	{"id":63719,"date":"2021-10-25T17:00:23","date_gmt":"2021-10-25T16:00:23","guid":{"rendered":"https:\/\/www.artefact.com\/?post_type=news&#038;p=63719"},"modified":"2024-09-20T17:45:46","modified_gmt":"2024-09-20T16:45:46","slug":"serving-ml-models-at-scale-using-mlflow-on-kubernetes-part-2","status":"publish","type":"blog","link":"https:\/\/www.artefact.com\/de\/blog\/serving-ml-models-at-scale-using-mlflow-on-kubernetes-part-2\/","title":{"rendered":"ML-Modelle in gro\u00dfem Ma\u00dfstab mit Mlflow auf Kubernetes bereitstellen - Teil 2"},"content":{"rendered":"<p><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-1 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling article-author\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-background-color:#ffffff;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-0 fusion_builder_column_1_2 1_2 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:50%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:50%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-title title fusion-title-1 fusion-sep-none fusion-title-text fusion-title-size-two\" style=\"--awb-margin-bottom-small:8px;\"><h2 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:50;line-height:1.2;\">Autor<\/h2><\/div><img decoding=\"async\" src=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27150%27%20height%3D%270%27%20viewBox%3D%270%200%20150%200%27%3E%3Crect%20width%3D%27150%27%20height%3D%270%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/kais-laribi.jpeg\" alt=\"Image\" class=\"lazyload artefact-elegant-image align-left article-author-image\" style=\"width: 150px; border-radius: 54% 46% 77% 23% \/ 74% 40% 60% 26%; overflow: hidden;\" width=\"150\" height=\"auto\" \/><div class=\"fusion-title title fusion-title-2 fusion-sep-none fusion-title-text fusion-title-size-three article-author-name-title\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">Kais Laribi<\/h3><\/div><div class=\"fusion-text fusion-text-1 article-author-description\"><p>Leitender Data-Wissenschaftler bei Artefact<\/p>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-2 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-margin-top:40px;--awb-margin-bottom:40px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-center fusion-flex-justify-content-center fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-1 fusion_builder_column_1_1 1_1 fusion-flex-column fusion-flex-align-self-center fusion-column-inner-bg-wrapper\" style=\"--awb-padding-top:20px;--awb-padding-right:20px;--awb-padding-bottom:20px;--awb-padding-left:20px;--awb-overflow:hidden;--awb-inner-bg-size:cover;--awb-border-color:rgba(10,17,40,0.1);--awb-border-top:1px;--awb-border-right:1px;--awb-border-bottom:1px;--awb-border-left:1px;--awb-border-style:solid;--awb-border-radius:4px 4px 4px 4px;--awb-inner-bg-border-radius:4px 4px 4px 4px;--awb-inner-bg-overflow:hidden;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><span class=\"fusion-column-inner-bg hover-type-none\"><a class=\"fusion-column-anchor\" href=\"https:\/\/medium.com\/artefact-engineering-and-data-science\/serving-ml-models-at-scale-using-mlflow-on-kubernetes-7a85c28d38e\" rel=\"noopener noreferrer\" target=\"_blank\"><span class=\"fusion-column-inner-bg-image\"><\/span><\/a><\/span><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-center fusion-content-layout-row fusion-flex-align-items-center\"><div class=\"fusion-text fusion-text-2\"><p><u>Lesen Sie unseren Artikel \u00fcber<\/u><\/p>\n<\/div><div class=\"fusion-image-element\" style=\"--awb-margin-right:20px;--awb-margin-left:20px;--awb-max-width:150px;--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-1 hover-type-none\"><img decoding=\"async\" width=\"4000\" height=\"992\" title=\"Mittel Blog\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog.png\" alt class=\"lazyload img-responsive wp-image-60582\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%274000%27%20height%3D%27992%27%20viewBox%3D%270%200%204000%20992%27%3E%3Crect%20width%3D%274000%27%20height%3D%27992%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-200x50.png 200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-400x99.png 400w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-600x149.png 600w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-800x198.png 800w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog-1200x298.png 1200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/04\/Medium-Blog.png 4000w\" data-sizes=\"auto\" data-orig-sizes=\"(max-width: 640px) 100vw, 4000px\" \/><\/span><\/div><div class=\"fusion-text fusion-text-3\"><p>.<\/p>\n<\/div><\/div><\/div><\/div><\/div><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-3 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-2 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-text fusion-text-4 description\"><p>Dieser Artikel ist der zweite Teil einer Serie, in der wir den Prozess der Protokollierung von Modellen mit Hilfe von Mlflow durchlaufen, sie als API-Endpunkt bereitstellen und sie schlie\u00dflich entsprechend den Anforderungen unserer Anwendung skalieren. Wir empfehlen Ihnen, unseren vorherigen Artikel zu lesen, in dem wir zeigen, wie Sie eine Tracking-Instanz auf k8s bereitstellen und die praktischen Voraussetzungen (Geheimnisse, Umgebungsvariablen...) \u00fcberpr\u00fcfen, da wir hier weiter darauf aufbauen werden.<br \/>\nIm Folgenden zeigen wir, wie Sie ein bereits in Mlflow registriertes Modell f\u00fcr maschinelles Lernen als API-Endpunkt auf k8s bereitstellen.<\/p>\n<\/div><\/div><\/div><\/div><\/div><article class=\"fusion-fullwidth fullwidth-box fusion-builder-row-4 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-flex-start fusion-flex-justify-content-center fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-3 fusion_builder_column_1_1 1_1 fusion-flex-column\" style=\"--awb-bg-size:cover;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper fusion-column-has-shadow fusion-flex-justify-content-flex-start fusion-content-layout-column\"><div class=\"fusion-title title fusion-title-3 fusion-sep-none fusion-title-text fusion-title-size-one\" style=\"--awb-margin-bottom-small:8px;\"><h1 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:70;line-height:1;\">Teil 2 - Wie kann ich ein Modell als API in Kubernetes bereitstellen?<\/h1><\/div><div class=\"fusion-title title fusion-title-4 fusion-sep-none fusion-title-text fusion-title-size-two\" style=\"--awb-margin-bottom-small:8px;\"><h2 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:50;line-height:1.2;\">Einf\u00fchrung<\/h2><\/div><div class=\"fusion-text fusion-text-5\"><p>Es liegt auf der Hand, dass die Verfolgung und Optimierung der Leistung von Modellen ein wichtiger Bestandteil der Erstellung von ML-Modellen ist. Die n\u00e4chste Herausforderung besteht darin, sie in eine Anwendung oder ein Produkt zu integrieren, um ihre Vorhersagen zu nutzen. Dies nennen wir die Verwendung der Modelle oder Inferenz. Es gibt verschiedene Frameworks und Techniken, die uns dies erm\u00f6glichen. Wir werden uns hier jedoch auf Mlflow konzentrieren und zeigen, wie effizient und einfach das sein kann.<\/p>\n<\/div><div class=\"fusion-title title fusion-title-5 fusion-sep-none fusion-title-text fusion-title-size-two\" style=\"--awb-margin-bottom-small:8px;\"><h2 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:50;line-height:1.2;\">Erstellen und verteilen Sie das Serving Image<\/h2><\/div><div class=\"fusion-text fusion-text-6\"><p>Die verschiedenen hier verwendeten Konfigurationsdateien sind Teil der\u00a0<a class=\"bv ig\" href=\"https:\/\/github.com\/artefactory-global\/mlflow-serving-example\" target=\"_blank\" rel=\"noopener ugc nofollow\">praktisches Repository<\/a>\u00a0Im Grunde genommen m\u00fcssen wir das:<\/p>\n<\/div><div class=\"fusion-title title fusion-title-6 fusion-sep-none fusion-title-text fusion-title-size-three\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">1. Bereiten Sie das Mlflow-Serving-Docker-Image vor und pushen Sie es in die Container-Registry auf GCP.<\/h3><\/div><div class=\"fusion-text fusion-text-7\"><pre class=\"hp hq hr hs ht kg gv be\"><span id=\"09f2\" class=\"ej kh ii dm ki b kj kk kl s km\" data-selectable-paragraph=\"\">cd mlflow-serving-beispiel<\/span><span id=\"eda1\" class=\"ej kh ii dm ki b kj kn ko kp kq kr kl s km\" data-selectable-paragraph=\"\">docker build --tag $\/mlflow_serving:v1 <br \/>--file docker_mlflow_serving .<\/span><span id=\"855d\" class=\"ej kh ii dm ki b kj kn ko kp kq kr kl s km\" data-selectable-paragraph=\"\">docker push $\/mlflow_serving:v1<\/span><\/pre>\n<\/div><div class=\"fusion-title title fusion-title-7 fusion-sep-none fusion-title-text fusion-title-size-three\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">2. Bereiten Sie die Kubernetes-Bereitstellungsdatei vor<\/h3><\/div><div class=\"fusion-text fusion-text-8\"><p>indem Sie den Container-Abschnitt \u00e4ndern und ihn dem\u00a0<strong>Docker-Image<\/strong>\u00a0die zuvor an GCR gesendet wurden,\u00a0<strong>der Modellpfad<\/strong>\u00a0und\u00a0<strong>den Serving Port.<\/strong><\/p>\n<\/div><div class=\"fusion-text fusion-text-9\"><div class=\"code\">\n<table class=\"highlight tab-size js-file-line-container\" data-tab-size=\"8\" data-paste-markdown-skip=\"\">\n<tbody>\n<tr>\n<td id=\"file-mlflow_serving-yaml-LC1\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">apiVersion<\/span>: <span class=\"pl-s\">apps\/v1<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L2\" class=\"blob-num js-line-number\" data-line-number=\"2\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC2\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">freundlich<\/span>: <span class=\"pl-s\">Einsatz<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L3\" class=\"blob-num js-line-number\" data-line-number=\"3\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC3\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">metadata<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L4\" class=\"blob-num js-line-number\" data-line-number=\"4\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC4\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">mlflow-serving<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L5\" class=\"blob-num js-line-number\" data-line-number=\"5\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC5\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Etiketten<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L6\" class=\"blob-num js-line-number\" data-line-number=\"6\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC6\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">app<\/span>: <span class=\"pl-s\">serve-ML-model-mlflow<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L7\" class=\"blob-num js-line-number\" data-line-number=\"7\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC7\" class=\"blob-code blob-code-inner js-file-line\">\u00a0<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L8\" class=\"blob-num js-line-number\" data-line-number=\"8\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC8\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">spec<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L9\" class=\"blob-num js-line-number\" data-line-number=\"9\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC9\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Repliken<\/span>: <span class=\"pl-c1\">1<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L10\" class=\"blob-num js-line-number\" data-line-number=\"10\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC10\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Selektor<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L11\" class=\"blob-num js-line-number\" data-line-number=\"11\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC11\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">matchLabels<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L12\" class=\"blob-num js-line-number\" data-line-number=\"12\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC12\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">app<\/span>: <span class=\"pl-s\">mlflow-serving<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L13\" class=\"blob-num js-line-number\" data-line-number=\"13\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC13\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Vorlage<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L14\" class=\"blob-num js-line-number\" data-line-number=\"14\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC14\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">metadata<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L15\" class=\"blob-num js-line-number\" data-line-number=\"15\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC15\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Etiketten<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L16\" class=\"blob-num js-line-number\" data-line-number=\"16\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC16\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">app<\/span>: <span class=\"pl-s\">mlflow-serving<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L17\" class=\"blob-num js-line-number\" data-line-number=\"17\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC17\" class=\"blob-code blob-code-inner js-file-line\">\u00a0<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L18\" class=\"blob-num js-line-number\" data-line-number=\"18\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC18\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">spec<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L19\" class=\"blob-num js-line-number\" data-line-number=\"19\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC19\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Container<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L20\" class=\"blob-num js-line-number\" data-line-number=\"20\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC20\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">mlflow-serving<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L21\" class=\"blob-num js-line-number\" data-line-number=\"21\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC21\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Bild<\/span>: <span class=\"pl-s\">GCR_REPO\/mlflow_serving:v1 <\/span><span class=\"pl-c\">#auschen Sie hier<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L22\" class=\"blob-num js-line-number\" data-line-number=\"22\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC22\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">env<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L23\" class=\"blob-num js-line-number\" data-line-number=\"23\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC23\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">MODELL_URI<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L24\" class=\"blob-num js-line-number\" data-line-number=\"24\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC24\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Wert<\/span>: <span class=\"pl-s\"><span class=\"pl-pds\">\u201c<\/span>gs:\/\/..\/artifacts\/..\/..\/artifacts\/..<span class=\"pl-pds\">\u201c<\/span><\/span> <span class=\"pl-c\">#auschen Sie hier<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L25\" class=\"blob-num js-line-number\" data-line-number=\"25\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC25\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">SERVING_PORT<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L26\" class=\"blob-num js-line-number\" data-line-number=\"26\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC26\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Wert<\/span>: <span class=\"pl-s\"><span class=\"pl-pds\">\u201c<\/span>8082<span class=\"pl-pds\">\u201c<\/span><\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L27\" class=\"blob-num js-line-number\" data-line-number=\"27\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC27\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">GOOGLE_APPLICATION_CREDENTIALS<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L28\" class=\"blob-num js-line-number\" data-line-number=\"28\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC28\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Wert<\/span>: <span class=\"pl-s\"><span class=\"pl-pds\">\u201c<\/span>\/etc\/geheimnisse\/keyfile.json<span class=\"pl-pds\">\u201c<\/span><\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L29\" class=\"blob-num js-line-number\" data-line-number=\"29\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC29\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">volumeMounts<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L30\" class=\"blob-num js-line-number\" data-line-number=\"30\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC30\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">gcsfs-creds<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L31\" class=\"blob-num js-line-number\" data-line-number=\"31\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC31\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">mountPfad<\/span>: <span class=\"pl-s\"><span class=\"pl-pds\">\u201c<\/span>\/etc\/secrets<span class=\"pl-pds\">\u201c<\/span><\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L32\" class=\"blob-num js-line-number\" data-line-number=\"32\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC32\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">readOnly<\/span>: <span class=\"pl-c1\">wahr<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L33\" class=\"blob-num js-line-number\" data-line-number=\"33\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC33\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Ressourcen<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L34\" class=\"blob-num js-line-number\" data-line-number=\"34\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC34\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Anfragen<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L35\" class=\"blob-num js-line-number\" data-line-number=\"35\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC35\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">cpu<\/span>: <span class=\"pl-s\"><span class=\"pl-pds\">\u201c<\/span>1000m<span class=\"pl-pds\">\u201c<\/span><\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L36\" class=\"blob-num js-line-number\" data-line-number=\"36\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC36\" class=\"blob-code blob-code-inner js-file-line\">\u00a0<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L37\" class=\"blob-num js-line-number\" data-line-number=\"37\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC37\" class=\"blob-code blob-code-inner js-file-line\">\u00a0<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L38\" class=\"blob-num js-line-number\" data-line-number=\"38\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC38\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">B\u00e4nde<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L39\" class=\"blob-num js-line-number\" data-line-number=\"39\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC39\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Name<\/span>: <span class=\"pl-s\">gcsfs-creds<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L40\" class=\"blob-num js-line-number\" data-line-number=\"40\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC40\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Geheimnis<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L41\" class=\"blob-num js-line-number\" data-line-number=\"41\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC41\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">secretName<\/span>: <span class=\"pl-s\">gcsfs-creds<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L42\" class=\"blob-num js-line-number\" data-line-number=\"42\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC42\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Artikel<\/span>:<\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L43\" class=\"blob-num js-line-number\" data-line-number=\"43\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC43\" class=\"blob-code blob-code-inner js-file-line\">\u2013 <span class=\"pl-ent\">Schl\u00fcssel<\/span>: <span class=\"pl-s\">keyfile.json<\/span><\/td>\n<\/tr>\n<tr>\n<td id=\"file-mlflow_serving-yaml-L44\" class=\"blob-num js-line-number\" data-line-number=\"44\">\u00a0<\/td>\n<td id=\"file-mlflow_serving-yaml-LC44\" class=\"blob-code blob-code-inner js-file-line\"><span class=\"pl-ent\">Pfad<\/span>: <span class=\"pl-s\">keyfile.json<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<\/div><div class=\"fusion-title title fusion-title-8 fusion-sep-none fusion-title-text fusion-title-size-three\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">3. Einsatzbefehle ausf\u00fchren<\/h3><\/div><div class=\"fusion-text fusion-text-10\"><p>kubectl create -f Bereitstellungen\/mlflow-serving\/mlflow_serving.yaml<\/p>\n<\/div><div class=\"fusion-title title fusion-title-9 fusion-sep-none fusion-title-text fusion-title-size-three\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">4. Die Bereitstellung f\u00fcr den externen Zugriff freigeben<\/h3><\/div><div class=\"fusion-text fusion-text-11\"><p>Mit dem folgenden Befehl wird eine neue Ressource erstellt, um den externen Datenverkehr auf unsere API umzuleiten.<\/p>\n<pre class=\"hp hq hr hs ht kg gv be\"><span id=\"d7d7\" class=\"ej kh ii dm ki b kj kk kl s km\" data-selectable-paragraph=\"\">kubectl expose deployment mlflow-serving --port 8082 --type=\"LoadBalancer\"<\/span><\/pre>\n<\/div><div class=\"fusion-title title fusion-title-10 fusion-sep-none fusion-title-text fusion-title-size-three\" style=\"--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">5. \u00dcberpr\u00fcfen Sie die Bereitstellung und fragen Sie den Endpunkt ab<\/h3><\/div><div class=\"fusion-text fusion-text-12\"><p>Wenn die Bereitstellung erfolgreich war, sollte mlflow-serving UP sein und ein Pod sollte verf\u00fcgbar sein. Sie k\u00f6nnen dies \u00fcberpr\u00fcfen, indem Sie Folgendes eingeben\u00a0<em><strong class=\"jh ka\">kubectl get pods<\/strong><\/em><\/p>\n<\/div><div class=\"fusion-image-element\" style=\"--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-2 hover-type-none\"><img decoding=\"async\" width=\"559\" height=\"60\" title=\"artikel-kais-teil2\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2.png\" alt class=\"lazyload img-responsive wp-image-63932\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27559%27%20height%3D%2760%27%20viewBox%3D%270%200%20559%2060%27%3E%3Crect%20width%3D%27559%27%20height%3D%2760%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-200x21.png 200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-400x43.png 400w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2.png 559w\" data-sizes=\"auto\" data-orig-sizes=\"(max-width: 640px) 100vw, 559px\" \/><\/span><\/div><div class=\"fusion-text fusion-text-13\"><p>Der letzte Schritt besteht darin, die externe IP-Adresse zu \u00fcberpr\u00fcfen, die dem Load Balancer zugewiesen wurde, der den Verkehr zu unserem API-Container umleitet.\u00a0<strong><em>kubectl get Dienste<\/em><\/strong>\u00a0und testen Sie die Antwort auf ein paar Abfragen.<\/p>\n<\/div><div class=\"fusion-image-element\" style=\"--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-3 hover-type-none\"><img decoding=\"async\" width=\"700\" height=\"100\" title=\"artikel-kais-teil2-2\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2.png\" alt class=\"lazyload img-responsive wp-image-63933\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27700%27%20height%3D%27100%27%20viewBox%3D%270%200%20700%20100%27%3E%3Crect%20width%3D%27700%27%20height%3D%27100%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2-200x29.png 200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2-400x57.png 400w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2-600x86.png 600w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-2.png 700w\" data-sizes=\"auto\" data-orig-sizes=\"(max-width: 640px) 100vw, 700px\" \/><\/span><\/div><div class=\"fusion-text fusion-text-14\"><p>Ein Beispielcode zur Durchf\u00fchrung dieser Abfragen finden Sie im Folgenden\u00a0<a class=\"bv ig\" href=\"https:\/\/github.com\/artefactory-global\/mlflow-serving-example\/blob\/main\/notebooks\/2.%20query%20mlflow%20serve%20API.ipynb\" target=\"_blank\" rel=\"noopener ugc nofollow\">Notizbuch<\/a>\u00a0in dem wir einige Zeilen von data laden, Merkmale ausw\u00e4hlen, sie in das JSON-Format konvertieren und sie in einer Post-Anfrage an die API senden.<br \/>\nWenn Sie nun die Schritte aus dem vorherigen und dem aktuellen Artikel kombinieren, w\u00fcrde unsere endg\u00fcltige Architektur wie folgt aussehen:<\/p>\n<\/div><div class=\"fusion-image-element\" style=\"--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-4 hover-type-none\"><img decoding=\"async\" width=\"700\" height=\"383\" title=\"artikel-kais-teil2-3\" src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3.png\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3.png\" alt class=\"lazyload img-responsive wp-image-63934\" srcset=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%27700%27%20height%3D%27383%27%20viewBox%3D%270%200%20700%20383%27%3E%3Crect%20width%3D%27700%27%20height%3D%27383%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-srcset=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3-200x109.png 200w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3-400x219.png 400w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3-600x328.png 600w, https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/10\/article-kais-part2-3.png 700w\" data-sizes=\"auto\" data-orig-sizes=\"(max-width: 640px) 100vw, 700px\" \/><\/span><\/div><div class=\"fusion-title title fusion-title-11 fusion-sep-none fusion-title-text fusion-title-size-two\" style=\"--awb-margin-bottom-small:8px;\"><h2 class=\"fusion-title-heading title-heading-left fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:50;line-height:1.2;\">Fazit<\/h2><\/div><div class=\"fusion-text fusion-text-15\"><p>In diesem Artikel haben wir gezeigt, dass wir Modelle f\u00fcr maschinelles Lernen mithilfe des Mlflow-Serving-Moduls ganz einfach als API-Endpunkt bereitstellen k\u00f6nnen.<\/p>\n<p>Wie Sie vielleicht bemerken, wurde in unserem aktuellen Einsatz nur ein Pod erstellt, um das Modell zu bedienen. Das funktioniert zwar gut f\u00fcr kleine Anwendungen, bei denen wir nicht mit mehreren parallelen Abfragen rechnen, k\u00f6nnte aber bei anderen Anwendungen schnell an seine Grenzen sto\u00dfen, da ein einzelner Pod nur \u00fcber begrenzte Ressourcen verf\u00fcgt. Au\u00dferdem kann die Anwendung auf diese Weise nicht mehr als die Rechenleistung eines Knotens nutzen. Im folgenden und letzten Artikel dieser Serie werden wir uns mit dem Problem der Skalierbarkeit befassen. Wir werden zun\u00e4chst die Engp\u00e4sse aufzeigen und versuchen, sie zu l\u00f6sen, um eine skalierbare Anwendung zu erhalten, die die Leistung unseres Kubernetes-Clusters nutzt.<\/p>\n<\/div><\/div><\/div><\/div><\/article><div class=\"fusion-fullwidth fullwidth-box fusion-builder-row-5 fusion-flex-container nonhundred-percent-fullwidth non-hundred-percent-height-scrolling\" style=\"--awb-border-radius-top-left:0px;--awb-border-radius-top-right:0px;--awb-border-radius-bottom-right:0px;--awb-border-radius-bottom-left:0px;--awb-margin-top:40px;--awb-margin-bottom:40px;--awb-flex-wrap:wrap;\" ><div class=\"fusion-builder-row fusion-row fusion-flex-align-items-center fusion-flex-justify-content-center fusion-flex-content-wrap\" style=\"max-width:calc( 1440px + 20px );margin-left: calc(-20px \/ 2 );margin-right: calc(-20px \/ 2 );\"><div class=\"fusion-layout-column fusion_builder_column fusion-builder-column-4 fusion_builder_column_1_1 1_1 fusion-flex-column fusion-flex-align-self-center\" style=\"--awb-padding-top:40px;--awb-padding-right:40px;--awb-padding-bottom:40px;--awb-padding-left:40px;--awb-overflow:hidden;--awb-bg-position:left center;--awb-bg-size:cover;--awb-border-color:rgba(10,17,40,0.1);--awb-border-style:solid;--awb-border-radius:4px 4px 4px 4px;--awb-width-large:100%;--awb-margin-top-large:0px;--awb-spacing-right-large:10px;--awb-margin-bottom-large:0px;--awb-spacing-left-large:10px;--awb-width-medium:100%;--awb-order-medium:0;--awb-spacing-right-medium:10px;--awb-spacing-left-medium:10px;--awb-width-small:100%;--awb-order-small:0;--awb-spacing-right-small:10px;--awb-spacing-left-small:10px;\"><div class=\"fusion-column-wrapper lazyload fusion-column-has-shadow fusion-flex-justify-content-center fusion-content-layout-column fusion-column-has-bg-image\" data-bg-url=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/03\/background.jpg\" data-bg=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/03\/background.jpg\"><div class=\"fusion-image-element\" style=\"text-align:center;--awb-margin-right:20px;--awb-margin-left:20px;--awb-max-width:150px;--awb-caption-title-font-family:var(--h2_typography-font-family);--awb-caption-title-font-weight:var(--h2_typography-font-weight);--awb-caption-title-font-style:var(--h2_typography-font-style);--awb-caption-title-size:var(--h2_typography-font-size);--awb-caption-title-transform:var(--h2_typography-text-transform);--awb-caption-title-line-height:var(--h2_typography-line-height);--awb-caption-title-letter-spacing:var(--h2_typography-letter-spacing);\"><span class=\"fusion-imageframe imageframe-none imageframe-5 hover-type-none\"><img decoding=\"async\" width=\"72\" height=\"41\" title=\"mittel\" src=\"data:image\/svg+xml,%3Csvg%20xmlns%3D%27http%3A%2F%2Fwww.w3.org%2F2000%2Fsvg%27%20width%3D%2772%27%20height%3D%2741%27%20viewBox%3D%270%200%2072%2041%27%3E%3Crect%20width%3D%2772%27%20height%3D%2741%27%20fill-opacity%3D%220%22%2F%3E%3C%2Fsvg%3E\" data-orig-src=\"https:\/\/www.artefact.com\/\/wp-content\/uploads\/2021\/03\/medium.png\" alt class=\"lazyload img-responsive wp-image-60927\"\/><\/span><\/div><div class=\"fusion-title title fusion-title-12 fusion-sep-none fusion-title-center fusion-title-text fusion-title-size-three\" style=\"--awb-margin-top:20px;--awb-margin-bottom:0px;--awb-margin-bottom-small:8px;\"><h3 class=\"fusion-title-heading title-heading-center fusion-responsive-typography-calculated\" style=\"margin:0;--fontSize:20;line-height:1.2;\">Mittel Blog von Artefact.<\/h3><\/div><div class=\"fusion-text fusion-text-16\" style=\"--awb-content-alignment:center;\"><p>Dieser Artikel wurde urspr\u00fcnglich ver\u00f6ffentlicht auf <strong>Medium.com<\/strong>.<br \/>\nFolgen Sie uns auf unserem Medium Blog !<\/p>\n<\/div><div style=\"text-align:center;\"><a class=\"fusion-button button-flat button-medium button-default fusion-button-default button-1 fusion-button-default-span fusion-button-default-type\" target=\"_blank\" rel=\"noopener noreferrer\" title=\"https:\/\/medium.com\/artefact-engineering-and-data-science\/serving-ml-models-at-scale-using-mlflow-on-kubernetes-7a85c28d38e\" aria-label=\"https:\/\/medium.com\/artefact-engineering-and-data-science\/serving-ml-models-at-scale-using-mlflow-on-kubernetes-7a85c28d38e\" href=\"https:\/\/medium.com\/artefact-engineering-and-data-science\/serving-ml-models-at-scale-using-mlflow-on-kubernetes-bf27258775e7\"><span class=\"fusion-button-text awb-button__text awb-button__text--default\">Unseren Artikel lesen<\/span><\/a><\/div><\/div><\/div><\/div><\/div><\/p>","protected":false},"excerpt":{"rendered":"<p>25. Oktober 2021<br \/>\nDieser Artikel ist der zweite Teil einer Serie, in der wir den Prozess der Protokollierung von Modellen mit Hilfe von Mlflow durchlaufen, sie als API-Endpunkt bereitstellen und sie schlie\u00dflich entsprechend den Anforderungen unserer Anwendung skalieren. Wir empfehlen Ihnen, unseren vorherigen Artikel zu lesen, in dem wir zeigen, wie Sie eine Tracking-Instanz auf k8s bereitstellen und die praktischen Voraussetzungen (Geheimnisse, Umgebungsvariablen...) \u00fcberpr\u00fcfen, da wir hier weiter darauf aufbauen werden.<br \/>\nIm Folgenden zeigen wir, wie Sie ein bereits in Mlflow registriertes Modell f\u00fcr maschinelles Lernen als API-Endpunkt auf k8s bereitstellen.<\/p>","protected":false},"featured_media":65014,"parent":0,"template":"","meta":{"_acf_changed":false,"ep_exclude_from_search":false},"blog-category":[21939],"blog-language":[2991],"class_list":["post-63719","blog","type-blog","status-publish","has-post-thumbnail","hentry","blog-category-medium","blog-language-en"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog\/63719","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog"}],"about":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/types\/blog"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/media\/65014"}],"wp:attachment":[{"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/media?parent=63719"}],"wp:term":[{"taxonomy":"blog-category","embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog-category?post=63719"},{"taxonomy":"blog-language","embeddable":true,"href":"https:\/\/www.artefact.com\/de\/wp-json\/wp\/v2\/blog-language?post=63719"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}