{"id":170302,"date":"2023-09-09T09:26:21","date_gmt":"2023-09-09T08:26:21","guid":{"rendered":"https:\/\/datascientest.com\/en\/?p=170302"},"modified":"2023-10-30T19:54:08","modified_gmt":"2023-10-30T18:54:08","slug":"unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","status":"publish","type":"post","link":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","title":{"rendered":"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel"},"content":{"rendered":"\t\t<div data-elementor-type=\"wp-post\" data-elementor-id=\"170302\" class=\"elementor elementor-170302\" data-elementor-post-type=\"post\">\n\t\t\t\t\t\t\t<div class=\"elementor-element elementor-element-19c859b e-flex e-con-boxed e-con\" data-id=\"19c859b\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-7d4ba69 e-con-full e-flex e-con\" data-id=\"7d4ba69\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;full&quot;}\">\n\t\t\t\t<div class=\"elementor-element elementor-element-1fa91b5 elementor-widget elementor-widget-heading\" data-id=\"1fa91b5\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-heading-title{padding:0;margin:0;line-height:1}.elementor-widget-heading .elementor-heading-title[class*=elementor-size-]>a{color:inherit;font-size:inherit;line-height:inherit}.elementor-widget-heading .elementor-heading-title.elementor-size-small{font-size:15px}.elementor-widget-heading .elementor-heading-title.elementor-size-medium{font-size:19px}.elementor-widget-heading .elementor-heading-title.elementor-size-large{font-size:29px}.elementor-widget-heading .elementor-heading-title.elementor-size-xl{font-size:39px}.elementor-widget-heading .elementor-heading-title.elementor-size-xxl{font-size:59px}<\/style><h2 class=\"elementor-heading-title elementor-size-default\">Daniel is back, the iconic face of our training courses who guides our learners through to their diploma. Today he's going to introduce you to a model frequently used in Computer Vision: VGG.<\/h2>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-15e0c3a e-flex e-con-boxed e-con\" data-id=\"15e0c3a\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-26080276 e-con-full e-flex e-con\" data-id=\"26080276\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;full&quot;}\">\n\t\t\t\t<div class=\"elementor-element elementor-element-28384279 elementor-widget elementor-widget-text-editor\" data-id=\"28384279\" data-element_type=\"widget\" data-widget_type=\"text-editor.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<style>\/*! elementor - v3.15.0 - 02-08-2023 *\/\n.elementor-widget-text-editor.elementor-drop-cap-view-stacked .elementor-drop-cap{background-color:#69727d;color:#fff}.elementor-widget-text-editor.elementor-drop-cap-view-framed .elementor-drop-cap{color:#69727d;border:3px solid;background-color:transparent}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap{margin-top:8px}.elementor-widget-text-editor:not(.elementor-drop-cap-view-default) .elementor-drop-cap-letter{width:1em;height:1em}.elementor-widget-text-editor .elementor-drop-cap{float:left;text-align:center;line-height:1;font-size:50px}.elementor-widget-text-editor .elementor-drop-cap-letter{display:inline-block}<\/style>\t\t\t\t<p>We&#8217;ve previously had the opportunity, some time ago, to delve into the concept of <strong>Transfer Learning.<\/strong> If you missed our article on the topic, I encourage you to revisit it here. What&#8217;s important to keep in mind is that Transfer Learning corresponds to the ability to leverage existing knowledge, developed to address specific issues, to solve a new problem.<\/p><p>\u00a0<\/p><p>\ud83d\udca1Also discover:<\/p><table dir=\"ltr\" border=\"1\" cellspacing=\"0\" cellpadding=\"0\"><colgroup><col width=\"250\" \/><\/colgroup><tbody><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Image Processing&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/image-processing-fundamental-principles-and-practical-uses\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/image-processing-fundamental-principles-and-practical-uses\" target=\"_blank\" rel=\"noopener\">Image Processing<\/a><\/td><\/tr><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Deep Learning - All you need to know&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/all-about-deep-learning\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/all-about-deep-learning\" target=\"_blank\" rel=\"noopener\">Deep Learning &#8211; All you need to know<\/a><\/td><\/tr><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Mushroom Recognition&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/mushroom-recognition\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/mushroom-recognition\" target=\"_blank\" rel=\"noopener\">Mushroom Recognition<\/a><\/td><\/tr><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Tensor Flow - Google's ML&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/tensor-flow-all-about-googles-machine-learning-framework\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/tensor-flow-all-about-googles-machine-learning-framework\" target=\"_blank\" rel=\"noopener\">Tensor Flow &#8211; Google&#8217;s ML<\/a><\/td><\/tr><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Dive into ML&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/unlock-your-future-dive-into-machine-learning-engineer-training\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/unlock-your-future-dive-into-machine-learning-engineer-training\" target=\"_blank\" rel=\"noopener\">Dive into ML<\/a><\/td><\/tr><tr><td data-sheets-value=\"{&quot;1&quot;:2,&quot;2&quot;:&quot;Data Poisoning&quot;}\" data-sheets-hyperlink=\"https:\/\/datascientest.com\/en\/data-poisoning-a-threat-to-machine-learning-models\"><a class=\"in-cell-link\" href=\"https:\/\/datascientest.com\/en\/data-poisoning-a-threat-to-machine-learning-models\" target=\"_blank\" rel=\"noopener\">Data Poisoning<\/a><\/td><\/tr><\/tbody><\/table>\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-16358410 elementor-widget elementor-widget-heading\" data-id=\"16358410\" 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\">A little history<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-5c5f0dca elementor-widget elementor-widget-text-editor\" data-id=\"5c5f0dca\" 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>VGG<\/strong> is a convolutional <a href=\"https:\/\/datascientest.com\/en\/deep-neural-network-what-is-it-and-how-is-it-working\">neural network<\/a> developed by K. Simonyan and A. Zisserman from the University of Oxford. It gained prominence by winning the ILSVRC (ImageNet Large Scale Visual Recognition Challenge) competition in 2014. The model achieved an impressive accuracy of 92.7% on ImageNet, which is one of the highest scores achieved at the time.<\/p><p>It marked an advancement over previous models by introducing smaller convolutional kernels (3&#215;3) in its convolutional layers, a departure from the norm. The training process for this model spanned several weeks and made use of cutting-edge graphics cards for computation.<\/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-63c74a34 elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"63c74a34\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/en\/data-courses\">\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\">Discover our courses<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-3113b025 elementor-widget elementor-widget-heading\" data-id=\"3113b025\" 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\">ImageNet<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-a5208a7 elementor-widget elementor-widget-text-editor\" data-id=\"a5208a7\" 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>ImageNet<\/strong> is an<a href=\"https:\/\/datascientest.com\/en\/database-what-is-it\"> immense database<\/a> containing over 14 million labeled images distributed across more than 1000 classes, as of 2014. In 2007, a researcher named Fei-Fei Li initiated the idea of creating such a dataset. While model design is undoubtedly crucial for achieving good performance, having high-quality data is equally essential for quality learning. The data was collected and labeled by humans from the web, making it open source and not owned by any particular company.<\/p><p>Since 2010, the<strong> ImageNet Large Scale Visual Recognition Challenge<\/strong> has been held annually. Its purpose is to challenge image processing models. The competition is conducted on a subset of ImageNet, consisting of 1.2 million training images, 50,000 validation images, and 150,000 images for testing the model&#8217;s performance.<\/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-74005d07 elementor-widget elementor-widget-heading\" data-id=\"74005d07\" 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\">The architecture <\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-700bd638 elementor-widget elementor-widget-text-editor\" data-id=\"700bd638\" 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; color: #000000;\">In practice, there are <a href=\"https:\/\/datascientest.com\/en\/algorithm-what-is-it\">two available algorithms:<\/a> VGG16 and VGG19. In this article, we will focus on the architecture of the first one, VGG16. While both architectures are very similar and follow the same principles, VGG19 has a larger number of convolutional layers.<\/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-2f01b7aa elementor-widget elementor-widget-image\" data-id=\"2f01b7aa\" 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<img decoding=\"async\" width=\"800\" height=\"439\" src=\"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG_Plan-de-travail-1.png\" class=\"attachment-large size-large wp-image-170306\" alt=\"\" srcset=\"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG_Plan-de-travail-1.png 1024w, https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG_Plan-de-travail-1-300x165.png 300w, https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG_Plan-de-travail-1-768x422.png 768w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Architecture VGG16 algorithm<\/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<div class=\"elementor-element elementor-element-5436e624 elementor-widget elementor-widget-image\" data-id=\"5436e624\" 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=\"800\" height=\"439\" src=\"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-02.png\" class=\"attachment-large size-large wp-image-170307\" alt=\"\" srcset=\"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-02.png 1024w, https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-02-300x165.png 300w, https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-02-768x422.png 768w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\">Structure Algoritme 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<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-bec2bc8 e-flex e-con-boxed e-con\" data-id=\"bec2bc8\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-ed00f6d e-con-full e-flex e-con\" data-id=\"ed00f6d\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;full&quot;}\">\n\t\t\t\t<div class=\"elementor-element elementor-element-631520a elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"631520a\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/en\/data-scientist-course\">\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\">Data Scientist Course<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-4345019 elementor-widget elementor-widget-text-editor\" data-id=\"4345019\" 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>The model requires only one specific preprocessing step, which involves subtracting the mean RGB value calculated across the entire training set from each pixel.<\/p><p>During the<strong> training of the model<\/strong>, the input to the first convolutional layer is an RGB image of size 224 x 224. For all convolutional layers, the convolution kernel is of size 3&#215;3 &#8211; the smallest dimension to capture concepts like top, bottom, left\/right, and center.<\/p><p>This was a <strong>unique feature<\/strong> of the model at the time of its publication. Up until VGG16, many models leaned towards larger convolutional kernels (e.g., 11&#215;11 or 5&#215;5). It&#8217;s important to note that these convolutional layers aim to filter the image, retaining only discriminative information such as unusual geometric shapes.<\/p><p>These convolutional layers are accompanied by Max-Pooling layers, each of size 2&#215;2, to reduce filter sizes during training.<\/p><p>Following the <strong>convolution and pooling layers,<\/strong> we have 3 Fully-Connected layers. The first two consist of 4096 neurons each, and the final layer consists of 1000 neurons with a softmax activation function to determine the image&#8217;s class.<\/p><p>As you&#8217;ve observed, the architecture is clear and straightforward to understand, which is one of the strengths of this model.<\/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-142ea4f elementor-widget elementor-widget-heading\" data-id=\"142ea4f\" 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\">ImageNet results<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-1e146c4 elementor-widget elementor-widget-image\" data-id=\"1e146c4\" 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=\"602\" height=\"259\" src=\"https:\/\/datascientest.com\/en\/files\/2021\/04\/image2-1.png\" class=\"attachment-large size-large wp-image-64955\" alt=\"\" srcset=\"https:\/\/datascientest.com\/en\/files\/2021\/04\/image2-1.png 602w, https:\/\/datascientest.com\/en\/files\/2021\/04\/image2-1-300x129.png 300w\" sizes=\"(max-width: 602px) 100vw, 602px\" \/>\t\t\t\t\t\t\t\t\t\t\t<figcaption class=\"widget-image-caption wp-caption-text\"><\/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<div class=\"elementor-element elementor-element-8cf4ff5 elementor-widget elementor-widget-text-editor\" data-id=\"8cf4ff5\" 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;\">The figure above compares the results of various models from 2014 and previous years. We can observe that<strong> VGG<\/strong> achieves the best results on both the validation and test datasets. It&#8217;s worth noting that the model performs significantly better than in the 2012 and 2013 sessions.<\/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-276347a elementor-widget elementor-widget-heading\" data-id=\"276347a\" 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\">And what about transfer learning?<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-6d1e445 elementor-widget elementor-widget-text-editor\" data-id=\"6d1e445\" 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>As we&#8217;ve mentioned before, training a model like VGG can be very time-consuming, especially if you have limited resources. Furthermore, since it has been trained on <strong>ImageNet<\/strong>, it can be advantageous to retrieve the weights of the pre-trained model, particularly the filters in the convolutional layers learned from ImageNet. In practice, this is a common approach: we retrieve the weights from the convolutional layers, and we only need to train the 3 additional layers we add. The principle remains the same: using the knowledge gained from<strong> ImageNet to address a similar problem.<\/strong><\/p><p>It&#8217;s notably possible to easily obtain the pre-trained model and apply the specific preprocessing requested by the model.<\/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-36318be elementor-widget elementor-widget-heading\" data-id=\"36318be\" 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\">A little practice<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-ff8bc9d elementor-widget elementor-widget-text-editor\" data-id=\"ff8bc9d\" 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>In practice, there are two available algorithms: VGG16 and VGG19.<\/p><p>Thanks to the Keras library in TensorFlow, it&#8217;s straightforward to retrieve the <strong>pre-trained model, which by default, is trained on ImageNet.<\/strong><\/p><p>First, we need to apply the same specific preprocessing as was done during the model&#8217;s training. Additionally, we incorporate data augmentation on the training data to prevent overfitting. It&#8217;s also crucial to ensure that the input images are RGB images with a size of 224&#215;224.<\/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-d4865f2 elementor-widget elementor-widget-shortcode\" data-id=\"d4865f2\" 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\/Thibault-venet\/ec44f996ba1ff384448564aeb6960735.js\"><\/script><\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-f66526b e-flex e-con-boxed e-con\" data-id=\"f66526b\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-2a45fed e-con-full e-flex e-con\" data-id=\"2a45fed\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;full&quot;}\">\n\t\t\t\t<div class=\"elementor-element elementor-element-c9ddb55 elementor-widget elementor-widget-text-editor\" data-id=\"c9ddb55\" 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;\">We can then recover the optimized weights from the convolution layers and train the 3 Dense layers that we add and compile:<\/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-0e676ab elementor-widget elementor-widget-shortcode\" data-id=\"0e676ab\" 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\/Thibault-venet\/2daa77bdb6f4ab8aacc1606908ce14fb.js\"><\/script><\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t<div class=\"elementor-element elementor-element-4997537 e-flex e-con-boxed e-con\" data-id=\"4997537\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;boxed&quot;}\">\n\t\t\t\t\t<div class=\"e-con-inner\">\n\t\t<div class=\"elementor-element elementor-element-8253e6b e-con-full e-flex e-con\" data-id=\"8253e6b\" data-element_type=\"container\" data-settings=\"{&quot;content_width&quot;:&quot;full&quot;}\">\n\t\t\t\t<div class=\"elementor-element elementor-element-5ada717 elementor-widget elementor-widget-text-editor\" data-id=\"5ada717\" 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;\">After training, if the results are not satisfactory, it&#8217;s possible to unfreeze the last convolutional layers and retrain them in an attempt to achieve better performance. However, this additional step can be time-consuming and resource-intensive, depending on the resources at your disposal. It&#8217;s a trade-off between potentially improving performance and the computational cost of retraining these layers.<\/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-acb71cc elementor-widget elementor-widget-heading\" data-id=\"acb71cc\" data-element_type=\"widget\" data-widget_type=\"heading.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t<h3 class=\"elementor-heading-title elementor-size-default\">Conclusion<\/h3>\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<div class=\"elementor-element elementor-element-2078a8b elementor-widget elementor-widget-text-editor\" data-id=\"2078a8b\" 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>VGG is a well-known <a href=\"https:\/\/datascientest.com\/en\/u-net-computer-visions-neural-network\">algorithm in Computer Vision<\/a>, often used for transfer learning to avoid the need for retraining and solve similar problems for which VGG has already been trained. There are many other algorithms of the same type as VGG, such as ResNet or even Xception, available in the Keras library.<\/p><p>If you&#8217;re interested in gaining expertise in Deep Learning and Computer Vision, consider joining us in a Bootcamp or continuous training program.<\/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-ce2433a elementor-align-center button-principale elementor-widget elementor-widget-button\" data-id=\"ce2433a\" data-element_type=\"widget\" data-widget_type=\"button.default\">\n\t\t\t\t<div class=\"elementor-widget-container\">\n\t\t\t\t\t<div class=\"elementor-button-wrapper\">\n\t\t\t<a class=\"elementor-button elementor-button-link elementor-size-sm elementor-animation-grow\" href=\"https:\/\/datascientest.com\/en\/data-scientist-course\">\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\">Become a Data Scientist<\/span>\n\t\t<\/span>\n\t\t\t\t\t<\/a>\n\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"<p>Daniel is back, the iconic face of our training courses who guides our learners through to their diploma. Today he&#8217;s going to introduce you to a model frequently used in Computer Vision: VGG. We&#8217;ve previously had the opportunity, some time ago, to delve into the concept of Transfer Learning. If you missed our article on [&hellip;]<\/p>\n","protected":false},"author":67,"featured_media":170303,"comment_status":"open","ping_status":"open","sticky":false,"template":"elementor_theme","format":"standard","meta":{"content-type":"","inline_featured_image":false},"categories":[2425],"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>Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel<\/title>\n<meta name=\"description\" content=\"Explore the VGG model in-depth with Daniel as he breaks down its components and functionalities. Discover what makes the VGG\" \/>\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\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\" \/>\n<meta property=\"og:locale\" content=\"en_GB\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel\" \/>\n<meta property=\"og:description\" content=\"Daniel is back, the iconic face of our training courses who guides our learners through to their diploma. Today he&#039;s going to introduce you to a model\" \/>\n<meta property=\"og:url\" content=\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\" \/>\n<meta property=\"og:site_name\" content=\"Data Science Courses | DataScientest\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/DataScientestEN\" \/>\n<meta property=\"article:published_time\" content=\"2023-09-09T08:26:21+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2023-10-30T18:54:08+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-03.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"562\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Melanie\" \/>\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=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Melanie\" \/>\n\t<meta name=\"twitter:label2\" content=\"Estimated reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"5 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel#article\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\"},\"author\":{\"name\":\"Melanie\",\"@id\":\"https:\/\/datascientest.com\/en\/#\/schema\/person\/19d5f7ade9407eb69af3bf6d1ae1731d\"},\"headline\":\"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel\",\"datePublished\":\"2023-09-09T08:26:21+00:00\",\"dateModified\":\"2023-10-30T18:54:08+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\"},\"wordCount\":984,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/datascientest.com\/en\/#organization\"},\"articleSection\":[\"Data Science\"],\"inLanguage\":\"en-GB\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel#respond\"]}],\"copyrightYear\":\"2023\",\"copyrightHolder\":{\"@id\":\"https:\/\/datascientest.com\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\",\"url\":\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\",\"name\":\"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel\",\"isPartOf\":{\"@id\":\"https:\/\/datascientest.com\/en\/#website\"},\"datePublished\":\"2023-09-09T08:26:21+00:00\",\"dateModified\":\"2023-10-30T18:54:08+00:00\",\"description\":\"Explore the VGG model in-depth with Daniel as he breaks down its components and functionalities. Discover what makes the VGG\",\"inLanguage\":\"en-GB\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel\"]}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/datascientest.com\/en\/#website\",\"url\":\"https:\/\/datascientest.com\/en\/\",\"name\":\"Data Science Courses | DataScientest\",\"description\":\"Leader in Data Science Courses\",\"publisher\":{\"@id\":\"https:\/\/datascientest.com\/en\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/datascientest.com\/en\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-GB\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/datascientest.com\/en\/#organization\",\"name\":\"DataScientest\",\"url\":\"https:\/\/datascientest.com\/en\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-GB\",\"@id\":\"https:\/\/datascientest.com\/en\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/datascientest.com\/en\/wp-content\/blogs.dir\/9\/files\/2024\/02\/DataScientest_icon_blue-2.png\",\"contentUrl\":\"https:\/\/datascientest.com\/en\/wp-content\/blogs.dir\/9\/files\/2024\/02\/DataScientest_icon_blue-2.png\",\"width\":301,\"height\":301,\"caption\":\"DataScientest\"},\"image\":{\"@id\":\"https:\/\/datascientest.com\/en\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/DataScientestEN\",\"https:\/\/twitter.com\/DataScientest__\",\"https:\/\/www.linkedin.com\/school\/datascientest\/mycompany\/\",\"https:\/\/www.tiktok.com\/@datascientest\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/datascientest.com\/en\/#\/schema\/person\/19d5f7ade9407eb69af3bf6d1ae1731d\",\"name\":\"Melanie\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-GB\",\"@id\":\"https:\/\/datascientest.com\/en\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/8d3d660a0698de06f13ecf4823519da1?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/8d3d660a0698de06f13ecf4823519da1?s=96&d=mm&r=g\",\"caption\":\"Melanie\"},\"url\":\"https:\/\/datascientest.com\/en\/author\/melaniekrause\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel","description":"Explore the VGG model in-depth with Daniel as he breaks down its components and functionalities. Discover what makes the VGG","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\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","og_locale":"en_GB","og_type":"article","og_title":"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel","og_description":"Daniel is back, the iconic face of our training courses who guides our learners through to their diploma. Today he's going to introduce you to a model","og_url":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","og_site_name":"Data Science Courses | DataScientest","article_publisher":"https:\/\/www.facebook.com\/DataScientestEN","article_published_time":"2023-09-09T08:26:21+00:00","article_modified_time":"2023-10-30T18:54:08+00:00","og_image":[{"width":1024,"height":562,"url":"https:\/\/datascientest.com\/en\/files\/2023\/09\/illu_VGG-03.png","type":"image\/png"}],"author":"Melanie","twitter_card":"summary_large_image","twitter_creator":"@DataScientest__","twitter_site":"@DataScientest__","twitter_misc":{"Written by":"Melanie","Estimated reading time":"5 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel#article","isPartOf":{"@id":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel"},"author":{"name":"Melanie","@id":"https:\/\/datascientest.com\/en\/#\/schema\/person\/19d5f7ade9407eb69af3bf6d1ae1731d"},"headline":"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel","datePublished":"2023-09-09T08:26:21+00:00","dateModified":"2023-10-30T18:54:08+00:00","mainEntityOfPage":{"@id":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel"},"wordCount":984,"commentCount":0,"publisher":{"@id":"https:\/\/datascientest.com\/en\/#organization"},"articleSection":["Data Science"],"inLanguage":"en-GB","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel#respond"]}],"copyrightYear":"2023","copyrightHolder":{"@id":"https:\/\/datascientest.com\/#organization"}},{"@type":"WebPage","@id":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","url":"https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel","name":"Unveiling the Secrets of the VGG Model: A Deep Dive with Daniel","isPartOf":{"@id":"https:\/\/datascientest.com\/en\/#website"},"datePublished":"2023-09-09T08:26:21+00:00","dateModified":"2023-10-30T18:54:08+00:00","description":"Explore the VGG model in-depth with Daniel as he breaks down its components and functionalities. Discover what makes the VGG","inLanguage":"en-GB","potentialAction":[{"@type":"ReadAction","target":["https:\/\/datascientest.com\/en\/unveiling-the-secrets-of-the-vgg-model-a-deep-dive-with-daniel"]}]},{"@type":"WebSite","@id":"https:\/\/datascientest.com\/en\/#website","url":"https:\/\/datascientest.com\/en\/","name":"Data Science Courses | DataScientest","description":"Leader in Data Science Courses","publisher":{"@id":"https:\/\/datascientest.com\/en\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/datascientest.com\/en\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-GB"},{"@type":"Organization","@id":"https:\/\/datascientest.com\/en\/#organization","name":"DataScientest","url":"https:\/\/datascientest.com\/en\/","logo":{"@type":"ImageObject","inLanguage":"en-GB","@id":"https:\/\/datascientest.com\/en\/#\/schema\/logo\/image\/","url":"https:\/\/datascientest.com\/en\/wp-content\/blogs.dir\/9\/files\/2024\/02\/DataScientest_icon_blue-2.png","contentUrl":"https:\/\/datascientest.com\/en\/wp-content\/blogs.dir\/9\/files\/2024\/02\/DataScientest_icon_blue-2.png","width":301,"height":301,"caption":"DataScientest"},"image":{"@id":"https:\/\/datascientest.com\/en\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/DataScientestEN","https:\/\/twitter.com\/DataScientest__","https:\/\/www.linkedin.com\/school\/datascientest\/mycompany\/","https:\/\/www.tiktok.com\/@datascientest"]},{"@type":"Person","@id":"https:\/\/datascientest.com\/en\/#\/schema\/person\/19d5f7ade9407eb69af3bf6d1ae1731d","name":"Melanie","image":{"@type":"ImageObject","inLanguage":"en-GB","@id":"https:\/\/datascientest.com\/en\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/8d3d660a0698de06f13ecf4823519da1?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/8d3d660a0698de06f13ecf4823519da1?s=96&d=mm&r=g","caption":"Melanie"},"url":"https:\/\/datascientest.com\/en\/author\/melaniekrause"}]}},"_links":{"self":[{"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/posts\/170302"}],"collection":[{"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/users\/67"}],"replies":[{"embeddable":true,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/comments?post=170302"}],"version-history":[{"count":0,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/posts\/170302\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/media\/170303"}],"wp:attachment":[{"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/media?parent=170302"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/categories?post=170302"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/datascientest.com\/en\/wp-json\/wp\/v2\/tags?post=170302"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}