradas de la línea 67: [1360, 1377, 1602, 1606, 1604, 1610, 1612, 1621, 1625, 1623, 1627, 1633, 4741, 1608, 1617, 1631, 1489, 1374, 1376, 1550, 1615, 1619, 1629, 1487, 3826, 5267, 5605]',\n", " 'Paradas de la línea 70: [215, 208, 218, 222, 220, 210, 236, 224, 230, 242, 4230, 4708, 203, 207, 232, 228, 253, 250, 212, 205, 30, 244, 246, 240, 234, 226, 5458, 5603, 5747]',\n", " 'Paradas de la línea 107: [208, 214, 210, 499, 1846, 1850, 5722, 203, 207, 205, 30, 501, 497, 503, 1839, 1845, 1852, 1837, 1841, 1843, 1848, 1851, 4752, 5337]',\n", " 'Paradas de la línea 129: [126, 146, 128, 130, 510, 2150, 3249, 5133, 5606, 138, 203, 134, 207, 132, 205, 30, 140, 145, 136, 509, 2148, 3245, 3247, 3594]',\n", " 'Paradas de la línea 134: [1022, 1548, 1732, 1743, 1745, 1749, 3623, 1025, 1489, 1739, 1741, 1753, 1544, 2864, 1546, 1734, 1736, 1751, 3279, 1760, 1762, 2968, 2970, 1487, 1021, 3634, 4968, 5329, 3826, 4651, 5610, 5982]',\n", " 'Paradas de la línea 135: [1602, 3269, 1489, 3272, 3270, 1487, 3826, 5604]',\n", " 'Paradas de la línea 173: [3261, 3603, 3605, 5518, 1488, 2653, 3253, 3252, 1487, 3566, 5517, 5612, 5927, 5926]',\n", " 'Paradas de la línea 174: [3260, 3261, 3267, 4266, 4268, 4366, 5887, 5911, 5917, 5919, 51022, 51023, 1488, 2653, 3253, 3258, 3252, 3664, 3254, 3256, 1487, 3544, 4033, 5395, 5397, 5892, 5612, 5894, 5899, 5915, 5921, 5399]',\n", " 'Paradas de la línea 175: [399, 402, 3870, 3918, 4382, 4493, 5608, 1488, 2653, 3252, 3769, 3771, 1487, 3796, 4502, 3670, 4501, 4508]',\n", " 'Paradas de la línea 176: [402, 271, 1866, 3261, 3778, 3794, 3874, 4493, 269, 1488, 2653, 3253, 3252, 3869, 1487, 3566, 3671, 3790, 3872, 5607, 4495, 5984]',\n", " 'Paradas de la línea 177: [1537, 3683, 5020, 5606, 5798, 5802, 5803, 5804, 5807, 5800, 5799, 5801, 5805, 5806]',\n", " 'Paradas de la línea 178: [1022, 3623, 3918, 1025, 1488, 2864, 2653, 3279, 1760, 1762, 2968, 2970, 3252, 1487, 1021, 3620, 3621, 3634, 3670, 5609]']" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "[f\"Paradas de la línea {line}: {list(df[df['line'] == line]['stop'].unique())}\" for line in lines]" ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Consistency and Validity Checks:\n", "✅ - QA verified by accessing https://www.emtmadrid.es/Bloques-EMT/EMT-BUS/Mi-linea-(1).aspx\n" ] } ], "source": [ "print(\"Consistency and Validity Checks:\")\n", "print(\"✅ - QA verified by accessing https://www.emtmadrid.es/Bloques-EMT/EMT-BUS/Mi-linea-(1).aspx\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "#### Estudio de paradas específicas\n", "Paradas con más líneas conectadas:" ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "stop\n", "1487 8\n", "1488 5\n", "3252 5\n", "2653 5\n", "3253 3\n", " ..\n", "1608 1\n", "1610 1\n", "1612 1\n", "1615 1\n", "51023 1\n", "Name: line, Length: 283, dtype: int64" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.groupby(\"stop\")[\"line\"].nunique().sort_values(ascending=False)" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PKlinestopisHeaddestinationdeviationbusestimateArriveDistanceBusdatetimedatepositionBusLonpositionBusLatdayTypeStartTimeStopTimeMinimunFrequencyMaximumFrequencystrike
13932024-03-13 08:01:21.907965_B2072_L174_S14871741487FalseVALDEBEBAS02072728-12024-03-13 08:01:21.9079652024-03-13-3.64768440.483790LA06:0023:457.022.0N
13942024-03-13 08:01:21.907965_B2147_L175_S14871751487FalseLAS TABLAS NORTE021471649082024-03-13 08:01:21.9079652024-03-13-3.68823840.467754LA06:0023:459.022.0N
13952024-03-13 08:01:21.907965_B2163_L174_S14871741487FalseVALDEBEBAS021631615792024-03-13 08:01:21.9079652024-03-13-3.68892440.468011LA06:0023:457.022.0N
13962024-03-13 08:01:21.907965_B2259_L178_S14871781487FalseMONTECARMELO022591989412024-03-13 08:01:21.9079652024-03-13-3.68829740.467753LA06:0023:456.022.0N
13972024-03-13 08:01:21.907965_B2263_L135_S14871351487FalseHOSPITAL RAMON Y CAJAL022635566552024-03-13 08:01:21.9079652024-03-13-3.68840540.467987LA06:3023:459.029.0N
............................................................
11340112024-03-13 22:59:02.692429_B5568_L67_S1487671487FalseBARRIO PEÑAGRANDE05568164078792024-03-13 22:59:02.6924292024-03-13-3.73237340.481259LA06:0023:459.030.0N
11340132024-03-13 22:59:02.692429_B5720_L67_S1487671487FalseBARRIO PEÑAGRANDE0572068134202024-03-13 22:59:02.6924292024-03-13-3.69960440.484061LA06:0023:459.030.0N
11340142024-03-13 22:59:02.692429_B8856_L176_S14871761487FalseLAS TABLAS SUR08856161289972024-03-13 22:59:02.6924292024-03-13-3.66488040.505486LA06:0023:456.020.0N
11340152024-03-13 22:59:02.692429_B8864_L176_S14871761487FalseLAS TABLAS SUR088644746322024-03-13 22:59:02.6924292024-03-13-3.68839040.470399LA06:0023:456.020.0N
11340162024-03-13 22:59:02.692429_B8879_L174_S14871741487FalseVALDEBEBAS0887944216282024-03-13 22:59:02.6924292024-03-13-3.67580940.471649LA06:0023:457.022.0N
\n", "

13539 rows × 19 columns

\n", "
" ], "text/plain": [ " PK line stop isHead \\\n", "1393 2024-03-13 08:01:21.907965_B2072_L174_S1487 174 1487 False \n", "1394 2024-03-13 08:01:21.907965_B2147_L175_S1487 175 1487 False \n", "1395 2024-03-13 08:01:21.907965_B2163_L174_S1487 174 1487 False \n", "1396 2024-03-13 08:01:21.907965_B2259_L178_S1487 178 1487 False \n", "1397 2024-03-13 08:01:21.907965_B2263_L135_S1487 135 1487 False \n", "... ... ... ... ... \n", "1134011 2024-03-13 22:59:02.692429_B5568_L67_S1487 67 1487 False \n", "1134013 2024-03-13 22:59:02.692429_B5720_L67_S1487 67 1487 False \n", "1134014 2024-03-13 22:59:02.692429_B8856_L176_S1487 176 1487 False \n", "1134015 2024-03-13 22:59:02.692429_B8864_L176_S1487 176 1487 False \n", "1134016 2024-03-13 22:59:02.692429_B8879_L174_S1487 174 1487 False \n", "\n", " destination deviation bus estimateArrive DistanceBus \\\n", "1393 VALDEBEBAS 0 2072 728 -1 \n", "1394 LAS TABLAS NORTE 0 2147 164 908 \n", "1395 VALDEBEBAS 0 2163 161 579 \n", "1396 MONTECARMELO 0 2259 198 941 \n", "1397 HOSPITAL RAMON Y CAJAL 0 2263 556 655 \n", "... ... ... ... ... ... \n", "1134011 BARRIO PEÑAGRANDE 0 5568 1640 7879 \n", "1134013 BARRIO PEÑAGRANDE 0 5720 681 3420 \n", "1134014 LAS TABLAS SUR 0 8856 1612 8997 \n", "1134015 LAS TABLAS SUR 0 8864 474 632 \n", "1134016 VALDEBEBAS 0 8879 442 1628 \n", "\n", " datetime date positionBusLon positionBusLat \\\n", "1393 2024-03-13 08:01:21.907965 2024-03-13 -3.647684 40.483790 \n", "1394 2024-03-13 08:01:21.907965 2024-03-13 -3.688238 40.467754 \n", "1395 2024-03-13 08:01:21.907965 2024-03-13 -3.688924 40.468011 \n", "1396 2024-03-13 08:01:21.907965 2024-03-13 -3.688297 40.467753 \n", "1397 2024-03-13 08:01:21.907965 2024-03-13 -3.688405 40.467987 \n", "... ... ... ... ... \n", "1134011 2024-03-13 22:59:02.692429 2024-03-13 -3.732373 40.481259 \n", "1134013 2024-03-13 22:59:02.692429 2024-03-13 -3.699604 40.484061 \n", "1134014 2024-03-13 22:59:02.692429 2024-03-13 -3.664880 40.505486 \n", "1134015 2024-03-13 22:59:02.692429 2024-03-13 -3.688390 40.470399 \n", "1134016 2024-03-13 22:59:02.692429 2024-03-13 -3.675809 40.471649 \n", "\n", " dayType StartTime StopTime MinimunFrequency MaximumFrequency strike \n", "1393 LA 06:00 23:45 7.0 22.0 N \n", "1394 LA 06:00 23:45 9.0 22.0 N \n", "1395 LA 06:00 23:45 7.0 22.0 N \n", "1396 LA 06:00 23:45 6.0 22.0 N \n", "1397 LA 06:30 23:45 9.0 29.0 N \n", "... ... ... ... ... ... ... \n", "1134011 LA 06:00 23:45 9.0 30.0 N \n", "1134013 LA 06:00 23:45 9.0 30.0 N \n", "1134014 LA 06:00 23:45 6.0 20.0 N \n", "1134015 LA 06:00 23:45 6.0 20.0 N \n", "1134016 LA 06:00 23:45 7.0 22.0 N \n", "\n", "[13539 rows x 19 columns]" ] }, "execution_count": 32, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stop = 1487\n", "stop_df = df[df[\"stop\"] == stop]\n", "stop_df" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['174', '175', '178', '135', '173', '134', '67', '176']\n", "Categories (15, object): ['107', '129', '134', '135', ..., '42', '49', '67', '70']" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stop_df[\"line\"].unique()" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[2072, 2147, 2163, 2259, 2263, ..., 5720, 2545, 2543, 8851, 2450]\n", "Length: 83\n", "Categories (182, int64): [501, 504, 505, 506, ..., 9120, 9124, 9125, 9130]" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "stop_df[\"bus\"].unique()" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(15, 6))\n", "sns.lineplot(data=stop_df, x=\"datetime\", y=\"estimateArrive\", ci=None);" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## PROFILING 📑" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "8fbc56abe2e64bdca879ed0a45417e7d", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Summarize dataset: 0%| | 0/5 [00:00_B_L_S\",\n", " \"date\": \"Fecha de la petición a la API\",\n", " \"datetime\": \"Fecha y hora de la petición a la API\",\n", " \"line\": \"Línea de bus\",\n", " \"stop\": \"Parada de bus\",\n", " \"bus\": \"Número identificador del bus\",\n", " \"positionBusLon\": \"Longitud de las coordenadas del bus\",\n", " \"positionBusLat\": \"Latitud de las coordenadas del bus\",\n", " \"DistanceBus\": \"Distancia del bus a la parada (en metros)\",\n", " \"destination\": \"Destino del itinerario\",\n", " \"deviation\": \"Desviación en el cálculo del ETA\",\n", " \"StartTime\": \"Hora de inicio de la línea\",\n", " \"StopTime\": \"Hora de fin de la línea\",\n", " \"MinimunFrequency\": \"Frecuencia mínima de la línea\",\n", " \"MaximumFrequency\": \"Frecuencia máxima de la línea\",\n", " \"isHead\": \"Variable booleana para indicar si la parada es la cabecera de la línea\",\n", " \"dayType\": \" Tipo de día (LA: laboral, FE: festivo, SA: sábado)\",\n", " \"strike\": \"Variable para indicar si ese día hay huelga (S) o no (N)\",\n", " \"estimateArrive\": \"Tiempo estimado de espera del bus (en segundos)\",\n", " }\n", " },\n", " interactions=None,\n", " explorative=True,\n", " dark_mode=True,\n", ")\n", "profile.to_file(os.path.join(ROOT_PATH, \"docs\", \"qa\", \"emt_report.html\"))\n", "# profile.to_notebook_iframe()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3.11.7 ('venv': venv)", "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": "3.11.7" }, "orig_nbformat": 4, "vscode": { "interpreter": { "hash": "7056b316d281e41af8bb677c2c15c6d2127a9166071d02d66c8bc69014a8260f" } } }, "nbformat": 4, "nbformat_minor": 2 }