{"id":54780,"date":"2020-01-04T00:00:00","date_gmt":"2020-01-03T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/nejroseti-kuda-eto-vse-dvizhetsya"},"modified":"2020-02-18T14:02:50","modified_gmt":"2020-02-18T11:02:50","slug":"nejroseti-kuda-eto-vse-dvizhetsya","status":"publish","type":"post","link":"https:\/\/prohoster.info\/et\/blog\/novosti-interneta\/nejroseti-kuda-eto-vse-dvizhetsya","title":{"rendered":"Neuraalsed v\u00f5rgud. Kuhu see k\u00f5ik liigub","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p>Artikkel koosneb kahest osast:<\/p>\n<p><\/p>\n<ol>\n<li>L\u00fchike \u00fclevaade m\u00f5ningatest arhitektuuridest, mis on m\u00f5eldud objektide tuvastamiseks piltidel ja piltide segmentimiseks, koos k\u00f5ige arusaadavamate linkidega ressurssidele. P\u00fc\u00fcdsin valida video selgitusi, eelistatavalt vene keeles.<\/li>\n<li>Teine osa keskendub p\u00fc\u00fcdele m\u00f5ista neuraalv\u00f5rkude arhitektuuride arengusuundi. Ja tehnoloogiaid, mis p\u00f5hinevad neil.<\/li>\n<\/ol>\n<p><\/p>\n<p><img decoding=\"async\" alt=\"Neuraalsed v\u00f5rgud. Kuhu see k\u00f5ik liigub\" src=\"\/wp-content\/uploads\/2020\/01\/3e0238547dc956dbb069c11241e4534f.png\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p><\/p>\n<p>Joonis 1 \u2013 Neuraalv\u00f5rkude arhitektuuride m\u00f5istmine pole lihtne<\/p>\n<p><\/p>\n<p>K\u00f5ik algas sellest, et tegin kaks demonstreerivat rakendust objektide klassifitseerimiseks ja tuvastamiseks Android telefonil:<\/p>\n<p><\/p>\n<ul>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/foobar167\/junkyard\/tree\/master\/object_classifier\">Tagumise osa demo<\/a><\/noindex>, kus andmed t\u00f6\u00f6deldakse serveris ja edastatakse telefonile. Piltide klassifitseerimine (image classification) kolmele karu t\u00fc\u00fcbile: pruuni, musta ja pehme karu.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/foobar167\/android\/tree\/master\/object_detection_demo\">Eesmise osa demo<\/a><\/noindex>, kus andmed t\u00f6\u00f6deldakse telefonil. Objektide tuvastamine (object detection) kolmele t\u00fc\u00fcbile: sarapuup\u00e4hkel, viigimarjad ja datlid.<\/li>\n<\/ul>\n<p><noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><\/p>\n<p>Klassifitseerimise, objektide tuvastamise ja <noindex><a rel=\"nofollow\" href=\"https:\/\/medium.com\/analytics-vidhya\/image-classification-vs-object-detection-vs-image-segmentation-f36db85fe81\">piltide segmentimise \u00fclesannete vahel on erinevusi<\/a><\/noindex>. Seet\u00f5ttu tekkis vajadus teada saada, millised tehisn\u00e4rviv\u00f5rgud tuvastavad objekte piltidel ja millised suudavad segmentida. Leidsin j\u00e4rgmised arhitektuurid, millel on minu jaoks k\u00f5ige arusaadavamad lingid ressurssidele:<\/p>\n<p><\/p>\n<ul>\n<li>R-CNN arhitektuuride seeria (<strong>R<\/strong>regionid, mis <strong>C<\/strong>konvolutsioon <strong>N<\/strong>neuraalsete <strong>N<\/strong>v\u00f5rkude omadused): R-CNN, Fast R-CNN, <noindex><a rel=\"nofollow\" href=\"https:\/\/medium.com\/@smallfishbigsea\/faster-r-cnn-explained-864d4fb7e3f8\">Faster R-CNN<\/a><\/noindex>, <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/0vt05rQqk_I\">Mask R-CNN<\/a><\/noindex>. Pildi objekti tuvastamiseks kasutatakse Region Proposal Network (RPN) mehhanismi, mis eraldab piiratud piirkonnad (bounding boxes). Alguses kasutati RPN asemel aeglasemat Selective Search mehhanismi. Seej\u00e4rel edastatakse eraldatud piiratud piirkonnad tavap\u00e4rasele tehisn\u00e4rviv\u00f5rgule klassifitseerimiseks. R-CNN arhitektuuris on selged \u201efor\u201d ts\u00fcklid, mis loovad piiratud piirkondade jaoks, kokku kuni 2000 korda l\u00e4bi sisemise v\u00f5rgu AlexNet. Selgete \u201efor\u201d ts\u00fcklite t\u00f5ttu aeglustub piltide t\u00f6\u00f6tlemise kiirus. Igakuuselt v\u00e4hendatakse selgete ts\u00fcklite arvu ja l\u00e4bidud ringe sisemise tehisn\u00e4rviv\u00f5rgu kaudu, samuti tehakse k\u00fcmneid teisi muudatusi kiirusete suurendamiseks ja objekti tuvastamise \u00fclesande asendamiseks objekti segmentimisega Mask R-CNN-s.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/L0tzmv--CGY\">YOLO<\/a><\/noindex> (<strong>Y<\/strong>ainult <strong>O<\/strong>vaatab <strong>L<\/strong>aga <strong>O<\/strong>nce) \u2013 esimene n\u00e4rviv\u00f5rk, mis tuvastas objekte reaalajas mobiilseadmetes. Erip\u00e4ra: objektide eristamine \u00fche l\u00e4bimisega (piisab korraks vaatamisest). See t\u00e4hendab, et YOLO arhitektuuris ei ole selgeid 'for' silmuseid, mist\u00f5ttu t\u00f6\u00f6tab v\u00f5rk kiiresti. N\u00e4iteks sarnasus: NumPy-s puuduvad ka silmused 'for' maatriksite operatsioonide puhul, kuna neid teostatakse NumPy-s madalamal arhitektuuritasemel C-programmeerimiskeele abil. YOLO kasutab eelm\u00e4\u00e4ratud akendega ruudustikku. Selleks, et sama objekti ei tuvastataks korduvalt, kasutatakse akende kattuvuse koefitsienti (IoU, <strong>I<\/strong>ntersection <strong>o<\/strong>ver <strong>U<\/strong>nion). See arhitektuur t\u00f6\u00f6tab laias vahemikus ja omab k\u00f5rget <noindex><a rel=\"nofollow\" href=\"https:\/\/ru.wikipedia.org\/wiki\/%D0%A0%D0%BE%D0%B1%D0%B0%D1%81%D1%82%D0%BD%D0%BE%D1%81%D1%82%D1%8C\">robustsust<\/a><\/noindex>: mudel saab olla \u00f5petatud fotodel, kuid t\u00f6\u00f6tab h\u00e4sti ka joonistatud piltidel.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/P8e-G-Mhx4k\">SSD<\/a><\/noindex> (<strong>S<\/strong>ingle <strong>S<\/strong>hot MultiBox <strong>D<\/strong>etector) \u2013 kasutatakse parimaid YOLO arhitektuuri \u201enippe\u201c (nt mitte-maximalne mahasurumine) ning lisatakse uusi, et n\u00e4rviv\u00f5rk t\u00f6\u00f6taks kiiremini ja t\u00e4psemalt. Erip\u00e4ra: objektide eristamine \u00fche l\u00e4bimise kaudu antud akna ruudustiku (default box) abil pildip\u00fcramiidis. Pildip\u00fcramiid on kodeeritud konvolutsioonitensorites j\u00e4rjestikuste konvolutsiooni ja max-poolingu operatsioonide k\u00e4igus (max-poolingu operatsiooni puhul v\u00e4heneb ruumiline m\u00f5\u00f5de). Nii m\u00e4\u00e4ratakse suured kui ka v\u00e4ikesed objektid \u00fche v\u00f5rgu l\u00e4bimisega.<\/li>\n<li>MobileSSD (<strong>Mobiilne<\/strong>NetV2 + <strong>SSD<\/strong>) \u2013 kahest n\u00e4rviv\u00f5rgu arhitektuurist koosnev kombinatsioon. Esimene v\u00f5rk <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/post\/352804\/\">MobileNetV2<\/a><\/noindex> t\u00f6\u00f6tleb andmeid kiiresti ja suurendab tuvastamise t\u00e4psust. MobileNetV2 asendab VGG-16, mida kasutati esialgselt <noindex><a rel=\"nofollow\" href=\"https:\/\/arxiv.org\/abs\/1512.02325\">originaalses artiklis<\/a><\/noindex>. Teine v\u00f5rk SSD m\u00e4\u00e4rab objektide asukoha pildil.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/ge_RT5wvHvY\">SqueezeNet<\/a><\/noindex> \u2013 v\u00e4ga v\u00e4ike, kuid t\u00e4pne tehisintellekt. \u00dcksinda ei lahenda see objektide tuvastamise probleemi. Kuid seda saab kasutada erinevate arhitektuuride kombinatsioonis ning mobiilsetes seadmetes. Iseloomulik omadus on see, et andmed kokkusurutakse nelja 1\u00d71 konvolutsioonifiltri abil, seej\u00e4rel laiendatakse neid nelja 1\u00d71 ja nelja 3\u00d73 konvolutsioonifiltri abil. \u00dchte sellist andmete kokkusurumise ja laiendamise iteratsiooni nimetatakse \u201eFire Module\u201c.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/b6jhopSMit8\">DeepLab<\/a><\/noindex> (Semantilise pildisegmentatsiooni teostamine s\u00fcgavate konvolutsiooniv\u00f5rkudega) \u2013 objektide segmentatsioon pildil. Arhitektuuri iseloomustab haruldane (dilated convolution) konvolutsioon, mis s\u00e4ilitab ruumilise eraldusv\u00f5ime. J\u00e4rgneb tulemuste post-protsessing etapp graafilise t\u00f5en\u00e4osusmudeliga (conditional random field), mis aitab eemaldada v\u00e4ikeseid segadusi segmentatsioonis ja parandada segmenteeritud pildi kvaliteeti. Kohutava nime \u201egraafiline t\u00f5en\u00e4osusmudel\u201c taga peitub tavaline Gaussi filter, mis on viie punkti kaupa l\u00e4henenud.<\/li>\n<li>P\u00fc\u00fcdsin v\u00e4lja selgitada, kuidas see t\u00f6\u00f6tab <noindex><a rel=\"nofollow\" href=\"https:\/\/arxiv.org\/abs\/1711.06897\">RefineDet<\/a><\/noindex> (Single-Shot <strong>Refine<\/strong>ment Neural Network for Object <strong>Det<\/strong>ja, kuid sain v\u00e4he aru.<\/li>\n<li>Vaatasin ka, kuidas t\u00f6\u00f6tab tehnoloogia \u201et\u00e4helepanu\u201c (attention): <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/W2rWgXJBZhU\">video1<\/a><\/noindex>, <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/iDulhoQ2pro\">video2<\/a><\/noindex>, <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/H6Qiegq_36c\">video3<\/a><\/noindex>. T\u00e4helepanu arhitektuuri iseloomustab automaatne t\u00e4helepanu piirkondade (RoI) esilet\u00f5stmine pildil, <strong>R<\/strong>huvide <strong>o<\/strong>f <strong>I<\/strong>mille v\u00f5imaldab n\u00e4rviv\u00f5rk nimega Attention Unit. T\u00e4helepanu piirkonnad sarnanevad piiritletud alade (bounding boxes) m\u00e4rkidega, kuid erinevalt neist ei ole nad pildile kinnitatud ja v\u00f5ivad omada h\u00e4guseid piire. Seej\u00e4rel eraldatakse t\u00e4helepanu piirkondadest omadused (features), mis \u201eyoni on\u201c rekurentsv\u00f5rkudele, mille arhitektuurid on <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/5lUUrREboSk\">LSTM, GRU v\u00f5i Vanilla RNN<\/a><\/noindex>. Rekurentsv\u00f5rgud suudavad anal\u00fc\u00fcsida omaduste omavahelisi seoseid j\u00e4rjestuses. Rekurentsv\u00f5rgud on algselt kasutatud tekstide t\u00f5lkimiseks teistesse keeltesse, kuid n\u00fc\u00fcd ka pildist <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/e-WB4lfg30M\">teksti<\/a><\/noindex> ja <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/rAbhypxs1qQ\">teksti pildiks<\/a><\/noindex>.<\/li>\n<\/ul>\n<p><\/p>\n<p>Nende arhitektuuride uurimise k\u00e4igus <strong>sain aru, et ma ei saa midagi aru.<\/strong>. Ja asi pole selles, et minu n\u00e4rviv\u00f5rgul on t\u00e4helepanu mehhanismiga probleeme. K\u00f5ikide nende arhitektuuride loomine sarnaneb m\u00f5ne tohutu hackathoniga, kus autorid v\u00f5istlevad h\u00e4kkimises. Hack (h\u00e4kk) on kiire lahendus keerulisele tarkvarak\u00fcsimusele. See t\u00e4hendab, et k\u00f5igi nende arhitektuuride vahel pole n\u00e4htavat ega arusaadavat loogilist seost. K\u00f5ike, mis neid \u00fchendab, peetakse parimateks h\u00e4kkideks, mida nad \u00fcksteiselt laenavad, pluss \u00fchine k\u00f5igile <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/Ilg3gGewQ5U\">tagasiside koondamisoperatsioon<\/a><\/noindex> (vea tagasikandmine, backpropagation). Ei <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/post\/272473\/\">s\u00fcsteemset m\u00f5tlemist<\/a><\/noindex>! Pole selge, mida muuta ja kuidas olemasolevaid saavutusi optimeerida.<\/p>\n<p><\/p>\n<p>Tulemuseks on see, et h\u00e4kkide vahelise loogilise seose puudumise t\u00f5ttu on need \u00e4\u00e4rmiselt raskesti meeldej\u00e4tmised ja praktikas rakendatavad. Need on killustatud teadmised. Parimal juhul j\u00e4\u00e4vad meelde m\u00f5ned huvitavad ja ootamatud hetked, kuid enamik arusaadust ja arusaamatust kaob m\u00e4lust juba paar p\u00e4eva p\u00e4rast. Oleks hea, kui n\u00e4dal p\u00e4rast m\u00e4letatakse v\u00e4hemalt arhitektuuri nime. Ja selle artiklite lugemise ja \u00fclevaatevideote vaatamise peale on kulutatud mitu tundi ja isegi p\u00e4eva t\u00f6\u00f6aega!<\/p>\n<p><\/p>\n<p><img decoding=\"async\" alt=\"Neuraalsed v\u00f5rgud. Kuhu see k\u00f5ik liigub\" src=\"\/wp-content\/uploads\/2020\/01\/86f0f24e3be1d0f1a210f3132897d981.png\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p><\/p>\n<p>Joonis 2 \u2013 <noindex><a rel=\"nofollow\" href=\"https:\/\/www.asimovinstitute.org\/neural-network-zoo\/\">Neuraalv\u00f5rkude loomaaed<\/a><\/noindex><\/p>\n<p><\/p>\n<p>Minu isikliku arvamuse kohaselt p\u00fc\u00fcavad enamik teadusartiklite autoreid teha k\u00f5ik, et isegi need killustatud teadmised ei j\u00e4\u00e4ks lugejale arusaadavaks. Kuid gerundite kasutamine k\u00fcmnes rida koos valemitega, mis on v\u00f5etud \u201elaest\u201c \u2013 see on teema eraldi artikli jaoks (probleem <noindex><a rel=\"nofollow\" href=\"https:\/\/en.wikipedia.org\/wiki\/Publish_or_perish\">publish or perish<\/a><\/noindex>).<\/p>\n<p><\/p>\n<p>Sellep\u00e4rast tekkis vajadus s\u00fcsteematiseerida teave neuraalv\u00f5rkude kohta ja seel\u00e4bi parandada arusaamise ja m\u00e4letamise kvaliteeti. Seega sai erinevate tehnoloogiate ja kunstlike neuraalv\u00f5rkude arhitektuuride anal\u00fc\u00fcsi p\u00f5hiteemaks j\u00e4rgmine \u00fclesanne: <strong>v\u00e4lja selgitada, kuhu see k\u00f5ik liigub<\/strong>, mitte millise konkreetse neuraalv\u00f5rgu \u00fclesehitusse.<\/p>\n<p><\/p>\n<p>Kuhu see k\u00f5ik liigub. Peamised tulemused:<\/p>\n<p><\/p>\n<ul>\n<li>Masin\u00f5ppe alustestimite arv viimase kahe aasta jooksul <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/company\/recognitor\/blog\/455676\/\">on j\u00e4rsult v\u00e4henenud<\/a><\/noindex>. V\u00f5imalik p\u00f5hjus: \u201eneuraalv\u00f5rgud ei ole enam millegi uue s\u00fcmboliks.\u201d<\/li>\n<li>Iga\u00fcks saab luua t\u00f6\u00f6tava neuraalv\u00f5rgu lihtsa \u00fclesande lahendamiseks. Selleks v\u00f5tab ta valmis mudeli \u201emudelite loomaaed\u201d (model zoo) ja treenib neuraalv\u00f5rgu viimast kihti (<noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/yofjFQddwHE\">transfer learning<\/a><\/noindex>) valmisteatud andmete p\u00f5hjal <noindex><a rel=\"nofollow\" href=\"https:\/\/toolbox.google.com\/datasetsearch\">Google'i andmeh\u00f5ive otsing<\/a><\/noindex> v\u00f5i <noindex><a rel=\"nofollow\" href=\"https:\/\/www.kaggle.com\/datasets\">25 000 Kaggle'i andmestikku<\/a><\/noindex> tasuta <noindex><a rel=\"nofollow\" href=\"https:\/\/www.dataschool.io\/cloud-services-for-jupyter-notebook\/\">Jupyter Notebook'i pilves<\/a><\/noindex>.<\/li>\n<li>Suurte tehisintellekti tootjate loomine <strong>modellide loomaaedade<\/strong> (model zoo). Nende abil on v\u00f5imalik kiiresti luua \u00e4rirakendus: <noindex><a rel=\"nofollow\" href=\"https:\/\/tfhub.dev\/\">TF Hub<\/a><\/noindex> TensorFlow'i jaoks, <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/open-mmlab\/mmdetection\">MMDetection<\/a><\/noindex> PyTorch'i jaoks, <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/facebookresearch\/Detectron\">Detectron<\/a><\/noindex> Caffe2 jaoks, <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/wkentaro\/chainer-modelzoo\">chainer-modelzoo<\/a><\/noindex> Chainer'i jaoks ja <noindex><a rel=\"nofollow\" href=\"https:\/\/modelzoo.co\/\">teised<\/a><\/noindex>.<\/li>\n<li>Tehisintellekt, mis t\u00f6\u00f6tab <strong>reaalses ajas<\/strong> (real-time) mobiilseadmetes. 10 kuni 50 kaadrit sekundis.<\/li>\n<li>Tehisintellekti rakendamine telefonides (TF Lite), brauserites (TF.js) ja <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/19ZNz2N79u4\">igap\u00e4evastes esemetes<\/a><\/noindex> (IoT, <strong>I<\/strong>asjade <strong>o<\/strong>f <strong>T<\/strong>internet). Eriti telefonides, mis juba toetavad tehisintellekti riistvara tasemel (neuroakseleraatorid).<\/li>\n<li>\u201eIga seade, r\u00f5ivad ja v\u00f5ib-olla isegi toit saavad <strong>IP-v6 aadressi<\/strong> ja suhtlevad omavahel\u201c \u2013 <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/GG7H8Xa4m8I?t=85\">Sebastian Thrun<\/a><\/noindex>.<\/li>\n<li>Masin\u00f5ppe publikatsioonide arv on hakanud <noindex><a rel=\"nofollow\" href=\"http:\/\/data-mining.philippe-fournier-viger.com\/too-many-machine-learning-papers\">\u00fcletama Moore'i seadust<\/a><\/noindex> (kahekordistumine iga kahe aasta tagant) alates 2015. aastast. On ilmne, et on vaja artiklite anal\u00fc\u00fcsimiseks tehisintellekti.<\/li>\n<li>J\u00e4rgmised tehnoloogiad saavad \u00fcha populaarsemaks:\n<ul>\n<li><strong>PyTorch<\/strong> \u2013 populaarsus kasvab kiiresti ja n\u00e4ib, et see \u00fcletab TensorFlow'd.<\/li>\n<li>Automaatne h\u00fcperparameetrite valik <strong>AutoML<\/strong> \u2013 populaarsus kasvab aeglaselt.<\/li>\n<li>T\u00e4pseuse j\u00e4rkj\u00e4rguline v\u00e4henemine ja arvutuskiirus: <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/rln_kZbYaWc\">ebamugav loogika<\/a><\/noindex>, algoritmid <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/MIPkK5ZAsms\">boosting<\/a><\/noindex>, ebat\u00e4psed (ligikaudsed) arvutused, kvantimine (kui n\u00e4rviv\u00f5rgu kaalu muudetakse t\u00e4isarvudeks ja kvantitakse), n\u00e4rviakseleraatorid.<\/li>\n<li>T\u00f5lge <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/e-WB4lfg30M\">teksti<\/a><\/noindex> ja <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/rAbhypxs1qQ\">teksti pildiks<\/a><\/noindex>.<\/li>\n<li>Loomine <noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/OrHLacCDZVQ\">kolmem\u00f5\u00f5tmeliste objektide video p\u00f5hjal<\/a><\/noindex>, n\u00fc\u00fcd juba reaalajas.<\/li>\n<li>Peamine DL-is on see, et andmeid on palju, kuid nende kogumine ja t\u00e4histamine ei ole kerge. Seet\u00f5ttu areneb automaatne t\u00e4histamine (<noindex><a rel=\"nofollow\" href=\"https:\/\/youtu.be\/NcKTn4C91Yc\">automated annotation<\/a><\/noindex>) n\u00e4rviv\u00f5rkude jaoks kasutades n\u00e4rviv\u00f5rke.<\/li>\n<\/ul>\n<\/li>\n<li>N\u00e4rviv\u00f5rkudega on arvutiteadus \u00e4kitselt muutunud <strong>eksperimentaalseks teaduseks<\/strong> ja tekkinud on <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/post\/480348\">reproducibility crisis<\/a><\/noindex>.<\/li>\n<li>IT-raha ja n\u00e4rviv\u00f5rkude populaarsus ilmusid samaaegselt, kui arvutused muutusid turuv\u00e4\u00e4rtuseks. Majandus on kuldvaluuta-arvutuslikuks muutumas <strong>. Vaadake minu artiklit<\/strong>\u00f6konofoorikas <noindex><a rel=\"nofollow\" href=\"https:\/\/ru.wikipedia.org\/wiki\/%D0%AD%D0%BA%D0%BE%D0%BD%D0%BE%D1%84%D0%B8%D0%B7%D0%B8%D0%BA%D0%B0\">ja IT-raha tekkimise p\u00f5hjuseid.<\/a><\/noindex> Ajanadaliselt tekib uus<\/li>\n<\/ul>\n<p><\/p>\n<p>ML\/DL programmeerimise metodoloogia <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/post\/481844\">(Machine Learning &amp; Deep Learning), mis p\u00f5hineb programmi k\u00e4sitlemisel, kui kogum treenitud n\u00e4rviv\u00f5rkude mudeleid.<\/a><\/noindex> Joonis 3 \u2013 ML\/DL kui uus programmeerimise metodoloogia<\/p>\n<p><\/p>\n<p><img decoding=\"async\" alt=\"Neuraalsed v\u00f5rgud. Kuhu see k\u00f5ik liigub\" src=\"\/wp-content\/uploads\/2020\/01\/17df76a613191522145abeba177c2685.png\" style=\"display:block;margin: 0 auto;\" \/><\/p>\n<p><\/p>\n<p>Kuid \"neurov\u00f5rkude teooriat\" pole nii kunagi tekkinud.<\/p>\n<p><\/p>\n<p>Kuid see ei ilmunud kunagi <strong>\u00abneuraalv\u00f5rkude teooria\u00bb<\/strong>, mille v\u00f5ivad m\u00f5elda ja t\u00f6\u00f6tada s\u00fcsteemselt. See, mis praegu nimetatakse 'teooriaks', on tegelikult eksperimentaalsed, heuristilised algoritmid.<\/p>\n<p><\/p>\n<p>Lingid minu ja mitte ainult ressurssidele:<\/p>\n<p><\/p>\n<ul>\n<li>Andmeteaduse uudiskiri. Peamiselt piltide t\u00f6\u00f6tlemisest. Kes soovib seda saada, saatke e-kiri (foobar167&lt;\u0433\u0430\u0444-\u0433\u0430\u0444&gt;gmail&lt;\u0442\u043e\u0447\u043a\u0430&gt;com). Linke artiklitele ja videotele saatn kogumise k\u00e4igus.<\/li>\n<li>\u00dcksikasjalik <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/foobar167\/articles\/blob\/master\/Machine_Learning\/courses_on_machine_learning.md\">kursuste ja artiklite loetelu<\/a><\/noindex>, mida olen l\u00e4binud ja mida sooviksin l\u00e4bida.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/foobar167\/articles\/blob\/master\/Ubuntu\/13_Keras_and_TensorFlow_how-tos.md#exercises\">Algajatele m\u00f5eldud kursused ja videod<\/a><\/noindex>, millega tasub alustada tehisn\u00e4rviv\u00f5rkude \u00f5ppimist. Pluss bro\u0161\u00fc\u00fcr <noindex><a rel=\"nofollow\" href=\"https:\/\/foobar167.github.io\/page\/vvedeniye-v-mashinnoye-obucheniye-i-iskusstvennyye-neyronnyye-seti.html\">\u201eSissejuhatus masin\u00f5ppesse ja tehisn\u00e4rviv\u00f5rkudesse\u201c<\/a><\/noindex>.<\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/foobar167\/articles\/blob\/master\/Ubuntu\/13_Keras_and_TensorFlow_how-tos.md#tools\">Kasulikud t\u00f6\u00f6riistad<\/a><\/noindex>, kust iga\u00fcks leiab endale midagi huvitavat.<\/li>\n<li>Eriti kasulikud on olnud <strong>videokanalid teadusartiklite anal\u00fc\u00fcsiks<\/strong> andmeteaduse valdkonnas. Leidke need, registreeruge neile ja jagage linke oma kolleegidega ning minuga ka. N\u00e4iteks:\n<ul>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/www.youtube.com\/user\/keeroyz\">Two Minute Papers<\/a><\/noindex><\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/www.youtube.com\/channel\/UCHB9VepY6kYvZjj0Bgxnpbw\">Henry AI Labs<\/a><\/noindex><\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/www.youtube.com\/channel\/UCZHmQk67mSJgfCCTn7xBfew\">Yannic Kilcher<\/a><\/noindex><\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/www.youtube.com\/channel\/UC5_6ZD6s8klmMu9TXEB_1IA\">CodeEmporium<\/a><\/noindex><\/li>\n<li><noindex><a rel=\"nofollow\" href=\"https:\/\/www.dlology.com\">Chengwei Zhangi blogi<\/a><\/noindex> tuntud ka kui <noindex><a rel=\"nofollow\" href=\"https:\/\/github.com\/Tony607\">Tony607<\/a><\/noindex> astep-by-step juhiste ja avatud l\u00e4htekoodiga.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<p><\/p>\n<p>Ait\u00e4h t\u00e4helepanu eest!<\/p>\n<p>Allikas: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/post\/482794\/\">habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u0421\u0442\u0430\u0442\u044c\u044f \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0438\u0437 \u0434\u0432\u0443\u0445 \u0447\u0430\u0441\u0442\u0435\u0439: \u041a\u0440\u0430\u0442\u043a\u043e\u0435 \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u0441\u0435\u0442\u0435\u0439 \u043f\u043e \u043e\u0431\u043d\u0430\u0440\u0443\u0436\u0435\u043d\u0438\u044e \u043e\u0431\u044a\u0435\u043a\u0442\u043e\u0432 \u043d\u0430 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0438 \u0438 \u0441\u0435\u0433\u043c\u0435\u043d\u0442\u0430\u0446\u0438\u0438 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439 \u0441 \u0441\u0430\u043c\u044b\u043c\u0438 \u043f\u043e\u043d\u044f\u0442\u043d\u044b\u043c\u0438 \u0434\u043b\u044f \u043c\u0435\u043d\u044f \u0441\u0441\u044b\u043b\u043a\u0430\u043c\u0438 \u043d\u0430 \u0440\u0435\u0441\u0443\u0440\u0441\u044b. \u0421\u0442\u0430\u0440\u0430\u043b\u0441\u044f \u0432\u044b\u0431\u0438\u0440\u0430\u0442\u044c \u0432\u0438\u0434\u0435\u043e \u043f\u043e\u044f\u0441\u043d\u0435\u043d\u0438\u044f \u0438 \u0436\u0435\u043b\u0430\u0442\u0435\u043b\u044c\u043d\u043e \u043d\u0430 \u0440\u0443\u0441\u0441\u043a\u043e\u043c \u044f\u0437\u044b\u043a\u0435. \u0412\u0442\u043e\u0440\u0430\u044f \u0447\u0430\u0441\u0442\u044c \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0432 \u043f\u043e\u043f\u044b\u0442\u043a\u0435 \u043e\u0441\u043e\u0437\u043d\u0430\u0442\u044c \u043d\u0430\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u044f \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u043d\u0435\u0439\u0440\u043e\u043d\u043d\u044b\u0445 \u0441\u0435\u0442\u0435\u0439. \u0418 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0439 \u043d\u0430 \u0438\u0445 \u043e\u0441\u043d\u043e\u0432\u0435. \u0420\u0438\u0441\u0443\u043d\u043e\u043a 1 \u2013 \u041f\u043e\u043d\u0438\u043c\u0430\u0442\u044c [&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-54780","post","type-post","status-publish","format-standard","hentry","category-novosti-interneta"],"aioseo_notices":[],"aioseo_head":"\n\t\t<!-- All in One SEO 4.9.10 - aioseo.com -->\n\t<meta name=\"description\" content=\"\u0421\u0442\u0430\u0442\u044c\u044f \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0438\u0437 \u0434\u0432\u0443\u0445 \u0447\u0430\u0441\u0442\u0435\u0439: \u041a\u0440\u0430\u0442\u043a\u043e\u0435 \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u0441\u0435\u0442\u0435\u0439 \u043f\u043e \u043e\u0431\u043d\u0430\u0440\u0443\u0436\u0435\u043d\u0438\u044e \u043e\u0431\u044a\u0435\u043a\u0442\u043e\u0432 \u043d\u0430 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0438 \u0438 \u0441\u0435\u0433\u043c\u0435\u043d\u0442\u0430\u0446\u0438\u0438 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439 \u0441 \u0441\u0430\u043c\u044b\u043c\u0438 \u043f\u043e\u043d\u044f\u0442\u043d\u044b\u043c\u0438 \u0434\u043b\u044f \u043c\u0435\u043d\u044f \u0441\u0441\u044b\u043b\u043a\u0430\u043c\u0438 \u043d\u0430 \u0440\u0435\u0441\u0443\u0440\u0441\u044b. \u0421\u0442\u0430\u0440\u0430\u043b\u0441\u044f \u0432\u044b\u0431\u0438\u0440\u0430\u0442\u044c \u0432\u0438\u0434\u0435\u043e \u043f\u043e\u044f\u0441\u043d\u0435\u043d\u0438\u044f \u0438 \u0436\u0435\u043b\u0430\u0442\u0435\u043b\u044c\u043d\u043e \u043d\u0430 \u0440\u0443\u0441\u0441\u043a\u043e\u043c \u044f\u0437\u044b\u043a\u0435. \u0412\u0442\u043e\u0440\u0430\u044f \u0447\u0430\u0441\u0442\u044c \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0432 \u043f\u043e\u043f\u044b\u0442\u043a\u0435 \u043e\u0441\u043e\u0437\u043d\u0430\u0442\u044c \u043d\u0430\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u044f \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u043d\u0435\u0439\u0440\u043e\u043d\u043d\u044b\u0445 \u0441\u0435\u0442\u0435\u0439. \u0418 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0439 \u043d\u0430 \u0438\u0445 \u043e\u0441\u043d\u043e\u0432\u0435. \u0420\u0438\u0441\u0443\u043d\u043e\u043a 1 \u2013 \u041f\u043e\u043d\u0438\u043c\u0430\u0442\u044c\" \/>\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\/et\/blog\/novosti-interneta\/nejroseti-kuda-eto-vse-dvizhetsya\" \/>\n\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 4.9.10\" \/>\n\t\t<meta property=\"og:locale\" content=\"et_EE\" \/>\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\u041d\u0435\u0439\u0440\u043e\u0441\u0435\u0442\u0438. \u041a\u0443\u0434\u0430 \u044d\u0442\u043e \u0432\u0441\u0435 \u0434\u0432\u0438\u0436\u0435\u0442\u0441\u044f | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u0421\u0442\u0430\u0442\u044c\u044f \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0438\u0437 \u0434\u0432\u0443\u0445 \u0447\u0430\u0441\u0442\u0435\u0439: \u041a\u0440\u0430\u0442\u043a\u043e\u0435 \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u0441\u0435\u0442\u0435\u0439 \u043f\u043e \u043e\u0431\u043d\u0430\u0440\u0443\u0436\u0435\u043d\u0438\u044e \u043e\u0431\u044a\u0435\u043a\u0442\u043e\u0432 \u043d\u0430 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0438 \u0438 \u0441\u0435\u0433\u043c\u0435\u043d\u0442\u0430\u0446\u0438\u0438 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439 \u0441 \u0441\u0430\u043c\u044b\u043c\u0438 \u043f\u043e\u043d\u044f\u0442\u043d\u044b\u043c\u0438 \u0434\u043b\u044f \u043c\u0435\u043d\u044f \u0441\u0441\u044b\u043b\u043a\u0430\u043c\u0438 \u043d\u0430 \u0440\u0435\u0441\u0443\u0440\u0441\u044b. \u0421\u0442\u0430\u0440\u0430\u043b\u0441\u044f \u0432\u044b\u0431\u0438\u0440\u0430\u0442\u044c \u0432\u0438\u0434\u0435\u043e \u043f\u043e\u044f\u0441\u043d\u0435\u043d\u0438\u044f \u0438 \u0436\u0435\u043b\u0430\u0442\u0435\u043b\u044c\u043d\u043e \u043d\u0430 \u0440\u0443\u0441\u0441\u043a\u043e\u043c \u044f\u0437\u044b\u043a\u0435. \u0412\u0442\u043e\u0440\u0430\u044f \u0447\u0430\u0441\u0442\u044c \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0432 \u043f\u043e\u043f\u044b\u0442\u043a\u0435 \u043e\u0441\u043e\u0437\u043d\u0430\u0442\u044c \u043d\u0430\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u044f \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u043d\u0435\u0439\u0440\u043e\u043d\u043d\u044b\u0445 \u0441\u0435\u0442\u0435\u0439. \u0418 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0439 \u043d\u0430 \u0438\u0445 \u043e\u0441\u043d\u043e\u0432\u0435. \u0420\u0438\u0441\u0443\u043d\u043e\u043a 1 \u2013 \u041f\u043e\u043d\u0438\u043c\u0430\u0442\u044c\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/et\/blog\/novosti-interneta\/nejroseti-kuda-eto-vse-dvizhetsya\" \/>\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=\"2020-01-03T21:00:00+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2020-02-18T11:02:50+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\udd47N\u00e4rviv\u00f5rgud. Kuhu see k\u00f5ik liigub | ProHoster","description":"Artikkel jaguneb kaheks osaks: \u00fclevaade m\u00f5nest arhitektuurist, mis on seotud objektide tuvastamise ja piltide segmentimisega, koos minule arusaadavamate allikatega. P\u00fc\u00fcdsin valida videot selgitusi, eelistatult vene keeles. Teine osa keskendub arusaamisele, millises suunas arenevad n\u00e4rviv\u00f5rgud ja nende aluseks olevad tehnoloogiad. Joonis 1 \u2013 Arusaamine","canonical_url":"https:\/\/prohoster.info\/et\/blog\/novosti-interneta\/nejroseti-kuda-eto-vse-dvizhetsya","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"et_EE","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\u041d\u0435\u0439\u0440\u043e\u0441\u0435\u0442\u0438. \u041a\u0443\u0434\u0430 \u044d\u0442\u043e \u0432\u0441\u0435 \u0434\u0432\u0438\u0436\u0435\u0442\u0441\u044f | ProHoster","og:description":"\u0421\u0442\u0430\u0442\u044c\u044f \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0438\u0437 \u0434\u0432\u0443\u0445 \u0447\u0430\u0441\u0442\u0435\u0439: \u041a\u0440\u0430\u0442\u043a\u043e\u0435 \u043e\u043f\u0438\u0441\u0430\u043d\u0438\u0435 \u043d\u0435\u043a\u043e\u0442\u043e\u0440\u044b\u0445 \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u0441\u0435\u0442\u0435\u0439 \u043f\u043e \u043e\u0431\u043d\u0430\u0440\u0443\u0436\u0435\u043d\u0438\u044e \u043e\u0431\u044a\u0435\u043a\u0442\u043e\u0432 \u043d\u0430 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0438 \u0438 \u0441\u0435\u0433\u043c\u0435\u043d\u0442\u0430\u0446\u0438\u0438 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u0439 \u0441 \u0441\u0430\u043c\u044b\u043c\u0438 \u043f\u043e\u043d\u044f\u0442\u043d\u044b\u043c\u0438 \u0434\u043b\u044f \u043c\u0435\u043d\u044f \u0441\u0441\u044b\u043b\u043a\u0430\u043c\u0438 \u043d\u0430 \u0440\u0435\u0441\u0443\u0440\u0441\u044b. \u0421\u0442\u0430\u0440\u0430\u043b\u0441\u044f \u0432\u044b\u0431\u0438\u0440\u0430\u0442\u044c \u0432\u0438\u0434\u0435\u043e \u043f\u043e\u044f\u0441\u043d\u0435\u043d\u0438\u044f \u0438 \u0436\u0435\u043b\u0430\u0442\u0435\u043b\u044c\u043d\u043e \u043d\u0430 \u0440\u0443\u0441\u0441\u043a\u043e\u043c \u044f\u0437\u044b\u043a\u0435. \u0412\u0442\u043e\u0440\u0430\u044f \u0447\u0430\u0441\u0442\u044c \u0441\u043e\u0441\u0442\u043e\u0438\u0442 \u0432 \u043f\u043e\u043f\u044b\u0442\u043a\u0435 \u043e\u0441\u043e\u0437\u043d\u0430\u0442\u044c \u043d\u0430\u043f\u0440\u0430\u0432\u043b\u0435\u043d\u0438\u0435 \u0440\u0430\u0437\u0432\u0438\u0442\u0438\u044f \u0430\u0440\u0445\u0438\u0442\u0435\u043a\u0442\u0443\u0440 \u043d\u0435\u0439\u0440\u043e\u043d\u043d\u044b\u0445 \u0441\u0435\u0442\u0435\u0439. \u0418 \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0439 \u043d\u0430 \u0438\u0445 \u043e\u0441\u043d\u043e\u0432\u0435. \u0420\u0438\u0441\u0443\u043d\u043e\u043a 1 \u2013 \u041f\u043e\u043d\u0438\u043c\u0430\u0442\u044c","og:url":"https:\/\/prohoster.info\/et\/blog\/novosti-interneta\/nejroseti-kuda-eto-vse-dvizhetsya","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":"2020-01-03T21:00:00+00:00","article:modified_time":"2020-02-18T11:02:50+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"54780","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":"2026-01-24 12:43:20","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-02-28 19:57:38","updated":"2026-01-24 12:43:20"},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/posts\/54780","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/comments?post=54780"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/posts\/54780\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/media?parent=54780"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/categories?post=54780"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/et\/wp-json\/wp\/v2\/tags?post=54780"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}