{"id":3993,"date":"2022-09-16T14:36:57","date_gmt":"2022-09-16T22:36:57","guid":{"rendered":"https:\/\/live-cometml.pantheonsite.io\/?p=3993"},"modified":"2025-04-29T14:08:40","modified_gmt":"2025-04-29T14:08:40","slug":"introducing-comets-new-image-panel","status":"publish","type":"post","link":"https:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/","title":{"rendered":"Introducing Comet’s New Image Panel"},"content":{"rendered":"\n

The Need for Visualization <\/b><\/h2>\n\n\n\n

For computer vision tasks such as segmentation, reconstruction, or object detection, it is imperative to visualize the model\u2019s performance on your data. While tracking metrics such as loss and mAP are important , those values do not tell the whole story. <\/span><\/p>\n\n\n\n

Comet has long provided its users the ability to log all relevant images to the graphics tab. The tab makes it easy to visualize model performance on your image data for a single experiment. Now, with our new feature, one can compare their model visualizations across multiple experiments!<\/span><\/p>\n\n\n\n

Visualizing Autoencoders<\/h2>\n\n\n\n

Auto-encoders are powerful feature extractors and can be applied to anomaly detection use-cases and state space dimensionality reduction in Reinforcement Learning. When building an Auto-encoder model, there are a  myriad of architectures, learning rates, loss calculations and other hyper-parameters to choose from. <\/span><\/p>\n\n\n\n

I see that in my new experiment, my reconstruction loss decreased when compared to my past two experiments. But to what extent did that make my reconstructed image similar to the original image? <\/span><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

With the new Image Panel, engineers can visualize how their choices affect the autoencoder model\u2019s ability to reconstruct the original side-by-side. In the image panel, search for your unique image names across your experiments, and create a grid spanning your different experiments in seconds. <\/span><\/p>\n\n\n\n

The Image Panel not only enables users to visualize model predictions across experiments, but also to visualize their predictions throughout the training process. Using the stepper  within the Image Panel allows one to see how multiple models are improving over time.<\/span><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

 <\/p>\n\n\n\n

Where to find the Image Panel?<\/b><\/h4>\n\n\n\n

The Image Panel is a Built-in Panel! They can be found easily by clicking on the Add+<\/strong> button and then selecting the New Panel<\/strong> option.<\/span><\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

  <\/span><\/p>\n\n\n\n

Want to see the new Image Panel? <\/span><\/h4>\n\n\n\n

You can check it out for yourself in the public project.<\/p>\n\n\n\n

\n
View the public Comet project<\/a><\/div>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

The Need for Visualization  For computer vision tasks such as segmentation, reconstruction, or object detection, it is imperative to visualize the model\u2019s performance on your data. While tracking metrics such as loss and mAP are important , those values do not tell the whole story.  Comet has long provided its users the ability to log […]<\/p>\n","protected":false},"author":21,"featured_media":4015,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"customer_name":"","customer_description":"","customer_industry":"","customer_technologies":"","customer_logo":"","_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[8,9],"tags":[],"coauthors":[134],"class_list":["post-3993","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-comet-community-hub","category-product"],"yoast_head":"\nIntroducing Comet's New Image Panel - Comet<\/title>\n<meta name=\"description\" content=\"This feature makes it easy to visualize model performance on your image data. Now it's even easier to compare across multiple experiments!\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Introducing Comet's New Image Panel\" \/>\n<meta property=\"og:description\" content=\"This feature makes it easy to visualize model performance on your image data. Now it's even easier to compare across multiple experiments!\" \/>\n<meta property=\"og:url\" content=\"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/\" \/>\n<meta property=\"og:site_name\" content=\"Comet\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/cometdotml\" \/>\n<meta property=\"article:published_time\" content=\"2022-09-16T22:36:57+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-04-29T14:08:40+00:00\" \/>\n<meta property=\"og:image\" content=\"http:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif\" \/>\n\t<meta property=\"og:image:width\" content=\"897\" \/>\n\t<meta property=\"og:image:height\" content=\"756\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/gif\" \/>\n<meta name=\"author\" content=\"Siddharth Mehta\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@Cometml\" \/>\n<meta name=\"twitter:site\" content=\"@Cometml\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Siddharth Mehta\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Introducing Comet's New Image Panel - Comet","description":"This feature makes it easy to visualize model performance on your image data. Now it's even easier to compare across multiple experiments!","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":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/","og_locale":"en_US","og_type":"article","og_title":"Introducing Comet's New Image Panel","og_description":"This feature makes it easy to visualize model performance on your image data. Now it's even easier to compare across multiple experiments!","og_url":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/","og_site_name":"Comet","article_publisher":"https:\/\/www.facebook.com\/cometdotml","article_published_time":"2022-09-16T22:36:57+00:00","article_modified_time":"2025-04-29T14:08:40+00:00","og_image":[{"width":897,"height":756,"url":"http:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","type":"image\/gif"}],"author":"Siddharth Mehta","twitter_card":"summary_large_image","twitter_creator":"@Cometml","twitter_site":"@Cometml","twitter_misc":{"Written by":"Siddharth Mehta","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#article","isPartOf":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/"},"author":{"name":"Siddharth Mehta","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/652eb7d782d18f295922f50ea3b9e54c"},"headline":"Introducing Comet’s New Image Panel","datePublished":"2022-09-16T22:36:57+00:00","dateModified":"2025-04-29T14:08:40+00:00","mainEntityOfPage":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/"},"wordCount":326,"publisher":{"@id":"https:\/\/www.comet.com\/site\/#organization"},"image":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#primaryimage"},"thumbnailUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","articleSection":["Comet Community Hub","Product"],"inLanguage":"en-US"},{"@type":"WebPage","@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/","url":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/","name":"Introducing Comet's New Image Panel - Comet","isPartOf":{"@id":"https:\/\/www.comet.com\/site\/#website"},"primaryImageOfPage":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#primaryimage"},"image":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#primaryimage"},"thumbnailUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","datePublished":"2022-09-16T22:36:57+00:00","dateModified":"2025-04-29T14:08:40+00:00","description":"This feature makes it easy to visualize model performance on your image data. Now it's even easier to compare across multiple experiments!","breadcrumb":{"@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#primaryimage","url":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","contentUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","width":897,"height":756,"caption":"Gif of Comet image panel"},{"@type":"BreadcrumbList","@id":"http:\/\/www.comet.com\/site\/blog\/introducing-comets-new-image-panel\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.comet.com\/site\/"},{"@type":"ListItem","position":2,"name":"Introducing Comet’s New Image Panel"}]},{"@type":"WebSite","@id":"https:\/\/www.comet.com\/site\/#website","url":"https:\/\/www.comet.com\/site\/","name":"Comet","description":"Build Better Models Faster","publisher":{"@id":"https:\/\/www.comet.com\/site\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.comet.com\/site\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.comet.com\/site\/#organization","name":"Comet ML, Inc.","alternateName":"Comet","url":"https:\/\/www.comet.com\/site\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.comet.com\/site\/#\/schema\/logo\/image\/","url":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2025\/01\/logo_comet_square.png","contentUrl":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2025\/01\/logo_comet_square.png","width":310,"height":310,"caption":"Comet ML, Inc."},"image":{"@id":"https:\/\/www.comet.com\/site\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/cometdotml","https:\/\/x.com\/Cometml","https:\/\/www.youtube.com\/channel\/UCmN63HKvfXSCS-UwVwmK8Hw"]},{"@type":"Person","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/652eb7d782d18f295922f50ea3b9e54c","name":"Siddharth Mehta","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.comet.com\/site\/#\/schema\/person\/image\/940c7280faea9e1b8b086c2ed7ec01db","url":"https:\/\/secure.gravatar.com\/avatar\/27a672e997fa7a66796e4be0503e0efeec6bd34daae185bb6de163227a5a0739?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/27a672e997fa7a66796e4be0503e0efeec6bd34daae185bb6de163227a5a0739?s=96&d=mm&r=g","caption":"Siddharth Mehta"},"description":"ML Growth Engineer @ Comet. Interested in Computer Vision, Robotics, and Reinforcement Learning","sameAs":["https:\/\/www.comet.com\/"],"url":"https:\/\/www.comet.com\/site\/blog\/author\/siddharthmcomet-com\/"}]}},"jetpack_featured_media_url":"https:\/\/www.comet.com\/site\/wp-content\/uploads\/2022\/09\/NewFeature_Blog_img_panel-1.gif","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/3993","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/users\/21"}],"replies":[{"embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/comments?post=3993"}],"version-history":[{"count":3,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/3993\/revisions"}],"predecessor-version":[{"id":15820,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/posts\/3993\/revisions\/15820"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/media\/4015"}],"wp:attachment":[{"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/media?parent=3993"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/categories?post=3993"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/tags?post=3993"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.comet.com\/site\/wp-json\/wp\/v2\/coauthors?post=3993"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}