25:["$","title",null,{"children":"Les tableaux de données | Nicolas Lambert | Observable"}] 2a:[{"id":"79fac01a413c7fdcf0c52a9174b61f1e9863b6658e55eb3fe420e1cbf5d0cec78ec9cf206fd693ce01b777e06e874f68bc8619a030c06cd77e4c04b10a393b8a","href":"https://static.observableusercontent.com/files/79fac01a413c7fdcf0c52a9174b61f1e9863b6658e55eb3fe420e1cbf5d0cec78ec9cf206fd693ce01b777e06e874f68bc8619a030c06cd77e4c04b10a393b8a","name":"mydata.zip","type":"application/zip","lastModified":1695915582961.889,"size":4637},{"id":"78f56445daa39b8fbf724e2529f4b3c83053f9de7ac2799a65845a5385372c33ac6497687f5dbbb0f7e456221ed5df6a0b4cfa3bd3e4738cec1c9d0444fffa0d","href":"https://static.observableusercontent.com/files/78f56445daa39b8fbf724e2529f4b3c83053f9de7ac2799a65845a5385372c33ac6497687f5dbbb0f7e456221ed5df6a0b4cfa3bd3e4738cec1c9d0444fffa0d","name":"json.png","type":"image/png","lastModified":1695915582961.889,"size":2576},{"id":"d1491f9213c096a28ad4a6f0c81c3757c9d6a360623220e355ca3485e5be78a1c77f6bf1559863aafbe5bca71321a2561cdffb46c8e64c4d9b3e1619c9ec2073","href":"https://static.observableusercontent.com/files/d1491f9213c096a28ad4a6f0c81c3757c9d6a360623220e355ca3485e5be78a1c77f6bf1559863aafbe5bca71321a2561cdffb46c8e64c4d9b3e1619c9ec2073","name":"mydata.xlsx","type":"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet","lastModified":1695915582961.889,"size":16991},{"id":"94f52eac307a69a1a6ff0a6eda071df78d78addc58dff4711ecf9469a513ec05811c284fa14e269279703faa4187de9acf0bbf505866ae9572dc43eb6c302301","href":"https://static.observableusercontent.com/files/94f52eac307a69a1a6ff0a6eda071df78d78addc58dff4711ecf9469a513ec05811c284fa14e269279703faa4187de9acf0bbf505866ae9572dc43eb6c302301","name":"clic.png","type":"image/png","lastModified":1695915582961.889,"size":42388},{"id":"75dd806efd5236e543e1584ae2d4931ee3add928a556ca87ad98410c53d3516bffb8bbb419601f3c4bd37819fda6468bb4b523cfc9d61b0b6aa8f0698447213a","href":"https://static.observableusercontent.com/files/75dd806efd5236e543e1584ae2d4931ee3add928a556ca87ad98410c53d3516bffb8bbb419601f3c4bd37819fda6468bb4b523cfc9d61b0b6aa8f0698447213a","name":"mydata2@2.csv","type":"text/csv","lastModified":1695915582961.889,"size":1305},{"id":"cc2e6671be495f9663c69d8b8ab07055158bf4679bd41c218e714478b420f257cf972ef45f4e7208d45f4a8e86913153569ab16a9bdda2189bcc563359e950e4","href":"https://static.observableusercontent.com/files/cc2e6671be495f9663c69d8b8ab07055158bf4679bd41c218e714478b420f257cf972ef45f4e7208d45f4a8e86913153569ab16a9bdda2189bcc563359e950e4","name":"mydata@2.csv","type":"text/csv","lastModified":1695915582961.889,"size":3815},{"id":"a3a6dc7b3f96508c27b73ac9d22c983211105fb3ec19de45884c88bc2e7329f3f70f80af617aa7d9df91fa6c575fbdbbd453be9c0b841556f4dc664b796911d3","href":"https://static.observableusercontent.com/files/a3a6dc7b3f96508c27b73ac9d22c983211105fb3ec19de45884c88bc2e7329f3f70f80af617aa7d9df91fa6c575fbdbbd453be9c0b841556f4dc664b796911d3","name":"mydata.json","type":"application/json","lastModified":1695915582961.889,"size":19968},{"id":"bc6ae557078c1abec7ae9d8ece38d93a6fbef673c3d02b0f1127d5bcda87a46f521712dc0d34325ee0a5c34a518a6c1ad3e571dee6d1369ac80533f49df35e7f","href":"https://static.observableusercontent.com/files/bc6ae557078c1abec7ae9d8ece38d93a6fbef673c3d02b0f1127d5bcda87a46f521712dc0d34325ee0a5c34a518a6c1ad3e571dee6d1369ac80533f49df35e7f","name":"attach.png","type":"image/png","lastModified":1695915582961.889,"size":34255},{"id":"f40751ed61486fd244bd68fa0b9f7961e192cedd7caf2457ee23e784f584208647a23af541bf0259425f2568a85c343a19393ee1abc8aab593aa86222fa64d56","href":"https://static.observableusercontent.com/files/f40751ed61486fd244bd68fa0b9f7961e192cedd7caf2457ee23e784f584208647a23af541bf0259425f2568a85c343a19393ee1abc8aab593aa86222fa64d56","name":"tableau.png","type":"image/png","lastModified":1700063782351.564,"size":163029},{"id":"219f79f194262aaf9cf5a1639574eecdc5f8dbc2244608e4dbbc04308e4730f01e34978dfb2fbe88e7c50b807e088455cc79346333ec52092f3b2707d9626849","href":"https://static.observableusercontent.com/files/219f79f194262aaf9cf5a1639574eecdc5f8dbc2244608e4dbbc04308e4730f01e34978dfb2fbe88e7c50b807e088455cc79346333ec52092f3b2707d9626849","name":"cities.json","type":"application/json","lastModified":1764149538003.796,"size":144791},{"id":"9499efa92eeb6259145349c236483338c8cc76128e1417f69f5e0ba4f3a5a147fc53f41d4643f307c2ba7f0191156cf6dfbe7ee7f4fce5f7b4a5713337b5056f","href":"https://static.observableusercontent.com/files/9499efa92eeb6259145349c236483338c8cc76128e1417f69f5e0ba4f3a5a147fc53f41d4643f307c2ba7f0191156cf6dfbe7ee7f4fce5f7b4a5713337b5056f","name":"maddison.csv","type":"text/csv","lastModified":1764149547335.014,"size":9810}] 28:{"id":"f8d6fe0ea1b78e70","slug":"les-tableaux-de-donnees","title":"Les tableaux de données","thumbnail":"2a408d2bc9fa914cb8f8c1ab009dedac02abc44f980efad8c07ac72099c36c6f","access_level":"public","create_time":"$D2023-09-28T15:39:42.961Z","update_time":"$D2025-11-26T09:33:12.432Z","view_time":null,"first_public_time":"$D2023-11-10T10:31:41.286Z","fork":{"id":"6d3dae2f68d78677","slug":"manipuler-les-tableau-de-donnees","title":"Manipuler les tableau de données","owner":{"id":"2f0d745704536963","login":"cartographie","name":"Master 1 GAED (Université Paris Cité)"}},"stars":1,"starred":false,"owner":{"type":"team","id":"6457aadd156df119","login":"neocarto","name":"Nicolas Lambert","avatar_url":"https://avatars-next.observableusercontent.com/c01b6664b447a3bb88f88dbfa170e3ef1d61f3a5b13c2fc613c8da5ef8bccb41","tier":"starter_2024"},"version":898,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"01f8081399df8778","login":"neocarto","name":"Nicolas Lambert","avatar_url":"https://avatars-next.observableusercontent.com/c01b6664b447a3bb88f88dbfa170e3ef1d61f3a5b13c2fc613c8da5ef8bccb41","tier":null}],"members":[{"type":"individual","id":"01f8081399df8778","login":"neocarto","name":"Nicolas Lambert","avatar_url":"https://avatars-next.observableusercontent.com/c01b6664b447a3bb88f88dbfa170e3ef1d61f3a5b13c2fc613c8da5ef8bccb41","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":true,"trashed":false,"subscribed":null,"forks":1,"collection":{"id":"021e4058de70d966","slug":"observable-cest-quoi","title":"Observable pour les geographes","owner":{"id":"6457aadd156df119","login":"neocarto","name":"Nicolas Lambert"}},"type":"notebook","body":{"title":"Les tableaux de données","cells":[{"id":0,"value":"# Les tableaux de données","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":15,"value":"Dans ce notebook, nous voyons comment ajouter et manipuler des tableaux de données.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":852,"value":"FileAttachment(\"tableau.png\").image({ width: 700 })","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":17,"value":"

FICHIERS ATTACHÉS

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":23,"value":"Le moyen de plus simple d'ajouter des données dans Observable, est de les attacher comme une piece jointe via l'épingle à droite de l'écran. ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":364,"value":"attach = FileAttachment(\"attach.png\").image()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":367,"value":"Puis, on utilise l'instruction [**`FileAttachment(\"nom-du-fichier\")`**](https://observablehq.com/@observablehq/file-attachments). Puis, en fonction du format du fichier, on utilisera différentes instructions.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":33,"value":"

FICHIERS .CSV

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":25,"value":"data = FileAttachment(\"mydata@2.csv\").csv({ typed: true })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":108,"value":"l'option `{typed: true}` est facultative. Elle permet de typer automatiquement les données.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":128,"value":"Si on a un autre délimiteur autre que la virgule, on pourra utiliser l'instruction `d3.dsv()` en lui passant le bon séparateur.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":157,"value":"data2 = d3.dsv(\";\", await FileAttachment(\"mydata2@2.csv\").url(), d3.autoType)","pinned":false,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":686,"value":"data2","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":59,"value":"

FICHIERS EXCEL

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":65,"value":"Pour importer un fichier excel, on fait ça en deux temps. D'abord on importe le classeur.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":62,"value":"Workbook = FileAttachment(\"mydata.xlsx\").xlsx()","pinned":false,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":67,"value":"Puis on sélectionne une feuille.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":69,"value":"data3 = Workbook.sheet(\"data\", { headers: true })","pinned":false,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":114,"value":"l'option `{header: true}` permet de dire que la première ligne contient le nom des variables. ","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":137,"value":"

FICHIERS ZIP

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":175,"value":"Si vos données sont contenues dans un fichier zip, pas de problème. Tout d'abord, on import le fichier zip.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":173,"value":"zip = FileAttachment(\"mydata.zip\").zip()","pinned":true,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":189,"value":"Puis, on va chercher le fichier qu'on souhaite à l'intérieur.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":182,"value":"data4 = zip.file(\"mydata.csv\").csv({ typed: true })","pinned":false,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":396,"value":"

FICHIERS JSON

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":398,"value":"Vous pouvez aussi utiliser bien évidemment directement un fichier JSON.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":400,"value":"data5 = FileAttachment(\"mydata.json\").json()","pinned":false,"hidden":true,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":77,"value":"json = FileAttachment(\"json.png\").image()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":5,"value":"En JavaScript, les tableaux de données n'ont pas vraiment la forme de tableaux de données en deux dimensions avec des lignes et des colonnes. Les données sont stockées au format **JSON** *(JavaScript Object Notation)*. Elles prennent la forme d'un tableau JavaScript **[**...**]** contenant des objets **{**...**}** avec des clés et des valeurs.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":13,"value":"data","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":84,"value":"

VISUALISER LES DONNÉES

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":99,"value":"On peut rapidement visualiser des données en utilisant [**`Inputs.table()`**](https://github.com/observablehq/inputs#Table).","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":101,"value":"Inputs.table(data, { rows: 5 })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":87,"value":"

MANIPULER LES DONNÉES

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":97,"value":"Manipuler les données en JavaScript est parfois source de confusion. Voici quelques opérations utiles.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":145,"value":"Partons du fichier data","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":139,"value":"- **Nom des colonnes**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":143,"value":"data.columns","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":212,"value":"- **Récupérer une variable**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":848,"value":"data.map((d) => d.name)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":850,"value":"Ou","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":222,"value":"data.map((d) => d[\"name\"])","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":216,"value":"- **Filtrer**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":224,"value":"data.filter((d) => d.IDH > 0.5)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":230,"value":"data.filter((d) => d.nom == \"Botswana\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":220,"value":"- **Trier**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":232,"value":"data.sort((a, b) => d3.ascending(a.POP, b.POP))","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":237,"value":"data.sort((a, b) => d3.descending(a.POP, b.POP))","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":462,"value":"- **Modifier le tableau**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":482,"value":"Avec la méthode **`map()`**, on peut créer un nouveau tableau en selectionnant des variables, en définissant leur ordre et leur nom. Cette méthode sert également à modifier les valeurs.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":816,"value":"newdata1 = data.map((d) => ({ ...d, PIBhab: d.PIB / d.POP }))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":465,"value":"newdata2 = data.map((d) => ({\n id: d.iso3,\n nom: d.name,\n population: Math.round(d.POP * 1000) // on convertit en milliers et on arrondi\n}))","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":476,"value":"Inputs.table(newdata2, { rows: 5, sort: \"id\", width: 500 })","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":89,"value":"

LA BIBLIOTHÈQUE ARQUERO

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":239,"value":"Un autre moyen de manipuler les données est d'utiliser la bibliothèque `Arquero`. Arquero est inspiré par le package R dplyr. Elle permet de manipuler et transformer les données comme si on avait un tableau en deux dimensions.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":243,"value":"La bibliothèque Arquero est directement disponible dans Observable sous le nom **`aq`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":246,"value":"De nombreux opérateurs sont également disponibles.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":249,"value":"op = aq.op","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":452,"value":"- **JSON -> Arquero**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":254,"value":"Pour travailler avec Arquero, il est nécessaire de transformer les données au format Arqueo. Pour cela, on utilise **`aq.from()`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":257,"value":"dt = aq.from(data)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":454,"value":"- **Arquero -> JSON**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":429,"value":"Dans ce format Arquero, nous pourrons alors effectuer de nombreuses opérations de tri, de selection, etc.Puis, avec l'instruction **`object()`** ou le spread operator de JavaScript, on peut récupérer un tableau JSON normal.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":432,"value":"dt.objects()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":445,"value":"ou","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":447,"value":"[...dt]","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":264,"value":"- **Infos sur le tableau**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":268,"value":"On peut récupérer quelques infos sur le tableau, comme le nom des variables et la taille du tableau","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":310,"value":"dt.columnNames()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":271,"value":"dt.numCols()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":274,"value":"dt.numRows()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":343,"value":"- **Visualiser le tableau**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":345,"value":"En HTML (les 5 premières lignes)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":347,"value":"html`${dt.slice(0, 5).toHTML()}`","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":350,"value":"En Markdown (les 5 premières lignes)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":355,"value":"md`${dt.slice(0, 5).toMarkdown()}`","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":438,"value":"Ou, comme précédemment, avec **`Inputs.table`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":441,"value":"Inputs.table(dt.objects(), { rows: 5 })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":281,"value":"- **Récupérer une variable**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":283,"value":"dt.array(\"name\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":298,"value":"dt.values(\"name\") // executez cette cellule","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":288,"value":"- **Récupérer une valeur du tableau**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":291,"value":"Avec **`dt.get`**, on peut récupérer n'importe quelle valeur du tableau. Par exemple, la valeur qui correspond à l'indice 10 de variable iso3 (la 11e valeur).","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":305,"value":"dt.get(\"iso3\", 10)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":422,"value":"Ou alors, on oeut utiliser **`dt.getter`**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":424,"value":"getiso = dt.getter(\"iso3\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":427,"value":"getiso(10)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":324,"value":"- **Trier**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":326,"value":"Pour trier, on utilise l'instruction **`orderby`**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":337,"value":"dt.orderby(\"POP\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":328,"value":"dt.orderby(aq.desc(\"POP\"))","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":652,"value":"- **Supprimer les doublons**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":657,"value":"avec **`dedupe`**, on peut supprimer des lignes en dédoublonnant une variable.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":654,"value":"dt.dedupe(\"SUBREG\").objects()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":516,"value":"- **Selectionner les colonnes**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":520,"value":"dt.select(\"name\", \"POP\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":540,"value":"- **Extraire des lignes**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":532,"value":"Ici, on récupère les lignes 2 à 5","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":527,"value":"dt.slice(2, 5)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":545,"value":"- **Echantillon**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":547,"value":"Quand on a un gros tableau de données, il peut être utile d'en afficher un échantillon. Ici, on récupère 10 lignes au hasard.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":549,"value":"dt.sample(10)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":557,"value":"- **Changer l'emplacement d'une colonne**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":568,"value":"On peut changer l'emplacement d'une colonne avec **`relocate`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":559,"value":"dt.relocate(\"SUBREG\", { after: \"nom\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":564,"value":"dt.relocate(\"name\", { before: \"iso3\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":566,"value":"- **Ajouter un nouveau champ**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":570,"value":"On peut ajouter un nouveau champ avec **`derive`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":574,"value":"dt.derive({ pdppercapita: (d) => d.PIB / d.POP }, { after: \"PIB\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":580,"value":"- **Renommer un champ**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":582,"value":"On peut renommer nouveau champ avec **`rename`**.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":584,"value":"dt.rename({ name: \"iso3\", iso3: \"id\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":598,"value":"- **Filtrer**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":601,"value":"dt.filter((d) => d[\"IDH\"] > 0.5)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":613,"value":"- **Résumé statistique**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":615,"value":"On peut utiliser **`rollup`** et les opérateurs d'Arquero pour résumer une variable.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":617,"value":"x = dt\n .rollup({\n min: (d) => op.min(d[\"POP\"]),\n max: (d) => op.max(d[\"POP\"]),\n mean: (d) => op.mean(d[\"POP\"]),\n median: (d) => op.median(d[\"POP\"]),\n sum: (d) => op.sum(d[\"POP\"]),\n stdev: (d) => op.stdev(d[\"POP\"]),\n count: (d) => op.count()\n })\n .objects()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":622,"value":"Inputs.table(x)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":624,"value":"- **`groupby`**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":631,"value":"la méthode **`groupby`** permet de regrouper les données par catégories.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":626,"value":"dt.groupby(\"SUBREG\").objects({ grouped: true })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":634,"value":"On peut alors combiner `rollup` et `groupby` pour avoir un résumé statistique par région.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":636,"value":"summary = dt\n .groupby(\"SUBREG\")\n .rollup({\n min: (d) => d3.min(d[\"POP\"]),\n max: (d) => d3.max(d[\"POP\"]),\n mean: (d) => d3.mean(d[\"POP\"]),\n median: (d) => d3.median(d[\"POP\"]),\n sum: (d) => d3.sum(d[\"POP\"]),\n stdev: (d) => d3.stdev(d[\"POP\"]),\n count: (d) => d3.count()\n })\n .objects()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":642,"value":"Inputs.table(summary)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":660,"value":"- **Jointure**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":662,"value":"Avec **`join`**, on peut joindre deux table en se basant sur un champ commun.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":664,"value":"Par exemple, on a **`dt1`**","pinned":true,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":680,"value":"dt1 = aq.from(data).select(\"iso3\", \"name\", \"POP\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":671,"value":"Inputs.table(dt1.objects(), {\n rows: 5,\n sort: \"iso3\",\n width: 500,\n layout: \"auto\"\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":669,"value":"Et **`dt2`**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":676,"value":"dt2 = aq.from(data2).select(\"iso3\", \"ESPVIE\", \"AGEMED\").rename({ iso3: \"id\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":689,"value":"Inputs.table(dt2.objects(), { rows: 5, sort: \"id\", width: 500, layout: \"auto\" })","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":695,"value":"Avec **`join`**, on peut joindre les 2 tables.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":698,"value":"dt3 = dt1.join(dt2, [\"iso3\", \"id\"])","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":701,"value":"Inputs.table(dt3.objects(), {\n rows: 5,\n sort: \"iso3\",\n maxWidth: 700,\n layout: \"auto\"\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":718,"value":"- **Chaîner les opérations**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":721,"value":"Avec Arquero, on peut chainer autant d'opérations qu'on le souhaite","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":723,"value":"dt4 = aq\n .from(data)\n .select(\"iso3\", \"name\", \"POP\", \"PIB\")\n .orderby(\"POP\")\n .filter((d) => d[\"POP\"] > 50)\n .relocate(\"PIB\", { before: \"POP\" })","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":743,"value":"Inputs.table(dt4.objects(), {\n rows: 10,\n sort: \"iso3\",\n maxWidth: 700,\n layout: \"auto\"\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":885,"value":"

DUCKDB

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":888,"value":"DuckDB est un système de gestion de bases de données relationnelles (SGBDR) en mémoire, conçu pour être ultra-rapide pour l'analyse de données. Elle fonctionne directement dans la mémoire de l'ordinateur sans serveur dédié.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":891,"value":"database = DuckDBClient.of({\n cities: FileAttachment(\"cities.json\"),\n maddison: FileAttachment(\"maddison.csv\")\n})","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":893,"value":"maincities = database.query(`\nSELECT *\nFROM cities\nWHERE population > 10000\n`)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":895,"value":"C'est vraiment très puissant si on maîtrise le langage SQL. On peut par exemple calculer la population par pays.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":897,"value":"PopulationByCountries = database.query(`\nSELECT\n iso3,\nSUM(TRY_CAST(population AS DOUBLE)) AS total_population\nFROM cities\nGROUP BY iso3;\n`)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":91,"value":"

LES DATA TABLE

","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":774,"value":"Une autre façon très pratique de manipuler les données, est d'utiliser les data table d'Observable. Avec un fichier csv, il suffit juste cliquer sur le bouton ci-dessous pour insérer les données dans le notebook.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":776,"value":"

","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined"},{"id":783,"value":"mydat = await FileAttachment(\"mydata@2.csv\").csv({typed: \"auto\"})","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":786,"value":"Puis, vous pouvez chercher dans le tableau, trier (**`sort`**) le tableau sur n'importe quelle colonne, choisir les colonnes à afficher (**`Column`**), changer le nom des variables, les typer (string, number, date, boolean...), filtrer (**`Filter`**) et sélectionner des lignes (**`Slice`**). C'est vraiment très pratique (mais néanmoins moins traçable que si vous effectuez ces opérations en écrivant du code).","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":806,"value":"Ensuite, vous pouvez utiliser ce tableau modifier dans la suite de votre code.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":808,"value":"mydat","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined"},{"id":830,"value":"## *Appendix*","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"},{"id":828,"value":"","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined"}],"files":"$W2a","stdlib":"1","titling":"auto"}} 29:"$28:body" 24:E{"digest":"NEXT_REDIRECT;replace;/@neocarto/les-tableaux-de-donnees;307;"}