{"id":38573,"date":"2019-10-31T22:24:37","date_gmt":"2019-10-31T19:24:37","guid":{"rendered":"https:\/\/prohoster.info\/blog\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0\/"},"modified":"2019-10-31T22:24:37","modified_gmt":"2019-10-31T19:24:37","slug":"vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0","status":"publish","type":"post","link":"https:\/\/prohoster.info\/da\/blog\/news\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0","title":{"rendered":"Udgivelsen af maskinl\u00e6ringssystemet TensorFlow 2.0","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p><noindex><a rel=\"nofollow\" href=\"https:\/\/medium.com\/tensorflow\/tensorflow-2-0-is-now-available-57d706c2a9ab\">Pr\u00e6senteret<\/a><\/noindex> betydelig udgivelse af maskinl\u00e6ringsplatformen <noindex><a rel=\"nofollow\" href=\"http:\/\/tensorflow.org\/\">TensorFlow 2.0<\/a><\/noindex>, der tilbyder f\u00e6rdige implementeringer af forskellige algoritmer til dyb l\u00e6ring, en enkel programmeringsgr\u00e6nseflade til modelbygning i Python samt et lavniveau-gr\u00e6nseflade til C++, som muligg\u00f8r styring af opbygning og udf\u00f8relse af beregningsgrafer. Systemkoden er skrevet i C++ og Python og <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/tensorflow\/tensorflow\">distribueres<\/a><\/noindex> under Apache-licensen. <\/p>\n<p>platformen blev oprindeligt udviklet af Google Brain-teamet og anvendes i Google-tjenester til talegenkendelse, ansigtsgenkendelse p\u00e5 billeder, bestemmelse af billedlighed, spamfiltrering i Gmail, <noindex><a rel=\"nofollow\" href=\"https:\/\/www.blog.google\/products\/news\/new-google-news-ai-meets-human-intelligence\/\">nyhedsudvalg<\/a><\/noindex> i Google News og organisering af overs\u00e6ttelser med hensyntagen til betydning. Fordelte maskinl\u00e6ringssystemer kan oprettes p\u00e5 standardhardware takket v\u00e6re den indbyggede underst\u00f8ttelse i TensorFlow af fordeling af beregninger p\u00e5 flere CPU'er eller GPU'er. <\/p>\n<p>TensorFlow tilbyder et bibliotek af f\u00e6rdige algoritmer til numeriske beregninger, implementeret gennem dataflow-grafer (data flow graphs). Noder i disse grafer udf\u00f8rer matematiske operationer eller ind-\/udgangspunkter, mens kanterne i grafen repr\u00e6senterer multidimensionelle datam\u00e6ngder (tensorer), der str\u00f8mmer mellem noderne.<br \/>\nNoder kan v\u00e6re bundet til beregningsenheder og udf\u00f8res asynkront, hvilket muligg\u00f8r samtidig behandling af alle passende tensorer, hvilket g\u00f8r det muligt at organisere samtidig aktivering af noder i et neuralt netv\u00e6rk ligesom samtidig aktivering af neuroner i hjernen.<\/p>\n<p>Der blev lagt s\u00e6rlig v\u00e6gt p\u00e5 forenkling og brugervenlighed i forberedelsen af den nye version. <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/tensorflow\/tensorflow\/releases\/tag\/v2.0.0\">Nogle<\/a><\/noindex> <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/effective_tf2\">nyheder<\/a><\/noindex>:<\/p>\n<ul>\n<li class=\"l\"> For opbygning og tr\u00e6ning af modeller er der tilbudt et nyt h\u00f8j-niveau API <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/beta\/guide\/keras\/overview\">Keras<\/a><\/noindex>, der tilbyder flere gr\u00e6nsefladeindstillinger til modelbygning (Sequential, Functional, Subclassing) med mulighed for <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/eager\">\u00f8jeblikkelig udf\u00f8relse<\/a><\/noindex> (uden forudg\u00e5ende kompilering) og med en enkel fejls\u00f8gningsmekanisme;\n<li class=\"l\"> API er blevet tilf\u00f8jet <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/beta\/guide\/distribute_strategy\">tf.distribute.Strategy<\/a><\/noindex> til organisering af <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/distributed_training\">fordelt l\u00e6ring<\/a><\/noindex> af modeller med minimal \u00e6ndring af eksisterende kode. Udover muligheden for at fordele beregninger p\u00e5 <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/gpu\">flere GPU'er<\/a><\/noindex>, er der eksperimentel underst\u00f8ttelse af fordeling af l\u00e6ringsprocessen p\u00e5 flere uafh\u00e6ngige bearbejdere samt mulighed for at ans\u00e6tte cloud-baserede <noindex><a rel=\"nofollow\" href=\"https:\/\/ru.wikipedia.org\/wiki\/%D0%A2%D0%B5%D0%BD%D0%B7%D0%BE%D1%80%D0%BD%D1%8B%D0%B9_%D0%BF%D1%80%D0%BE%D1%86%D0%B5%D1%81%D1%81%D0%BE%D1%80_Google\">TPU<\/a><\/noindex> (Tensor processing unit);\n<li class=\"l\"> I stedet for en deklarativ model til grafopbygning med udf\u00f8relse gennem tf.Session er der nu mulighed for at skrive almindelige funktioner i Python, som ved hj\u00e6lp af kaldet tf.function kan omdannes til grafer og derefter k\u00f8res eksternt, serialiseres eller optimeres for at forbedre ydeevnen;\n<li class=\"l\"> Overs\u00e6tter tilf\u00f8jet <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/function\">AutoGraph<\/a><\/noindex>, der omdanner str\u00f8mmen af Python-kommandoer til TensorFlow-udtryk, hvilket muligg\u00f8r brugen af Python-kode inden for funktioner dekoreret med tf.function, tf.data, tf.distribute og tf.keras;\n<li class=\"l\"> I SavedModel er formatet til udveksling af modeller blevet ensartet, og der er tilf\u00f8jet st\u00f8tte til at gemme og genskabe modelzustand. Samlede modeller til TensorFlow kan nu bruges i <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/lite\">TensorFlow Lite<\/a><\/noindex> (p\u00e5 mobile enheder), <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/js\">TensorFlow JS<\/a><\/noindex>  (i browseren eller Node.js), <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/tensorflow\/serving\">TensorFlow Serving<\/a><\/noindex> og <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/hub\">TensorFlow Hub<\/a><\/noindex>;\n<li class=\"l\"> API'erne tf.train.Optimizers og tf.keras.Optimizers er blevet ensartet, og i stedet for compute_gradients til beregning af gradienter er der introduceret en ny klasse  <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/api_docs\/python\/tf\/GradientTape\">GradientTape<\/a><\/noindex>;\n<li class=\"l\"> Ydeevnen ved brug af GPU er betydeligt forbedret.<br \/>\nIndl\u00e6ringshastigheden for modeller p\u00e5 systemer med NVIDIA Volta og Turing GPU'er er steget op til tre gange;<\/p>\n<li class=\"l\"> <noindex><a rel=\"nofollow\" href=\"https:\/\/www.tensorflow.org\/guide\/effective_tf2\">Der er foretaget<\/a><\/noindex> en stor oprydning af API'et, mange kald er blevet omd\u00f8bt eller fjernet, og underst\u00f8ttelse af globale variabler i hj\u00e6lpefunktioner er blevet stoppet. I stedet for tf.app, tf.flags, tf.logging er der introduceret et nyt API kaldet absl-py. For fortsat brug af det gamle API er der forberedt et modul ved navn compat.v1.\n<\/ul>\n<p><noindex><a rel=\"nofollow\" name=\"link\"><\/a><\/noindex><\/p>\n<p>Kilde: <a \ncontent=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/www.opennet.ru\/opennews\/art.shtml?num=51595\">opennet.ru<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u041f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d \u0437\u043d\u0430\u0447\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0432\u044b\u043f\u0443\u0441\u043a \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f TensorFlow 2.0, \u043f\u0440\u0435\u0434\u043e\u0441\u0442\u0430\u0432\u043b\u044f\u044e\u0449\u0435\u0439 \u0433\u043e\u0442\u043e\u0432\u044b\u0435 \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u0438 \u0440\u0430\u0437\u043b\u0438\u0447\u043d\u044b\u0445 \u0430\u043b\u0433\u043e\u0440\u0438\u0442\u043c\u043e\u0432 \u0433\u043b\u0443\u0431\u043e\u043a\u043e\u0433\u043e \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f, \u043f\u0440\u043e\u0441\u0442\u043e\u0439 \u043f\u0440\u043e\u0433\u0440\u0430\u043c\u043c\u043d\u044b\u0439 \u0438\u043d\u0442\u0435\u0440\u0444\u0435\u0439\u0441 \u0434\u043b\u044f \u043f\u043e\u0441\u0442\u0440\u043e\u0435\u043d\u0438\u044f \u043c\u043e\u0434\u0435\u043b\u0435\u0439 \u043d\u0430 \u044f\u0437\u044b\u043a\u0435 Python \u0438 \u043d\u0438\u0437\u043a\u043e\u0443\u0440\u043e\u0432\u043d\u0435\u0432\u044b\u0439 \u0438\u043d\u0442\u0435\u0440\u0444\u0435\u0439\u0441 \u0434\u043b\u044f \u044f\u0437\u044b\u043a\u0430 \u0421++, \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u044e\u0449\u0438\u0439 \u0443\u043f\u0440\u0430\u0432\u043b\u044f\u0442\u044c \u043f\u043e\u0441\u0442\u0440\u043e\u0435\u043d\u0438\u0435\u043c \u0438 \u0432\u044b\u043f\u043e\u043b\u043d\u0435\u043d\u0438\u0435\u043c \u0432\u044b\u0447\u0438\u0441\u043b\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0445 \u0433\u0440\u0430\u0444\u043e\u0432. \u041a\u043e\u0434 \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043d\u0430\u043f\u0438\u0441\u0430\u043d \u043d\u0430 \u044f\u0437\u044b\u043a\u0430\u0445 \u0421++ \u0438 Python \u0438 \u0440\u0430\u0441\u043f\u0440\u043e\u0441\u0442\u0440\u0430\u043d\u044f\u0435\u0442\u0441\u044f \u043f\u043e\u0434 \u043b\u0438\u0446\u0435\u043d\u0437\u0438\u0435\u0439 Apache. \u041f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u0430 \u0438\u0437\u043d\u0430\u0447\u0430\u043b\u044c\u043d\u043e \u0440\u0430\u0437\u0440\u0430\u0431\u043e\u0442\u0430\u043d\u0430 \u043a\u043e\u043c\u0430\u043d\u0434\u043e\u0439 [&hellip;]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-38573","post","type-post","status-publish","format-standard","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t<!-- All in One SEO 5.0.3 - aioseo.com -->\n\t<meta name=\"description\" content=\"\u041f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d \u0437\u043d\u0430\u0447\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0432\u044b\u043f\u0443\u0441\u043a \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f\" \/>\n\t<meta name=\"robots\" content=\"max-image-preview:large\" \/>\n\t<meta name=\"author\" content=\"Yuri Gagarin\"\/>\n\t<link rel=\"canonical\" href=\"https:\/\/prohoster.info\/da\/blog\/news\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0\" \/>\n\t\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 5.0.3\" \/>\n\t\t<meta property=\"og:locale\" content=\"da_DK\" \/>\n\t\t<meta property=\"og:site_name\" content=\"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b\" \/>\n\t\t<meta property=\"og:type\" content=\"article\" \/>\n\t\t<meta property=\"og:title\" content=\"\ud83e\udd47\u0412\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f TensorFlow 2.0 | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u041f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d \u0437\u043d\u0430\u0447\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0432\u044b\u043f\u0443\u0441\u043a \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/da\/blog\/news\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0\" \/>\n\t\t<meta property=\"og:image\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:secure_url\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:width\" content=\"350\" \/>\n\t\t<meta property=\"og:image:height\" content=\"350\" \/>\n\t\t<meta property=\"article:published_time\" content=\"2019-10-31T19:24:37+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2019-10-31T19:24:37+00:00\" \/>\n\t\t<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<meta property=\"article:author\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<!-- All in One SEO -->\n\n","aioseo_head_json":{"title":"\ud83e\udd47Udgivelse af maskinl\u00e6ringssystemet TensorFlow 2.0 | ProHoster","description":"En betydelig udgivelse af maskinl\u00e6ringsplatformen er blevet pr\u00e6senteret","canonical_url":"https:\/\/prohoster.info\/da\/blog\/news\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"da_DK","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u0412\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f TensorFlow 2.0 | ProHoster","og:description":"\u041f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d \u0437\u043d\u0430\u0447\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u0439 \u0432\u044b\u043f\u0443\u0441\u043a \u043f\u043b\u0430\u0442\u0444\u043e\u0440\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f","og:url":"https:\/\/prohoster.info\/da\/blog\/news\/vypusk-sistemy-mashinnogo-obucheniya-tensorflow-2-0","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-10-31T19:24:37+00:00","article:modified_time":"2019-10-31T19:24:37+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"38573","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-23 22:35:19","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-03-01 01:07:38","updated":"2026-01-23 22:35:19","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts\/38573","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/comments?post=38573"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/posts\/38573\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/media?parent=38573"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/categories?post=38573"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/da\/wp-json\/wp\/v2\/tags?post=38573"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}