{"id":15805,"date":"2022-04-23T20:38:24","date_gmt":"2022-04-23T19:38:24","guid":{"rendered":"https:\/\/complex-systems-ai.com\/?page_id=15805"},"modified":"2022-04-23T20:54:04","modified_gmt":"2022-04-23T19:54:04","slug":"tutoriel-sur-le-t-sne","status":"publish","type":"page","link":"https:\/\/complex-systems-ai.com\/en\/data-analysis\/tutorial-on-the-t-sne\/","title":{"rendered":"t-SNE Tutorial"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-page\" data-elementor-id=\"15805\" class=\"elementor elementor-15805\">\n\t\t\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-9505f93 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"9505f93\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-top-column elementor-element elementor-element-23165db\" data-id=\"23165db\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-829b0e4 elementor-align-justify elementor-widget elementor-widget-button\" data-id=\"829b0e4\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/complex-systems-ai.com\/analyse-des-donnees\/\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">Analyse des donn\u00e9es<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-top-column elementor-element elementor-element-7c9d279\" data-id=\"7c9d279\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-15d508e elementor-align-justify elementor-widget elementor-widget-button\" data-id=\"15d508e\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/complex-systems-ai.com\/\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">Page d'accueil<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t<div class=\"elementor-column elementor-col-33 elementor-top-column elementor-element elementor-element-ee0c9f6\" data-id=\"ee0c9f6\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-02a8a59 elementor-align-justify elementor-widget elementor-widget-button\" data-id=\"02a8a59\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm\" href=\"https:\/\/en.wikipedia.org\/wiki\/Data_analysis\" target=\"_blank\" rel=\"noopener\">\n\t\t\t\t\t\t<span class=\"elementor-button-content-wrapper\">\n\t\t\t\t\t\t\t\t\t<span class=\"elementor-button-text\">Wiki<\/span>\n\t\t\t\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-9d13a7a elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"9d13a7a\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-3fa187d\" data-id=\"3fa187d\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-013de74 elementor-widget elementor-widget-text-editor\" data-id=\"013de74\" data-element_type=\"widget\" data-e-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\t\t<p>Apr\u00e8s avoir r\u00e9alis\u00e9 une <a href=\"https:\/\/complex-systems-ai.com\/en\/descriptive-analysis\/\">analyse descriptive<\/a> des donn\u00e9es, rempli les vides et s\u00e9lectionner les premi\u00e8res colonnes. Il est important de continuer de r\u00e9duire les dimensions, pour cela, ce tutoriel sur le t-SNE pr\u00e9sente la r\u00e9duction de dimensions par analyse non-lin\u00e9aire.<\/p><p><img decoding=\"async\" class=\"aligncenter wp-image-11096 size-full\" src=\"https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2020\/09\/cropped-Capture.png\" alt=\"tutoriel sur le t-SNE\" width=\"97\" height=\"97\" title=\"\"><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-36b2b02 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"36b2b02\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-0531bd7\" data-id=\"0531bd7\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-1e81f05 elementor-widget elementor-widget-heading\" data-id=\"1e81f05\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_82_2 counter-hierarchy ez-toc-counter ez-toc-grey ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Contenus<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/complex-systems-ai.com\/en\/data-analysis\/tutorial-on-the-t-sne\/#Tutoriel-sur-le-t-SNE-et-reduction-de-dimensions-non-lineaire\" >Tutoriel sur le t-SNE et r\u00e9duction de dimensions non-lin\u00e9aire<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/complex-systems-ai.com\/en\/data-analysis\/tutorial-on-the-t-sne\/#Representation-du-t-SNE\" >Repr\u00e9sentation du t-SNE<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/complex-systems-ai.com\/en\/data-analysis\/tutorial-on-the-t-sne\/#Multi-t-SNE\" >Multi t-SNE<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/complex-systems-ai.com\/en\/data-analysis\/tutorial-on-the-t-sne\/#Desavantages-du-t-SNE\" >D\u00e9savantages du t-SNE<\/a><\/li><\/ul><\/nav><\/div>\n<h2 class=\"elementor-heading-title elementor-size-default\"><span class=\"ez-toc-section\" id=\"Tutoriel-sur-le-t-SNE-et-reduction-de-dimensions-non-lineaire\"><\/span>Tutoriel sur le t-SNE et r\u00e9duction de dimensions non-lin\u00e9aire<span class=\"ez-toc-section-end\"><\/span><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-25670c2 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"25670c2\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-3bbe9e6\" data-id=\"3bbe9e6\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-86da373 elementor-widget elementor-widget-text-editor\" data-id=\"86da373\" data-element_type=\"widget\" data-e-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\t\t<p>La plupart des ensembles de donn\u00e9es du monde r\u00e9el ont de nombreuses fonctionnalit\u00e9s, parfois plusieurs milliers. Chacun d&rsquo;eux peut \u00eatre consid\u00e9r\u00e9 comme une dimension dans l&rsquo;espace des points de donn\u00e9es. Par cons\u00e9quent, le plus souvent, nous traitons des ensembles de donn\u00e9es de grande dimension, o\u00f9 la visualisation compl\u00e8te est assez difficile.<\/p><p>Pour examiner un ensemble de donn\u00e9es dans son ensemble, nous devons r\u00e9duire le nombre de dimensions utilis\u00e9es dans la visualisation sans perdre beaucoup d&rsquo;informations sur les donn\u00e9es. Cette t\u00e2che est appel\u00e9e r\u00e9duction de dimensionnalit\u00e9 et est un exemple de probl\u00e8me d&rsquo;apprentissage non supervis\u00e9 car nous devons d\u00e9river de nouvelles caract\u00e9ristiques de faible dimension \u00e0 partir des donn\u00e9es elles-m\u00eames, sans aucune entr\u00e9e supervis\u00e9e.<\/p><p>L&rsquo;une des m\u00e9thodes bien connues de r\u00e9duction de la dimensionnalit\u00e9 est l&rsquo;analyse en composantes principales (ACP), que nous \u00e9tudierons plus loin dans ce cours. Sa limitation est qu&rsquo;il s&rsquo;agit d&rsquo;un <a href=\"https:\/\/complex-systems-ai.com\/en\/algorithmic\/\">algorithme<\/a> lin\u00e9aire qui implique certaines restrictions sur les donn\u00e9es.<\/p><p>Il existe \u00e9galement de nombreuses m\u00e9thodes non lin\u00e9aires, appel\u00e9es collectivement Manifold Learning. L&rsquo;un des plus connus d&rsquo;entre eux est le t-SNE, d&rsquo;o\u00f9 ce tutoriel sur le t-SNE \ud83d\ude42<\/p><p>Le nom de la m\u00e9thode semble complexe et un peu intimidant\u00a0: t-distributed Stohastic Neighbor Embedding. Ses <a href=\"https:\/\/complex-systems-ai.com\/en\/logic-math-27\/\">math\u00e9matiques<\/a> sont \u00e9galement impressionnantes (nous ne nous y attarderons pas ici, mais, si vous vous sentez courageux, voici l&rsquo;article original de Laurens van der Maaten et Geoffrey Hinton de JMLR).<\/p><p>Son id\u00e9e de base est simple : trouver une projection pour un espace de caract\u00e9ristiques de grande dimension sur un plan (ou un hyperplan 3D, mais il est presque toujours 2D) de telle sorte que les points qui \u00e9taient \u00e9loign\u00e9s dans l&rsquo;espace initial \u00e0 n dimensions se termineront loin l&rsquo;un de l&rsquo;autre dans le plan. Ceux qui \u00e9taient proches \u00e0 l&rsquo;origine resteront proches les uns des autres.<\/p><p>Essentiellement, l&rsquo;incorporation de voisins est une recherche d&rsquo;une repr\u00e9sentation de donn\u00e9es nouvelle et moins dimensionnelle qui pr\u00e9serve le voisinage des exemples.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-f93b2db elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"f93b2db\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-b109edd\" data-id=\"b109edd\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-68642a4 elementor-widget elementor-widget-heading\" data-id=\"68642a4\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\"><span class=\"ez-toc-section\" id=\"Representation-du-t-SNE\"><\/span>Repr\u00e9sentation du t-SNE<span class=\"ez-toc-section-end\"><\/span><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-efbafd2 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"efbafd2\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-ea22f62\" data-id=\"ea22f62\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-062a340 elementor-widget elementor-widget-text-editor\" data-id=\"062a340\" data-element_type=\"widget\" data-e-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\t\t<p id=\"6498\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Maintenant, pratiquons un peu. Tout d&rsquo;abord, nous devons importer des classes suppl\u00e9mentaires\u00a0:<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"a0ae\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">from sklearn.manifold import TSNE<br \/>from sklearn.preprocessing import StandardScaler<\/span><\/pre><p id=\"dff0\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Nous laisserons de c\u00f4t\u00e9 les fonctionnalit\u00e9s\u00a0State\u00a0et\u00a0Churn\u00a0et convertirons les valeurs \u00ab\u00a0Oui\u00a0\u00bb\/\u00a0\u00bbNon\u00a0\u00bb des fonctionnalit\u00e9s binaires en valeurs num\u00e9riques \u00e0 l&rsquo;aide de\u00a0pandas.Series.map()\u00a0:<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"f4f7\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">X = df.<strong class=\"nb ix\">drop<\/strong>(['Churn', 'State'], axis=1)<br \/>X['International plan'] = X['International plan'].<br \/>                          <strong class=\"nb ix\">map<\/strong>({'Yes': 1, 'No': 0})<br \/>X['Voice mail plan'] = X['Voice mail plan'].<br \/>                       <strong class=\"nb ix\">map<\/strong>({'Yes': 1, 'No': 0})<\/span><\/pre><p id=\"dc36\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Nous devons \u00e9galement normaliser les donn\u00e9es. Pour cela, nous allons soustraire la moyenne de chaque variable et la diviser par son \u00e9cart-type. Tout cela peut \u00eatre fait avec\u00a0StandardScaler.<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"bc16\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">scaler = StandardScaler()<br \/>X_scaled = scaler.<strong class=\"nb ix\">fit_transform<\/strong>(X)<\/span><\/pre><p id=\"c2d9\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Construisons maintenant une repr\u00e9sentation t-SNE\u00a0:<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"c4c1\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">%%time tsne = TSNE(random_state=17)<br \/>tsne_repr = tsne.<strong class=\"nb ix\">fit_transform<\/strong>(X_scaled)<\/span><span id=\"e374\" class=\"fy nl lg iw nb b do xs xt xu xv xw nn l no\" data-selectable-paragraph=\"\"><em class=\"kz\">CPU times: user 1min 32s, sys: 7.86 s, total: 1min 39s<br \/>Wall time: 1min 39s<\/em><\/span><\/pre><p id=\"71c9\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Colorons cette repr\u00e9sentation t-SNE en fonction du churn (vert pour les clients fid\u00e8les, et rouge pour ceux qui sont partis).<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"b63c\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">plt.<strong class=\"nb ix\">scatter<\/strong>(tsne_repr[:, 0], tsne_repr[:, 1], <br \/>            c=df['Churn'].map({False: 'green', True: 'red'}));<\/span><\/pre><figure class=\"lb lc ld le gv jx gj gk paragraph-image\"><div class=\"nf ng dq nh cf ni\" tabindex=\"0\" role=\"button\"><div class=\"gj gk yz\"><img fetchpriority=\"high\" decoding=\"async\" class=\"aligncenter size-medium wp-image-15811\" src=\"https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_gNEzT6kjj2KQaKgza9yN_A-300x232.png\" alt=\"\" width=\"300\" height=\"232\" title=\"\" srcset=\"https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_gNEzT6kjj2KQaKgza9yN_A-300x232.png 300w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_gNEzT6kjj2KQaKgza9yN_A-16x12.png 16w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_gNEzT6kjj2KQaKgza9yN_A-600x464.png 600w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_gNEzT6kjj2KQaKgza9yN_A.png 700w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><\/div><\/div><\/figure><p id=\"0a9a\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Nous pouvons voir que les clients qui se sont d\u00e9tourn\u00e9s sont concentr\u00e9s dans quelques zones de l&rsquo;espace des caract\u00e9ristiques de dimension inf\u00e9rieure.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-0cce231 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"0cce231\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-a706938\" data-id=\"a706938\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-8a8db24 elementor-widget elementor-widget-heading\" data-id=\"8a8db24\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\"><span class=\"ez-toc-section\" id=\"Multi-t-SNE\"><\/span>Multi t-SNE<span class=\"ez-toc-section-end\"><\/span><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-f0363e9 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"f0363e9\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-3fedb9f\" data-id=\"3fedb9f\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-62abc35 elementor-widget elementor-widget-text-editor\" data-id=\"62abc35\" data-element_type=\"widget\" data-e-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\t\t<p id=\"5b4b\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Pour mieux comprendre l&rsquo;image, nous pouvons \u00e9galement la colorer avec les fonctionnalit\u00e9s binaires restantes : Forfait international et Messagerie vocale. Les points verts indiquent ici les objets qui sont positifs pour la caract\u00e9ristique binaire correspondante.<\/p><pre class=\"lb lc ld le gv nj bt nk\"><span id=\"f2f9\" class=\"fy nl lg iw nb b do nm nn l no\" data-selectable-paragraph=\"\">_, axes = plt.subplots(1, 2, sharey=True, figsize=(12, 5))<\/span><span id=\"e2fe\" class=\"fy nl lg iw nb b do xs xt xu xv xw nn l no\" data-selectable-paragraph=\"\">for i, name in enumerate(['International plan', 'Voice mail plan']):<br \/>    axes[i].<strong class=\"nb ix\">scatter<\/strong>(tsne_repr[:, 0], tsne_repr[:, 1],<br \/>                    c=df[name].<strong class=\"nb ix\">map<\/strong>({'Yes': 'green', 'No': 'red'}))<br \/>    axes[i].set_title(name)<\/span><\/pre><figure class=\"lb lc ld le gv jx gj gk paragraph-image\"><div class=\"nf ng dq nh cf ni\" tabindex=\"0\" role=\"button\"><div class=\"gj gk za\"><img decoding=\"async\" class=\"aligncenter wp-image-15812 size-full\" src=\"https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_FHLtGK9mM583wZiUrx1XFQ.png\" alt=\"\" width=\"700\" height=\"313\" title=\"\" srcset=\"https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_FHLtGK9mM583wZiUrx1XFQ.png 700w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_FHLtGK9mM583wZiUrx1XFQ-300x134.png 300w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_FHLtGK9mM583wZiUrx1XFQ-18x8.png 18w, https:\/\/complex-systems-ai.com\/wp-content\/uploads\/2022\/04\/1_FHLtGK9mM583wZiUrx1XFQ-600x268.png 600w\" sizes=\"(max-width: 700px) 100vw, 700px\" \/><\/div><\/div><\/figure><p id=\"708e\" class=\"pw-post-body-paragraph ka kb iw kc b kd ke kf kg kh ki kj kk kl km kn ko kp kq kr ks kt ku kv kw kx ip fy\" data-selectable-paragraph=\"\">Or force est de constater que, par exemple, de nombreux clients m\u00e9contents qui ont r\u00e9sili\u00e9 leur abonnement sont entass\u00e9s dans le cluster le plus au sud-ouest qui repr\u00e9sente les personnes ayant le forfait international mais pas de messagerie vocale.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-f4a651e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"f4a651e\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-e4b512b\" data-id=\"e4b512b\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-d57f398 elementor-widget elementor-widget-heading\" data-id=\"d57f398\" data-element_type=\"widget\" data-e-type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<h2 class=\"elementor-heading-title elementor-size-default\"><span class=\"ez-toc-section\" id=\"Desavantages-du-t-SNE\"><\/span>D\u00e9savantages du t-SNE<span class=\"ez-toc-section-end\"><\/span><\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-91840d5 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"91840d5\" data-element_type=\"section\" data-e-type=\"section\">\n\t\t\t\t\t\t<div class=\"elementor-container elementor-column-gap-default\">\n\t\t\t\t\t<div class=\"elementor-column elementor-col-100 elementor-top-column elementor-element elementor-element-6008163\" data-id=\"6008163\" data-element_type=\"column\" data-e-type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3d47c19 elementor-widget elementor-widget-text-editor\" data-id=\"3d47c19\" data-element_type=\"widget\" data-e-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\t\t<p>Enfin, notons quelques inconv\u00e9nients dans ce tutoriel sur le t-SNE :<\/p><p>Grande <a href=\"https:\/\/complex-systems-ai.com\/en\/algorithmic\/complexity-in-time\/\">complexit\u00e9<\/a> de calcul. Il est peu probable que l&rsquo;impl\u00e9mentation dans scikit-learn soit r\u00e9alisable dans une t\u00e2che r\u00e9elle. Si vous avez un grand nombre d&rsquo;\u00e9chantillons, vous devriez plut\u00f4t essayer Multicore-TSNE.<\/p><p>L&rsquo;intrigue peut beaucoup changer en fonction de la graine al\u00e9atoire, ce qui complique l&rsquo;interpr\u00e9tation. Voici un bon tutoriel sur t-SNE. En g\u00e9n\u00e9ral, vous ne devriez pas tirer de conclusions de grande envergure sur la base de tels graphiques, car cela peut \u00e9quivaloir \u00e0 de simples suppositions. Bien s\u00fbr, certaines d\u00e9couvertes dans les images t-SNE peuvent inspirer une id\u00e9e et \u00eatre confirm\u00e9es par des recherches plus approfondies sur toute la ligne, mais cela n&rsquo;arrive pas tr\u00e8s souvent.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Data Analysis Wiki Home Page After performing a descriptive analysis of the data, fill in the blanks and select the first columns. It is important \u2026 <\/p>","protected":false},"author":1,"featured_media":0,"parent":15503,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"class_list":["post-15805","page","type-page","status-publish","hentry"],"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/pages\/15805","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/comments?post=15805"}],"version-history":[{"count":3,"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/pages\/15805\/revisions"}],"predecessor-version":[{"id":15815,"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/pages\/15805\/revisions\/15815"}],"up":[{"embeddable":true,"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/pages\/15503"}],"wp:attachment":[{"href":"https:\/\/complex-systems-ai.com\/en\/wp-json\/wp\/v2\/media?parent=15805"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}