{"id":134409,"date":"2021-12-16T17:26:03","date_gmt":"2021-12-16T16:26:03","guid":{"rendered":"https:\/\/multi.datascientest.com\/?p=134409"},"modified":"2023-10-30T17:38:23","modified_gmt":"2023-10-30T16:38:23","slug":"convolutional-neural-network-es","status":"publish","type":"post","link":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es","title":{"rendered":"Convolutional Neural Network : definici\u00f3n y funcionamiento"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"134409\" class=\"elementor elementor-134409\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-4693eef elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"4693eef\" data-element_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-1a13d96\" data-id=\"1a13d96\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-60e5fe6 elementor-widget elementor-widget-text-editor\" data-id=\"60e5fe6\" 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<h2><b>En esta parte, vamos a centrarnos en uno de los algoritmos m\u00e1s eficaces del Deep Learning, las Convolutional Neural Network o CNN: redes neuronales convolucionales, son modelos de programaci\u00f3n potentes que permiten principalmente el reconocimiento de im\u00e1genes atribuyendo autom\u00e1ticamente a cada imagen proporcionada en la entrada, una etiqueta correspondiente a la clase a la que pertenece.<\/b><\/h2>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d2f4b37 elementor-widget elementor-widget-text-editor\" data-id=\"d2f4b37\" 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>Te damos la bienvenida a este tercer episodio de nuestro dossier de Deep Learning. Despu\u00e9s de la introducci\u00f3n al Deep Learning y sus aplicaciones en la primera parte, abordamos ahora la estructura y el funcionamiento de las redes neuronales en esta segunda parte.<\/p><p>\u00a1All\u00e1 vamos!<\/p>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-37d0b59 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"37d0b59\" data-element_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-d09b7e6\" data-id=\"d09b7e6\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2b1a243 elementor-widget elementor-widget-video\" data-id=\"2b1a243\" data-element_type=\"widget\" data-settings=\"{&quot;youtube_url&quot;:&quot;https:\\\/\\\/www.youtube.com\\\/watch?v=DazUaVu5MO0&amp;ab_channel=DataScientest&quot;,&quot;video_type&quot;:&quot;youtube&quot;,&quot;controls&quot;:&quot;yes&quot;}\" data-widget_type=\"video.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-video .elementor-widget-container{overflow:hidden;transform:translateZ(0)}.elementor-widget-video .elementor-wrapper{aspect-ratio:var(--video-aspect-ratio)}.elementor-widget-video .elementor-wrapper iframe,.elementor-widget-video .elementor-wrapper video{height:100%;width:100%;display:flex;border:none;background-color:#000}@supports not (aspect-ratio:1\/1){.elementor-widget-video .elementor-wrapper{position:relative;overflow:hidden;height:0;padding-bottom:calc(100% \/ var(--video-aspect-ratio))}.elementor-widget-video .elementor-wrapper iframe,.elementor-widget-video .elementor-wrapper video{position:absolute;top:0;right:0;bottom:0;left:0}}.elementor-widget-video .elementor-open-inline .elementor-custom-embed-image-overlay{position:absolute;top:0;right:0;bottom:0;left:0;background-size:cover;background-position:50%}.elementor-widget-video .elementor-custom-embed-image-overlay{cursor:pointer;text-align:center}.elementor-widget-video .elementor-custom-embed-image-overlay:hover .elementor-custom-embed-play i{opacity:1}.elementor-widget-video .elementor-custom-embed-image-overlay img{display:block;width:100%;aspect-ratio:var(--video-aspect-ratio);-o-object-fit:cover;object-fit:cover;-o-object-position:center center;object-position:center center}@supports not (aspect-ratio:1\/1){.elementor-widget-video .elementor-custom-embed-image-overlay{position:relative;overflow:hidden;height:0;padding-bottom:calc(100% \/ var(--video-aspect-ratio))}.elementor-widget-video .elementor-custom-embed-image-overlay img{position:absolute;top:0;right:0;bottom:0;left:0}}.elementor-widget-video .e-hosted-video .elementor-video{-o-object-fit:cover;object-fit:cover}.e-con-inner>.elementor-widget-video,.e-con>.elementor-widget-video{width:var(--container-widget-width);--flex-grow:var(--container-widget-flex-grow)}<\/style>\t\t<div class=\"elementor-wrapper elementor-open-inline\">\n\t\t\t<div class=\"elementor-video\"><\/div>\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<\/div>\n\t\t\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-257b3b8 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"257b3b8\" data-element_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-823b3b9\" data-id=\"823b3b9\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-d783a11 elementor-widget elementor-widget-text-editor\" data-id=\"d783a11\" 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<strong>\u00cdndice:<\/strong>\n<ol>\n \t<li style=\"font-weight: 400;\"><span style=\"color: #6e4ceb;\"><a style=\"color: #6e4ceb;\" href=\"#architecturecnn\"><span style=\"font-weight: 400;\">Arquitectura de una Convolutional Neural Network-CNN<\/span><\/a><\/span><\/li>\n \t<li style=\"font-weight: 400;\"><span style=\"color: #6e4ceb;\"><a style=\"color: #6e4ceb;\" href=\"#partieconvolutive\"><span style=\"font-weight: 400;\">Parte convolucional<\/span><\/a><\/span><\/li>\n \t<li><span style=\"color: #6e4ceb;\"><a style=\"color: #6e4ceb;\" href=\"#maxpooling\">El Max-Pooling<\/a><\/span><\/li>\n \t<li><span style=\"color: #6e4ceb;\"><a style=\"color: #6e4ceb;\" href=\"#miseenapplicationcnn\">Aplicaci\u00f3n de la CNN<\/a><\/span><\/li>\n<\/ol>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-d57a369 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"d57a369\" data-element_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-9972d5e\" data-id=\"9972d5e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-5b09b7d elementor-widget elementor-widget-image\" data-id=\"5b09b7d\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/datascientest.com\/catalog\/module\/151-introduction-au-deeplearning-avec-le-framework-keras\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"370\" height=\"264\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-1.png\" class=\"attachment-large size-large wp-image-4469\" alt=\"modele cnn\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-1.png 370w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-1-300x214.png 300w\" sizes=\"(max-width: 370px) 100vw, 370px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Reconocimiento de un Pap\u00e1 Noel con un modelo de CNN<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-14ea57f elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"14ea57f\" data-element_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-db37e72\" data-id=\"db37e72\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-ba14e68 elementor-widget elementor-widget-heading\" data-id=\"ba14e68\" 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><h3 class=\"elementor-heading-title elementor-size-default\">Arquitectura de una Convolutional Neural Network (CNN)<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-e4bc4c0 elementor-widget elementor-widget-menu-anchor\" data-id=\"e4bc4c0\" data-element_type=\"widget\" data-widget_type=\"menu-anchor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\nbody.elementor-page .elementor-widget-menu-anchor{margin-bottom:0}<\/style>\t\t<div id=\"architecturecnn\" class=\"elementor-menu-anchor\"><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-50e452d elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"50e452d\" data-element_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-9537378\" data-id=\"9537378\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-9f33b3c elementor-widget elementor-widget-text-editor\" data-id=\"9f33b3c\" 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<blockquote><p><i><span style=\"font-weight: 400;\">Las <\/span><\/i><b><i>CNN<\/i><\/b><i><span style=\"font-weight: 400;\"> designan una subcategor\u00eda de redes neuronales y hoy en d\u00eda son uno de los <\/span><\/i><b><i>modelos de clasificaci\u00f3n de im\u00e1genes<\/i><\/b><i><span style=\"font-weight: 400;\"> considerados como m\u00e1s eficaces.<\/span><\/i><\/p><\/blockquote>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5059733 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"5059733\" 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\" href=\"https:\/\/datascientest.com\/es\/formacion-data-scientist\">\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\">M\u00e1s informaci\u00f3n sobre las CNN<\/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<\/div>\n\t\t\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-fba700d elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"fba700d\" data-element_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-0ceff5f\" data-id=\"0ceff5f\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-579859c elementor-widget elementor-widget-text-editor\" data-id=\"579859c\" 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\tA primera vista, su modo de funcionamiento es sencillo: el usuario proporciona una imagen de entrada en forma de matriz de p\u00edxeles.\nEsta dispone de 3 dimensiones:\n<ul>\n \t<li><strong>Dos dimensiones<\/strong> para una imagen en niveles de gris.<\/strong><\/li>\n \t<li>Una tercera dimensi\u00f3n, de profundidad 3 para representar los colores fundamentales (rojo, verde, azul).<\/li>\n<\/ul>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-da3c8a1 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"da3c8a1\" data-element_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-e9a151a\" data-id=\"e9a151a\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3abe7f5 elementor-widget elementor-widget-text-editor\" data-id=\"3abe7f5\" 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>Al contrario que un modelo MLP (Multi Layers Perceptron) cl\u00e1sico que solo contiene una parte de clasificaci\u00f3n, la arquitectura de la Convolutional Neural Network dispone previamente de una parte convolucional y consta por tanto de dos partes bien diferenciadas:<\/p><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><b>Un parte convolucional<\/b><span style=\"font-weight: 400;\">: Su objetivo final es <\/span><b>extraer caracter\u00edsticas propias de cada imagen<\/b> <b>comprimi\u00e9ndolas<\/b><span style=\"font-weight: 400;\"> para <\/span><b>reducir su tama\u00f1o inicial<\/b><span style=\"font-weight: 400;\">. En resumen, la imagen proporcionada de entrada pasa a trav\u00e9s de una <\/span><b>sucesi\u00f3n de filtros<\/b><span style=\"font-weight: 400;\">, creando a la vez nuevas im\u00e1genes llamadas <\/span><b>tarjetas de convoluciones<\/b><span style=\"font-weight: 400;\">. Por \u00faltimo, las <\/span><b>tarjetas de convoluciones<\/b> <b>obtenidas se concatenan<\/b><span style=\"font-weight: 400;\"> en un vector de caracter\u00edsticas llamado <\/span><b>c\u00f3digo CNN<\/b><span style=\"font-weight: 400;\">.<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">Una <\/span><b>parte de clasificaci\u00f3n<\/b><span style=\"font-weight: 400;\">: El <\/span><b>c\u00f3digo CNN obtenido en la salida<\/b><span style=\"font-weight: 400;\"> de la parte convolucional <\/span><b>se suministra como entrada en una segunda parte<\/b><span style=\"font-weight: 400;\">, compuesta de capas totalmente conectadas llamadas perceptr\u00f3n multicapa (MLP o Multi Layers Perceptron). El papel de esta parte es <\/span><b>combinar las caracter\u00edsticas del c\u00f3digo CNN para clasificar la imagen<\/b><span style=\"font-weight: 400;\">. Para volver a esta parte, no dudes en <\/span><span style=\"font-weight: 400;\">consultar el art\u00edculo sobre este tema<\/span><span style=\"font-weight: 400;\">.<\/span><\/li><\/ul>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-be1c9b1 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"be1c9b1\" data-element_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-16ba27d\" data-id=\"16ba27d\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-961ded1 elementor-widget elementor-widget-image\" data-id=\"961ded1\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/blog.octo.com\/classification-dimages-les-reseaux-de-neurones-convolutifs-en-toute-simplicite\/\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"125\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-2.png\" class=\"attachment-large size-large wp-image-4468\" alt=\"convolutional neural network\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-2.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-2-300x73.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Sch\u00e9ma repr\u00e9sentant l\u2019architecture d\u2019un CNN<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-7dbfda2 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"7dbfda2\" data-element_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-a4ec15e\" data-id=\"a4ec15e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-f421e61 elementor-widget elementor-widget-heading\" data-id=\"f421e61\" 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\">Parte convolucional\n<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ca5e0f5 elementor-widget elementor-widget-menu-anchor\" data-id=\"ca5e0f5\" data-element_type=\"widget\" data-widget_type=\"menu-anchor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div id=\"partieconvolutive\" class=\"elementor-menu-anchor\"><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-9cc2afe elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"9cc2afe\" data-element_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-2756347\" data-id=\"2756347\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c45282b elementor-widget elementor-widget-heading\" data-id=\"c45282b\" 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\">En primer lugar, \u00bfpara qu\u00e9 sirve la convoluci\u00f3n?\n<\/h3>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-c91f422 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"c91f422\" data-element_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-45915f5\" data-id=\"45915f5\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-42318c8 elementor-widget elementor-widget-text-editor\" data-id=\"42318c8\" 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;\">La <strong>convoluci\u00f3n<\/strong> es una operaci\u00f3n matem\u00e1tica sencilla que se suele utilizar para el tratamiento y el <strong>reconocimiento de im\u00e1genes<\/strong>. En una imagen, su efecto se asimila a un filtrado, aqu\u00ed puedes ver su funcionamiento:<\/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<\/div>\n\t\t\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-85cc246 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"85cc246\" data-element_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-7d0ed7b\" data-id=\"7d0ed7b\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-faf8b45 elementor-widget elementor-widget-image\" data-id=\"faf8b45\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/www.natural-solutions.eu\/blog\/la-reconnaissance-dimage-avec-les-rseaux-de-neurones-convolutifs#:~:text=Qu&#039;est%2Dce%20que%20la,%C3%A0%20la%20reconnaissance%20d&#039;image.\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"471\" height=\"302\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-3.png\" class=\"attachment-large size-large wp-image-4467\" alt=\"neural network\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-3.png 471w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-3-300x192.png 300w\" sizes=\"(max-width: 471px) 100vw, 471px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Sch\u00e9ma du parcours de la fen\u00eatre de filtre sur l'image<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-1774dd2 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"1774dd2\" data-element_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-c340912\" data-id=\"c340912\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3034022 elementor-widget elementor-widget-text-editor\" data-id=\"3034022\" 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<ol><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">En primer lugar, <\/span><b>se define el tama\u00f1o de la ventana de filtro<\/b><span style=\"font-weight: 400;\"> situada en la parte superior izquierda.<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">La ventana de filtro, que representa la caracter\u00edstica, <\/span><b>se desplaza progresivamente<\/b><span style=\"font-weight: 400;\"> de izquierda a derecha un determinado n\u00famero de casillas definido previamente (<\/span><b>el paso)<\/b><span style=\"font-weight: 400;\"> hasta llegar al final de la imagen.<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><span style=\"font-weight: 400;\">En cada porci\u00f3n de imagen que encuentra, se efect\u00faa el c\u00e1lculo de convoluci\u00f3n permitiendo obtener en la salida una tarjeta de activaci\u00f3n o feature map que indica d\u00f3nde est\u00e1n localizadas las features en la imagen: cuanto m\u00e1s arriba est\u00e9 la feature, m\u00e1s se parecer\u00e1 a ella la porci\u00f3n de imagen barrida.<\/span><\/li><\/ol>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-d3635a8 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"d3635a8\" 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\" href=\"https:\/\/datascientest.com\/es\/formacion-data-scientist\">\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\">M\u00e1s informaci\u00f3n sobre la convoluci\u00f3n<\/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<\/div>\n\t\t\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-4824bae elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"4824bae\" data-element_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-445ac93\" data-id=\"445ac93\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-e382902 elementor-widget elementor-widget-heading\" data-id=\"e382902\" 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\">Ejemplo de un filtro de convoluci\u00f3n cl\u00e1sica<\/h3>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-2c03b12 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"2c03b12\" data-element_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-93b60e7\" data-id=\"93b60e7\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-8d38ee8 elementor-widget elementor-widget-text-editor\" data-id=\"8d38ee8\" 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;\">Durante la parte convolucional de una Convolutional Neural Network, la imagen proporcionada de entrada pasa a trav\u00e9s de una <\/span><b>serie de filtros de convoluci\u00f3n<\/b><span style=\"font-weight: 400;\">. Por ejemplo existen filtros de convoluci\u00f3n que se suelen utilizar y que permiten extraer caracter\u00edsticas m\u00e1s pertinentes que p\u00edxeles como la detecci\u00f3n de bordes (<\/span><b>filtro derivador<\/b><span style=\"font-weight: 400;\">) o formas geom\u00e9tricas. La elecci\u00f3n y la aplicaci\u00f3n de los filtros la hace de forma <\/span><b>autom\u00e1tica<\/b><span style=\"font-weight: 400;\"> el modelo.<\/span><\/p><p><span style=\"font-weight: 400;\">Entre los filtros m\u00e1s conocidos, encontramos en particular el <\/span><b>filtro promediador<\/b><span style=\"font-weight: 400;\"> (calcula para cada p\u00edxel la media del p\u00edxel con sus 8 vecinos m\u00e1s cercanos) o incluso el <\/span><b>filtro gaussiano<\/b><span style=\"font-weight: 400;\"> que permite reducir el ruido de una imagen proporcionada de entrada:<\/span><\/p><p><span style=\"font-weight: 400;\">Aqu\u00ed puedes ver un ejemplo de esos dos filtros diferentes en una imagen que incluye un gran ruido (se puede pensar en una fotograf\u00eda tomada con poca luminosidad por ejemplo). Sin embargo, uno de los inconvenientes de la reducci\u00f3n del ruido es que por lo general va acompa\u00f1ada de una reducci\u00f3n de la nitidez:<\/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<\/div>\n\t\t\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-8a8e100 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"8a8e100\" data-element_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-6541140\" data-id=\"6541140\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-36152f0 elementor-widget elementor-widget-image\" data-id=\"36152f0\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"306\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-4.png\" class=\"attachment-large size-large wp-image-4466\" alt=\"CNN Deep learning\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-4.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-4-300x179.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Effet des filtres moyenneur et gaussien - DataScientest<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-35db3e3 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"35db3e3\" data-element_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-202dec2\" data-id=\"202dec2\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2194949 elementor-widget elementor-widget-text-editor\" data-id=\"2194949\" 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;\">Como se puede observar, al contrario del filtro promediador, el filtro gaussiano reduce el ruido sin reducir significativamente la nitidez.\u00a0<\/span><\/p><blockquote><p><i><span style=\"font-weight: 400;\">Adem\u00e1s de su funci\u00f3n de filtrado, el inter\u00e9s de la <\/span><\/i><b><i>parte convolucional de una CNN<\/i><\/b><i><span style=\"font-weight: 400;\"> es que permite <\/span><\/i><b><i>extraer caracter\u00edsticas propias de cada imagen <\/i><\/b><i><span style=\"font-weight: 400;\">comprimi\u00e9ndolas para reducir su tama\u00f1o inicial, a trav\u00e9s de m\u00e9todos de submuestreo como el Max-Pooling.<\/span><\/i><\/p><\/blockquote>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-4b1b32e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"4b1b32e\" data-element_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-7bb509d\" data-id=\"7bb509d\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-8409d84 elementor-widget elementor-widget-heading\" data-id=\"8409d84\" 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\">M\u00e9todo de submuestreo : el Max-Pooling\n<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1568211 elementor-widget elementor-widget-menu-anchor\" data-id=\"1568211\" data-element_type=\"widget\" data-widget_type=\"menu-anchor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div id=\"maxpoooling\" class=\"elementor-menu-anchor\"><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-410250b elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"410250b\" data-element_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-519070a\" data-id=\"519070a\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-dc35588 elementor-widget elementor-widget-text-editor\" data-id=\"dc35588\" 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;\">El Max-Pooling es un proceso de discretizaci\u00f3n basado en muestras. Su objetivo es submuestrear una representaci\u00f3n de entrada (imagen, matriz de salida de capa escondida, etc.) reduciendo su tama\u00f1o. Adem\u00e1s, su inter\u00e9s es que reduce el coste de c\u00e1lculo reduciendo el n\u00famero de par\u00e1metros que tiene que aprender y proporciona una invariancia por peque\u00f1as translaciones (si una peque\u00f1a translaci\u00f3n no modifica el m\u00e1ximo de la regi\u00f3n barrida, el m\u00e1ximo de cada regi\u00f3n seguir\u00e1 siendo el mismo y por tanto la nueva matriz creada ser\u00e1 id\u00e9ntica).<\/span><\/p><p><span style=\"font-weight: 400;\">Para que la acci\u00f3n del Max-Pooling sea m\u00e1s concreta, aqu\u00ed tienes un ejemplo: imagina que tenemos una matriz de 4\u00d74 que representa nuestra entrada inicial y un filtro de una ventana de 2\u00d72 que aplicaremos en nuestra entrada. Para cada regi\u00f3n barrida por el filtro, el Max-pooling tomar\u00e1 el m\u00e1ximo y de ese modo crear\u00e1 a la vez una nueva matriz de salida en la que cada elemento corresponder\u00e1 a los m\u00e1ximos de cada regi\u00f3n detectada.<\/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<\/div>\n\t\t\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-3c537ab elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"3c537ab\" data-element_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-7996df4\" data-id=\"7996df4\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-0875309 elementor-widget elementor-widget-text-editor\" data-id=\"0875309\" 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;\">Ilustremos el proceso :<\/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-fe70b45 elementor-widget elementor-widget-image\" data-id=\"fe70b45\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/cs231n.github.io\/convolutional-networks\/\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"236\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-5.png\" class=\"attachment-large size-large wp-image-4465\" alt=\"convolutional neural network\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-5.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-5-300x138.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Procesamiento del Max-Pooling<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-3cedbf8 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"3cedbf8\" data-element_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-ec9661a\" data-id=\"ec9661a\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c225c39 elementor-widget elementor-widget-text-editor\" data-id=\"c225c39\" 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;\">La ventana de filtro se desplaza dos p\u00edxeles hacia la derecha (stride\/paso = 2) y recupera en cada paso \u201cel argmax\u201d que corresponde al valor m\u00e1s alto de los 4 valores de p\u00edxeles.<\/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<\/div>\n\t\t\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-739f869 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"739f869\" data-element_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-3f202c6\" data-id=\"3f202c6\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3caf6de elementor-widget elementor-widget-image\" data-id=\"3caf6de\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/cs231n.github.io\/convolutional-networks\/\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"386\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-6.png\" class=\"attachment-large size-large wp-image-4474\" alt=\"reseau de neurones convolutif\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-6.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-6-300x226.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Exemple d\u2019effet du Max-Pooling<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-3e2f893 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"3e2f893\" data-element_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-d4839e5\" data-id=\"d4839e5\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-74005e7 elementor-widget elementor-widget-text-editor\" data-id=\"74005e7\" 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><b>Se comprende mejor la utilidad de la parte convolucional de una CNN<\/b><span style=\"font-weight: 400;\">: al contrario que un modelo MLP cl\u00e1sico, a\u00f1adir previamente la parte convolucional permite <\/span><b>obtener en la salida una \u201ctarjeta de caracter\u00edsticas\u201d o \u201cc\u00f3digo CNN\u201d<\/b><span style=\"font-weight: 400;\"> (matriz de p\u00edxeles situada a la derecha en el ejemplo) cuyas <\/span><b>dimensiones<\/b><span style=\"font-weight: 400;\"> son <\/span><b>m\u00e1s peque\u00f1as<\/b><span style=\"font-weight: 400;\"> que las de la imagen inicial lo cual va a tener la ventaja de <\/span><b>disminuir<\/b><span style=\"font-weight: 400;\"> en gran medida el n\u00famero de <\/span><b>par\u00e1metros que tiene que calcular el modelo<\/b><span style=\"font-weight: 400;\">.<\/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<\/div>\n\t\t\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-26c055e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"26c055e\" data-element_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-fc5949e\" data-id=\"fc5949e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2122224 elementor-align-center elementor-widget elementor-widget-button\" data-id=\"2122224\" 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\" href=\"https:\/\/datascientest.com\/es\/formacion-data-scientist\">\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\"> Aprende a hacer Max-Pooling<\/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<\/div>\n\t\t\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-28729d8 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"28729d8\" data-element_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-07d11d8\" data-id=\"07d11d8\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-8e00637 elementor-widget elementor-widget-heading\" data-id=\"8e00637\" 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\">Ejemplo de una arquitectura CNN y su salida\n<\/h3>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-8693802 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"8693802\" data-element_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-51d2faf\" data-id=\"51d2faf\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-ca003c0 elementor-widget elementor-widget-text-editor\" data-id=\"ca003c0\" 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;\">Despu\u00e9s de la parte convolucional de una CNN, viene la parte de clasificaci\u00f3n. Esa parte de clasificaci\u00f3n, com\u00fan para todos los modelos de redes neuronales, corresponde a un <\/span>modelo de perceptr\u00f3n multicapas (MLP)<span style=\"font-weight: 400;\">.\u00a0 <\/span><\/p><blockquote><p><i><span style=\"font-weight: 400;\">Su objetivo es atribuir a cada muestra de datos una etiqueta que describa la clase a la que pertenezca.\u00a0<\/span><\/i><\/p><\/blockquote>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-14c38f9 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"14c38f9\" data-element_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-55c286d\" data-id=\"55c286d\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-8cbe06e elementor-widget elementor-widget-image\" data-id=\"8cbe06e\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/www.natural-solutions.eu\/blog\/histoire-du-deep-learning#:~:text=Les%20perceptrons%20multicouches%3A%20MLP%20(1986)&#038;text=Les%20perceptrons%20multicouches%20ont%20pour,mieux%20cette%20base%20de%20donn%C3%A9es.\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"445\" height=\"310\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-7.png\" class=\"attachment-large size-large wp-image-4473\" alt=\"reseau de neurones\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-7.png 445w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-7-300x209.png 300w\" sizes=\"(max-width: 445px) 100vw, 445px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Representaci\u00f3n de un perceptr\u00f3n multicapas<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-fff7dc6 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"fff7dc6\" data-element_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-80bcd15\" data-id=\"80bcd15\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-e485c6b elementor-widget elementor-widget-text-editor\" data-id=\"e485c6b\" 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;\">El algoritmo que utilizan los perceptrones para actualizar sus pesos (o coeficientes de redes) se llama la <\/span><b>retropropagaci\u00f3n del gradiente del error<\/b><span style=\"font-weight: 400;\">, famoso algoritmo de descenso de gradiente que veremos m\u00e1s en detalle despu\u00e9s.\u00a0\u00a0<\/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<\/div>\n\t\t\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-11a5c5e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"11a5c5e\" data-element_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-9d737ff\" data-id=\"9d737ff\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-5cacc93 elementor-widget elementor-widget-image\" data-id=\"5cacc93\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"http:\/\/dspace.univ-tlemcen.dz\/bitstream\/112\/12235\/1\/Classification-des-images-avec-les-reseaux-de-neurones.pdf\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"131\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-8.png\" class=\"attachment-large size-large wp-image-4472\" alt=\"reseau de neurones\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-8.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-8-300x77.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Ejemplo de arquitectura de una CNN<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-c9a9c28 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"c9a9c28\" data-element_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-75ab1fe\" data-id=\"75ab1fe\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-5579c36 elementor-widget elementor-widget-text-editor\" data-id=\"5579c36\" 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;\">Por lo general, la <\/span><b>arquitectura de una Convolutional Neural Network<\/b><span style=\"font-weight: 400;\"> es pr\u00e1cticamente igual:\u00a0<\/span><\/p><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><b>Capa de convoluci\u00f3n (CONV)<\/b><span style=\"font-weight: 400;\">: El papel de esta primera capa es <\/span><b>analizar las im\u00e1genes proporcionadas en la entrada<\/b><span style=\"font-weight: 400;\"> y <\/span><b>detectar la presencia de un conjunto de features<\/b><span style=\"font-weight: 400;\">. A la salida de esa capa se obtiene un conjunto de features maps (<\/span><i><span style=\"font-weight: 400;\">ver m\u00e1s arriba: \u00bfpara qu\u00e9 sirve la convoluci\u00f3n?<\/span><\/i><span style=\"font-weight: 400;\">).<\/span><\/li><li style=\"font-weight: 400;\" aria-level=\"1\"><b>Capa de Pooling (POOL)<\/b><span>: La capa de Pooling es una operaci\u00f3n que por lo general se aplica entre dos capas de convoluci\u00f3n. Esta recibe en la entrada las features maps formadas en la salida de la capa de convoluci\u00f3n y su papel es<\/span><b> reducir el tama\u00f1o de las im\u00e1genes y, a la vez, preservar sus caracter\u00edsticas m\u00e1s esenciales<\/b><span>. Entre las m\u00e1s utilizadas, encontramos el <\/span><b>Max-pooling<\/b><span> mencionado m\u00e1s arriba o incluso el <\/span><b>Average pooling<\/b><span> cuyo manejo consiste en conservar en cada paso, el valor medio de la ventana de filtro.<\/span><\/li><\/ul><blockquote><p><i><span style=\"font-weight: 400;\">Finalmente, se obtiene en la salida de esa capa de Pooling el mismo n\u00famero de feature maps que en la salida, pero considerablemente comprimidas.<\/span><\/i><\/p><\/blockquote><ul><li style=\"font-weight: 400;\" aria-level=\"1\"><b>La capa de activaci\u00f3n ReLU (Rectified Linear Units)<\/b><span style=\"font-weight: 400;\">: Esa capa <\/span><b>sustituye todos los valores negativos recibidos en la entrada por ceros<\/b><span style=\"font-weight: 400;\">. El inter\u00e9s de esas capas de activaci\u00f3n es hacer que el modelo sea no lineal y por tanto, m\u00e1s complejo.<\/span><\/li><\/ul>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-e904e28 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"e904e28\" data-element_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-7de78de\" data-id=\"7de78de\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-14a63ba elementor-widget elementor-widget-image\" data-id=\"14a63ba\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/ml4a.github.io\/ml4a\/fr\/neural_networks\/\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"274\" height=\"228\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-9.png\" class=\"attachment-large size-large wp-image-4471\" alt=\"ReLU Rectified Linear Units\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Funci\u00f3n de activaci\u00f3n ReLU<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-2ba61ec elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"2ba61ec\" data-element_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-ff37bb2\" data-id=\"ff37bb2\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-faff565 elementor-widget elementor-widget-text-editor\" data-id=\"faff565\" 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<ul><li style=\"font-weight: 400;\" aria-level=\"1\"><b>Capa Fully Connected (FC)<\/b><span style=\"font-weight: 400;\">: Estas capas se colocan al final de la arquitectura de la CNN y <\/span><b>se conectan por completo a todas las neuronas de salida<\/b><span style=\"font-weight: 400;\"> (de ah\u00ed el t\u00e9rmino fully-connected). Despu\u00e9s de haber recibido un vector en la entrada, la capa FC aplica sucesivamente una <\/span><b>combinaci\u00f3n lineal<\/b><span style=\"font-weight: 400;\"> y una <\/span><b>funci\u00f3n de activaci\u00f3n<\/b><span style=\"font-weight: 400;\"> con la finalidad de <\/span><b>clasificar el input imagen<\/b><span style=\"font-weight: 400;\"> (<\/span><i><span style=\"font-weight: 400;\">ver el siguiente esquema<\/span><\/i><span style=\"font-weight: 400;\">). Finalmente en la salida devuelve un <\/span><b>vector de tama\u00f1o correspondiente al n\u00famero de clases<\/b><span style=\"font-weight: 400;\"> en el que cada componente representa la probabilidad de que el input imagen pertenezca a una clase.<\/span><\/li><\/ul>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-76b0287 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"76b0287\" data-element_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-22b647b\" data-id=\"22b647b\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-9413892 elementor-widget elementor-widget-image\" data-id=\"9413892\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"512\" height=\"295\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-10.png\" class=\"attachment-large size-large wp-image-4470\" alt=\"couche fully connected\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-10.png 512w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-10-300x173.png 300w\" sizes=\"(max-width: 512px) 100vw, 512px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Funcionamiento de una red neuronal con 2 capas escondidas<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-d03691f elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"d03691f\" data-element_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-caf5cfe\" data-id=\"caf5cfe\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-35e108d elementor-align-center elementor-widget elementor-widget-button\" data-id=\"35e108d\" 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\" href=\"https:\/\/datascientest.com\/es\/formacion-data-scientist\">\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\">Aprende a crear arquitecturas CNN<\/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<\/div>\n\t\t\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-23dc1ce elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"23dc1ce\" data-element_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-ccc656f\" data-id=\"ccc656f\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c0f205a elementor-widget elementor-widget-heading\" data-id=\"c0f205a\" 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\">Implementaci\u00f3n de una CNN preentrenada en Python :\n<\/h3>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-695284e elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"695284e\" data-element_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-e57fc50\" data-id=\"e57fc50\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3c6a601 elementor-widget elementor-widget-text-editor\" data-id=\"3c6a601\" 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;\">Para usos pr\u00e1cticos y considerando la complejidad de creaci\u00f3n de las <\/span><b>eficaces<\/b><span style=\"font-weight: 400;\"> CNN \u201chechas a mano\u201d, vamos a utilizar las redes preentrenadas disponibles en el m\u00f3dulo <\/span><a href=\"https:\/\/pytorch.org\/docs\/stable\/torchvision\/index.html\"><span style=\"font-weight: 400;\">Torchvision<\/span><\/a><span style=\"font-weight: 400;\">. Veamos c\u00f3mo se puede implementar en Python:<\/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<\/div>\n\t\t\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-ecedcff elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"ecedcff\" data-element_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-490a3f0\" data-id=\"490a3f0\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-c1aef0a elementor-widget elementor-widget-heading\" data-id=\"c1aef0a\" 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\">Aplicaci\u00f3n de una CNN : Identificaci\u00f3n de im\u00e1genes gen\u00e9ricas a partir del dataset ImageNet: diez millones de im\u00e1genes etiquetadas\n<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-199db5e elementor-widget elementor-widget-menu-anchor\" data-id=\"199db5e\" data-element_type=\"widget\" data-widget_type=\"menu-anchor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div id=\"miseenapplicationcnn\" class=\"elementor-menu-anchor\"><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-6e54b6b elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"6e54b6b\" data-element_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-cb7cab8\" data-id=\"cb7cab8\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-d13e04d elementor-widget elementor-widget-text-editor\" data-id=\"d13e04d\" 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;\">ImageNet es una base de datos de m\u00e1s de diez millones de im\u00e1genes etiquetadas creada por la organizaci\u00f3n del mismo nombre, y destinada a trabajos de investigaci\u00f3n de visi\u00f3n por computador.<\/span><\/p><p><span style=\"font-weight: 400;\">Aqu\u00ed tienes un extracto de ese voluminoso dataset:<\/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<\/div>\n\t\t\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-cfbe4ca elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"cfbe4ca\" data-element_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-edab4ea\" data-id=\"edab4ea\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-4ea9e21 elementor-widget elementor-widget-image\" data-id=\"4ea9e21\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t\t<a href=\"https:\/\/machinelearningmastery.com\/use-pre-trained-vgg-model-classify-objects-photographs\/\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"442\" height=\"272\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-11.png\" class=\"attachment-large size-large wp-image-4479\" alt=\"Extrait de la base de donn\u00e9es ImageNet\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-11.png 442w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-11-300x185.png 300w\" sizes=\"(max-width: 442px) 100vw, 442px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Extrait de la base de donn\u00e9es ImageNet<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-041a00a elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"041a00a\" data-element_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-be0e20f\" data-id=\"be0e20f\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-aab94b7 elementor-widget elementor-widget-text-editor\" data-id=\"aab94b7\" 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><b>Etapa 0 <\/b><span style=\"font-weight: 400;\">: Primero, importamos todas las bibliotecas que ser\u00e1n necesarias para los siguientes pasos<\/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-0807104 elementor-widget elementor-widget-shortcode\" data-id=\"0807104\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/fa94546ddc37a047de6c3135b9d726ef.js\"><\/script>\n<\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-986d1d3 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"986d1d3\" data-element_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-25b7998\" data-id=\"25b7998\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-60ff47b elementor-widget elementor-widget-text-editor\" data-id=\"60ff47b\" 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><b>Etapa 1 <\/b><span style=\"font-weight: 400;\">: Entrenamiento del modelo preentrenado VGG16<\/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-5fa27c3 elementor-widget elementor-widget-shortcode\" data-id=\"5fa27c3\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/4b4bf05a7373eb02de7279fd09c1c065.js\"><\/script><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-0cf6deb elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"0cf6deb\" data-element_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-7f3dde5\" data-id=\"7f3dde5\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-a5150ba elementor-widget elementor-widget-text-editor\" data-id=\"a5150ba\" 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><b>Etapa 2 <\/b><span style=\"font-weight: 400;\">: Importaci\u00f3n de las 3 im\u00e1genes que hay que predecir<\/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-086233a elementor-widget elementor-widget-shortcode\" data-id=\"086233a\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/cef25c1c0944feefdff693be31c9d70e.js\"><\/script><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-6a05dc3 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"6a05dc3\" data-element_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-dd58b01\" data-id=\"dd58b01\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-a3fdef0 elementor-widget elementor-widget-image\" data-id=\"a3fdef0\" 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\t<a href=\"https:\/\/www.amazon.fr\/Ballon-football-Taille-standard-blanc\/dp\/B018URJZPK\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"275\" height=\"183\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-12.png\" class=\"attachment-large size-large wp-image-4478\" alt=\"\" \/>\t\t\t\t\t\t\t\t<\/a>\n\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\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-97e464d\" data-id=\"97e464d\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-6c90ec0 elementor-widget elementor-widget-image\" data-id=\"6c90ec0\" 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\t<a href=\"https:\/\/fr.wikipedia.org\/wiki\/Canis_lupus\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"290\" height=\"193\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-13.png\" class=\"attachment-large size-large wp-image-4477\" alt=\"\" \/>\t\t\t\t\t\t\t\t<\/a>\n\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\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-00c823e\" data-id=\"00c823e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3ba997e elementor-widget elementor-widget-image\" data-id=\"3ba997e\" 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\t<a href=\"https:\/\/fr.wikipedia.org\/wiki\/General_Atomics_MQ-9_Reaper\">\n\t\t\t\t\t\t\t<img decoding=\"async\" width=\"150\" height=\"150\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-14-150x150.png\" class=\"attachment-thumbnail size-thumbnail wp-image-4476\" alt=\"\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-14-150x150.png 150w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-14-300x300.png 300w, https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-14.png 329w\" sizes=\"(max-width: 150px) 100vw, 150px\" \/>\t\t\t\t\t\t\t\t<\/a>\n\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\t<\/div>\n\t\t<\/div>\n\t\t\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-8953aec elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"8953aec\" data-element_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-292a98e\" data-id=\"292a98e\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-6c50679 elementor-widget elementor-widget-text-editor\" data-id=\"6c50679\" 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><b>Etapa 3 <\/b><span style=\"font-weight: 400;\">: Pretratamiento de las im\u00e1genes<\/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-1d6c78b elementor-widget elementor-widget-shortcode\" data-id=\"1d6c78b\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/2d4c94833f521261a9aec7bdcf64d186.js\"><\/script><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-f5e27b4 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"f5e27b4\" data-element_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-44d2aae\" data-id=\"44d2aae\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-485a3e6 elementor-widget elementor-widget-text-editor\" data-id=\"485a3e6\" 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><b>Etapa 4 <\/b><span style=\"font-weight: 400;\">: Predicci\u00f3n del modelo<\/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-0071d3a elementor-widget elementor-widget-shortcode\" data-id=\"0071d3a\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/1cf7fdd70574e0f61ff286dc00ed9c50.js\"><\/script><\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-c620bde elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"c620bde\" data-element_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-5faa4b3\" data-id=\"5faa4b3\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-2b57f28 elementor-widget elementor-widget-image\" data-id=\"2b57f28\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"592\" height=\"60\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/cnn.png\" class=\"attachment-large size-large wp-image-4480\" alt=\"prediction algorithme\" srcset=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/cnn.png 592w, https:\/\/datascientest.com\/es\/files\/2020\/06\/cnn-300x30.png 300w\" sizes=\"(max-width: 592px) 100vw, 592px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\"> Visualizaci\u00f3n de las predicciones del modelo VGG16<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-d88e116 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"d88e116\" data-element_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-adf0df1\" data-id=\"adf0df1\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-4a61045 elementor-widget elementor-widget-text-editor\" data-id=\"4a61045\" 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;\">Tambi\u00e9n se pueden visualizar las 5 etiquetas que VGG16 estime como m\u00e1s probables :<\/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-5e683f3 elementor-widget elementor-widget-shortcode\" data-id=\"5e683f3\" data-element_type=\"widget\" data-widget_type=\"shortcode.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-shortcode\"><script src=\"https:\/\/gist.github.com\/garybedoucha\/5c0c858cc543f0f3f4a9babc4149fa93.js\"><\/script>\n<\/div>\n\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-1715fd1 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"1715fd1\" data-element_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-9c94eb2\" data-id=\"9c94eb2\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-5e621c7 elementor-widget elementor-widget-image\" data-id=\"5e621c7\" 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<figure class=\"wp-caption\">\n\t\t\t\t\t\t\t\t\t\t<img decoding=\"async\" width=\"226\" height=\"168\" src=\"https:\/\/datascientest.com\/es\/files\/2020\/06\/CNN-15-.png\" class=\"attachment-large size-large wp-image-4475\" alt=\"prediction algorithme\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\"> Top 5 des labels jug\u00e9s les plus probables par VGG16<\/figcaption>\n\t\t\t\t\t\t\t\t\t\t<\/figure>\n\t\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-dbc4739 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"dbc4739\" data-element_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-e900fe4\" data-id=\"e900fe4\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-fdc907b elementor-widget elementor-widget-text-editor\" data-id=\"fdc907b\" 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><b>\u2192 Por tanto VGG16 ha conseguido predecir con gran confianza (95,6\u00a0%) no solo que la imagen proporcionada en la entrada era un lobo, sino que ha ido un paso m\u00e1s all\u00e1 y ha precisado su raza, es decir un lobo gris del Este (timber wolf). Impresionante, \u00bfno?<\/b><\/p>\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\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<section class=\"elementor-section elementor-top-section elementor-element elementor-element-9e6adec elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"9e6adec\" data-element_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-9d23601\" data-id=\"9d23601\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-3a63a95 elementor-widget elementor-widget-text-editor\" data-id=\"3a63a95\" 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;\">Por \u00faltimo, el principio de funcionamiento de una CNN es bastante f\u00e1cil de comprender, pero parad\u00f3jicamente, la implementaci\u00f3n de un procedimiento como este para clasificar im\u00e1genes sigue siendo muy complejo debido al n\u00famero considerable de par\u00e1metros que hay que definir: n\u00famero, tama\u00f1o, desplazamiento de filtros, elecci\u00f3n del m\u00e9todo de pooling, elecci\u00f3n del n\u00famero de capas neuronales, n\u00famero de neuronas por capas, etc.\u00a0<\/span><\/p><p><span style=\"font-weight: 400;\">Para paliar ese obst\u00e1culo, Python ofrece, a trav\u00e9s del m\u00f3dulo de Torchvision, la posibilidad de explotar modelos de CNN preentrenados eficaces como VGG16, Resnet101, etc.<\/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<\/div>\n\t\t\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-59df323 elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"59df323\" data-element_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-82c0203\" data-id=\"82c0203\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-259b7f4 elementor-widget elementor-widget-text-editor\" data-id=\"259b7f4\" 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 este art\u00edculo hemos definido el <\/span><b>funcionamiento<\/b><span style=\"font-weight: 400;\"> y la <\/span><b>arquitectura<\/b> <b>de las Convolutional Neural Network<\/b><span style=\"font-weight: 400;\">, centr\u00e1ndonos en su especificidad: la parte convolucional.\u00a0<\/span><\/p><p><span style=\"font-weight: 400;\">Todav\u00eda nos queda por descubrir una etapa de la clasificaci\u00f3n: la retropropagaci\u00f3n del gradiente del error, famoso algoritmo de descenso de gradiente. No te preocupes, pronto podr\u00e1s leer el pr\u00f3ximo cap\u00edtulo sobre el tema en nuestro blog.<\/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<\/div>\n\t\t\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-2285c6c elementor-section-boxed elementor-section-height-default elementor-section-height-default\" data-id=\"2285c6c\" data-element_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-b504c68\" data-id=\"b504c68\" data-element_type=\"column\">\n\t\t\t<div class=\"elementor-widget-wrap elementor-element-populated\">\n\t\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-72cc3d2 elementor-widget elementor-widget-text-editor\" data-id=\"72cc3d2\" 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><strong>Art\u00edculo 1 :\u00a0<\/strong><a href=\"https:\/\/datascientest.com\/es\/deep-learning-definicion\"><span style=\"color: #6e4ceb;\">Introducci\u00f3n al Deep Learning<\/span><br \/><\/a><\/p><p><strong>Art\u00edculo 2 : <span style=\"font-weight: 400;\">Redes neuronales: biol\u00f3gicas vs. Artificiales <\/span><\/strong><\/p><p><strong>Art\u00edculo 3 : <\/strong><span style=\"font-weight: 400;\">Convolutional Neural Network<\/span><\/p><p><strong>Art\u00edculo 4 :<\/strong> <span style=\"font-weight: 400;\">\u00a0Funcionamientos de las redes neuronales<\/span><\/p><p><strong>Art\u00edculo 5 :<\/strong><a href=\"https:\/\/datascientest.com\/es\/que-es-el-transfer-learning\"> <span style=\"color: #6e4ceb;\">Transfer learning<\/span><br \/><\/a><\/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-970742f elementor-widget elementor-widget-html\" data-id=\"970742f\" 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\": \"\u00bfPara qu\u00e9 sirve la convoluci\u00f3n?\",\r\n    \"acceptedAnswer\": {\r\n      \"@type\": \"Answer\",\r\n      \"text\": \"La convoluci\u00f3n es una operaci\u00f3n matem\u00e1tica sencilla que se suele utilizar para el tratamiento y el reconocimiento de im\u00e1genes. En una imagen, su efecto se asimila a un filtrado\"\r\n    }\r\n  }\r\n}\r\n<\/script>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-c28db3f elementor-align-center elementor-widget elementor-widget-button\" data-id=\"c28db3f\" 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-md\" href=\"https:\/\/datascientest.com\/es\/nuestros-cursos-de-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\">Descubrir nuestras formaciones<\/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<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>En esta parte, vamos a centrarnos en uno de los algoritmos m\u00e1s eficaces del Deep Learning, las Convolutional Neural Network o CNN: redes neuronales convolucionales, son modelos de programaci\u00f3n potentes que permiten principalmente el reconocimiento de im\u00e1genes atribuyendo autom\u00e1ticamente a cada imagen proporcionada en la entrada, una etiqueta correspondiente a la clase a la que [&hellip;]<\/p>\n","protected":false},"author":45,"featured_media":23759,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2427],"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>Convolutional Neural Network : definici\u00f3n y funcionamiento<\/title>\n<meta name=\"description\" content=\"la CNN o Convolutional Neural Network : Gracias a nuestros expertos, descubre uno de los algoritmos m\u00e1s potentes del Deep Learning\" \/>\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\/es\/convolutional-neural-network-es\" \/>\n<meta property=\"og:locale\" content=\"es_ES\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Convolutional Neural Network : definici\u00f3n y funcionamiento\" \/>\n<meta property=\"og:description\" content=\"En esta parte, vamos a centrarnos en uno de los algoritmos m\u00e1s eficaces del Deep Learning, las Convolutional Neural Network o CNN: redes neuronales\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\" \/>\n<meta property=\"og:site_name\" content=\"DataScientest\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/DataScientestEN\" \/>\n<meta property=\"article:published_time\" content=\"2021-12-16T16:26:03+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-30T16:38:23+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2020\/06\/illu_convulational_blog-32.png\" \/>\n\t<meta property=\"og:image:width\" content=\"4267\" \/>\n\t<meta property=\"og:image:height\" content=\"2342\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Daniel\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@DataScientest__\" \/>\n<meta name=\"twitter:site\" content=\"@DataScientest__\" \/>\n<meta name=\"twitter:label1\" content=\"Escrito por\" \/>\n\t<meta name=\"twitter:data1\" content=\"Daniel\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tiempo de lectura\" \/>\n\t<meta name=\"twitter:data2\" content=\"12 minutos\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es#article\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\"},\"author\":{\"name\":\"Daniel\",\"@id\":\"https:\/\/datascientest.com\/es\/#\/schema\/person\/5a86e30c5090234fb43dd8baa83c44b0\"},\"headline\":\"Convolutional Neural Network : definici\u00f3n y funcionamiento\",\"datePublished\":\"2021-12-16T16:26:03+00:00\",\"dateModified\":\"2023-10-30T16:38:23+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\"},\"wordCount\":2136,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/datascientest.com\/es\/#organization\"},\"articleSection\":[\"Data Science\"],\"inLanguage\":\"es\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es#respond\"]}],\"copyrightYear\":\"2021\",\"copyrightHolder\":{\"@id\":\"https:\/\/datascientest.com\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\",\"url\":\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\",\"name\":\"Convolutional Neural Network : definici\u00f3n y funcionamiento\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/es\/#website\"},\"datePublished\":\"2021-12-16T16:26:03+00:00\",\"dateModified\":\"2023-10-30T16:38:23+00:00\",\"description\":\"la CNN o Convolutional Neural Network : Gracias a nuestros expertos, descubre uno de los algoritmos m\u00e1s potentes del Deep Learning\",\"inLanguage\":\"es\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/es\/convolutional-neural-network-es\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/es\/#website\",\"url\":\"https:\/\/datascientest.com\/es\/\",\"name\":\"DataScientest\",\"description\":\"L\u00edder en formaci\u00f3n en Ciencia de Datos\",\"publisher\":{\"@id\":\"https:\/\/datascientest.com\/es\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/es\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"es\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/datascientest.com\/es\/#organization\",\"name\":\"DataScientest\",\"url\":\"https:\/\/datascientest.com\/es\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\/\/datascientest.com\/es\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2024\/02\/DataScientest_icon_blue-2.png\",\"contentUrl\":\"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2024\/02\/DataScientest_icon_blue-2.png\",\"width\":301,\"height\":301,\"caption\":\"DataScientest\"},\"image\":{\"@id\":\"https:\/\/datascientest.com\/es\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/DataScientestEN\",\"https:\/\/twitter.com\/DataScientest__\",\"https:\/\/www.linkedin.com\/company\/datascientest-spain\/\",\"https:\/\/www.tiktok.com\/@datascientest\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/es\/#\/schema\/person\/5a86e30c5090234fb43dd8baa83c44b0\",\"name\":\"Daniel\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"es\",\"@id\":\"https:\/\/datascientest.com\/es\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/837acad4f2dc47e85d81cce837261670?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/837acad4f2dc47e85d81cce837261670?s=96&d=mm&r=g\",\"caption\":\"Daniel\"},\"sameAs\":[\"https:\/\/www.linkedin.com\/school\/datascientest\/\",\"user\"],\"jobTitle\":\"ing\u00e9nieur logiciel\",\"worksFor\":\"DataScientest\",\"url\":\"https:\/\/datascientest.com\/es\/author\/team-redac\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Convolutional Neural Network : definici\u00f3n y funcionamiento","description":"la CNN o Convolutional Neural Network : Gracias a nuestros expertos, descubre uno de los algoritmos m\u00e1s potentes del Deep Learning","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\/es\/convolutional-neural-network-es","og_locale":"es_ES","og_type":"article","og_title":"Convolutional Neural Network : definici\u00f3n y funcionamiento","og_description":"En esta parte, vamos a centrarnos en uno de los algoritmos m\u00e1s eficaces del Deep Learning, las Convolutional Neural Network o CNN: redes neuronales","og_url":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es","og_site_name":"DataScientest","article_publisher":"https:\/\/www.facebook.com\/DataScientestEN","article_published_time":"2021-12-16T16:26:03+00:00","article_modified_time":"2023-10-30T16:38:23+00:00","og_image":[{"width":4267,"height":2342,"url":"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2020\/06\/illu_convulational_blog-32.png","type":"image\/png"}],"author":"Daniel","twitter_card":"summary_large_image","twitter_creator":"@DataScientest__","twitter_site":"@DataScientest__","twitter_misc":{"Escrito por":"Daniel","Tiempo de lectura":"12 minutos"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es#article","isPartOf":{"@id":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es"},"author":{"name":"Daniel","@id":"https:\/\/datascientest.com\/es\/#\/schema\/person\/5a86e30c5090234fb43dd8baa83c44b0"},"headline":"Convolutional Neural Network : definici\u00f3n y funcionamiento","datePublished":"2021-12-16T16:26:03+00:00","dateModified":"2023-10-30T16:38:23+00:00","mainEntityOfPage":{"@id":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es"},"wordCount":2136,"commentCount":0,"publisher":{"@id":"https:\/\/datascientest.com\/es\/#organization"},"articleSection":["Data Science"],"inLanguage":"es","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/datascientest.com\/es\/convolutional-neural-network-es#respond"]}],"copyrightYear":"2021","copyrightHolder":{"@id":"https:\/\/datascientest.com\/#organization"}},{"@type":"WebPage","@id":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es","url":"https:\/\/datascientest.com\/es\/convolutional-neural-network-es","name":"Convolutional Neural Network : definici\u00f3n y funcionamiento","isPartOf":{"@id":"https:\/\/datascientest.com\/es\/#website"},"datePublished":"2021-12-16T16:26:03+00:00","dateModified":"2023-10-30T16:38:23+00:00","description":"la CNN o Convolutional Neural Network : Gracias a nuestros expertos, descubre uno de los algoritmos m\u00e1s potentes del Deep Learning","inLanguage":"es","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/es\/convolutional-neural-network-es"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/es\/#website","url":"https:\/\/datascientest.com\/es\/","name":"DataScientest","description":"L\u00edder en formaci\u00f3n en Ciencia de Datos","publisher":{"@id":"https:\/\/datascientest.com\/es\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/es\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"es"},{"@type":"Organization","@id":"https:\/\/datascientest.com\/es\/#organization","name":"DataScientest","url":"https:\/\/datascientest.com\/es\/","logo":{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/datascientest.com\/es\/#\/schema\/logo\/image\/","url":"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2024\/02\/DataScientest_icon_blue-2.png","contentUrl":"https:\/\/datascientest.com\/es\/wp-content\/blogs.dir\/7\/files\/2024\/02\/DataScientest_icon_blue-2.png","width":301,"height":301,"caption":"DataScientest"},"image":{"@id":"https:\/\/datascientest.com\/es\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/DataScientestEN","https:\/\/twitter.com\/DataScientest__","https:\/\/www.linkedin.com\/company\/datascientest-spain\/","https:\/\/www.tiktok.com\/@datascientest"]},{"@type":"Person","@id":"https:\/\/datascientest.com\/es\/#\/schema\/person\/5a86e30c5090234fb43dd8baa83c44b0","name":"Daniel","image":{"@type":"ImageObject","inLanguage":"es","@id":"https:\/\/datascientest.com\/es\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/837acad4f2dc47e85d81cce837261670?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/837acad4f2dc47e85d81cce837261670?s=96&d=mm&r=g","caption":"Daniel"},"sameAs":["https:\/\/www.linkedin.com\/school\/datascientest\/","user"],"jobTitle":"ing\u00e9nieur logiciel","worksFor":"DataScientest","url":"https:\/\/datascientest.com\/es\/author\/team-redac"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/posts\/134409"}],"collection":[{"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/users\/45"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/comments?post=134409"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/posts\/134409\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/media\/23759"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/media?parent=134409"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/categories?post=134409"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/es\/wp-json\/wp\/v2\/tags?post=134409"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}