{"id":997,"date":"2026-04-28T12:10:47","date_gmt":"2026-04-28T12:10:47","guid":{"rendered":"https:\/\/pythonia.fr\/?page_id=997"},"modified":"2026-04-28T12:14:56","modified_gmt":"2026-04-28T12:14:56","slug":"rag-en-prodi","status":"publish","type":"page","link":"https:\/\/pythonia.fr\/index.php\/rag-en-prodi\/","title":{"rendered":"RAG en prodi"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-page\" data-elementor-id=\"997\" class=\"elementor elementor-997\">\n\t\t\t\t<div class=\"elementor-element elementor-element-7312b2c e-grid e-con-boxed e-con e-parent\" data-id=\"7312b2c\" data-element_type=\"container\" data-e-type=\"container\" data-settings=\"{&quot;background_background&quot;:&quot;gradient&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-ad2445a elementor-widget elementor-widget-heading\" data-id=\"ad2445a\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\">Agents RAG en production<\/h2>\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-46d93df e-grid e-con-boxed e-con e-parent\" data-id=\"46d93df\" data-element_type=\"container\" data-e-type=\"container\" data-settings=\"{&quot;background_background&quot;:&quot;gradient&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-48986f5 elementor-widget elementor-widget-heading\" data-id=\"48986f5\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\"><div style=\"display:flex;gap:40px;flex-wrap:wrap;font-family:sans-serif\">\n  <!-- Colonne gauche : Objectifs -->\n  <div style=\"flex:2;min-width:300px\">\n    <h3 style=\"color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px;font-size:16px\">\n      Objectifs :\n    <\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:2;margin-left:20px\">\n      <li>Comprendre l'architecture d'un syst\u00e8me RAG et identifier ses cas d'usage en entreprise<\/li>\n      <li>Construire un pipeline RAG complet : embeddings, chunking, vector store et g\u00e9n\u00e9ration<\/li>\n      <li>Mettre en \u0153uvre des techniques de retrieval avanc\u00e9es : recherche hybride, reranking, HyDE<\/li>\n      <li>Concevoir des architectures RAG agentiques et multimodales (PDF, tableaux, images)<\/li>\n      <li>\u00c9valuer la qualit\u00e9 d'un syst\u00e8me RAG avec RAGAS et mettre en place l'observabilit\u00e9<\/li>\n      <li>D\u00e9ployer un RAG en production en ma\u00eetrisant s\u00e9curit\u00e9, co\u00fbts et latence<\/li>\n    <\/ul>\n  <\/div>\n  <!-- Colonne droite : Tarif + liens -->\n  <div style=\"flex:1;min-width:200px;background:#0B1D33;padding:25px;border-radius:12px;height:fit-content\">\n    <div style=\"padding:12px 0;border-bottom:1px solid #1a3a5c\">\n      <span style=\"color:#8899AA;font-size:12px;text-transform:uppercase;letter-spacing:0.5px\">Tarif inter \/ participant<\/span>\n      <div style=\"color:#FFFFFF;font-size:22px;font-weight:bold;margin-top:4px\">\n        2 500 \u20ac <span style=\"font-size:13px;font-weight:normal;color:#8899AA\">HT<\/span>\n      <\/div>\n    <\/div>\n    <a href=\"http:\/\/pythonia.fr\/wp-content\/uploads\/2026\/04\/Programme_RAG_Production_Pythonia.pdf\" target=\"_blank\" style=\"display:block;color:#FFFFFF;font-size:15px;text-decoration:none;padding:12px 0;border-bottom:1px solid #1a3a5c\">\n      \ud83d\udcc4 Programme (PDF)\n    <\/a>\n    <a href=\"#\" id=\"btn-dates\" style=\"display:block;color:#FFFFFF;font-size:15px;text-decoration:none;padding:12px 0;border-bottom:1px solid #1a3a5c\">\n      \ud83d\udcc5 Voir les dates\n    <\/a>\n    <button id=\"btn-contact\" style=\"display:block;width:100%;margin-top:20px;padding:15px 20px;background:#5DADE2;color:#FFFFFF;font-size:15px;font-weight:bold;border:none;border-radius:8px;cursor:pointer\">\n      \u2709\ufe0f Demande d'information\n    <\/button>\n  <\/div>\n<\/div><\/h2>\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-95c5e86 e-grid e-con-boxed e-con e-parent\" data-id=\"95c5e86\" data-element_type=\"container\" data-e-type=\"container\" data-settings=\"{&quot;background_background&quot;:&quot;gradient&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-512795e elementor-widget elementor-widget-heading\" data-id=\"512795e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\"><div style=\"display:flex;gap:40px;flex-wrap:wrap\">\n\n  <!-- Colonne gauche : Contenu -->\n  <div style=\"flex:2;min-width:300px\">\n\n    <h3 style=\"font-size:24px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">Public vis\u00e9<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>D\u00e9veloppeurs Python souhaitant ma\u00eetriser la conception de syst\u00e8mes RAG en production<\/li>\n      <li>Data scientists et ML engineers charg\u00e9s de mettre en \u0153uvre des solutions IA documentaires<\/li>\n      <li>Tech leads et architectes logiciels int\u00e9grant des capacit\u00e9s IA dans leurs applications<\/li>\n      <li>Consultants et freelances accompagnant des entreprises sur leurs projets IA g\u00e9n\u00e9rative<\/li>\n    <\/ul>\n\n    <h3 style=\"font-size:24px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">Objectifs p\u00e9dagogiques<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Comprendre l'architecture d'un syst\u00e8me RAG et identifier ses cas d'usage en entreprise<\/li>\n      <li>Choisir et impl\u00e9menter une strat\u00e9gie d'embedding et de chunking adapt\u00e9e au contexte<\/li>\n      <li>S\u00e9lectionner et configurer un vector store pertinent (Chroma, Qdrant, pgvector)<\/li>\n      <li>Mettre en \u0153uvre des techniques de retrieval avanc\u00e9es : recherche hybride, reranking, HyDE<\/li>\n      <li>Concevoir des architectures RAG agentiques avec routing et raisonnement multi-\u00e9tapes<\/li>\n      <li>Traiter des documents complexes : PDF, tableaux, images (RAG multimodal)<\/li>\n      <li>\u00c9valuer objectivement la qualit\u00e9 d'un syst\u00e8me RAG avec RAGAS et DeepEval<\/li>\n      <li>Mettre en place l'observabilit\u00e9 en production avec LangSmith ou Langfuse<\/li>\n      <li>S\u00e9curiser un syst\u00e8me RAG : prompt injection, gestion des PII, conformit\u00e9 RGPD<\/li>\n      <li>Optimiser la latence et les co\u00fbts : caching s\u00e9mantique, batching, mod\u00e8les hybrides<\/li>\n      <li>D\u00e9ployer un RAG complet avec FastAPI et Docker dans une architecture scalable<\/li>\n    <\/ul>\n\n    <h3 style=\"font-size:24px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">Pr\u00e9requis<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Ma\u00eetrise de Python interm\u00e9diaire : fonctions, classes, gestion de d\u00e9pendances<\/li>\n      <li>Exp\u00e9rience pr\u00e9alable avec au moins une API LLM (OpenAI, Anthropic Claude, Mistral)<\/li>\n      <li>Notions de base en NLP : tokenisation, embeddings, similarit\u00e9 s\u00e9mantique<\/li>\n      <li>Connaissance d'un environnement de d\u00e9veloppement (VS Code, Git, terminal)<\/li>\n      <li>Anglais technique en lecture (documentation, articles de recherche)<\/li>\n    <\/ul>\n\n    <h3 style=\"font-size:24px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">M\u00e9thodes p\u00e9dagogiques<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Alternance de th\u00e9orie (30%) et de pratique (70%)<\/li>\n      <li>D\u00e9monstrations live sur API r\u00e9elles (Claude, OpenAI, Cohere) et frameworks de production<\/li>\n      <li>P\u00e9dagogie par la comparaison : chaque concept d\u00e9montr\u00e9 sur plusieurs approches<\/li>\n      <li>Projet fil rouge sur les 4 jours avec un dataset m\u00e9tier repr\u00e9sentatif<\/li>\n      <li>Soutenance individuelle du projet final avec retour personnalis\u00e9 du formateur<\/li>\n      <li>Code source, notebooks et template RAG production-ready remis aux stagiaires<\/li>\n    <\/ul>\n\n    <h1 style=\"font-size:24px;color:#FFFFFF;font-weight:bold;margin-top:40px;margin-bottom:20px\">Programme d\u00e9taill\u00e9<\/h1>\n\n    <h3 style=\"font-size:16px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">JOUR 1 \u2014 Fondamentaux du RAG et premier pipeline<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Pourquoi le RAG : limites des LLM, hallucinations, knowledge cutoff<\/li>\n      <li>RAG vs fine-tuning : co\u00fbts, maintenance, cas d'usage en entreprise<\/li>\n      <li>Typologie des cas d'usage : Q&amp;A documentaire, support client, copilote m\u00e9tier<\/li>\n      <li>Architecture d'un pipeline RAG : ingestion, indexation, retrieval, g\u00e9n\u00e9ration<\/li>\n      <li>Panorama des embeddings : OpenAI, Cohere, BGE, E5, mod\u00e8les open source<\/li>\n      <li>Crit\u00e8res de choix : dimension, langue, co\u00fbt, latence, performance MTEB<\/li>\n      <li>Comparatif des vector stores : Chroma, Qdrant, Pinecone, pgvector, Weaviate<\/li>\n      <li>Crit\u00e8res de s\u00e9lection : self-hosted vs managed, scalabilit\u00e9, filtres m\u00e9tier<\/li>\n      <li>Strat\u00e9gies de chunking : fixe, r\u00e9cursif, s\u00e9mantique, par structure de document<\/li>\n      <li>Pi\u00e8ges du chunking : perte de contexte, overlap, granularit\u00e9 optimale<\/li>\n      <li>Construction d'un pipeline avec LangChain et LlamaIndex : comparaison<\/li>\n    <\/ul>\n    <p style=\"font-size:14px;color:#FFFFFF;margin-left:20px;margin-top:15px\"><strong>\u25a0 Travaux pratiques :<\/strong><br>\n    \u2192 TP1 : G\u00e9n\u00e9rer et comparer des embeddings (OpenAI vs mod\u00e8le open source) sur un corpus m\u00e9tier<br>\n    \u2192 TP2 : Impl\u00e9menter trois strat\u00e9gies de chunking et mesurer leur impact qualitatif<br>\n    \u2192 TP3 (projet fil rouge) : Premier RAG fonctionnel sur un corpus de documents internes<\/p>\n\n    <h3 style=\"font-size:16px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">JOUR 2 \u2014 Retrieval avanc\u00e9 : la diff\u00e9rence POC \/ production<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Limites de la recherche purement vectorielle : recall, entit\u00e9s, jargon m\u00e9tier<\/li>\n      <li>Recherche lexicale avec BM25 : principes et impl\u00e9mentation<\/li>\n      <li>Recherche hybride : fusion BM25 + dense, pond\u00e9ration, RRF (Reciprocal Rank Fusion)<\/li>\n      <li>Reranking avec cross-encoders : pourquoi et quand l'utiliser<\/li>\n      <li>Int\u00e9gration de Cohere Rerank et de mod\u00e8les open source (bge-reranker)<\/li>\n      <li>Mesure du gain : pr\u00e9cision, MRR, NDCG<\/li>\n      <li>Query transformation : pourquoi r\u00e9\u00e9crire la requ\u00eate utilisateur<\/li>\n      <li>HyDE (Hypothetical Document Embeddings) : principe et mise en \u0153uvre<\/li>\n      <li>Multi-query retrieval : g\u00e9n\u00e9rer plusieurs variantes et fusionner<\/li>\n      <li>Step-back prompting : remonter \u00e0 la question g\u00e9n\u00e9rale<\/li>\n      <li>Contextual Retrieval (technique Anthropic) : enrichir chaque chunk de son contexte<\/li>\n      <li>Metadata filtering : combiner recherche s\u00e9mantique et filtres structur\u00e9s<\/li>\n    <\/ul>\n    <p style=\"font-size:14px;color:#FFFFFF;margin-left:20px;margin-top:15px\"><strong>\u25a0 Travaux pratiques :<\/strong><br>\n    \u2192 TP1 : Impl\u00e9menter une recherche hybride BM25 + dense et mesurer le gain<br>\n    \u2192 TP2 : Ajouter un reranker Cohere et comparer les r\u00e9sultats<br>\n    \u2192 TP3 (challenge) : Benchmarker 5 strat\u00e9gies de retrieval sur un dataset annot\u00e9 avec rapport de performance<\/p>\n\n    <h3 style=\"font-size:16px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">JOUR 3 \u2014 RAG agentique et multimodal<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Du RAG lin\u00e9aire au RAG agentique : quand et pourquoi<\/li>\n      <li>Routers : diriger la requ\u00eate vers la bonne source de donn\u00e9es<\/li>\n      <li>Self-querying : l'agent g\u00e9n\u00e8re ses propres filtres de metadata<\/li>\n      <li>Multi-step reasoning : d\u00e9composition de requ\u00eates complexes<\/li>\n      <li>RAG conversationnel : gestion de la m\u00e9moire et du contexte multi-tours<\/li>\n      <li>Outils d'orchestration : LangGraph, LlamaIndex Workflows<\/li>\n      <li>G\u00e9n\u00e9ration avec citations structur\u00e9es et tra\u00e7abilit\u00e9<\/li>\n      <li>Documents complexes : PDF avec tableaux, images, sch\u00e9mas<\/li>\n      <li>Outils de parsing avanc\u00e9 : Unstructured, LlamaParse, Docling<\/li>\n      <li>RAG multimodal : embeddings d'images, description automatique par VLM<\/li>\n      <li>Introduction au Graph RAG : principes et cas d'usage<\/li>\n      <li>Knowledge graphs : extraction d'entit\u00e9s et de relations<\/li>\n    <\/ul>\n    <p style=\"font-size:14px;color:#FFFFFF;margin-left:20px;margin-top:15px\"><strong>\u25a0 Travaux pratiques :<\/strong><br>\n    \u2192 TP1 : Construire un agent RAG avec routing entre plusieurs sources (documentation, base de donn\u00e9es, FAQ)<br>\n    \u2192 TP2 : Traiter un corpus de PDF techniques contenant tableaux et sch\u00e9mas<br>\n    \u2192 TP3 (mini-projet) : Assembler un RAG conversationnel multi-tours avec m\u00e9moire et citations<\/p>\n\n    <h3 style=\"font-size:16px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">JOUR 4 \u2014 \u00c9valuation, production, s\u00e9curit\u00e9 et projet final<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Pourquoi l'\u00e9valuation est critique : sans m\u00e9triques, pas d'am\u00e9lioration<\/li>\n      <li>Construire un dataset d'\u00e9valuation : annotation manuelle vs synth\u00e9tique<\/li>\n      <li>RAGAS : faithfulness, answer relevancy, context precision, context recall<\/li>\n      <li>DeepEval et Phoenix : alternatives et sp\u00e9cificit\u00e9s<\/li>\n      <li>\u00c9valuation LLM-as-a-judge : forces, limites, biais<\/li>\n      <li>Observabilit\u00e9 en production : LangSmith, Langfuse, Arize<\/li>\n      <li>Traces, spans, feedback utilisateur, d\u00e9tection de r\u00e9gression<\/li>\n      <li>S\u00e9curit\u00e9 : prompt injection, jailbreaks, exfiltration de donn\u00e9es sensibles<\/li>\n      <li>Gestion des PII : d\u00e9tection, anonymisation, conformit\u00e9 RGPD<\/li>\n      <li>Optimisation des co\u00fbts : caching s\u00e9mantique, batching, mod\u00e8les en cascade<\/li>\n      <li>Latence : streaming, parall\u00e9lisation du retrieval, pr\u00e9-calcul<\/li>\n      <li>D\u00e9ploiement : FastAPI + Docker, architecture scalable, CI\/CD<\/li>\n    <\/ul>\n    <p style=\"font-size:14px;color:#FFFFFF;margin-left:20px;margin-top:15px\"><strong>\u25a0 Travaux pratiques :<\/strong><br>\n    \u2192 TP1 : Mettre en place un pipeline d'\u00e9valuation RAGAS reproductible<br>\n    \u2192 TP2 : Configurer l'observabilit\u00e9 et le monitoring avec LangSmith ou Langfuse<br>\n    \u2192 TP3 (Projet final) : RAG complet d\u00e9ploy\u00e9 en API FastAPI avec endpoint s\u00e9curis\u00e9, soutenance individuelle<\/p>\n\n    <h3 style=\"font-size:16px;color:#FFFFFF;font-weight:bold;margin-top:25px;margin-bottom:10px\">\u00c9valuation<\/h3>\n    <ul style=\"font-size:14px;color:#FFFFFF;line-height:1.8;margin-left:20px\">\n      <li>Test de positionnement technique en d\u00e9but de formation<\/li>\n      <li>\u00c9valuations formatives : exercices pratiques corrig\u00e9s, benchmarks, QCM interm\u00e9diaires<\/li>\n      <li>QCM mi-parcours de 20 questions (Jour 2)<\/li>\n      <li>QCM final de 30 questions (Jour 4) \u2014 crit\u00e8re de r\u00e9ussite : 60%<\/li>\n      <li>Soutenance du projet final avec d\u00e9monstration et analyse des choix techniques<\/li>\n      <li>Attestation de fin de formation d\u00e9livr\u00e9e<\/li>\n    <\/ul>\n\n  <\/div>\n<\/div><\/h2>\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Agents RAG en production Objectifs : Comprendre l&#8217;architecture d&#8217;un syst\u00e8me RAG et identifier ses cas d&#8217;usage en entreprise Construire un [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"site-sidebar-layout":"no-sidebar","site-content-layout":"","ast-site-content-layout":"full-width-container","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"disabled","ast-breadcrumbs-content":"","ast-featured-img":"disabled","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"class_list":["post-997","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/pages\/997","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/comments?post=997"}],"version-history":[{"count":7,"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/pages\/997\/revisions"}],"predecessor-version":[{"id":1007,"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/pages\/997\/revisions\/1007"}],"wp:attachment":[{"href":"https:\/\/pythonia.fr\/index.php\/wp-json\/wp\/v2\/media?parent=997"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}