{"id":477506,"date":"2023-08-09T09:15:57","date_gmt":"2023-08-09T09:15:57","guid":{"rendered":""},"modified":"2023-09-05T11:14:50","modified_gmt":"2023-09-05T11:14:50","slug":"hugging-face","status":"publish","type":"wiki","link":"https:\/\/oneproxy.pro\/vn\/wiki\/hugging-face\/","title":{"rendered":"\u00f4m m\u1eb7t"},"content":{"rendered":"<p>\u00d4m M\u1eb7t l\u00e0 m\u1ed9t c\u00f4ng ty ti\u00ean phong v\u00e0 c\u1ed9ng \u0111\u1ed3ng ngu\u1ed3n m\u1edf chuy\u00ean v\u1ec1 x\u1eed l\u00fd ng\u00f4n ng\u1eef t\u1ef1 nhi\u00ean (NLP) v\u00e0 tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o (AI). N\u1ed5i ti\u1ebfng nh\u1ea5t v\u1edbi c\u00e1c m\u00f4 h\u00ecnh Transformer v\u00e0 c\u00e1c th\u01b0 vi\u1ec7n PyTorch v\u00e0 TensorFlow li\u00ean quan, Hugging Face \u0111\u00e3 n\u1ed5i l\u00ean nh\u01b0 m\u1ed9t l\u1ef1c l\u01b0\u1ee3ng h\u00e0ng \u0111\u1ea7u trong nghi\u00ean c\u1ee9u v\u00e0 ph\u00e1t tri\u1ec3n NLP.<\/p>\n<h2>Ngu\u1ed3n g\u1ed1c c\u1ee7a vi\u1ec7c \u00f4m m\u1eb7t<\/h2>\n<p>Hugging Face, Inc. \u0111\u01b0\u1ee3c \u0111\u1ed3ng s\u00e1ng l\u1eadp b\u1edfi Clement Delangue v\u00e0 Julien Chaumond t\u1ea1i th\u00e0nh ph\u1ed1 New York v\u00e0o n\u0103m 2016. Ban \u0111\u1ea7u, c\u00f4ng ty t\u1eadp trung ph\u00e1t tri\u1ec3n m\u1ed9t chatbot c\u00f3 t\u00ednh c\u00e1ch ri\u00eang bi\u1ec7t, t\u01b0\u01a1ng t\u1ef1 nh\u01b0 Siri v\u00e0 Alexa. Tuy nhi\u00ean, tr\u1ecdng t\u00e2m c\u1ee7a h\u1ecd \u0111\u00e3 thay \u0111\u1ed5i v\u00e0o n\u0103m 2018 khi h\u1ecd tung ra m\u1ed9t th\u01b0 vi\u1ec7n ngu\u1ed3n m\u1edf, c\u00f3 t\u00ean l\u00e0 Transformers, \u0111\u1ec3 \u0111\u00e1p \u1ee9ng v\u1edbi l\u0129nh v\u1ef1c m\u00f4 h\u00ecnh d\u1ef1a tr\u00ean m\u00e1y bi\u1ebfn \u00e1p \u0111ang ph\u00e1t tri\u1ec3n, \u0111ang c\u00e1ch m\u1ea1ng h\u00f3a l\u0129nh v\u1ef1c NLP.<\/p>\n<h2>L\u00e0m s\u00e1ng t\u1ecf khu\u00f4n m\u1eb7t \u00f4m<\/h2>\n<p>V\u1ec1 c\u1ed1t l\u00f5i, \u00d4m m\u1eb7t cam k\u1ebft d\u00e2n ch\u1ee7 h\u00f3a AI v\u00e0 cung c\u1ea5p cho c\u1ed9ng \u0111\u1ed3ng c\u00e1c c\u00f4ng c\u1ee5 gi\u00fap t\u1ea5t c\u1ea3 m\u1ecdi ng\u01b0\u1eddi \u0111\u1ec1u c\u00f3 th\u1ec3 ti\u1ebfp c\u1eadn NLP hi\u1ec7n \u0111\u1ea1i. Nh\u00f3m \u00d4m M\u1eb7t duy tr\u00ec m\u1ed9t th\u01b0 vi\u1ec7n c\u00f3 t\u00ean Transformers, n\u01a1i cung c\u1ea5p h\u00e0ng ngh\u00ecn m\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc \u0111\u1ec3 th\u1ef1c hi\u1ec7n c\u00e1c t\u00e1c v\u1ee5 tr\u00ean v\u0103n b\u1ea3n, ch\u1eb3ng h\u1ea1n nh\u01b0 ph\u00e2n lo\u1ea1i v\u0103n b\u1ea3n, tr\u00edch xu\u1ea5t th\u00f4ng tin, t\u00f3m t\u1eaft t\u1ef1 \u0111\u1ed9ng, d\u1ecbch thu\u1eadt v\u00e0 t\u1ea1o v\u0103n b\u1ea3n.<\/p>\n<p>N\u1ec1n t\u1ea3ng \u00d4m M\u1eb7t c\u0169ng bao g\u1ed3m m\u00f4i tr\u01b0\u1eddng \u0111\u00e0o t\u1ea1o c\u1ed9ng t\u00e1c, API suy lu\u1eadn v\u00e0 trung t\u00e2m m\u00f4 h\u00ecnh. Trung t\u00e2m m\u00f4 h\u00ecnh cho ph\u00e9p c\u00e1c nh\u00e0 nghi\u00ean c\u1ee9u v\u00e0 nh\u00e0 ph\u00e1t tri\u1ec3n chia s\u1ebb v\u00e0 c\u1ed9ng t\u00e1c tr\u00ean c\u00e1c m\u00f4 h\u00ecnh, g\u00f3p ph\u1ea7n t\u1ea1o n\u00ean t\u00ednh ch\u1ea5t m\u1edf c\u1ee7a n\u1ec1n t\u1ea3ng.<\/p>\n<h2>Ho\u1ea1t \u0111\u1ed9ng b\u00ean trong c\u1ee7a vi\u1ec7c \u00f4m m\u1eb7t<\/h2>\n<p>\u00d4m M\u1eb7t ho\u1ea1t \u0111\u1ed9ng tr\u00ean n\u1ec1n t\u1ea3ng c\u1ee7a ki\u1ebfn tr\u00fac m\u00e1y bi\u1ebfn \u00e1p, s\u1eed d\u1ee5ng c\u01a1 ch\u1ebf t\u1ef1 ch\u00fa \u00fd \u0111\u1ec3 hi\u1ec3u m\u1ee9c \u0111\u1ed9 li\u00ean quan theo ng\u1eef c\u1ea3nh c\u1ee7a c\u00e1c t\u1eeb trong c\u00e2u. C\u00e1c m\u00f4 h\u00ecnh m\u00e1y bi\u1ebfn \u00e1p \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc tr\u00ean c\u00e1c t\u1eadp d\u1eef li\u1ec7u v\u0103n b\u1ea3n l\u1edbn v\u00e0 c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c tinh ch\u1ec9nh cho m\u1ed9t nhi\u1ec7m v\u1ee5 c\u1ee5 th\u1ec3.<\/p>\n<p>Trong ph\u1ea7n ph\u1ee5 tr\u1ee3, th\u01b0 vi\u1ec7n Transformers h\u1ed7 tr\u1ee3 c\u1ea3 PyTorch v\u00e0 TensorFlow, hai trong s\u1ed1 c\u00e1c khung h\u1ecdc s\u00e2u \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng r\u1ed9ng r\u00e3i nh\u1ea5t. \u0110i\u1ec1u n\u00e0y l\u00e0m cho n\u00f3 c\u1ef1c k\u1ef3 linh ho\u1ea1t v\u00e0 cho ph\u00e9p ng\u01b0\u1eddi d\u00f9ng chuy\u1ec3n \u0111\u1ed5i gi\u1eefa hai khung n\u00e0y m\u1ed9t c\u00e1ch li\u1ec1n m\u1ea1ch.<\/p>\n<h2>\u0110\u1eb7c \u0111i\u1ec3m ch\u00ednh c\u1ee7a \u00f4m m\u1eb7t<\/h2>\n<ul>\n<li><strong>M\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc \u0111a d\u1ea1ng<\/strong>: Th\u01b0 vi\u1ec7n Transformers c\u1ee7a Hugging Face cung c\u1ea5p m\u1ed9t lo\u1ea1t c\u00e1c m\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc, ch\u1eb3ng h\u1ea1n nh\u01b0 BERT, GPT-2, T5 v\u00e0 RoBERTa, c\u00f9ng v\u1edbi c\u00e1c m\u00f4 h\u00ecnh kh\u00e1c.<\/li>\n<li><strong>H\u1ed7 tr\u1ee3 ng\u00f4n ng\u1eef r\u1ed9ng<\/strong>: C\u00e1c m\u00f4 h\u00ecnh c\u00f3 th\u1ec3 x\u1eed l\u00fd nhi\u1ec1u ng\u00f4n ng\u1eef, v\u1edbi c\u00e1c m\u00f4 h\u00ecnh c\u1ee5 th\u1ec3 \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u00ean c\u00e1c b\u1ed9 d\u1eef li\u1ec7u kh\u00f4ng ph\u1ea3i ti\u1ebfng Anh.<\/li>\n<li><strong>Kh\u1ea3 n\u0103ng tinh ch\u1ec9nh<\/strong>: C\u00e1c m\u00f4 h\u00ecnh c\u00f3 th\u1ec3 d\u1ec5 d\u00e0ng tinh ch\u1ec9nh cho c\u00e1c nhi\u1ec7m v\u1ee5 c\u1ee5 th\u1ec3, mang l\u1ea1i t\u00ednh linh ho\u1ea1t trong nhi\u1ec1u tr\u01b0\u1eddng h\u1ee3p s\u1eed d\u1ee5ng kh\u00e1c nhau.<\/li>\n<li><strong>H\u01b0\u1edbng t\u1edbi c\u1ed9ng \u0111\u1ed3ng<\/strong>: Hugging Face ph\u00e1t tri\u1ec3n m\u1ea1nh nh\u1edd c\u1ed9ng \u0111\u1ed3ng c\u1ee7a m\u00ecnh. N\u00f3 khuy\u1ebfn kh\u00edch ng\u01b0\u1eddi d\u00f9ng \u0111\u00f3ng g\u00f3p v\u00e0o c\u00e1c m\u00f4 h\u00ecnh, n\u00e2ng cao ch\u1ea5t l\u01b0\u1ee3ng t\u1ed5ng th\u1ec3 v\u00e0 s\u1ef1 \u0111a d\u1ea1ng c\u1ee7a c\u00e1c m\u00f4 h\u00ecnh c\u00f3 s\u1eb5n.<\/li>\n<\/ul>\n<h2>C\u00e1c ki\u1ec3u m\u1eabu \u00f4m m\u1eb7t<\/h2>\n<p>D\u01b0\u1edbi \u0111\u00e2y l\u00e0 danh s\u00e1ch m\u1ed9t s\u1ed1 m\u1eabu m\u00e1y bi\u1ebfn \u00e1p ph\u1ed5 bi\u1ebfn nh\u1ea5t hi\u1ec7n c\u00f3 trong th\u01b0 vi\u1ec7n Transformers c\u1ee7a Hugging Face:<\/p>\n<table>\n<thead>\n<tr>\n<th>T\u00ean m\u1eabu<\/th>\n<th>S\u1ef1 mi\u00eau t\u1ea3<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>BERT<\/td>\n<td>Bi\u1ec3u di\u1ec5n b\u1ed9 m\u00e3 h\u00f3a hai chi\u1ec1u t\u1eeb Transformers \u0111\u1ec3 \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc c\u00e1c bi\u1ec3u di\u1ec5n hai chi\u1ec1u s\u00e2u t\u1eeb v\u0103n b\u1ea3n kh\u00f4ng \u0111\u01b0\u1ee3c g\u1eafn nh\u00e3n<\/td>\n<\/tr>\n<tr>\n<td>GPT-2<\/td>\n<td>Generative Pretraining Transformer 2 cho c\u00e1c nhi\u1ec7m v\u1ee5 t\u1ea1o ng\u00f4n ng\u1eef<\/td>\n<\/tr>\n<tr>\n<td>T5<\/td>\n<td>Bi\u1ebfn \u00e1p chuy\u1ec3n v\u0103n b\u1ea3n th\u00e0nh v\u0103n b\u1ea3n cho c\u00e1c t\u00e1c v\u1ee5 NLP kh\u00e1c nhau<\/td>\n<\/tr>\n<tr>\n<td>roberta<\/td>\n<td>Phi\u00ean b\u1ea3n BERT \u0111\u01b0\u1ee3c t\u1ed1i \u01b0u h\u00f3a m\u1ea1nh m\u1ebd \u0111\u1ec3 c\u00f3 k\u1ebft qu\u1ea3 ch\u00ednh x\u00e1c h\u01a1n<\/td>\n<\/tr>\n<tr>\n<td>ch\u01b0ng c\u1ea5tBERT<\/td>\n<td>Phi\u00ean b\u1ea3n ch\u01b0ng c\u1ea5t c\u1ee7a BERT nh\u1eb9 h\u01a1n v\u00e0 nhanh h\u01a1n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>T\u1eadn d\u1ee5ng c\u00e1ch \u00f4m m\u1eb7t v\u00e0 gi\u1ea3i quy\u1ebft c\u00e1c th\u00e1ch th\u1ee9c<\/h2>\n<p>M\u00f4 h\u00ecnh Khu\u00f4n m\u1eb7t \u00f4m c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng cho nhi\u1ec1u nhi\u1ec7m v\u1ee5 kh\u00e1c nhau, t\u1eeb ph\u00e2n t\u00edch c\u1ea3m x\u00fac v\u00e0 ph\u00e2n lo\u1ea1i v\u0103n b\u1ea3n \u0111\u1ebfn d\u1ecbch m\u00e1y v\u00e0 t\u00f3m t\u1eaft v\u0103n b\u1ea3n. Tuy nhi\u00ean, gi\u1ed1ng nh\u01b0 t\u1ea5t c\u1ea3 c\u00e1c m\u00f4 h\u00ecnh AI, ch\u00fang c\u00f3 th\u1ec3 \u0111\u1eb7t ra nh\u1eefng th\u00e1ch th\u1ee9c, ch\u1eb3ng h\u1ea1n nh\u01b0 y\u00eau c\u1ea7u l\u01b0\u1ee3ng l\u1edbn d\u1eef li\u1ec7u \u0111\u1ec3 \u0111\u00e0o t\u1ea1o v\u00e0 nguy c\u01a1 sai l\u1ec7ch trong m\u00f4 h\u00ecnh. \u00d4m M\u1eb7t gi\u1ea3i quy\u1ebft nh\u1eefng th\u00e1ch th\u1ee9c n\u00e0y b\u1eb1ng c\u00e1ch cung c\u1ea5p h\u01b0\u1edbng d\u1eabn chi ti\u1ebft \u0111\u1ec3 tinh ch\u1ec9nh c\u00e1c m\u00f4 h\u00ecnh v\u00e0 nhi\u1ec1u m\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc \u0111\u1ec3 b\u1ea1n l\u1ef1a ch\u1ecdn.<\/p>\n<h2>So s\u00e1nh v\u1edbi c\u00e1c c\u00f4ng c\u1ee5 t\u01b0\u01a1ng t\u1ef1<\/h2>\n<p>M\u1eb7c d\u00f9 \u00d4m m\u1eb7t l\u00e0 m\u1ed9t n\u1ec1n t\u1ea3ng ph\u1ed5 bi\u1ebfn r\u1ed9ng r\u00e3i cho c\u00e1c nhi\u1ec7m v\u1ee5 NLP, nh\u01b0ng v\u1eabn c\u00f3 s\u1eb5n c\u00e1c c\u00f4ng c\u1ee5 kh\u00e1c, nh\u01b0 spaCy, NLTK v\u00e0 StanfordNLP. Tuy nhi\u00ean, \u0111i\u1ec1u khi\u1ebfn Hugging Face tr\u1edf n\u00ean kh\u00e1c bi\u1ec7t l\u00e0 ph\u1ea1m vi r\u1ed9ng l\u1edbn c\u1ee7a c\u00e1c m\u00f4 h\u00ecnh \u0111\u01b0\u1ee3c \u0111\u00e0o t\u1ea1o tr\u01b0\u1edbc v\u00e0 kh\u1ea3 n\u0103ng t\u00edch h\u1ee3p li\u1ec1n m\u1ea1ch v\u1edbi PyTorch v\u00e0 TensorFlow.<\/p>\n<h2>T\u01b0\u01a1ng lai c\u1ee7a vi\u1ec7c \u00f4m m\u1eb7t<\/h2>\n<p>V\u1edbi s\u1ef1 nh\u1ea5n m\u1ea1nh v\u00e0o c\u1ed9ng \u0111\u1ed3ng, \u00d4m M\u1eb7t ti\u1ebfp t\u1ee5c v\u01b0\u1ee3t qua ranh gi\u1edbi c\u1ee7a nghi\u00ean c\u1ee9u NLP v\u00e0 AI. Tr\u1ecdng t\u00e2m g\u1ea7n \u0111\u00e2y c\u1ee7a h\u1ecd l\u00e0 v\u1ec1 l\u0129nh v\u1ef1c m\u00f4 h\u00ecnh ng\u00f4n ng\u1eef l\u1edbn nh\u01b0 GPT-4 v\u00e0 vai tr\u00f2 c\u1ee7a nh\u1eefng m\u00f4 h\u00ecnh n\u00e0y trong c\u00e1c nhi\u1ec7m v\u1ee5 c\u00f3 m\u1ee5c \u0111\u00edch chung. H\u1ecd c\u0169ng \u0111ang \u0111\u00e0o s\u00e2u v\u00e0o c\u00e1c l\u0129nh v\u1ef1c nh\u01b0 h\u1ecdc m\u00e1y tr\u00ean thi\u1ebft b\u1ecb v\u00e0 b\u1ea3o v\u1ec7 quy\u1ec1n ri\u00eang t\u01b0.<\/p>\n<h2>M\u00e1y ch\u1ee7 proxy v\u00e0 khu\u00f4n m\u1eb7t \u00f4m \u1ea5p<\/h2>\n<p>M\u00e1y ch\u1ee7 proxy c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng c\u00f9ng v\u1edbi Hugging Face cho c\u00e1c t\u00e1c v\u1ee5 nh\u01b0 qu\u00e9t web, trong \u0111\u00f3 vi\u1ec7c xoay v\u00f2ng IP l\u00e0 r\u1ea5t quan tr\u1ecdng \u0111\u1ec3 \u1ea9n danh. Vi\u1ec7c s\u1eed d\u1ee5ng m\u00e1y ch\u1ee7 proxy cho ph\u00e9p c\u00e1c nh\u00e0 ph\u00e1t tri\u1ec3n truy c\u1eadp v\u00e0 truy xu\u1ea5t d\u1eef li\u1ec7u t\u1eeb web, d\u1eef li\u1ec7u n\u00e0y c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c \u0111\u01b0a v\u00e0o c\u00e1c m\u00f4 h\u00ecnh \u00d4m M\u1eb7t cho c\u00e1c t\u00e1c v\u1ee5 NLP kh\u00e1c nhau.<\/p>\n<h2>Li\u00ean k\u1ebft li\u00ean quan<\/h2>\n<ul>\n<li>Trang web \u00f4m m\u1eb7t: <a href=\"https:\/\/huggingface.co\/\" target=\"_new\" rel=\"noopener nofollow\">https:\/\/huggingface.co\/<\/a><\/li>\n<li>Th\u01b0 vi\u1ec7n Transformers tr\u00ean GitHub: <a href=\"https:\/\/github.com\/huggingface\/transformers\" target=\"_new\" rel=\"noopener nofollow\">https:\/\/github.com\/huggingface\/transformers<\/a><\/li>\n<li>Trung t\u00e2m m\u00f4 h\u00ecnh \u00f4m m\u1eb7t: <a href=\"https:\/\/huggingface.co\/models\" target=\"_new\" rel=\"noopener nofollow\">https:\/\/huggingface.co\/models<\/a><\/li>\n<li>Kh\u00f3a h\u1ecdc \u00f4m m\u1eb7t ch\u00ednh th\u1ee9c: <a href=\"https:\/\/huggingface.co\/course\/chapter1\" target=\"_new\" rel=\"noopener nofollow\">https:\/\/huggingface.co\/course\/chapter1<\/a><\/li>\n<\/ul>","protected":false},"featured_media":0,"menu_order":0,"template":"","meta":{"_acf_changed":false,"content-type":"","inline_featured_image":false,"footnotes":""},"class_list":["post-477506","wiki","type-wiki","status-publish","hentry"],"acf":{"faq_title":"Frequently Asked Questions about <mark>Hugging Face: An In-Depth Guide to the Transformer Revolution<\/mark>","faq_items":[{"question":"What is Hugging Face?","answer":"<p>Hugging Face is a company and open-source community specializing in natural language processing (NLP) and artificial intelligence (AI). They are known for their Transformers library, which offers a vast array of pre-trained models for various NLP tasks.<\/p>"},{"question":"Who founded Hugging Face and when?","answer":"<p>Hugging Face was co-founded by Clement Delangue and Julien Chaumond in 2016 in New York City. Initially, the company focused on developing a chatbot, but their focus shifted towards transformer-based models for NLP in 2018.<\/p>"},{"question":"What are some key features of Hugging Face?","answer":"<p>Hugging Face offers diverse pre-trained models, broad language support, fine-tuning capabilities for specific tasks, and a thriving community-driven approach. These features make Hugging Face a leading platform for NLP tasks.<\/p>"},{"question":"What types of Hugging Face models exist?","answer":"<p>Hugging Face's Transformers library provides many transformer models, such as BERT, GPT-2, T5, RoBERTa, and DistilBERT, which can be used for a range of NLP tasks like text classification, information extraction, automatic summarization, translation, and text generation.<\/p>"},{"question":"What challenges can occur when using Hugging Face?","answer":"<p>Some challenges when using Hugging Face models may include the requirement of large amounts of data for training and the risk of bias in the models. Hugging Face addresses these challenges by providing detailed guides for fine-tuning models and a diverse range of pre-trained models.<\/p>"},{"question":"How does Hugging Face compare to similar tools?","answer":"<p>While other NLP tools like spaCy, NLTK, and StanfordNLP exist, Hugging Face stands out due to its extensive range of pre-trained models and its seamless integration with popular deep learning frameworks like PyTorch and TensorFlow.<\/p>"},{"question":"What is the future perspective of Hugging Face?","answer":"<p>Hugging Face continues to push the boundaries of NLP and AI research. They are focusing on the development and use of large language models like GPT-4 and exploring fields such as on-device and privacy-preserving machine learning.<\/p>"},{"question":"How can proxy servers be used with Hugging Face?","answer":"<p>Proxy servers can be used with Hugging Face for tasks like web scraping. The use of proxy servers allows for IP rotation for anonymity and facilitates the retrieval of web data, which can be processed using Hugging Face models for various NLP tasks.<\/p>"}]},"_links":{"self":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki\/477506","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki"}],"about":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/types\/wiki"}],"version-history":[{"count":0,"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/wiki\/477506\/revisions"}],"wp:attachment":[{"href":"https:\/\/oneproxy.pro\/vn\/wp-json\/wp\/v2\/media?parent=477506"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}