Compania Jina a lansat sub licen\u021ba Apache 2.0 un model de \u00eenv\u0103\u021bare automat\u0103 pentru reprezentarea vectorial\u0103 a textului \u2014 jina-embeddings-v2. Modelul permite transformarea unui text arbitrar, care include p\u00e2n\u0103 la 8192 de caractere, \u00eentr-o mic\u0103 succesiune de numere reale, form\u00e2nd un vector asociat cu textul ini\u021bial \u0219i reproduc\u00e2ndu-i semantica. Jina Embedding a devenit primul model de \u00eenv\u0103\u021bare automat\u0103 deschis, av\u00e2nd caracteristici care nu cedeaz\u0103 modelului proprietar de vectorizare a textului de la OpenAI (text-embedding-ada-002), capabil \u0219i el s\u0103 proceseze texte cu p\u00e2n\u0103 la 8192 de tokeni. <\/p>\n
Distan\u021ba dintre cele dou\u0103 vectori forma\u021bi poate fi utilizat\u0103 pentru a determina rela\u021bia semnificativ\u0103 dintre textele originale. \u00cen practic\u0103, vectorii forma\u021bi pot fi folosi\u021bi pentru analizarea similitudinii textelor, organizarea c\u0103ut\u0103rii materialelor similare \u00een tem\u0103 (clasificarea rezultatelor \u00een func\u021bie de apropierea semantic\u0103), gruparea textelor \u00een func\u021bie de sens, formarea recomand\u0103rilor (sugestie de liste cu \u0219iruri textuale similare), identificarea anomaliilor, detectarea plagiatului \u0219i clasificarea testelor. Exemple de domenii de utilizare includ analiza documentelor juridice, analiza de afaceri, cercet\u0103ri medicale pentru procesarea articolelor \u0219tiin\u021bifice, critic\u0103 literar\u0103, analiza rapoartelor financiare \u0219i \u00eembun\u0103t\u0103\u021birea calit\u0103\u021bii proces\u0103rii \u00eentreb\u0103rilor complexe de c\u0103tre chatbot-uri. <\/p>\n
Sunt disponibile dou\u0103 variante ale modelului jina-embeddings pentru desc\u0103rcare (baza \u2014 0.27 GB \u0219i o variant\u0103 redus\u0103 \u2014 0.07 GB), antrenate pe 400 de milioane de perechi de secven\u021be textuale \u00een limba englez\u0103, acoperind diverse domenii de cuno\u0219tin\u021be. \u00cen timpul antren\u0103rii, au fost folosite secven\u021be de 512 tokeni, care au fost extrapolate la dimensiunea de 8192 folosind metoda ALiBi (Attention with Linear Biases). <\/p>\n
Modelul de baz\u0103 include 137 milioane de parametri \u0219i este destinat utiliz\u0103rii pe sisteme desktop cu GPU. Modelul redus include 33 milioane de parametri, ofer\u0103 o precizie mai mic\u0103 \u0219i este destinat utiliz\u0103rii pe dispozitive mobile \u0219i pe sisteme cu capacitate limitat\u0103 de memorie. \u00cen cur\u00e2nd se preconizeaz\u0103 publicarea unui model mare care va acoperi 435 milioane de parametri. De asemenea, se dezvolt\u0103 o variant\u0103 multilingv\u0103 a modelului, care se concentreaz\u0103 \u00een prezent pe suportul limbilor german\u0103 \u0219i spaniol\u0103. Un plugin a fost preg\u0103tit separat pentru utilizarea modelului jina-embeddings prin intermediul instrumentarului LLM. \u041a\u043e\u043c\u043f\u0430\u043d\u0438\u044f Jina \u043e\u0442\u043a\u0440\u044b\u043b\u0430 \u043f\u043e\u0434 \u043b\u0438\u0446\u0435\u043d\u0437\u0438\u0435\u0439 Apache 2.0 \u043c\u043e\u0434\u0435\u043b\u044c \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0434\u043b\u044f \u0432\u0435\u043a\u0442\u043e\u0440\u043d\u043e\u0433\u043e \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d\u0438\u044f \u0442\u0435\u043a\u0441\u0442\u0430 — jina-embeddings-v2. \u041c\u043e\u0434\u0435\u043b\u044c \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u0435\u0442 \u043f\u0440\u0435\u043e\u0431\u0440\u0430\u0437\u043e\u0432\u0430\u0442\u044c \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u043b\u044c\u043d\u044b\u0439 \u0442\u0435\u043a\u0441\u0442, \u0432\u043a\u043b\u044e\u0447\u0430\u044e\u0449\u0438\u0439 \u0434\u043e 8192 \u0437\u043d\u0430\u043a\u043e\u0432, \u0432 \u043d\u0435\u0431\u043e\u043b\u044c\u0448\u0443\u044e \u043f\u043e\u0441\u043b\u0435\u0434\u043e\u0432\u0430\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u044c \u0432\u0435\u0449\u0435\u0441\u0442\u0432\u0435\u043d\u043d\u044b\u0445 \u0447\u0438\u0441\u0435\u043b, \u043e\u0431\u0440\u0430\u0437\u0443\u044e\u0449\u0438\u0445 \u0432\u0435\u043a\u0442\u043e\u0440, \u0441\u043e\u043f\u043e\u0441\u0442\u0430\u0432\u043b\u0435\u043d\u043d\u044b\u0439 \u0441 \u0438\u0441\u0445\u043e\u0434\u043d\u044b\u043c \u0442\u0435\u043a\u0441\u0442\u043e\u043c \u0438 \u0432\u043e\u0441\u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u044f\u0449\u0438\u0439 \u0435\u0433\u043e \u0441\u0435\u043c\u0430\u043d\u0442\u0438\u043a\u0443 (\u0441\u043c\u044b\u0441\u043b). Jina Embedding \u0441\u0442\u0430\u043b\u0430 \u043f\u0435\u0440\u0432\u043e\u0439 \u043e\u0442\u043a\u0440\u044b\u0442\u043e\u0439 \u043c\u043e\u0434\u0435\u043b\u044c\u044e \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f, \u043e\u0431\u043b\u0430\u0434\u0430\u044e\u0449\u0435\u0439 \u0445\u0430\u0440\u0430\u043a\u0442\u0435\u0440\u0438\u0441\u0442\u0438\u043a\u0430\u043c\u0438, \u043d\u0435 \u0443\u0441\u0442\u0443\u043f\u0430\u044e\u0449\u0438\u043c\u0438 \u043f\u0440\u043e\u043f\u0440\u0438\u0435\u0442\u0430\u0440\u043d\u043e\u0439 \u043c\u043e\u0434\u0435\u043b\u0438 […]<\/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-111103","post","type-post","status-publish","format-standard","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Codul Jina Embedding, model pentru reprezentarea vectorial\u0103 a semnifica\u021biei textului, este deschis | ProHoster","description":"Compania Jina a lansat sub licen\u021ba Apache 2.0 un model de \u00eenv\u0103\u021bare automat\u0103 pentru reprezentarea vectorial\u0103 a textului - jina-embeddings-v2.","canonical_url":"https:\/\/prohoster.info\/ro\/blog\/news\/otkryt-kod-jina-embedding-modeli-dlya-vektornogo-predstavleniya-smysla-teksta","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"ro_RO","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\u041e\u0442\u043a\u0440\u044b\u0442 \u043a\u043e\u0434 Jina Embedding, \u043c\u043e\u0434\u0435\u043b\u0438 \u0434\u043b\u044f \u0432\u0435\u043a\u0442\u043e\u0440\u043d\u043e\u0433\u043e \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d\u0438\u044f \u0441\u043c\u044b\u0441\u043b\u0430 \u0442\u0435\u043a\u0441\u0442\u0430 | ProHoster","og:description":"\u041a\u043e\u043c\u043f\u0430\u043d\u0438\u044f Jina \u043e\u0442\u043a\u0440\u044b\u043b\u0430 \u043f\u043e\u0434 \u043b\u0438\u0446\u0435\u043d\u0437\u0438\u0435\u0439 Apache 2.0 \u043c\u043e\u0434\u0435\u043b\u044c \u043c\u0430\u0448\u0438\u043d\u043d\u043e\u0433\u043e \u043e\u0431\u0443\u0447\u0435\u043d\u0438\u044f \u0434\u043b\u044f \u0432\u0435\u043a\u0442\u043e\u0440\u043d\u043e\u0433\u043e \u043f\u0440\u0435\u0434\u0441\u0442\u0430\u0432\u043b\u0435\u043d\u0438\u044f \u0442\u0435\u043a\u0441\u0442\u0430 - jina-embeddings-v2.","og:url":"https:\/\/prohoster.info\/ro\/blog\/news\/otkryt-kod-jina-embedding-modeli-dlya-vektornogo-predstavleniya-smysla-teksta","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":"2023-10-26T13:10:31+00:00","article:modified_time":"2023-10-26T13:10:31+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\/ro\/wp-json\/wp\/v2\/posts\/111103","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/comments?post=111103"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/posts\/111103\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/media?parent=111103"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/categories?post=111103"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/ro\/wp-json\/wp\/v2\/tags?post=111103"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n
Sursa: opennet.ro<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"