{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Analyse de la base Sirene via SQLite \n", "\n", "Voici la correction de l'analyse de la base sirène. \n", "Nous allons réaliser l'analyse en utilisant SQLite et Pandas simultanément. \n", "Ce n'est pas obligatoire pour cette taille de fichier mais ça permet d'illustrer la mise en étoile \n", "d'un schéma. \n", "\n", "## Sources\n", "Ce notebook suppose que le fichier a été télécharger [ici](https://links-biblio.lille.inria.fr/paperman/datas/base-sirene.csv.gz) et décompresser. Attention, ce fichier pèse 500mo décompressé. " ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import pandas\n", "import sqlite3\n", "import csv\n", "import matplotlib.pyplot as plot" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exploration du fichier. \n", "\n", "Il est possible de réaliser la phase d'exploration du fichier, en chargeant qu'une petite partie du fichier en mémoire. Le fichier en question se nomme `base-siren.csv`. On execute simplement la commande `head` en ligne de commande pour tronquer le fichier (ici à la ligne 100). " ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "!head -n100 base-sirene.csv > short.csv " ] }, { "cell_type": "code", "execution_count": 3, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
SIRENNICsiretStatut de diffusion de l'établissementDate de création de l'établissementTranche de l'effectif de l'établissementAnnée de la tranche d'effectif de l'établissementActivité principale de l'établissementDate de la dernière mise à jour de l'établissementEtablissement siège...Sous-section de l'unité légaleDivision de l'unité légaleGroupe de l'unité légaleClasse de l'unité légaleSIRET du siège de l'unité légaleNature juridique de l'unité légalePremière ligne de l'adressageDate de fermeture de l'unité légaleFilenameGéolocalisation de l'établissement
095650682830395650682800303O1990-06-016 à 9 salariés2017.046.69B2019-06-24T16:13:08+02:00non...Fabrication de machines et equipements n.c.a.Fabrication d'autres machines d'usage generalFabrication d'equipements aerauliques et frigo...Fabrication d'equipements aerauliques et frigo...95650682800196SAS, société par actions simplifiéeALDES AERAULIQUENaNsirene_v3_29050.664368,2.984585
195650846938795650846900387O2006-03-016 à 9 salariés2017.052.29B2019-06-24T16:13:08+02:00non...Entreposage et services auxiliaires des transp...Services auxiliaires des transportsAutres services auxiliaires des transportsAffretement et organisation des transports95650846900452SA à directoire (s.a.i.)TRANSPORTS P. FATTONNaNsirene_v3_290NaN
295651066324195651066300241O1998-01-01Etablissement non employeurNaN31.1C2007-05-22T04:47:38+02:00non...Activites immobilieresLocation et exploitation de biens immobiliers ...Location et exploitation de biens immobiliers ...Location de terrains et d'autres biens immobil...95651066300027SA à conseil d'administration (s.a.i.)CEGELEC MAINTENANCE ET SERVICES2009-06-30sirene_v3_29050.669662,3.208746
35006723572650067235700026O2012-10-11NaNNaN64.30Z2012-11-08T10:24:28+01:00oui...Activites des services financiers, hors assura...Fonds de placement et entites financieres simi...Fonds de placement et entites financieres simi...Fonds de placement et entites financieres simi...50067235700026Autre société civileAH CAPITALNaNsirene_v3_17450.678707,3.182991
45006723812650067238100026O2012-06-29NaNNaN41.10D2019-02-20T12:42:47+01:00oui...Construction de batimentsPromotion immobilierePromotion immobiliereSupports juridiques de programmes50067238100026Société civile immobilière de construction-venteSCCV ERQUY-CAROUALNaNsirene_v3_17450.635277,3.078823
..................................................................
945009520561750095205600017O2007-10-15NaNNaN70.22Z2009-02-17T05:11:58+01:00oui...Activites des sieges sociaux ; conseil de gestionConseil de gestionConseil pour les affaires et autres conseils d...Conseil pour les affaires et autres conseils d...50095205600017SARL unipersonnelleL'HOMME DE MISSION2008-12-31sirene_v3_17450.623222,2.929786
955009522543450095225400034O2011-07-01Etablissement non employeurNaN47.71Z2019-02-20T12:42:47+01:00non...Commerce de detail, a l'exception des automobi...Commerce de detail sur eventaires et marchesCommerce de detail de textiles, d'habillement ...Commerce de detail de textiles, d'habillement ...50095225400018Entrepreneur individuelMonsieur GUILLAUME LEFEBVRENaNsirene_v3_17450.635421,3.066737
965009544091650095440900016O2007-05-14NaNNaN93.12Z2016-07-29T15:03:27+02:00non...Activites sportives, recreatives et de loisirsActivites liees au sportActivites de clubs de sportsActivites de clubs de sports50095440900024Association déclaréeTEAM VTT ATTICHESNaNsirene_v3_17450.545826,3.03764
975009557942850095579400028O2016-05-03NaNNaN47.99A2016-05-25T12:30:48+02:00non...Commerce de detail, a l'exception des automobi...Commerce de detail d'equipements de l'informat...Commerce de detail de materiels de telecommuni...Commerce de detail de materiels de telecommuni...50095579400010Entrepreneur individuelMadame CARINE DELEFOSSENaNsirene_v3_17450.559949,2.838323
985009718901650097189000016O2007-10-25NaNNaN46.19B2011-07-12T21:54:19+02:00oui...Commerce de gros, a l'exception des automobile...Intermediaires du commerce de grosIntermediaires du commerce en produits diversAutres intermediaires du commerce en produits ...50097189000016SARL unipersonnelleAGENCE DU PLAT PAYS2011-03-30sirene_v3_17450.663025,3.033027
\n", "

99 rows × 108 columns

\n", "
" ], "text/plain": [ " SIREN NIC siret Statut de diffusion de l'établissement \\\n", "0 956506828 303 95650682800303 O \n", "1 956508469 387 95650846900387 O \n", "2 956510663 241 95651066300241 O \n", "3 500672357 26 50067235700026 O \n", "4 500672381 26 50067238100026 O \n", ".. ... ... ... ... \n", "94 500952056 17 50095205600017 O \n", "95 500952254 34 50095225400034 O \n", "96 500954409 16 50095440900016 O \n", "97 500955794 28 50095579400028 O \n", "98 500971890 16 50097189000016 O \n", "\n", " Date de création de l'établissement \\\n", "0 1990-06-01 \n", "1 2006-03-01 \n", "2 1998-01-01 \n", "3 2012-10-11 \n", "4 2012-06-29 \n", ".. ... \n", "94 2007-10-15 \n", "95 2011-07-01 \n", "96 2007-05-14 \n", "97 2016-05-03 \n", "98 2007-10-25 \n", "\n", " Tranche de l'effectif de l'établissement \\\n", "0 6 à 9 salariés \n", "1 6 à 9 salariés \n", "2 Etablissement non employeur \n", "3 NaN \n", "4 NaN \n", ".. ... \n", "94 NaN \n", "95 Etablissement non employeur \n", "96 NaN \n", "97 NaN \n", "98 NaN \n", "\n", " Année de la tranche d'effectif de l'établissement \\\n", "0 2017.0 \n", "1 2017.0 \n", "2 NaN \n", "3 NaN \n", "4 NaN \n", ".. ... \n", "94 NaN \n", "95 NaN \n", "96 NaN \n", "97 NaN \n", "98 NaN \n", "\n", " Activité principale de l'établissement \\\n", "0 46.69B \n", "1 52.29B \n", "2 31.1C \n", "3 64.30Z \n", "4 41.10D \n", ".. ... \n", "94 70.22Z \n", "95 47.71Z \n", "96 93.12Z \n", "97 47.99A \n", "98 46.19B \n", "\n", " Date de la dernière mise à jour de l'établissement Etablissement siège \\\n", "0 2019-06-24T16:13:08+02:00 non \n", "1 2019-06-24T16:13:08+02:00 non \n", "2 2007-05-22T04:47:38+02:00 non \n", "3 2012-11-08T10:24:28+01:00 oui \n", "4 2019-02-20T12:42:47+01:00 oui \n", ".. ... ... \n", "94 2009-02-17T05:11:58+01:00 oui \n", "95 2019-02-20T12:42:47+01:00 non \n", "96 2016-07-29T15:03:27+02:00 non \n", "97 2016-05-25T12:30:48+02:00 non \n", "98 2011-07-12T21:54:19+02:00 oui \n", "\n", " ... Sous-section de l'unité légale \\\n", "0 ... Fabrication de machines et equipements n.c.a. \n", "1 ... Entreposage et services auxiliaires des transp... \n", "2 ... Activites immobilieres \n", "3 ... Activites des services financiers, hors assura... \n", "4 ... Construction de batiments \n", ".. ... ... \n", "94 ... Activites des sieges sociaux ; conseil de gestion \n", "95 ... Commerce de detail, a l'exception des automobi... \n", "96 ... Activites sportives, recreatives et de loisirs \n", "97 ... Commerce de detail, a l'exception des automobi... \n", "98 ... Commerce de gros, a l'exception des automobile... \n", "\n", " Division de l'unité légale \\\n", "0 Fabrication d'autres machines d'usage general \n", "1 Services auxiliaires des transports \n", "2 Location et exploitation de biens immobiliers ... \n", "3 Fonds de placement et entites financieres simi... \n", "4 Promotion immobiliere \n", ".. ... \n", "94 Conseil de gestion \n", "95 Commerce de detail sur eventaires et marches \n", "96 Activites liees au sport \n", "97 Commerce de detail d'equipements de l'informat... \n", "98 Intermediaires du commerce de gros \n", "\n", " Groupe de l'unité légale \\\n", "0 Fabrication d'equipements aerauliques et frigo... \n", "1 Autres services auxiliaires des transports \n", "2 Location et exploitation de biens immobiliers ... \n", "3 Fonds de placement et entites financieres simi... \n", "4 Promotion immobiliere \n", ".. ... \n", "94 Conseil pour les affaires et autres conseils d... \n", "95 Commerce de detail de textiles, d'habillement ... \n", "96 Activites de clubs de sports \n", "97 Commerce de detail de materiels de telecommuni... \n", "98 Intermediaires du commerce en produits divers \n", "\n", " Classe de l'unité légale \\\n", "0 Fabrication d'equipements aerauliques et frigo... \n", "1 Affretement et organisation des transports \n", "2 Location de terrains et d'autres biens immobil... \n", "3 Fonds de placement et entites financieres simi... \n", "4 Supports juridiques de programmes \n", ".. ... \n", "94 Conseil pour les affaires et autres conseils d... \n", "95 Commerce de detail de textiles, d'habillement ... \n", "96 Activites de clubs de sports \n", "97 Commerce de detail de materiels de telecommuni... \n", "98 Autres intermediaires du commerce en produits ... \n", "\n", " SIRET du siège de l'unité légale \\\n", "0 95650682800196 \n", "1 95650846900452 \n", "2 95651066300027 \n", "3 50067235700026 \n", "4 50067238100026 \n", ".. ... \n", "94 50095205600017 \n", "95 50095225400018 \n", "96 50095440900024 \n", "97 50095579400010 \n", "98 50097189000016 \n", "\n", " Nature juridique de l'unité légale \\\n", "0 SAS, société par actions simplifiée \n", "1 SA à directoire (s.a.i.) \n", "2 SA à conseil d'administration (s.a.i.) \n", "3 Autre société civile \n", "4 Société civile immobilière de construction-vente \n", ".. ... \n", "94 SARL unipersonnelle \n", "95 Entrepreneur individuel \n", "96 Association déclarée \n", "97 Entrepreneur individuel \n", "98 SARL unipersonnelle \n", "\n", " Première ligne de l'adressage Date de fermeture de l'unité légale \\\n", "0 ALDES AERAULIQUE NaN \n", "1 TRANSPORTS P. FATTON NaN \n", "2 CEGELEC MAINTENANCE ET SERVICES 2009-06-30 \n", "3 AH CAPITAL NaN \n", "4 SCCV ERQUY-CAROUAL NaN \n", ".. ... ... \n", "94 L'HOMME DE MISSION 2008-12-31 \n", "95 Monsieur GUILLAUME LEFEBVRE NaN \n", "96 TEAM VTT ATTICHES NaN \n", "97 Madame CARINE DELEFOSSE NaN \n", "98 AGENCE DU PLAT PAYS 2011-03-30 \n", "\n", " Filename Géolocalisation de l'établissement \n", "0 sirene_v3_290 50.664368,2.984585 \n", "1 sirene_v3_290 NaN \n", "2 sirene_v3_290 50.669662,3.208746 \n", "3 sirene_v3_174 50.678707,3.182991 \n", "4 sirene_v3_174 50.635277,3.078823 \n", ".. ... ... \n", "94 sirene_v3_174 50.623222,2.929786 \n", "95 sirene_v3_174 50.635421,3.066737 \n", "96 sirene_v3_174 50.545826,3.03764 \n", "97 sirene_v3_174 50.559949,2.838323 \n", "98 sirene_v3_174 50.663025,3.033027 \n", "\n", "[99 rows x 108 columns]" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pandas.read_csv(\"short.csv\", delimiter=\";\")\n", "df" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## SQLite et Python\n", "\n", "On peut communiquer avec une base de donnée SQLite en Python directement. Il faut faire attention à ne jamais tout charger en mémoire afin de ne pas la saturer." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "!rm siren.db\n", "db = sqlite3.connect(\"siren.db\") # création de la base de donnée " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "L'intérêt d'utiliser Python et de pouvoir se limiter à seulement les éléments qui vont nous intéresser et de ne pas importer les colonnes inutiles. \n", "Nous allons garder uniquement les colonnes suivantes, que l'on stock dans le dictionnaire `important_keys`. \n", "Dans ce dictionnaire, on associe des labels plus agréable à manipuler aux vrais labels utiliser dans le fichier." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "important_keys = {\n", "\"siren\" : \"SIREN\",\n", "\"effectif\" : \"Tranche de l'effectif de l'établissement\",\n", "\"code_postal\" : \"Code postal de l'établissement\",\n", "\"commune\" : \"Commune de l'établissement\",\n", "\"état_administratif\": \"Etat administratif de l'établissement\",\n", "\"dénomination\" : \"Dénomination usuelle de l'établissement\",\n", "\"section\" : \"Section de l'établissement\",\n", "\"sous_section\" : \"Sous-section de l'établissement\",\n", "\"division\" : \"Division de l'établissement\",\n", "\"groupe\" : \"Groupe de l'établissement\",\n", "\"classe\" : \"Classe de l'établissement\",\n", "\"addresse\" : \"Adresse de l'établissement\",\n", "\"date_fermeture\" : \"Date de fermeture de l'établissement\",\n", "\"date_creation\" : \"Date de création de l'établissement\",\n", "\"position\" : \"Géolocalisation de l'établissement\"\n", "}" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous aurons également besoin de l'index où apparait chacune de ses colonnes dans le fichier. \n", "On le construit à partir de l'entête calculer par pandas." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'siren': 0, 'effectif': 5, 'code_postal': 16, 'commune': 17, 'état_administratif': 40, 'dénomination': 44, 'section': 58, 'sous_section': 59, 'division': 60, 'groupe': 61, 'classe': 62, 'addresse': 63, 'date_fermeture': 64, 'date_creation': 4, 'position': 107}\n" ] } ], "source": [ "keys = list(df.keys())\n", "important_keys_index = {k:keys.index(v) for k,v in important_keys.items()}\n", "print(important_keys_index)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Approche naïve\n", "\n", "Nous allons maintenant créer un table SQLite qui contient les données qui nous intéresse.\n", "Nous pouvons générer la requête en Python pour ne pas la taper à la main. \n", "L'objectif est de stocker dans une seule table SQLite la table des faits et d'executer des requêtes SQL pour générer les tableaux croiser que l'on transmet à Pandas." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Query que nous allons executer:\n", " CREATE TABLE table_des_faits (siren,\n", "\teffectif,\n", "\tcode_postal,\n", "\tcommune,\n", "\tétat_administratif,\n", "\tdénomination,\n", "\tsection,\n", "\tsous_section,\n", "\tdivision,\n", "\tgroupe,\n", "\tclasse,\n", "\taddresse,\n", "\tdate_fermeture,\n", "\tdate_creation,\n", "\tposition)\n" ] } ], "source": [ "query = \"CREATE TABLE table_des_faits ({})\".format(\",\\n\\t\".join(important_keys.keys()))\n", "print(\"Query que nous allons executer:\\n\", query)\n", "c = db.execute(query, important_keys)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous devons maintenant faire ingérer à SQLite les données du fichier CSV.\n", "Nous devons lire le fichier (en Python) et filtrer les données inutiles. \n" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{0, 64, 4, 5, 40, 107, 44, 16, 17, 58, 59, 60, 61, 62, 63}\n" ] } ], "source": [ "indices = set(important_keys_index.values())\n", "print(indices)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous aurons également besoin de la liste des colonnes de la table trié par leur index. On stock ça une fois pour toute dans une variable. " ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "['siren', 'date_creation', 'effectif', 'code_postal', 'commune', 'état_administratif', 'dénomination', 'section', 'sous_section', 'division', 'groupe', 'classe', 'addresse', 'date_fermeture', 'position']\n" ] } ], "source": [ "important_keys_trie = sorted(important_keys, key=lambda e:important_keys_index[e])\n", "print(important_keys_trie)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut maintenant préparer la requête d'insertion. On utilise des *placeholder* afin d'appeler la fonction executemany et d'accélerer sensiblement l'ingestion. Pour plus de détails, vous pouvez lire la documentation du module [sqlite3](https://docs.python.org/3/library/sqlite3.html)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "INSERT INTO table_des_faits (siren,date_creation,effectif,code_postal,commune,état_administratif,dénomination,section,sous_section,division,groupe,classe,addresse,date_fermeture,position) VALUES (?,?,?,?,?,?,?,?,?,?,?,?,?,?,?)\n" ] } ], "source": [ "query = \"INSERT INTO table_des_faits ({}) VALUES ({})\".format(\n", " \",\".join(important_keys_trie), \n", " \",\".join([\"?\" for _ in range(len(important_keys))]))\n", "print(query) #Le nombre de ? est égale au nombre de colonne. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut maintenant parcourir le fichier et transmettre chaque ligne à la base de donnée via \n", "la commande `executemany`. On doit faire attention à ne pas stocker en mémoire toute les lignes. \n", "Le plus simple pour ça consiste à simplement utiliser des opérations fonctionnelles. \n", "Nous allons simplifier la lecture du fichier CSV à l'aide module `csv` dont l'utilisation est aisée." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "def filtre_ligne(e):\n", " \"\"\"\n", " Fonction qui prend une ligne de text\n", " et qui retourne un tuple contenant uniquement les colonnes\n", " importantes\n", " \"\"\"\n", " return tuple(map(lambda e:e[1], filter(lambda e:e[0] in indices, enumerate(e))))\n", "\n", "with open(\"base-sirene.csv\") as fichier_raw:\n", " fichier_parse = csv.reader(fichier_raw, delimiter=';')\n", " next(fichier_parse) # on dépile la première ligne\n", " lignes = map(filtre_ligne, fichier_parse) # l'execution ne se lance pas ici\n", " db.executemany(query, lignes) # déplie l'itérateur et insère les lignes \n", " db.commit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut vérifier que le nombre de lignes inséré est le bon. " ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "468466" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "list(db.execute(\"SELECT COUNT(*) FROM table_des_faits\"))[0][0]" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "468467 base-sirene.csv\r\n" ] } ], "source": [ "!wc -l base-sirene.csv" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Quelques tableaux croisés simple\n", "\n", "On peut immédiatement utiliser pandas via SQL pour produire des tableaux croisés. " ] }, { "cell_type": "code", "execution_count": 14, "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", "
état_administratifActifFermé
commune
ALLENNES LES MARAIS303.0407.0
ALLENNES-LES-MARAIS10.04.0
ANNOEULLIN979.01560.0
ANSTAING162.0194.0
ARMENTIERES2807.05045.0
.........
WAVRIN878.01236.0
WERVICQ SUD574.0694.0
WERVICQ-SUD16.011.0
WICRES57.058.0
WILLEMS379.0508.0
\n", "

121 rows × 2 columns

\n", "
" ], "text/plain": [ "état_administratif Actif Fermé\n", "commune \n", "ALLENNES LES MARAIS 303.0 407.0\n", "ALLENNES-LES-MARAIS 10.0 4.0\n", "ANNOEULLIN 979.0 1560.0\n", "ANSTAING 162.0 194.0\n", "ARMENTIERES 2807.0 5045.0\n", "... ... ...\n", "WAVRIN 878.0 1236.0\n", "WERVICQ SUD 574.0 694.0\n", "WERVICQ-SUD 16.0 11.0\n", "WICRES 57.0 58.0\n", "WILLEMS 379.0 508.0\n", "\n", "[121 rows x 2 columns]" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "communes_etats = pandas.read_sql(\"SELECT commune, état_administratif, COUNT(*) as Count FROM table_des_faits GROUP BY commune, état_administratif\", db)\n", "communes_etats.pivot_table(\"Count\", index=\"commune\", columns=\"état_administratif\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit des soucis de formattage au niveau des communes, ça peut aisément se régler via une simple requête SQL. " ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "db.execute(\"UPDATE table_des_faits SET commune = REPLACE(commune, '-',' ')\")\n", "db.commit()" ] }, { "cell_type": "code", "execution_count": 16, "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", "
état_administratifActifFermé
commune
ALLENNES LES MARAIS313.0411.0
ANNOEULLIN979.01560.0
ANSTAING162.0194.0
ARMENTIERES2807.05045.0
AUBERS271.0249.0
.........
WATTRELOS2926.05709.0
WAVRIN878.01236.0
WERVICQ SUD590.0705.0
WICRES57.058.0
WILLEMS379.0508.0
\n", "

96 rows × 2 columns

\n", "
" ], "text/plain": [ "état_administratif Actif Fermé\n", "commune \n", "ALLENNES LES MARAIS 313.0 411.0\n", "ANNOEULLIN 979.0 1560.0\n", "ANSTAING 162.0 194.0\n", "ARMENTIERES 2807.0 5045.0\n", "AUBERS 271.0 249.0\n", "... ... ...\n", "WATTRELOS 2926.0 5709.0\n", "WAVRIN 878.0 1236.0\n", "WERVICQ SUD 590.0 705.0\n", "WICRES 57.0 58.0\n", "WILLEMS 379.0 508.0\n", "\n", "[96 rows x 2 columns]" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "communes_etats = pandas.read_sql(\"SELECT commune, état_administratif, COUNT(*) as Count FROM table_des_faits GROUP BY commune, état_administratif\", db)\n", "communes_etats.pivot_table(\"Count\", index=\"commune\", columns=\"état_administratif\")" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 383 ms, sys: 52.4 ms, total: 435 ms\n", "Wall time: 434 ms\n" ] } ], "source": [ "%%time\n", "communes_etats = pandas.read_sql(\"SELECT commune, effectif, COUNT(*) as Count FROM table_des_faits GROUP BY commune, effectif\", db)" ] }, { "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", "
effectif0 salarié1 ou 2 salariés10 à 19 salariés100 à 199 salariés1000 à 1999 salariés20 à 49 salariés200 à 249 salariés2000 à 4999 salariés250 à 499 salariés3 à 5 salariés50 à 99 salariés500 à 999 salariés5000 à 9999 salariés6 à 9 salariésEtablissement non employeur
commune
ALLENNES LES MARAIS390.010.023.03.00.00.02.00.00.00.07.01.00.00.07.0281.0
ANNOEULLIN1313.035.095.018.00.00.015.00.00.01.038.04.00.00.020.01000.0
ANSTAING214.03.019.01.00.00.00.00.00.00.04.00.00.00.01.0114.0
ARMENTIERES3681.0114.0262.060.011.02.041.03.00.01.0126.013.01.00.047.03490.0
AUBERS267.07.025.02.00.00.01.00.00.00.07.00.00.00.04.0207.0
...................................................
WATTRELOS4388.0118.0248.056.07.01.037.01.00.02.0120.010.01.00.063.03583.0
WAVRIN1091.025.068.016.00.00.09.00.00.00.035.08.00.00.023.0839.0
WERVICQ SUD724.017.046.06.00.00.08.00.00.00.013.02.00.00.06.0473.0
WICRES68.03.07.00.00.00.00.00.00.00.03.00.00.00.00.034.0
WILLEMS480.020.035.05.00.00.05.00.00.01.013.01.00.00.07.0320.0
\n", "

96 rows × 16 columns

\n", "
" ], "text/plain": [ "effectif 0 salarié 1 ou 2 salariés 10 à 19 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 390.0 10.0 23.0 3.0 \n", "ANNOEULLIN 1313.0 35.0 95.0 18.0 \n", "ANSTAING 214.0 3.0 19.0 1.0 \n", "ARMENTIERES 3681.0 114.0 262.0 60.0 \n", "AUBERS 267.0 7.0 25.0 2.0 \n", "... ... ... ... ... \n", "WATTRELOS 4388.0 118.0 248.0 56.0 \n", "WAVRIN 1091.0 25.0 68.0 16.0 \n", "WERVICQ SUD 724.0 17.0 46.0 6.0 \n", "WICRES 68.0 3.0 7.0 0.0 \n", "WILLEMS 480.0 20.0 35.0 5.0 \n", "\n", "effectif 100 à 199 salariés 1000 à 1999 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 0.0 0.0 \n", "ANNOEULLIN 0.0 0.0 \n", "ANSTAING 0.0 0.0 \n", "ARMENTIERES 11.0 2.0 \n", "AUBERS 0.0 0.0 \n", "... ... ... \n", "WATTRELOS 7.0 1.0 \n", "WAVRIN 0.0 0.0 \n", "WERVICQ SUD 0.0 0.0 \n", "WICRES 0.0 0.0 \n", "WILLEMS 0.0 0.0 \n", "\n", "effectif 20 à 49 salariés 200 à 249 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 2.0 0.0 \n", "ANNOEULLIN 15.0 0.0 \n", "ANSTAING 0.0 0.0 \n", "ARMENTIERES 41.0 3.0 \n", "AUBERS 1.0 0.0 \n", "... ... ... \n", "WATTRELOS 37.0 1.0 \n", "WAVRIN 9.0 0.0 \n", "WERVICQ SUD 8.0 0.0 \n", "WICRES 0.0 0.0 \n", "WILLEMS 5.0 0.0 \n", "\n", "effectif 2000 à 4999 salariés 250 à 499 salariés 3 à 5 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 0.0 0.0 7.0 \n", "ANNOEULLIN 0.0 1.0 38.0 \n", "ANSTAING 0.0 0.0 4.0 \n", "ARMENTIERES 0.0 1.0 126.0 \n", "AUBERS 0.0 0.0 7.0 \n", "... ... ... ... \n", "WATTRELOS 0.0 2.0 120.0 \n", "WAVRIN 0.0 0.0 35.0 \n", "WERVICQ SUD 0.0 0.0 13.0 \n", "WICRES 0.0 0.0 3.0 \n", "WILLEMS 0.0 1.0 13.0 \n", "\n", "effectif 50 à 99 salariés 500 à 999 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 1.0 0.0 \n", "ANNOEULLIN 4.0 0.0 \n", "ANSTAING 0.0 0.0 \n", "ARMENTIERES 13.0 1.0 \n", "AUBERS 0.0 0.0 \n", "... ... ... \n", "WATTRELOS 10.0 1.0 \n", "WAVRIN 8.0 0.0 \n", "WERVICQ SUD 2.0 0.0 \n", "WICRES 0.0 0.0 \n", "WILLEMS 1.0 0.0 \n", "\n", "effectif 5000 à 9999 salariés 6 à 9 salariés \\\n", "commune \n", "ALLENNES LES MARAIS 0.0 7.0 \n", "ANNOEULLIN 0.0 20.0 \n", "ANSTAING 0.0 1.0 \n", "ARMENTIERES 0.0 47.0 \n", "AUBERS 0.0 4.0 \n", "... ... ... \n", "WATTRELOS 0.0 63.0 \n", "WAVRIN 0.0 23.0 \n", "WERVICQ SUD 0.0 6.0 \n", "WICRES 0.0 0.0 \n", "WILLEMS 0.0 7.0 \n", "\n", "effectif Etablissement non employeur \n", "commune \n", "ALLENNES LES MARAIS 281.0 \n", "ANNOEULLIN 1000.0 \n", "ANSTAING 114.0 \n", "ARMENTIERES 3490.0 \n", "AUBERS 207.0 \n", "... ... \n", "WATTRELOS 3583.0 \n", "WAVRIN 839.0 \n", "WERVICQ SUD 473.0 \n", "WICRES 34.0 \n", "WILLEMS 320.0 \n", "\n", "[96 rows x 16 columns]" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pivot = communes_etats.pivot_table(\"Count\", index=\"commune\", columns=\"effectif\", aggfunc=sum).fillna(0)\n", "pivot" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On voit que le temps de calcul est raisonnable mais le dataset est plutôt petit. On peut quand même accélérer les requêtes en ajoutant des indexes." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "db.execute(\"CREATE INDEX commune_effectifs ON table_des_faits (commune, effectif)\")" ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 45.9 ms, sys: 115 µs, total: 46.1 ms\n", "Wall time: 45.6 ms\n" ] } ], "source": [ "%%time\n", "communes_etats = pandas.read_sql(\"SELECT commune, effectif, COUNT(*) as Count FROM table_des_faits GROUP BY commune, effectif\", db)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On gagne un facteur 10 immédiatement. " ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On peut regarder aisément le rapport entre la taille et le nombre de fermeture addministrative. " ] }, { "cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [], "source": [ "effectif_etats = pandas.read_sql(\"SELECT effectif, état_administratif, COUNT(*) as Count FROM table_des_faits GROUP BY effectif, état_administratif\", db)\n", "pivot = effectif_etats.pivot_table(\"Count\", index=\"effectif\", columns=\"état_administratif\", aggfunc=sum).fillna(0)" ] }, { "cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "(pivot[\"Actif\"]/pivot[\"Fermé\"]).plot(kind=\"bar\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "La visualisation à trié par ordre alphabétique et non pas par nombre de salarié, ce qui la rend inutilisable. On peut arranger ça aisément. " ] }, { "cell_type": "code", "execution_count": 23, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "['',\n", " '0 salarié',\n", " '1 ou 2 salariés',\n", " '10 à 19 salariés',\n", " '100 à 199 salariés',\n", " '1000 à 1999 salariés',\n", " '20 à 49 salariés',\n", " '200 à 249 salariés',\n", " '2000 à 4999 salariés',\n", " '250 à 499 salariés',\n", " '3 à 5 salariés',\n", " '50 à 99 salariés',\n", " '500 à 999 salariés',\n", " '5000 à 9999 salariés',\n", " '6 à 9 salariés',\n", " 'Etablissement non employeur']" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "list(pivot.index)" ] }, { "cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [], "source": [ "reindex = sorted(list(pivot.index)[:-1], key=lambda e:int(e.split(\" \")[0]) if e else -1)\n", "reindex.insert(1, 'Etablissement non employeur')" ] }, { "cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pivot = pivot.reindex(reindex)\n", "(pivot[\"Actif\"]/pivot[\"Fermé\"]).plot(kind=\"bar\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Modélisation en cube et schéma en étoile\n", "\n", "La modélisation précédante est relativement limité. Il est compliqué de faire des analyses par granularité \n", "et les performances peuvent être mauvaises quand les données sont très grosses. \n", "Pour éviter ça, on peut introduire un schéma en étoile qui va pré-agrégé les données en fonction de leur granularité. \n", "\n", "Il faut dans un premier temps identifié les dimensions d'intérêts. On les modélise en associant à chaque dimension une liste de colonne ordonnée par granularité." ] }, { "cell_type": "code", "execution_count": 26, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'siren': 'SIREN',\n", " 'effectif': \"Tranche de l'effectif de l'établissement\",\n", " 'code_postal': \"Code postal de l'établissement\",\n", " 'commune': \"Commune de l'établissement\",\n", " 'état_administratif': \"Etat administratif de l'établissement\",\n", " 'dénomination': \"Dénomination usuelle de l'établissement\",\n", " 'section': \"Section de l'établissement\",\n", " 'sous_section': \"Sous-section de l'établissement\",\n", " 'division': \"Division de l'établissement\",\n", " 'groupe': \"Groupe de l'établissement\",\n", " 'classe': \"Classe de l'établissement\",\n", " 'addresse': \"Adresse de l'établissement\",\n", " 'date_fermeture': \"Date de fermeture de l'établissement\",\n", " 'date_creation': \"Date de création de l'établissement\",\n", " 'position': \"Géolocalisation de l'établissement\"}" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "important_keys" ] }, { "cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [], "source": [ "dimensions = {\n", " 'effectif':['effectif'], \n", " 'localisation':['commune', 'code_postal', 'addresse', 'position'],\n", " 'creation': ['date_creation'],\n", " 'fermeture': ['date_fermeture'],\n", " 'etat':[\"état_administratif\"],\n", " 'description':['section', 'sous_section', 'division', 'groupe', 'classe', 'dénomination'] # ordonné par nombre de valeur !=\n", "}" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Nous allons traiter les dates différemment puisque nous voulons agréger par année, par mois et pas seulement par date précise. Nous allons construire directement depuis la table des faits les tables de dimensions qui nous intéresse. " ] }, { "cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CREATE TABLE effectif (id INTEGER PRIMARY KEY, effectif)\n", "INSERT INTO effectif (effectif) SELECT DISTINCT effectif FROM table_des_faits\n", "CREATE TABLE localisation (id INTEGER PRIMARY KEY, commune, code_postal, addresse, position)\n", "INSERT INTO localisation (commune, code_postal, addresse, position) SELECT DISTINCT commune, code_postal, addresse, position FROM table_des_faits\n", "CREATE TABLE creation (id INTEGER PRIMARY KEY, date_creation)\n", "INSERT INTO creation (date_creation) SELECT DISTINCT date_creation FROM table_des_faits\n", "CREATE TABLE fermeture (id INTEGER PRIMARY KEY, date_fermeture)\n", "INSERT INTO fermeture (date_fermeture) SELECT DISTINCT date_fermeture FROM table_des_faits\n", "CREATE TABLE etat (id INTEGER PRIMARY KEY, état_administratif)\n", "INSERT INTO etat (état_administratif) SELECT DISTINCT état_administratif FROM table_des_faits\n", "CREATE TABLE description (id INTEGER PRIMARY KEY, section, sous_section, division, groupe, classe, dénomination)\n", "INSERT INTO description (section, sous_section, division, groupe, classe, dénomination) SELECT DISTINCT section, sous_section, division, groupe, classe, dénomination FROM table_des_faits\n", "CPU times: user 2.18 s, sys: 582 ms, total: 2.76 s\n", "Wall time: 3.52 s\n" ] } ], "source": [ "%%time \n", "for dimension, colonnes in dimensions.items():\n", " colonnes = colonnes\n", " colonnes = \", \".join(colonnes)\n", " query = \"CREATE TABLE {} (id INTEGER PRIMARY KEY, {})\".format(dimension, colonnes)\n", " print(query) # Affiche la requête executée\n", " db.execute(query)\n", " query = \"INSERT INTO {} ({colonnes}) SELECT DISTINCT {colonnes} FROM table_des_faits\"\n", " query = query.format(dimension, colonnes=colonnes)\n", " print(query) # Affiche l'insertion dans la table\n", " db.execute(query)\n", " query = \"CREATE INDEX {dimension}_dim ON {dimension}({})\".format(colonnes, dimension=dimension)\n", " query = query.format(dimension, colonnes=colonnes)\n", " db.execute(query)\n", "db.commit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Pour créer le Cube, il faut maintenant transformer chaque ligne de la table des faits en une ligne utilisant \n", "les id de dimensions avec les bonnes valeurs. On peut aisément créer la table en utilisant `SIREN` comme identifiant de ligne. On pourrait ajouter des clefs étrangère pour garantir la cohérence de la base, mais ce n'est pas non plus indispensable. " ] }, { "cell_type": "code", "execution_count": 29, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CREATE TABLE cube (siren, effectif, localisation, creation, fermeture, etat, description)\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "query = \"CREATE TABLE cube (siren, {})\".format(\", \".join(dimensions))\n", "print(query)\n", "db.execute(query)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "La requête suivante est compliqué à écrire, on le fait à la main pour ne pas se tromper. " ] }, { "cell_type": "code", "execution_count": 30, "metadata": {}, "outputs": [], "source": [ "query = \"\"\"\n", "INSERT INTO cube(siren, effectif, localisation, creation, fermeture, etat, description)\n", "SELECT t.siren, e.id, l.id, c.id, f.id, etat.id, d.id \n", "FROM table_des_faits t, effectif e, localisation l, creation c, fermeture f, etat, description d \n", "WHERE \n", " t.effectif = e.effectif AND\n", " t.commune = l.commune AND \n", " t.code_postal= l.code_postal AND\n", " t.addresse = l.addresse AND \n", " t.position = l.position AND\n", " t.date_creation = c.date_creation AND \n", " t.date_fermeture = f.date_fermeture AND \n", " t.section = d.section AND\n", " t.sous_section = d.sous_section AND\n", " t.état_administratif = etat.état_administratif AND\n", " t.division = d.division AND\n", " t.groupe = d.groupe AND\n", " t.classe = d.classe AND \n", " t.dénomination = d.dénomination \n", "\"\"\"\n", "# Ouf. " ] }, { "cell_type": "code", "execution_count": 31, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 2.23 s, sys: 1.21 s, total: 3.44 s\n", "Wall time: 3.85 s\n" ] } ], "source": [ "%%time \n", "db.execute(query) # La requête serait excessivement longue sans les index.\n", "db.commit()" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [], "source": [ "db.commit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Gestion des tables temporelles\n", "\n", "Les deux tables temporelles contiennent des données désagrégé. Il peut être pertinent de les traité pour permettre d'avoir une vue par année et par mois. \n" ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "db.execute(\"ALTER TABLE creation ADD COLUMN année\")\n", "db.execute(\"ALTER TABLE creation ADD COLUMN mois\")\n", "db.execute(\"ALTER TABLE fermeture ADD COLUMN année\")\n", "db.execute(\"ALTER TABLE fermeture ADD COLUMN mois\")\n" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [], "source": [ "db.execute(\"UPDATE creation SET année=SUBSTR(date_creation, 0, 5),mois=SUBSTR(date_creation, 0, 8)\")\n", "db.execute(\"UPDATE fermeture SET année=SUBSTR(date_fermeture, 0, 5),mois=SUBSTR(date_fermeture, 0, 8)\")\n", "db.commit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Ajouter des index permet de retrouver rapidement les colonnes dont on a besoin en cas de changement de granularité. " ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [], "source": [ "db.execute(\"CREATE INDEX _create ON creation (année, mois, date_creation)\")\n", "db.execute(\"CREATE INDEX _fermeture ON fermeture (année, mois, date_fermeture)\")\n", "db.commit()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Quelques tableaux croisés\n", "On peut reprendre les tableaux croisés de la version naïve. " ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [], "source": [ "query =\"\"\" \n", "SELECT e.effectif, e2.état_administratif, COUNT(*) as Count \n", "FROM cube, etat e2, effectif e \n", "WHERE \n", " cube.etat=e2.id AND\n", " cube.effectif=e.id \n", "GROUP BY cube.etat, cube.effectif\n", "\"\"\"\n", "effectif_etats = pandas.read_sql(query, db)\n", "pivot = effectif_etats.pivot_table(\"Count\", index=\"effectif\", columns=\"état_administratif\").fillna(0)" ] }, { "cell_type": "code", "execution_count": 37, "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", "
état_administratifActifFermé
effectif
113994.0130498.0
0 salarié2851.04462.0
1 ou 2 salariés11752.02015.0
10 à 19 salariés3234.0267.0
100 à 199 salariés429.028.0
1000 à 1999 salariés23.03.0
20 à 49 salariés2212.0141.0
200 à 249 salariés78.04.0
2000 à 4999 salariés10.01.0
250 à 499 salariés158.011.0
3 à 5 salariés6278.0754.0
50 à 99 salariés880.064.0
500 à 999 salariés57.03.0
5000 à 9999 salariés2.00.0
6 à 9 salariés3637.0358.0
Etablissement non employeur34410.0149852.0
\n", "
" ], "text/plain": [ "état_administratif Actif Fermé\n", "effectif \n", " 113994.0 130498.0\n", "0 salarié 2851.0 4462.0\n", "1 ou 2 salariés 11752.0 2015.0\n", "10 à 19 salariés 3234.0 267.0\n", "100 à 199 salariés 429.0 28.0\n", "1000 à 1999 salariés 23.0 3.0\n", "20 à 49 salariés 2212.0 141.0\n", "200 à 249 salariés 78.0 4.0\n", "2000 à 4999 salariés 10.0 1.0\n", "250 à 499 salariés 158.0 11.0\n", "3 à 5 salariés 6278.0 754.0\n", "50 à 99 salariés 880.0 64.0\n", "500 à 999 salariés 57.0 3.0\n", "5000 à 9999 salariés 2.0 0.0\n", "6 à 9 salariés 3637.0 358.0\n", "Etablissement non employeur 34410.0 149852.0" ] }, "execution_count": 37, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pivot" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 38, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "pivot = pivot.reindex(reindex)\n", "(pivot[\"Actif\"]/pivot[\"Fermé\"]).plot(kind=\"bar\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "On retrouve bien la même chose ! Faisons varier l'année de création maintenant ! \n", "On récupère la liste des années:" ] }, { "cell_type": "code", "execution_count": 148, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 243 ms, sys: 8.35 ms, total: 251 ms\n", "Wall time: 251 ms\n" ] } ], "source": [ "%%time \n", "query =\"\"\" \n", "SELECT e.effectif, c.année, COUNT(*) as Count \n", "FROM cube, etat e2, effectif e, creation c\n", "WHERE \n", " cube.etat=e2.id AND\n", " cube.effectif=e.id AND\n", " e.effectif != \"\" AND\n", " cube.creation =c.id AND \n", " cast(c.année as number) > 1960 AND \n", " cast(c.année as number) < 2018 AND\n", " e2.état_administratif = \"Actif\"\n", "GROUP BY cube.etat, cube.effectif, cube.creation\n", "\"\"\"\n", "df = pandas.read_sql(query, db)\n", "df" ] }, { "cell_type": "code", "execution_count": 149, "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", "
effectifannéeCount
00 salarié20062
10 salarié19985
20 salarié20121
30 salarié20073
40 salarié20153
............
226932000 à 4999 salariés19832
226942000 à 4999 salariés19841
226952000 à 4999 salariés19841
226965000 à 9999 salariés20161
226975000 à 9999 salariés19831
\n", "

22698 rows × 3 columns

\n", "
" ], "text/plain": [ " effectif année Count\n", "0 0 salarié 2006 2\n", "1 0 salarié 1998 5\n", "2 0 salarié 2012 1\n", "3 0 salarié 2007 3\n", "4 0 salarié 2015 3\n", "... ... ... ...\n", "22693 2000 à 4999 salariés 1983 2\n", "22694 2000 à 4999 salariés 1984 1\n", "22695 2000 à 4999 salariés 1984 1\n", "22696 5000 à 9999 salariés 2016 1\n", "22697 5000 à 9999 salariés 1983 1\n", "\n", "[22698 rows x 3 columns]" ] }, "execution_count": 149, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df" ] }, { "cell_type": "code", "execution_count": 150, "metadata": {}, "outputs": [], "source": [ "pivot = df.pivot_table(\"Count\", index='année', columns=\"effectif\", aggfunc=\"sum\").fillna(0)\n", "pivot = pivot[reindex[1:]]" ] }, { "cell_type": "code", "execution_count": 151, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
effectif Etablissement non employeur 0 salarié 1 ou 2 salariés 3 à 5 salariés 6 à 9 salariés 10 à 19 salariés 20 à 49 salariés 50 à 99 salariés 100 à 199 salariés 200 à 249 salariés 250 à 499 salariés 500 à 999 salariés 1000 à 1999 salariés 2000 à 4999 salariés 5000 à 9999 salariés
année
19611421030100000000
19621702111000000000
19632012011100000000
19643103216100000000
19654006412000000000
19662403201000100000
19673003101200000000
19683222211000000000
19693712131010000000
19704304000000000000
19714905051010110000
19725101311200000000
19735007852000000000
19744515152100000000
19757715221210000000
19767432222201000000
19778027423100000000
1978142014836130000000
1979235116423400100000
1980303412113111012101000
198127762010158221012000
19822943309126920100000
198340598711516635213194404136821
19843828423320261487131020
198547177436384123104120110
1986548286633192624193111000
1987636166731244231179231000
198880117744122251942251000
198910212618151333019113102010
19901251202306538352566231000
19919221373612920251210170000
1992956268146263930138211000
199311032910752323027104110000
199411862511573394027137040000
199512142710284514036166012100
199611922712367584235105220000
199713713514383634350710247210
199814494115590686844248221100
19991578511831165761321712220100
2000165336161106516042147361000
2001177438205100575751174111100
20021913451941286871492410220100
20032020512001317682451114210100
20042271852641758863542616211100
20054916628317911575552012231000
20065371043241859991642811161000
200751111038721211297593716242000
2008559118439215134107832214162110
2009523127393224137121742916055000
201060512952027215611211431204111200
20115841485993481871411123924362010
20125662036183451891441105910092200
20133452267764082251581174413671000
20142002379034782631741253820164000
20156126411445292702211544724683000
2016339113095842751921083515570001
20170985639918614470166020010
" ], "text/plain": [ "" ] }, "execution_count": 151, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pivot.style.background_gradient(cmap='Blues')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Création d'entreprise par ville (importantes)" ] }, { "cell_type": "code", "execution_count": 174, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 922 ms, sys: 188 ms, total: 1.11 s\n", "Wall time: 1.11 s\n" ] } ], "source": [ "%%time \n", "query =\"\"\" \n", "SELECT c2.commune, c.année, COUNT(*) as Count \n", "FROM cube, localisation c2, creation c\n", "WHERE\n", " cube.localisation = c2.id AND\n", " cube.creation =c.id AND \n", " cast(c.année as number) > 1950 AND\n", " cast(c.année as number) < 2018 \n", "GROUP BY cube.localisation, cube.creation\n", "\"\"\"\n", "df = pandas.read_sql(query, db)\n", "df" ] }, { "cell_type": "code", "execution_count": 175, "metadata": {}, "outputs": [], "source": [ "pivot = df.pivot_table(\"Count\", index=\"année\", columns=\"commune\", aggfunc=sum).fillna(0)" ] }, { "cell_type": "code", "execution_count": 176, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 176, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "sommes = pivot.apply(sum)\n", "total = sum(sommes)\n", "pivot[sommes[sommes > total/30].keys()].plot(figsize=(20,10))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Affinée sur Lille" ] }, { "cell_type": "code", "execution_count": 177, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "CPU times: user 957 ms, sys: 208 ms, total: 1.17 s\n", "Wall time: 1.16 s\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "%%time \n", "query =\"\"\" \n", "SELECT c2.code_postal, c.année, COUNT(*) as Count \n", "FROM cube, localisation c2, creation c\n", "WHERE\n", " cube.localisation = c2.id AND\n", " cube.creation =c.id AND \n", " cast(c.année as number) > 1950 AND\n", " cast(c.année as number) < 2018 \n", "GROUP BY cube.localisation, cube.creation\n", "\"\"\"\n", "df = pandas.read_sql(query, db)\n", "pivot = df.pivot_table(\"Count\", index=\"année\", columns=\"code_postal\", aggfunc=sum).fillna(0)\n", "sommes = pivot.apply(sum)\n", "total = sum(sommes)\n", "pivot[sommes[sommes > total/30].keys()].plot(figsize=(20,10))" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.3" } }, "nbformat": 4, "nbformat_minor": 2 }