{"id":2209,"date":"2019-07-31T10:33:51","date_gmt":"2019-07-31T08:33:51","guid":{"rendered":"http:\/\/www.datavaloris.com\/?p=2209"},"modified":"2020-06-26T17:52:13","modified_gmt":"2020-06-26T15:52:13","slug":"neural-network-optimization","status":"publish","type":"post","link":"https:\/\/www.datavaloris.com\/en\/neural-network-optimization\/","title":{"rendered":"Neural Network Optimization"},"content":{"rendered":"<h5>What is a good neural network?<\/h5>\n<p class=\" translation-block\">There are <strong>two criteria<\/strong> that will be <strong>decisive<\/strong>, depending on the objective we set :<\/p>\n<ul>\n<li class=\" translation-block\">Either we try to make the neural network <strong> more efficient <\/strong>, and we will want to improve its <strong> precision rate <\/strong> to maximize the right answers. We will therefore try to obtain a neuron network with the highest accuracy rate.<\/li>\n<li class=\" translation-block\">Either the precision rate is satisfactory, but we want <strong> to reduce the size of the neural network <\/strong>, to be able to embed it in IoT for example. So we want to <strong> reduce the number of parameters <\/strong> (number of neurons per layer, and number of connections between them), to the maximum.<\/li>\n<\/ul>\n<p>Evidemment, il n&rsquo;est&nbsp; pas facile d&rsquo;atteindre un objectif sans que cela fasse bouger l&rsquo;autre crit\u00e8re. Si on r\u00e9duit le nombre de neurones, on va souvent voir le taux de pr\u00e9cision chuter. Et si on augmente la pr\u00e9cision, cela se fait souvent en ajoutant des param\u00e8tres&#8230;<\/p>\n<h5>Build the best neural network<\/h5>\n<p class=\" translation-block\">There is a platform called <a href=\"https:\/\/www.kaggle.com\/\" target=\"_self\"> Kaggle <\/a> that offers tools and competitions for datascientists. One of these contests is called <a href=\"https:\/\/en.wikipedia.org\/wiki\/MNIST_destination_base\" target=\"_self\"> MNIST <\/a> and the object is to build a network of neurons capable of recognizing handwritten numeric characters.<\/p>\n<figure id=\"attachment_2212\" aria-describedby=\"caption-attachment-2212\" style=\"width: 594px\" class=\"wp-caption alignnone\"><img loading=\"lazy\" decoding=\"async\" class=\"size-full wp-image-2212\" src=\"http:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/07\/MnistExamples.png\" alt=\"Exemple MNIST - Source Wikipedia\" width=\"594\" height=\"361\"><figcaption id=\"caption-attachment-2212\" class=\"wp-caption-text\">Example MNIST - Source Wikipedia<\/figcaption><\/figure>\n<p class=\" translation-block\">Hundreds of datascientist compete in these contests and when a champion emerges, we touch the <strong>state of the art in terms of neural network model.<\/strong>\u00a0That champion is the best neural network built by a human being to date. But could a \"non-human intelligence\" do better?<\/p>\n<h5><strong>Optimize a neural network with NNTO<\/strong><\/h5>\n<p><a href=\"https:\/\/www.nnto.ai\/\">NNTO <\/a>is a SAAS platform that optimizes neural networks through genetic mutation algorithms. No need to provide your data. Optimization is done on <a href=\"http:\/\/www.datavaloris.com\/en\/services\/nnto\/\">the topology<\/a> of the neural network. <a href=\"https:\/\/www.youtube.com\/watch?v=sBaaWiwM2Xc&amp;t=2s\">Mutants are generated<\/a>they are selected according to their results to reach the chosen objective.<\/p>\n<p>So we took a MNIST Kaggle champion in order to <a href=\"http:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/01\/benchmark.pdf\">benchmark<\/a>, et nous avons r\u00e9ussi \u00e0 l&rsquo;am\u00e9liorer gr\u00e2ce \u00e0 NNTO.&nbsp;Comme l&rsquo;enjeu est de montrer la pertinence du mod\u00e8le dans ce contexte, nous avons d\u00e9cid\u00e9 d&rsquo;utiliser les donn\u00e9es d&rsquo;origine du repo de KERAS <strong>without any modifications<\/strong>.We evaluated the champion's score with this data, which gave us a score base (a bit below the one announced by the champion's author). All our model optimizations exceeded the calculated score. In addition, our best champion exceeded the score of the author:<\/p>\n<p>[embedyt] https:\/\/www.youtube.com\/watch?v=7ynU7jvJzBM[\/embedyt]<\/p>\n<p>NNTO has managed to improve the neuron champion network in two ways with:<\/p>\n<ul>\n<li><strong>A more efficient neural network<\/strong> : 99.50% after 10 generations of mutants (implying an 8% decrease in error) compared to the original model.<\/li>\n<li><strong>A lighter neural network<\/strong> : a parameter reduction of 76% at 175K with another mutant.<\/li>\n<\/ul>\n<h5>Optimize your neural network to break a deadlock<\/h5>\n<p>When the results are not there, it is sometimes very difficult to know where the problem comes from. Does this come from the neural network model? Does the problem come from the data? Without the ability to clearly identify the cause, solving the problem is almost impossible, and despite the hours of work invested, we remain stuck in this rut.<\/p>\n<p>The optimization of the neural network with NNTO makes it possible to get out of this impasse. By mutating the model, one solves the problem if the model of the neural network was responsible for it. And if, despite the number of mutations, we are still faced with the lack of results, then, we know for sure that the problem lies with the data. So we can work towards a solution.<\/p>\n<p class=\" translation-block\">If you're curious about the details of the experience, <a href=\"http:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/01\/benchmark.pdf\" target=\"_self\"> everything is available <\/a>. And your are welcome to <a href=\"http:\/\/www.datavaloris.com\/en\/nnto_pre-inscription\/\" target=\"_self\"> test our mutant neural networks <\/a>, do not hesitate!<\/p>","protected":false},"excerpt":{"rendered":"<p>Qu&rsquo;est-ce qu&rsquo;un bon r\u00e9seau de neurones ? Il y a deux crit\u00e8res qui vont \u00eatre d\u00e9terminants, selon l&rsquo;objectif que l&rsquo;on se fixe : Soit on cherche \u00e0 rendre le r\u00e9seau de neurones plus performant, et on va vouloir am\u00e9liorer son taux de pr\u00e9cision pour maximiser les bonnes r\u00e9ponses. On va donc tenter d&rsquo;obtenir un r\u00e9seau [&hellip;]<\/p>","protected":false},"author":6,"featured_media":2177,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_exactmetrics_skip_tracking":false,"_exactmetrics_sitenote_active":false,"_exactmetrics_sitenote_note":"","_exactmetrics_sitenote_category":0,"footnotes":""},"categories":[15],"tags":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v22.7 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Optimisation de r\u00e9seaux de neurones - DATAVALORIS<\/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:\/\/www.datavaloris.com\/en\/neural-network-optimization\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Optimisation de r\u00e9seaux de neurones - DATAVALORIS\" \/>\n<meta property=\"og:description\" content=\"Qu&rsquo;est-ce qu&rsquo;un bon r\u00e9seau de neurones ? Il y a deux crit\u00e8res qui vont \u00eatre d\u00e9terminants, selon l&rsquo;objectif que l&rsquo;on se fixe : Soit on cherche \u00e0 rendre le r\u00e9seau de neurones plus performant, et on va vouloir am\u00e9liorer son taux de pr\u00e9cision pour maximiser les bonnes r\u00e9ponses. On va donc tenter d&rsquo;obtenir un r\u00e9seau [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.datavaloris.com\/en\/neural-network-optimization\/\" \/>\n<meta property=\"og:site_name\" content=\"DATAVALORIS\" \/>\n<meta property=\"article:published_time\" content=\"2019-07-31T08:33:51+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2020-06-26T15:52:13+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1191\" \/>\n\t<meta property=\"og:image:height\" content=\"908\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Claire Chabas\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DataValoris\" \/>\n<meta name=\"twitter:site\" content=\"@DataValoris\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Claire Chabas\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/\"},\"author\":{\"name\":\"Claire Chabas\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/fa1837c9e9b0656da555eb9b426ad982\"},\"headline\":\"Optimisation de r\u00e9seaux de neurones\",\"datePublished\":\"2019-07-31T08:33:51+00:00\",\"dateModified\":\"2020-06-26T15:52:13+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/\"},\"wordCount\":743,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#organization\"},\"image\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png\",\"articleSection\":[\"IA\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/\",\"url\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/\",\"name\":\"Optimisation de r\u00e9seaux de neurones - DATAVALORIS\",\"isPartOf\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png\",\"datePublished\":\"2019-07-31T08:33:51+00:00\",\"dateModified\":\"2020-06-26T15:52:13+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage\",\"url\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png\",\"contentUrl\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png\",\"width\":1191,\"height\":908},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.datavaloris.com\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Optimisation de r\u00e9seaux de neurones\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#website\",\"url\":\"https:\/\/www.datavaloris.com\/fr\/\",\"name\":\"DATAVALORIS\",\"description\":\"YOUR AI, OUR MISSION\",\"publisher\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.datavaloris.com\/fr\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#organization\",\"name\":\"DataValoris\",\"url\":\"https:\/\/www.datavaloris.com\/fr\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/05\/dv_hp_datavaloris_com_BD.png\",\"contentUrl\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/05\/dv_hp_datavaloris_com_BD.png\",\"width\":200,\"height\":128,\"caption\":\"DataValoris\"},\"image\":{\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/DataValoris\",\"https:\/\/www.linkedin.com\/company-beta\/10550917\/\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/fa1837c9e9b0656da555eb9b426ad982\",\"name\":\"Claire Chabas\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/12\/Me-small.jpg\",\"contentUrl\":\"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/12\/Me-small.jpg\",\"caption\":\"Claire Chabas\"},\"url\":\"https:\/\/www.datavaloris.com\/en\/author\/clairechabas\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Optimisation de r\u00e9seaux de neurones - DATAVALORIS","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:\/\/www.datavaloris.com\/en\/neural-network-optimization\/","og_locale":"en_US","og_type":"article","og_title":"Optimisation de r\u00e9seaux de neurones - DATAVALORIS","og_description":"Qu&rsquo;est-ce qu&rsquo;un bon r\u00e9seau de neurones ? Il y a deux crit\u00e8res qui vont \u00eatre d\u00e9terminants, selon l&rsquo;objectif que l&rsquo;on se fixe : Soit on cherche \u00e0 rendre le r\u00e9seau de neurones plus performant, et on va vouloir am\u00e9liorer son taux de pr\u00e9cision pour maximiser les bonnes r\u00e9ponses. On va donc tenter d&rsquo;obtenir un r\u00e9seau [&hellip;]","og_url":"https:\/\/www.datavaloris.com\/en\/neural-network-optimization\/","og_site_name":"DATAVALORIS","article_published_time":"2019-07-31T08:33:51+00:00","article_modified_time":"2020-06-26T15:52:13+00:00","og_image":[{"width":1191,"height":908,"url":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png","type":"image\/png"}],"author":"Claire Chabas","twitter_card":"summary_large_image","twitter_creator":"@DataValoris","twitter_site":"@DataValoris","twitter_misc":{"Written by":"Claire Chabas","Est. reading time":"4 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#article","isPartOf":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/"},"author":{"name":"Claire Chabas","@id":"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/fa1837c9e9b0656da555eb9b426ad982"},"headline":"Optimisation de r\u00e9seaux de neurones","datePublished":"2019-07-31T08:33:51+00:00","dateModified":"2020-06-26T15:52:13+00:00","mainEntityOfPage":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/"},"wordCount":743,"commentCount":0,"publisher":{"@id":"https:\/\/www.datavaloris.com\/fr\/#organization"},"image":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage"},"thumbnailUrl":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png","articleSection":["IA"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/","url":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/","name":"Optimisation de r\u00e9seaux de neurones - DATAVALORIS","isPartOf":{"@id":"https:\/\/www.datavaloris.com\/fr\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage"},"image":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage"},"thumbnailUrl":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png","datePublished":"2019-07-31T08:33:51+00:00","dateModified":"2020-06-26T15:52:13+00:00","breadcrumb":{"@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#primaryimage","url":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png","contentUrl":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/06\/Schema_opti_efficacite_BD.png","width":1191,"height":908},{"@type":"BreadcrumbList","@id":"https:\/\/www.datavaloris.com\/fr\/neural-network-optimization\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.datavaloris.com\/"},{"@type":"ListItem","position":2,"name":"Optimisation de r\u00e9seaux de neurones"}]},{"@type":"WebSite","@id":"https:\/\/www.datavaloris.com\/fr\/#website","url":"https:\/\/www.datavaloris.com\/fr\/","name":"DATAVALORIS","description":"YOUR AI, OUR MISSION","publisher":{"@id":"https:\/\/www.datavaloris.com\/fr\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.datavaloris.com\/fr\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.datavaloris.com\/fr\/#organization","name":"DataValoris","url":"https:\/\/www.datavaloris.com\/fr\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.datavaloris.com\/fr\/#\/schema\/logo\/image\/","url":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/05\/dv_hp_datavaloris_com_BD.png","contentUrl":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2019\/05\/dv_hp_datavaloris_com_BD.png","width":200,"height":128,"caption":"DataValoris"},"image":{"@id":"https:\/\/www.datavaloris.com\/fr\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/DataValoris","https:\/\/www.linkedin.com\/company-beta\/10550917\/"]},{"@type":"Person","@id":"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/fa1837c9e9b0656da555eb9b426ad982","name":"Claire Chabas","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.datavaloris.com\/fr\/#\/schema\/person\/image\/","url":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/12\/Me-small.jpg","contentUrl":"https:\/\/www.datavaloris.com\/wp-content\/uploads\/2018\/12\/Me-small.jpg","caption":"Claire Chabas"},"url":"https:\/\/www.datavaloris.com\/en\/author\/clairechabas\/"}]}},"_links":{"self":[{"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/posts\/2209"}],"collection":[{"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/comments?post=2209"}],"version-history":[{"count":17,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/posts\/2209\/revisions"}],"predecessor-version":[{"id":4009,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/posts\/2209\/revisions\/4009"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/media\/2177"}],"wp:attachment":[{"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/media?parent=2209"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/categories?post=2209"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.datavaloris.com\/en\/wp-json\/wp\/v2\/tags?post=2209"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}