{"id":228216,"date":"2024-02-27T06:07:00","date_gmt":"2024-02-27T06:07:00","guid":{"rendered":"https:\/\/datascientest.com\/?p=228216"},"modified":"2024-02-20T10:07:07","modified_gmt":"2024-02-20T10:07:07","slug":"imbalanced-learn-tout-savoir","status":"publish","type":"post","link":"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir","title":{"rendered":"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"228216\" class=\"elementor elementor-228216\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-ef03a62 e-flex e-con-boxed e-con\" data-id=\"ef03a62\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-d39fe12 elementor-widget elementor-widget-heading\" data-id=\"d39fe12\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-heading-title{padding:0;margin:0;line-height:1}.elementor-widget-heading .elementor-heading-title[class*=elementor-size-]>a{color:inherit;font-size:inherit;line-height:inherit}.elementor-widget-heading .elementor-heading-title.elementor-size-small{font-size:15px}.elementor-widget-heading .elementor-heading-title.elementor-size-medium{font-size:19px}.elementor-widget-heading .elementor-heading-title.elementor-size-large{font-size:29px}.elementor-widget-heading .elementor-heading-title.elementor-size-xl{font-size:39px}.elementor-widget-heading .elementor-heading-title.elementor-size-xxl{font-size:59px}<\/style><h2 class=\"elementor-heading-title elementor-size-default\">Imbalanced Learn est une biblioth\u00e8que Python open source permettant de rem\u00e9dier au probl\u00e8me de d\u00e9s\u00e9quilibre des classes dans les ensembles de donn\u00e9es de Machine Learning. D\u00e9couvrez pourquoi elle s\u2019av\u00e8re particuli\u00e8rement utile, et comment apprendre \u00e0 la ma\u00eetriser !<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3224b95 elementor-widget elementor-widget-text-editor\" data-id=\"3224b95\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-text-editor.elementor-drop-cap-view-stacked .elementor-drop-cap{background-color:#69727d;color:#fff}.elementor-widget-text-editor.elementor-drop-cap-view-framed .elementor-drop-cap{color:#69727d;border:3px solid;background-color:transparent}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap{margin-top:8px}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap-letter{width:1em;height:1em}.elementor-widget-text-editor .elementor-drop-cap{float:left;text-align:center;line-height:1;font-size:50px}.elementor-widget-text-editor .elementor-drop-cap-letter{display:inline-block}<\/style>\t\t\t\t<p><span style=\"font-weight: 400;\">\u00c0 moins d\u2019\u00eatre dot\u00e9 d\u2019un sixi\u00e8me sens ou d\u2019avoir des pr\u00e9monitions, il est essentiel de disposer d\u2019informations fiables pour tenter d\u2019<strong>anticiper le futur<\/strong>. Pour la m\u00eame raison, dans le domaine du <a href=\"https:\/\/datascientest.com\/le-machine-learning-au-service-de-la-cybersecurite-pour-la-detection-de-spams-dans-les-e-mails\">Machine Learning<\/a>, la qualit\u00e9 des donn\u00e9es d\u2019entra\u00eenement est un imp\u00e9ratif.<\/span><\/p><p><span style=\"font-weight: 400;\">Pour cause, elle impacte fortement la <strong>qualit\u00e9 des pr\u00e9dictions d\u2019un mod\u00e8le<\/strong>. Si les donn\u00e9es sont erron\u00e9es d\u00e8s le d\u00e9part, les r\u00e9sultats des calculs le seront aussi.<\/span><\/p><p><span style=\"font-weight: 400;\">Cependant, de nombreux ensembles de donn\u00e9es r\u00e9els ont un grave d\u00e9faut : elles pr\u00e9sentent des <strong>classes d\u00e9s\u00e9quilibr\u00e9es<\/strong>. Cela signifie tout simplement qu\u2019une classe est significativement plus repr\u00e9sent\u00e9e que les autres.<\/span><\/p><p><span style=\"font-weight: 400;\">Or, ce d\u00e9s\u00e9quilibre peut entra\u00eener des biais dans la <strong>performance des mod\u00e8les<\/strong>. Les classes minoritaires seront souvent n\u00e9glig\u00e9es, tandis que les classes majoritaires seront privil\u00e9gi\u00e9es.<\/span><\/p><p><span style=\"font-weight: 400;\">Pour pallier ce probl\u00e8me, diverses techniques et outils ont \u00e9t\u00e9 d\u00e9velopp\u00e9s. Parmi elles, une <strong>biblioth\u00e8que Python enti\u00e8rement d\u00e9di\u00e9e \u00e0 la gestion efficace de ce probl\u00e8me de d\u00e9s\u00e9quilibre de classe<\/strong> : Imbalanced Learn.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\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-b0a9cd6 e-flex e-con-boxed e-con\" data-id=\"b0a9cd6\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-c0c7415 elementor-widget elementor-widget-heading\" data-id=\"c0c7415\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Qu\u2019est-ce que le d\u00e9s\u00e9quilibre de classe ? Pourquoi est-ce un probl\u00e8me ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-48319d9 elementor-widget elementor-widget-text-editor\" data-id=\"48319d9\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">On parle de <strong>d\u00e9s\u00e9quilibre de classe<\/strong> lorsqu\u2019il y a une distribution in\u00e9gale des classes dans un ensemble de donn\u00e9es, avec une classe dominante et une ou plusieurs classes minoritaires.<\/span><\/p><p><span style=\"font-weight: 400;\"><strong>Les causes de ce d\u00e9s\u00e9quilibre peuvent \u00eatre multiples<\/strong>, allant de la collecte de donn\u00e9es biais\u00e9es \u00e0 la nature intrins\u00e8que du ph\u00e9nom\u00e8ne \u00e9tudi\u00e9. Par exemple, dans le domaine m\u00e9dical, certaines maladies peuvent \u00eatre beaucoup plus rares que d\u2019autres.<\/span><\/p><p><span style=\"font-weight: 400;\">C\u2019est ce qui peut conduire \u00e0 un d\u00e9s\u00e9quilibre dans les <a href=\"https:\/\/datascientest.com\/public-datasets-machine-learning\">datasets<\/a>. Alors, quelles sont les cons\u00e9quences potentielles ?<\/span><\/p><p><span style=\"font-weight: 400;\">L\u2019impact sur les <strong>performances des mod\u00e8les de Machine Learning<\/strong> peut \u00eatre significatif. En raison de la pr\u00e9dominance des classes majoritaires, les mod\u00e8les ont tendance \u00e0 d\u00e9velopper des biais et favoriser la pr\u00e9diction de ces classes au d\u00e9triment des minoritaires.<\/span><\/p><p><span style=\"font-weight: 400;\">Cela peut conduire \u00e0 des erreurs co\u00fbteuses dans des domaines tels que la d\u00e9tection de fraudes, le diagnostic m\u00e9dical, ou encore la d\u00e9tection d\u2019anomalies. Pour cause, la <strong>pr\u00e9diction pr\u00e9cise des classes minoritaires<\/strong> y est cruciale.<\/span><\/p><p><span style=\"font-weight: 400;\">Ainsi, g\u00e9rer efficacement ce d\u00e9s\u00e9quilibre est imp\u00e9ratif pour <strong>garantir des pr\u00e9dictions fiables et pr\u00e9cises dans diverses applications du Machine Learning<\/strong>. C\u2019est ce qui permet d\u2019am\u00e9liorer la sensibilit\u00e9, la sp\u00e9cificit\u00e9, et d\u2019autres m\u00e9triques d\u2019\u00e9valuation permettant aux mod\u00e8les de mieux g\u00e9n\u00e9raliser et de r\u00e9pondre aux besoins r\u00e9els des applications.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2a86180 elementor-widget elementor-widget-image\" data-id=\"2a86180\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-image{text-align:center}.elementor-widget-image a{display:inline-block}.elementor-widget-image a img[src$=\".svg\"]{width:48px}.elementor-widget-image img{vertical-align:middle;display:inline-block}<\/style>\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"900\" height=\"600\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-1.jpg\" class=\"attachment-full size-full wp-image-228270\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-1.jpg 900w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-1-300x200.jpg 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-1-768x512.jpg 768w\" sizes=\"(max-width: 900px) 100vw, 900px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-e49c93b elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"e49c93b\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formations-data\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Tout savoir sur Imbalanced Learn<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\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-e85a9ca e-flex e-con-boxed e-con\" data-id=\"e85a9ca\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-41fd607 elementor-widget elementor-widget-heading\" data-id=\"41fd607\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Imbalanced Learn : une biblioth\u00e8que Python pour surmonter cet obstacle<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ff6bad2 elementor-widget elementor-widget-text-editor\" data-id=\"ff6bad2\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Con\u00e7ue pour traiter les probl\u00e8mes li\u00e9s au d\u00e9s\u00e9quilibre de classe dans les ensembles de donn\u00e9es de Machine Learning, Imbalanced Learn est une <strong>biblioth\u00e8que Python<\/strong> <a href=\"https:\/\/datascientest.com\/open-source\">open-source<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Elle offre une gamme compl\u00e8te de <strong>techniques de r\u00e9\u00e9chantillonnage<\/strong>, de <strong>sous-\u00e9chantillonnage<\/strong>, de <strong>sur\u00e9chantillonnage<\/strong> et de <strong>combinaisons hybrides<\/strong> pour \u00e9quilibrer les classes et am\u00e9liorer les performances des mod\u00e8les.<\/span><\/p><p><span style=\"font-weight: 400;\">En guise d\u2019exemples, on peut citer <strong>SMOTE <\/strong>(Synthetic Minority Over-Sampling Technique), <strong>Random Under-Sampling<\/strong>, et d\u2019autres techniques avanc\u00e9es pour traiter les classes minoritaires et majoritaires.<\/span><\/p><p><span style=\"font-weight: 400;\">Gr\u00e2ce \u00e0 son <strong>int\u00e9gration fluide avec d\u2019autres biblioth\u00e8ques populaires<\/strong> comme <a href=\"https:\/\/datascientest.com\/tout-savoir-sur-scikit-learn\">Scikit-learn<\/a>, elle facilite la r\u00e9solution du probl\u00e8me. Il s\u2019agit d\u2019ailleurs d\u2019une extension de Scikit-learn.<\/span><\/p><p><span style=\"font-weight: 400;\">Elle s\u2019int\u00e8gre donc parfaitement avec les pipelines, les estimateurs et les outils de pr\u00e9traitement, offrant une solution holistique pour la <strong>gestion du d\u00e9s\u00e9quilibre de classe<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Cette biblioth\u00e8que s\u2019accompagne d\u2019une documentation exhaustive, d\u2019exemples de code et de tutoriels pour aider les utilisateurs \u00e0 comprendre et \u00e0 utiliser efficacement ses fonctionnalit\u00e9s.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\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-f52d4ba e-flex e-con-boxed e-con\" data-id=\"f52d4ba\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-d40e070 elementor-widget elementor-widget-heading\" data-id=\"d40e070\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">\u00c0 quoi \u00e7a sert ? Quels sont les cas d\u2019usage ?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c0a6855 elementor-widget elementor-widget-text-editor\" data-id=\"c0a6855\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">L\u2019un des exemples embl\u00e9matiques d\u2019application d\u2019Imbalanced Learn est <strong>la d\u00e9tection de fraudes financi\u00e8res<\/strong>. En utilisant les techniques de r\u00e9\u00e9chantillonnage, les institutions financi\u00e8res sont en mesure d\u2019am\u00e9liorer la capacit\u00e9 de leurs mod\u00e8les \u00e0 identifier les transactions frauduleuses tout en r\u00e9duisant les faux positifs. Ceci garantit une s\u00e9curit\u00e9 accrue.<\/span><\/p><p><span style=\"font-weight: 400;\">Dans le domaine m\u00e9dical, <strong>l\u2019imagerie diagnostique et la d\u00e9tection pr\u00e9coce de maladies<\/strong> peuvent b\u00e9n\u00e9ficier grandement des fonctionnalit\u00e9s de cette biblioth\u00e8que Python.<\/span><\/p><p><span style=\"font-weight: 400;\">En \u00e9quilibrant les classes et en optimisant les mod\u00e8les ML, les cliniciens peuvent obtenir des r\u00e9sultats plus fiables. Ceci permet d\u2019<strong>acc\u00e9l\u00e9rer les diagnostics<\/strong>, d\u2019<strong>am\u00e9liorer les taux de d\u00e9tection et d\u2019optimiser les soins aux patients<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">De m\u00eame, dans le secteur industriel, Imbalanced Learn peut \u00eatre appliqu\u00e9 \u00e0 <strong>la pr\u00e9vision de d\u00e9faillances et \u00e0 la maintenance pr\u00e9dictive<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Ceci offre des avantages majeurs, car<strong> l\u2019identification des risques et des anomalies dans les syst\u00e8mes et \u00e9quipements<\/strong> aide \u00e0 minimiser les temps d\u2019arr\u00eat. Les co\u00fbts op\u00e9rationnels s\u2019en trouvent r\u00e9duits, tandis que la fiabilit\u00e9 et la durabilit\u00e9 des infrastructures et des actifs augmentent.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-923f271 elementor-widget elementor-widget-image\" data-id=\"923f271\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"740\" height=\"600\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn2.jpg\" class=\"attachment-full size-full wp-image-228271\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn2.jpg 740w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn2-300x243.jpg 300w\" sizes=\"(max-width: 740px) 100vw, 740px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-da8b571 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"da8b571\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formations-data\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Apprendre \u00e0 utiliser Imbalanced Learn<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\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-9c7dcd1 e-flex e-con-boxed e-con\" data-id=\"9c7dcd1\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-946ff26 elementor-widget elementor-widget-heading\" data-id=\"946ff26\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Zoom sur les techniques de r\u00e9\u00e9chantillonnage<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-f5c3a66 elementor-widget elementor-widget-text-editor\" data-id=\"f5c3a66\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Penchons-nous maintenant plus en d\u00e9tail sur les diff\u00e9rentes techniques de r\u00e9\u00e9chantillonnage propos\u00e9es par Imbalanced Learn, et la fa\u00e7on dont elles permettent le <strong>r\u00e9\u00e9quilibrage des datasets<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Le sur\u00e9chantillonnage implique la g\u00e9n\u00e9ration de donn\u00e9es synth\u00e9tiques pour les classes minoritaires, afin d\u2019augmenter leur repr\u00e9sentation dans l\u2019ensemble de donn\u00e9es.<\/span><\/p><p><span style=\"font-weight: 400;\">Des techniques comme SMOTE cr\u00e9ent de nouveaux \u00e9chantillons en interpolant entre les instances existantes de la classe minoritaire, am\u00e9liorant ainsi la <strong>capacit\u00e9 du mod\u00e8le \u00e0 g\u00e9n\u00e9raliser et \u00e0 pr\u00e9dire avec pr\u00e9cision les classes minoritaires<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Au contraire, le sous-\u00e9chantillonnage vise \u00e0 r\u00e9duire le nombre d\u2019\u00e9chantillons de la classe majoritaire. Cette approche peut \u00eatre utile dans les cas o\u00f9 le sur\u00e9chantillonnage n\u2019est pas souhaitable ou possible en raison de contraintes de calcul ou de risques de surajustement.<\/span><\/p><p><span style=\"font-weight: 400;\">Les techniques comme Random Under-Sampling suppriment al\u00e9atoirement des \u00e9chantillons de la classe majoritaire pour <strong>\u00e9quilibrer la distribution des classes<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Afin de maximiser les avantages des deux approches, des techniques hybrides peuvent \u00eatre utilis\u00e9es. Elles combinent \u00e0 la fois le sur\u00e9chantillonnage et le sous-\u00e9chantillonnage pour atteindre un <strong>\u00e9quilibre optimal entre les classes<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Ces m\u00e9thodes hybrides offrent une flexibilit\u00e9 et permettent de mieux adapter les strat\u00e9gies de r\u00e9\u00e9chantillonnage aux caract\u00e9ristiques sp\u00e9cifiques des datasets et aux exigences des applications.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\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-86609e7 e-flex e-con-boxed e-con\" data-id=\"86609e7\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-b338175 elementor-widget elementor-widget-heading\" data-id=\"b338175\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Les m\u00e9thodes d\u2019\u00e9valuation et de comparaison<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5b86a46 elementor-widget elementor-widget-text-editor\" data-id=\"5b86a46\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Le r\u00e9\u00e9quilibrage est une premi\u00e8re \u00e9tape essentielle, mais comment \u00e9valuer sa r\u00e9ussite ? Plusieurs m\u00e9triques d\u2019\u00e9valuation existent, mais il convient de choisir les plus pertinentes qui tiennent compte de la distribution in\u00e9gale des classes.<\/span><\/p><p><span style=\"font-weight: 400;\">La <strong>Pr\u00e9cision<\/strong> (<strong>Precision<\/strong>) mesure l\u2019exactitude des pr\u00e9dictions positives, tandis que le <strong>Rappel<\/strong> (<a href=\"https:\/\/datascientest.com\/recall-tout-savoir\">Recall<\/a>) mesure la capacit\u00e9 du mod\u00e8le \u00e0 identifier correctement les instances positives.<\/span><\/p><p><span style=\"font-weight: 400;\">De son c\u00f4t\u00e9, le <strong>F-Score<\/strong> est la moyenne harmonique entre la pr\u00e9cision et le rappel. La <strong>courbe ROC et AUC<\/strong>, quant \u00e0 elle, \u00e9value la performance du mod\u00e8le en fonction de diff\u00e9rents seuils de classification.<\/span><\/p><p><span style=\"font-weight: 400;\">De m\u00eame, pour \u00e9valuer l\u2019efficacit\u00e9 des techniques de r\u00e9\u00e9chantillonnage impl\u00e9ment\u00e9es avec Imbalanced Learn, suivre des m\u00e9thodologies rigoureuses est indispensable.<\/span><\/p><p><span style=\"font-weight: 400;\">Cela peut inclure la <strong>division stratifi\u00e9e des ensembles de donn\u00e9es<\/strong>, la <a href=\"https:\/\/datascientest.com\/cross-validation\">validation crois\u00e9e<\/a>, et la <strong>comparaison des m\u00e9triques d\u2019\u00e9valuation<\/strong> sur plusieurs exp\u00e9rimentations. Le but est de d\u00e9terminer la meilleure strat\u00e9gie.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-217f55e elementor-widget elementor-widget-image\" data-id=\"217f55e\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"1080\" height=\"600\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-3.jpg\" class=\"attachment-full size-full wp-image-228272\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-3.jpg 1080w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-3-300x167.jpg 300w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-3-1024x569.jpg 1024w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-3-768x427.jpg 768w\" sizes=\"(max-width: 1080px) 100vw, 1080px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-8936194 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"8936194\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formations-data\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">Ma\u00eetriser Imbalanced Learn<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\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-90f4bd0 e-flex e-con-boxed e-con\" data-id=\"90f4bd0\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-3fd2b15 elementor-widget elementor-widget-heading\" data-id=\"3fd2b15\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Imbalanced Learn et l\u2019int\u00e9gration avec d\u2019autres outils et biblioth\u00e8ques<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5268ae4 elementor-widget elementor-widget-text-editor\" data-id=\"5268ae4\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">Comme nous l\u2019\u00e9voquions plus t\u00f4t, <strong>l\u2019un des grands avantages d\u2019Imbalanced Learn est son int\u00e9gration native avec scikit-learn<\/strong> : une biblioth\u00e8que Python couramment utilis\u00e9e pour l\u2019<a href=\"https:\/\/datascientest.com\/apprentissage-automatique\">apprentissage automatique<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Cette int\u00e9gration permet aux utilisateurs d\u2019incorporer tr\u00e8s facilement les fonctionnalit\u00e9s d\u2019Imbalanced Learn dans leurs pipelines d\u2019apprentissage, en combinant des techniques de r\u00e9\u00e9chantillonnage avec des estimateurs de scikit-learn pour construire des <strong>mod\u00e8les robustes et \u00e9quilibr\u00e9s<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Par ailleurs, elle peut \u00e9galement \u00eatre int\u00e9gr\u00e9e avec d\u2019autres <a href=\"https:\/\/datascientest.com\/quest-ce-quun-framework\">frameworks<\/a> et outils de Machine Learning comme <a href=\"https:\/\/datascientest.com\/tensorflow\">TensorFlow<\/a>, <a href=\"https:\/\/datascientest.com\/pytorch-tout-savoir\">PyTorch<\/a>, et d\u2019autres biblioth\u00e8ques populaires.<\/span><\/p><p><span style=\"font-weight: 400;\">Cette compatibilit\u00e9 \u00e9tendue permet aux utilisateurs d\u2019<strong>exploiter les fonctionnalit\u00e9s avanc\u00e9es dans divers environnements et architectures<\/strong>, offrant une flexibilit\u00e9 et une adaptabilit\u00e9 largement accrues.<\/span><\/p><p><span style=\"font-weight: 400;\">Les chercheurs et ing\u00e9nieurs en Machine Learning ont la possibilit\u00e9 d\u2019appliquer des techniques de r\u00e9\u00e9chantillonnage avanc\u00e9es dans des domaines tels que la <a href=\"https:\/\/datascientest.com\/meta-presente-sam-la-revolution-de-la-computer-vision\">vision par ordinateur<\/a>, le <a href=\"https:\/\/datascientest.com\/chapsvision-acquiert-qwan-et-devient-le-leader-europeen-du-traitement-du-langage-naturel\">traitement du langage naturel<\/a>, et d\u2019autres applications n\u00e9cessitant des architectures de <a href=\"https:\/\/datascientest.com\/deep-neural-network\">r\u00e9seau neuronal profond<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Avec l\u2019\u00e9volution vers des architectures distribu\u00e9es et des <strong>environnements edge computing<\/strong>, l\u2019int\u00e9gration d\u2019Imbalanced Learn dans des solutions cloud et edge est aussi devenue primordiale.<\/span><\/p><p><span style=\"font-weight: 400;\">Les biblioth\u00e8ques compatibles et les <strong>outils d\u2019orchestration<\/strong> <a href=\"https:\/\/datascientest.com\/kubernetes-data-science\">Kubernetes<\/a> peuvent faciliter le d\u00e9ploiement et la gestion de mod\u00e8les \u00e9quilibr\u00e9s, permettant ainsi une mise \u00e0 l\u2019\u00e9chelle efficace et une ex\u00e9cution en temps r\u00e9el dans des environnements diversifi\u00e9s et dynamiques.<\/span><\/p>\t\t\t\t\t\t<\/div>\n\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-89d5f9b e-flex e-con-boxed e-con\" data-id=\"89d5f9b\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t\t\t<div class=\"elementor-element elementor-element-5ae835e elementor-widget elementor-widget-heading\" data-id=\"5ae835e\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Conclusion : Imbalanced Learn, un pr\u00e9cieux atout contre le d\u00e9s\u00e9quilibre des classes<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1eecf16 elementor-widget elementor-widget-text-editor\" data-id=\"1eecf16\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t<p><span style=\"font-weight: 400;\">En permettant d\u2019am\u00e9liorer les performances des mod\u00e8les de Machine Learning sur des ensembles de donn\u00e9es d\u00e9s\u00e9quilibr\u00e9s, <strong>Imbalanced Learn repr\u00e9sente une ressource pr\u00e9cieuse et inestimable<\/strong>.<\/span><\/p><p><span style=\"font-weight: 400;\">Elle permet aux chercheurs, aux professionnels et aux organisations cherchant \u00e0 maximiser leur pr\u00e9cision et leur efficacit\u00e9 dans des contextes complexes et exigeants.<\/span><\/p><p><span style=\"font-weight: 400;\">Afin de devenir expert en apprentissage automatique, vous pouvez <strong>choisir DataScientest<\/strong>. Nos diff\u00e9rents cursus <a href=\"https:\/\/datascientest.com\/formation-data-scientist-belgique\">Data Scientist<\/a>, <a href=\"https:\/\/datascientest.com\/formation-data-analyst-pole-emploi\">Data Analyst<\/a>, ou encore <a href=\"https:\/\/datascientest.com\/article-formation-machine-learning-engineer\">ML Engineer<\/a> comportent tous un module enti\u00e8rement d\u00e9di\u00e9 au Machine Learning !<\/span><\/p><p><span style=\"font-weight: 400;\">Vous d\u00e9couvrirez notamment les <strong>techniques de classification<\/strong>, <strong>r\u00e9gression<\/strong> et <strong>clustering<\/strong>, les <a href=\"https:\/\/datascientest.com\/series-temporelles\">s\u00e9ries temporelles<\/a> et la r\u00e9duction de dimension, ou encore les applications telles que le <a href=\"https:\/\/datascientest.com\/text-mining-definition\">Text Mining<\/a> et le <a href=\"https:\/\/datascientest.com\/web-scraping-tout-savoir\">Web Scraping<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Ces formations vous conf\u00e8rent toutes les comp\u00e9tences requises pour travailler dans le domaine de la <a href=\"https:\/\/datascientest.com\/data-science-definition\">Data Science<\/a> et du <a href=\"https:\/\/datascientest.com\/machine-learning-tout-savoir\">Machine Learning<\/a>, telles que la ma\u00eetrise du <a href=\"https:\/\/datascientest.com\/python-tout-savoir\">langage Python<\/a>, des <a href=\"https:\/\/datascientest.com\/base-de-donnees-definition\">bases de donn\u00e9es<\/a> et de la <a href=\"https:\/\/datascientest.com\/business-intelligence\">Business Intelligence<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Nos formations permettent non seulement d\u2019<strong>obtenir un dipl\u00f4me reconnu par l\u2019\u00c9ta<\/strong>t, mais aussi un certificat des Mines ParisTech PSL Executive Education et une certification cloud <a href=\"https:\/\/datascientest.com\/amazon-web-services-tout-savoir\">AWS<\/a> ou <a href=\"https:\/\/datascientest.com\/tout-savoir-sur-microsoft-azure\">Microsoft Azure<\/a>.<\/span><\/p><p><span style=\"font-weight: 400;\">Tous nos programmes se compl\u00e8tent int\u00e9gralement \u00e0 distance en <a href=\"https:\/\/datascientest.com\/definition-bootcamp\">BootCamp<\/a>, <strong>formation continue<\/strong> ou <strong>alternance<\/strong>, et notre organisme est \u00e9ligible au <a href=\"https:\/\/datascientest.com\/cpf-comment-utiliser-son-compte-personnel-de-formation\">CPF<\/a> pour le financement. <strong>D\u00e9couvrez vite DataScientest !<\/strong><\/span><\/p>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-31ee509 elementor-widget elementor-widget-image\" data-id=\"31ee509\" data-element_type=\"widget\" data-widget_type=\"image.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"736\" height=\"600\" src=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-4.jpg\" class=\"attachment-full size-full wp-image-228276\" alt=\"\" loading=\"lazy\" srcset=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-4.jpg 736w, https:\/\/datascientest.com\/wp-content\/uploads\/2024\/01\/imbalanced-learn-4-300x245.jpg 300w\" sizes=\"(max-width: 736px) 100vw, 736px\" \/>\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-692cc39 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"692cc39\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/formations-data\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t<span class=\"elementor-button-text\">S'inscrire \u00e0 une formation sur DataScientest<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6bb14b0 elementor-widget elementor-widget-html\" data-id=\"6bb14b0\" data-element_type=\"widget\" data-widget_type=\"html.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<script type=\"application\/ld+json\">\r\n{\r\n  \"@context\": \"https:\/\/schema.org\",\r\n  \"@type\": \"FAQPage\",\r\n  \"mainEntity\": [{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Qu\u2019est-ce que le d\u00e9s\u00e9quilibre de classe ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"On parle de d\u00e9s\u00e9quilibre de classe lorsqu\u2019il y a une distribution in\u00e9gale des classes dans un ensemble de donn\u00e9es, avec une classe dominante et une ou plusieurs classes minoritaires.\"\r\n    }\r\n  },{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Qu'est-ce que Imbalanced Learn ?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Con\u00e7ue pour traiter les probl\u00e8mes li\u00e9s au d\u00e9s\u00e9quilibre de classe dans les ensembles de donn\u00e9es de Machine Learning, Imbalanced Learn est une biblioth\u00e8que Python open-source. Elle offre une gamme compl\u00e8te de techniques de r\u00e9\u00e9chantillonnage, de sous-\u00e9chantillonnage, de sur\u00e9chantillonnage et de combinaisons hybrides pour \u00e9quilibrer les classes et am\u00e9liorer les performances des mod\u00e8les.\"\r\n    }\r\n  },{\r\n    \"@type\": \"Question\",\r\n    \"name\": \"Imbalanced Learn et l\u2019int\u00e9gration avec d\u2019autres outils et biblioth\u00e8ques\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"Comme nous l\u2019\u00e9voquions plus t\u00f4t, l\u2019un des grands avantages d\u2019Imbalanced Learn est son int\u00e9gration native avec scikit-learn : une biblioth\u00e8que Python couramment utilis\u00e9e pour l\u2019apprentissage automatique. Elle peut \u00e9galement \u00eatre int\u00e9gr\u00e9e avec d\u2019autres frameworks et outils de Machine Learning comme TensorFlow, PyTorch, et d\u2019autres biblioth\u00e8ques populaires.\"\r\n    }\r\n  }]\r\n}\r\n<\/script>\r\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Imbalanced Learn est une biblioth\u00e8que Python open source permettant de rem\u00e9dier au probl\u00e8me de d\u00e9s\u00e9quilibre des classes dans les ensembles de donn\u00e9es de Machine Learning. D\u00e9couvrez pourquoi elle s\u2019av\u00e8re particuli\u00e8rement utile, et comment apprendre \u00e0 la ma\u00eetriser ! \u00c0 moins d\u2019\u00eatre dot\u00e9 d\u2019un sixi\u00e8me sens ou d\u2019avoir des pr\u00e9monitions, il est essentiel de disposer d\u2019informations [&hellip;]<\/p>\n","protected":false},"author":78,"featured_media":232793,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2497],"tags":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v21.7 (Yoast SEO v21.7) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir\" \/>\n<meta property=\"og:site_name\" content=\"Formation Data Science | DataScientest.com\" \/>\n<meta property=\"article:published_time\" content=\"2024-02-27T06:07:00+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-02-20T10:07:07+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/02\/imbalanced_learn_python.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"600\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"J\u00e9r\u00e9my Robert\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"J\u00e9r\u00e9my Robert\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"9 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir\",\"url\":\"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir\",\"name\":\"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/#website\"},\"datePublished\":\"2024-02-27T06:07:00+00:00\",\"dateModified\":\"2024-02-20T10:07:07+00:00\",\"author\":{\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\"},\"inLanguage\":\"fr-FR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/#website\",\"url\":\"https:\/\/datascientest.com\/\",\"name\":\"Formation Data Science | DataScientest.com\",\"description\":\"Leader de la formation en Data Science\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"fr-FR\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925\",\"name\":\"J\u00e9r\u00e9my Robert\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"fr-FR\",\"@id\":\"https:\/\/datascientest.com\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"contentUrl\":\"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg\",\"caption\":\"J\u00e9r\u00e9my Robert\"},\"description\":\"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.\",\"sameAs\":[\"https:\/\/www.linkedin.com\/in\/jeremyrobert1\/\",\"Data Science\"],\"url\":\"https:\/\/datascientest.com\/author\/jeremyrobert\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir","og_locale":"fr_FR","og_type":"article","og_title":"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML","og_url":"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir","og_site_name":"Formation Data Science | DataScientest.com","article_published_time":"2024-02-27T06:07:00+00:00","article_modified_time":"2024-02-20T10:07:07+00:00","og_image":[{"width":1200,"height":600,"url":"https:\/\/datascientest.com\/wp-content\/uploads\/2024\/02\/imbalanced_learn_python.png","type":"image\/png"}],"author":"J\u00e9r\u00e9my Robert","twitter_card":"summary_large_image","twitter_misc":{"\u00c9crit par":"J\u00e9r\u00e9my Robert","Dur\u00e9e de lecture estim\u00e9e":"9 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir","url":"https:\/\/datascientest.com\/imbalanced-learn-tout-savoir","name":"Imbalanced Learn : la biblioth\u00e8que Python de r\u00e9\u00e9quilbrage des datasets ML","isPartOf":{"@id":"https:\/\/datascientest.com\/#website"},"datePublished":"2024-02-27T06:07:00+00:00","dateModified":"2024-02-20T10:07:07+00:00","author":{"@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/imbalanced-learn-tout-savoir"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/#website","url":"https:\/\/datascientest.com\/","name":"Formation Data Science | DataScientest.com","description":"Leader de la formation en Data Science","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"fr-FR"},{"@type":"Person","@id":"https:\/\/datascientest.com\/#\/schema\/person\/7af2361c48e24e79a8aa2360e6e74925","name":"J\u00e9r\u00e9my Robert","image":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/datascientest.com\/#\/schema\/person\/image\/","url":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","contentUrl":"https:\/\/datascientest.com\/wp-content\/uploads\/2023\/11\/jeremyrobert.jpeg","caption":"J\u00e9r\u00e9my Robert"},"description":"J\u00e9r\u00e9my Robert est actuellement Directeur p\u00e9dagogique B2B chez DataScientest.com. Avant de rejoindre DataScientest, il a travaill\u00e9 en tant que Consultant Business Intelligence chez DEAPDATA, o\u00f9 il a g\u00e9r\u00e9 des projets dans les domaines de l'assurance, bancaire et industrie. Il a \u00e9galement \u00e9t\u00e9 Co-fondateur et pr\u00e9sident de Tripzer. En tant que consultant ind\u00e9pendant, J\u00e9r\u00e9my a propos\u00e9 des formations sur MicroStrategy pour divers niveaux professionnels. Il est dipl\u00f4m\u00e9 de l'Universit\u00e9 Paris Cit\u00e9 avec un Master MIAGE, combinant comp\u00e9tences en informatique et gestion.","sameAs":["https:\/\/www.linkedin.com\/in\/jeremyrobert1\/","Data Science"],"url":"https:\/\/datascientest.com\/author\/jeremyrobert"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/228216"}],"collection":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/users\/78"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/comments?post=228216"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/posts\/228216\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media\/232793"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/media?parent=228216"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/categories?post=228216"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/wp-json\/wp\/v2\/tags?post=228216"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}