{"id":106018,"date":"2022-12-13T09:36:49","date_gmt":"2022-12-13T07:36:49","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30"},"modified":"2022-12-13T09:36:49","modified_gmt":"2022-12-13T07:36:49","slug":"vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30","status":"publish","type":"post","link":"https:\/\/prohoster.info\/hu\/blog\/news\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30","title":{"rendered":"Megjelent az OpenNMT-tf 2.30 g\u00e9pi ford\u00edt\u00f3 rendszer","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p>Megjelent az OpenNMT-tf 2.30.0 verzi\u00f3ja, amely a g\u00e9pi ford\u00edt\u00e1sokr\u00f3l sz\u00f3l (Open Neural Machine Translation), \u00e9s g\u00e9pi tanul\u00e1si m\u00f3dszereket alkalmaz. Az OpenNMT-tf projekt keretein bel\u00fcl fejlesztett modulok k\u00f3dja Python nyelven k\u00e9sz\u00fclt, a TensorFlow k\u00f6nyvt\u00e1rat haszn\u00e1lja \u00e9s MIT licenc alatt terjesztik.     <\/p>\n<p>P\u00e1rhuzamosan fejl\u0151dik a PyTorch k\u00f6nyvt\u00e1rra \u00e9p\u00fcl\u0151 OpenNMT verzi\u00f3, amely k\u00fcl\u00f6nb\u00f6z\u0151 k\u00e9pess\u00e9gekkel rendelkezik. Ezen t\u00falmen\u0151en, a PyTorch alap\u00fa OpenNMT-t egyszer\u0171bben haszn\u00e1lhat\u00f3nak \u00e9s multimod\u00e1lisnak tekintik, m\u00edg a TensorFlow alap\u00fa verzi\u00f3t modul\u00e1risnak \u00e9s stabilnak eml\u00edtik, amely lehet\u0151v\u00e9 teszi a GPU teljes\u00edtm\u00e9ny\u00e9nek kihaszn\u00e1l\u00e1s\u00e1t a neur\u00e1lis h\u00e1l\u00f3 tanul\u00e1si folyamat\u00e1nak felgyors\u00edt\u00e1s\u00e1ra. A term\u00e9k egyszer\u0171s\u00edtett terjeszt\u00e9se \u00e9rdek\u00e9ben a projekt egy \u00f6n\u00e1ll\u00f3 C++ nyelven \u00edrt ford\u00edt\u00f3t is fejleszt &#8212; a CTranslate2-t, amely el\u0151re betan\u00edtott modelleket haszn\u00e1l, an\u00e9lk\u00fcl, hogy tov\u00e1bbi f\u00fcgg\u0151s\u00e9gekre lenne sz\u00fcks\u00e9ge.         <\/p>\n<p>A modellek angol, n\u00e9met \u00e9s katal\u00e1n nyelvekhez k\u00e9sz\u00fcltek, a t\u00f6bbi nyelvhez \u00f6n\u00e1ll\u00f3an hozhat\u00f3 l\u00e9tre modell az OPUS projekt adathalmaza alapj\u00e1n (a rendszer betan\u00edt\u00e1s\u00e1hoz k\u00e9t f\u00e1jlt kell \u00e1tadni &#8212; az egyiket az eredeti nyelven \u00edrt mondatokkal, a m\u00e1sikat pedig az ezeknek a mondatoknak a c\u00e9lnyelven t\u00f6rt\u00e9n\u0151 min\u0151s\u00e9gi ford\u00edt\u00e1s\u00e1val).    <\/p>\n<p>A projekt a g\u00e9pi ford\u00edt\u00f3 rendszerek fejleszt\u00e9s\u00e9re specializ\u00e1l\u00f3dott SYSTRAN c\u00e9g \u00e9s a Harvard egyetem kutat\u00f3csoportja k\u00f6zrem\u0171k\u00f6d\u00e9s\u00e9vel fejl\u0151dik, amely emberi nyelvi modelleket fejleszt a g\u00e9pi tanul\u00e1si rendszerekhez. A felhaszn\u00e1l\u00f3i fel\u00fclet maxim\u00e1lisan egyszer\u0171s\u00edtett, \u00e9s csak a sz\u00f6veget tartalmaz\u00f3 bemeneti f\u00e1jl \u00e9s a ford\u00edt\u00e1si eredm\u00e9ny ment\u00e9s\u00e9re szolg\u00e1l\u00f3 f\u00e1jl megad\u00e1s\u00e1t ig\u00e9nyli. A b\u0151v\u00edt\u0151 rendszer lehet\u0151s\u00e9get biztos\u00edt az OpenNMT alap\u00fa kieg\u00e9sz\u00edt\u0151 funkci\u00f3k megval\u00f3s\u00edt\u00e1s\u00e1ra, p\u00e9ld\u00e1ul automatikus hivatkoz\u00e1sra, sz\u00f6vegek oszt\u00e1lyoz\u00e1s\u00e1ra \u00e9s feliratok gener\u00e1l\u00e1s\u00e1ra.     <\/p>\n<p>Az \u00faj verzi\u00f3ban:  <\/p>\n<ul>\n<li class=\"l\"> B\u0151v\u00fclt a TensorFlow 2.11 k\u00f6nyvt\u00e1r t\u00e1mogat\u00e1sa, de az \u00faj Keras optimaliz\u00e1l\u00f3k m\u00e9g nem t\u00e1mogatottak (a tf.keras.optimizers.legacy m\u00f3d haszn\u00e1lata sz\u00fcks\u00e9ges).\n<li class=\"l\"> Meg\u00e9rkezett az \u00faj CTranslate2 3.x motor\u00e1gazat t\u00e1mogat\u00e1sa, amely a 'transformer' architekt\u00far\u00e1val rendelkez\u0151 modellek hat\u00e9kony v\u00e9grehajt\u00e1s\u00e1ra szolg\u00e1l.\n<li class=\"l\"> Bevezet\u00e9sre ker\u00fclt a pad_to_bucket_boundary modellek k\u00e9pz\u00e9si param\u00e9ter, amely lehet\u0151v\u00e9 teszi a blokkok m\u00e9ret\u00e9nek kit\u00f6lt\u00e9s\u00e9t, hogy azok a length_bucket_width \u00e9rt\u00e9kek t\u00f6bbsz\u00f6r\u00f6sei legyenek.\n<li class=\"l\"> Integr\u00e1ltuk a SacreBLEU projekt chrf \u00e9s chrf++ metrik\u00e1it, amelyek a g\u00e9pi ford\u00edt\u00e1st az emberi referenciaford\u00edt\u00e1ssal hasonl\u00edtj\u00e1k \u00f6ssze.\n<li class=\"l\"> Elt\u00e1vol\u00edtottuk a ctranslate2_spec modell attrib\u00fatumot, amely m\u00e1r nem haszn\u00e1latos a CTranslate2-ben.  <\/ul>\n<p>Forr\u00e1s: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/www.opennet.ru\/opennews\/art.shtml?num=58316\">opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d \u0432\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30.0 (Open Neural Machine Translation), \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0449\u0435\u0439 \u043c\u0435\u0442\u043e\u0434\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f. \u041a\u043e\u0434 \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u0435\u043c\u044b\u0445 \u043f\u0440\u043e\u0435\u043a\u0442\u043e\u043c OpenNMT-tf \u043c\u043e\u0434\u0443\u043b\u0435\u0439 \u043d\u0430\u043f\u0438\u0441\u0430\u043d \u043d\u0430 \u044f\u0437\u044b\u043a\u0435 Python, \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u0435\u0442 \u0431\u0438\u0431\u043b\u0438\u043e\u0442\u0435\u043a\u0443 TensorFlow \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 MIT. \u041f\u0430\u0440\u0430\u043b\u043b\u0435\u043b\u044c\u043d\u043e \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u0435\u0442\u0441\u044f \u0432\u0430\u0440\u0438\u0430\u043d\u0442 OpenNMT \u043d\u0430 \u0431\u0430\u0437\u0435 \u0431\u0438\u0431\u043b\u0438\u043e\u0442\u0435\u043a\u0438 PyTorch, \u043a\u043e\u0442\u043e\u0440\u044b\u0439 \u043e\u0442\u043b\u0438\u0447\u0430\u0435\u0442\u0441\u044f \u043d\u0430 \u0443\u0440\u043e\u0432\u043d\u0435 \u043f\u043e\u0434\u0434\u0435\u0440\u0436\u0438\u0432\u0430\u0435\u043c\u044b\u0445 \u0432\u043e\u0437\u043c\u043e\u0436\u043d\u043e\u0441\u0442\u0435\u0439. \u041a\u0440\u043e\u043c\u0435 \u0442\u043e\u0433\u043e, OpenNMT \u043d\u0430 \u0431\u0430\u0437\u0435 PyTorch \u043f\u0440\u0435\u043f\u043e\u0434\u043d\u043e\u0441\u0438\u0442\u0441\u044f \u043a\u0430\u043a \u0431\u043e\u043b\u0435\u0435 [&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-106018","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=\"\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d \u0432\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30.0 (Open Neural Machine Translation), \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0449\u0435\u0439 \u043c\u0435\u0442\u043e\u0434\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\/hu\/blog\/news\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30\" \/>\n\t\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 5.0.3\" \/>\n\t\t<meta property=\"og:locale\" content=\"hu_HU\" \/>\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 \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30 | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d \u0432\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30.0 (Open Neural Machine Translation), \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0449\u0435\u0439 \u043c\u0435\u0442\u043e\u0434\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\/hu\/blog\/news\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30\" \/>\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=\"2022-12-13T07:36:49+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2022-12-13T07:36:49+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\udd47Az OpenNMT-tf 2.30 g\u00e9pi ford\u00edt\u00f3 rendszer kiad\u00e1sa | ProHoster","description":"Megjelent az OpenNMT-tf 2.30.0 g\u00e9pi ford\u00edt\u00f3 rendszer, amely g\u00e9pi tanul\u00e1si m\u00f3dszereket alkalmaz.","canonical_url":"https:\/\/prohoster.info\/hu\/blog\/news\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"hu_HU","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 \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30 | ProHoster","og:description":"\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d \u0432\u044b\u043f\u0443\u0441\u043a \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0430 OpenNMT-tf 2.30.0 (Open Neural Machine Translation), \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0449\u0435\u0439 \u043c\u0435\u0442\u043e\u0434\u044b \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f.","og:url":"https:\/\/prohoster.info\/hu\/blog\/news\/vypusk-sistemy-mashinnogo-perevoda-opennmt-tf-2-30","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":"2022-12-13T07:36:49+00:00","article:modified_time":"2022-12-13T07:36:49+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":[],"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts\/106018","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/comments?post=106018"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/posts\/106018\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/media?parent=106018"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/categories?post=106018"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/hu\/wp-json\/wp\/v2\/tags?post=106018"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}