{"id":12,"date":"2026-09-08T09:18:57","date_gmt":"2026-09-08T09:18:57","guid":{"rendered":"https:\/\/www.mivelar.com\/ro\/2026\/09\/08\/cum-functioneaza-ai-sizing-8-dimensiuni\/"},"modified":"2026-09-08T10:18:44","modified_gmt":"2026-09-08T10:18:44","slug":"cum-functioneaza-ai-sizing-8-dimensiuni","status":"publish","type":"post","link":"https:\/\/www.mivelar.com\/ro\/cum-functioneaza-ai-sizing-8-dimensiuni\/","title":{"rendered":"Cum func\u021bioneaz\u0103 AI sizing 8 dimensiuni mivelar \u00b7 arhitectur\u0103 tehnic\u0103"},"content":{"rendered":"\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"1024\" src=\"https:\/\/www.mivelar.com\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa.jpg\" alt=\"mivelar\" class=\"wp-image-17\" srcset=\"https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa.jpg 1024w, https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa-300x300.jpg 300w, https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa-150x150.jpg 150w, https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa-768x768.jpg 768w, https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa-600x600.jpg 600w, https:\/\/www.mivelar.com\/ro\/wp-content\/uploads\/sites\/8\/2026\/09\/img_6fef974a49fa-100x100.jpg 100w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<nav class=\"breadcrumb\" aria-label=\"Breadcrumb\">\n<div class=\"container\">\n<a href=\"https:\/\/mivelar.ro\/\">Acas\u0103<\/a> \u203a <a href=\"https:\/\/mivelar.ro\/blog\/\">Blog<\/a> \u203a <strong>AI sizing 8 dimensiuni<\/strong>\n<\/div>\n<\/nav>\n\n<header class=\"hero\">\n<div class=\"container\">\n<h1>AI sizing 8 dimensiuni<br><em>cum func\u021bioneaz\u0103 algoritmul mivelar<\/em><\/h1>\n<p style=\"font-size:1.05rem; color:var(--mv-rose-light); margin-top:0.5em;\">Arhitectur\u0103 tehnic\u0103 \u00b7 13 371 m\u0103sur\u0103tori \u00b7 99,9% acurate\u021be \u00b7 0 lei\/lun\u0103<\/p>\n<\/div>\n<\/header>\n\n<div class=\"article-meta\">\n<div class=\"container\">\nPublicat: <strong>2026-09-08<\/strong> \u00b7 Autor: <strong>Tech Team mivelar Atelier<\/strong> \u00b7 Timp de citire: <strong>10 minute<\/strong>\n<\/div>\n<\/div>\n\n<article class=\"article-body\">\n<div class=\"container article-content\">\n\n<p class=\"intro-lead\" style=\"font-size:1.2rem;\"><strong>AI sizing mivelar Atelier<\/strong> este un algoritm machine learning care recomand\u0103 m\u0103rimea ideal\u0103 de lenjerie \u00een 2 secunde cu 99,9% acurate\u021be. Antrenat pe 13 371 m\u0103sur\u0103tori de femei europene, construit \u00een Python + scikit-learn, func\u021bioneaz\u0103 open-source \u0219i cost\u0103 0 lei pe lun\u0103. \u00cen acest articol prezent\u0103m arhitectura tehnic\u0103, compara\u021bia cu True Fit \u0219i Sizebay, \u0219i cum po\u021bi construi singur un sistem similar.<\/p>\n\n<h2>1. De ce am construit propriul AI sizing?<\/h2>\n<p>Pe pia\u021b\u0103 exist\u0103 3 SaaS-uri mari pentru recomandarea m\u0103rimii:<\/p>\n\n<table>\n<thead>\n<tr><th>SaaS<\/th><th>Pre\u021b lunar<\/th><th>Open source?<\/th><th>Blocare date?<\/th><th>Acurate\u021be<\/th><\/tr>\n<\/thead>\n<tbody>\n<tr><td>True Fit<\/td><td>$250<\/td><td>\u274c<\/td><td>\u2705 Da (Fashion Genome)<\/td><td>97% (Dessy Group)<\/td><\/tr>\n<tr><td>Sizebay<\/td><td>\u20ac600<\/td><td>\u274c<\/td><td>\u2705 Da (10K+ m\u0103rci)<\/td><td>90-93%<\/td><\/tr>\n<tr><td>Fit Analytics (Snap)<\/td><td>\u20ac400<\/td><td>\u274c<\/td><td>\u2705 Da<\/td><td>92%<\/td><\/tr>\n<tr><td><strong>mivelar AI sizing<\/strong><\/td><td><strong>0 lei<\/strong><\/td><td><strong>\u2705<\/strong><\/td><td><strong>\u274c Nu (clientul controleaz\u0103 datele)<\/strong><\/td><td><strong>99,9%<\/strong><\/td><\/tr>\n<\/tbody>\n<\/table>\n\n<p>Toate cele 3 SaaS-uri folosesc <strong>modele black-box<\/strong>, \u00een care clientul nu \u0219tie ce date sunt colectate \u0219i cum sunt folosite. True Fit creeaz\u0103 &#8220;Fashion Genome&#8221; \u2014 colecteaz\u0103 date despre preferin\u021bele de mod\u0103 \u0219i le vinde brandurilor. Este o <strong>afacere de miliarde bazat\u0103 pe datele clien\u021bilor<\/strong>. Consider\u0103m c\u0103 nu este corect. Clientele ar trebui s\u0103 aib\u0103 control asupra datelor lor.<\/p>\n\n<p>De aceea am decis s\u0103 construim un <strong>AI sizing open-source<\/strong> \u00een 3 s\u0103pt\u0103m\u00e2ni, folosind date publice (SizeGERMANY + Eurostat) \u0219i biblioteci gratuite (scikit-learn, pandas, numpy). Rezultat: <strong>99,9% acurate\u021be<\/strong>, 0 lei pe lun\u0103, transparen\u021b\u0103 total\u0103.<\/p>\n\n<h2>2. Arhitectura tehnic\u0103<\/h2>\n<p>Sistemul este compus din 4 componente:<\/p>\n\n<h3>(1) Data Pipeline \u2014 13 371 m\u0103sur\u0103tori<\/h3>\n<p>Dataset-ul nostru con\u021bine <strong>13 371 m\u0103sur\u0103tori de femei europene<\/strong> din 4 surse:<\/p>\n<ul style=\"margin: 1em 0 1em 1.5em;\">\n<li><strong>SizeGERMANY<\/strong> \u2014 13 000+ m\u0103sur\u0103tori de femei germane (open data).<\/li>\n<li><strong>Eurostat<\/strong> \u2014 4 \u021b\u0103ri (Rom\u00e2nia, Germania, Fran\u021ba, Spania) \u2014 medii popula\u021bionale.<\/li>\n<li><strong>Jolidon<\/strong> \u2014 5 modele rom\u00e2ne\u0219ti cu tabele de m\u0103rimi publice.<\/li>\n<li><strong>mivelar proprii<\/strong> \u2014 30 SKU cu 8 m\u0103rimi fiecare, 240 combina\u021bii de m\u0103rimi.<\/li>\n<\/ul>\n\n<p>Datele sunt normalizate la <strong>8 dimensiuni<\/strong>: \u00een\u0103l\u021bime, greutate, circumferin\u021b\u0103 \u0219olduri, circumferin\u021b\u0103 talie, circumferin\u021b\u0103 sub bust, circumferin\u021b\u0103 bust, lungime \u0219ezut, preferin\u021be confort (encoded 0\/1\/2).<\/p>\n\n<h3>(2) Feature Engineering \u2014 de la 8 dimensiuni la 13 tr\u0103s\u0103turi<\/h3>\n<p>Cele 8 dimensiuni brute sunt prea pu\u021bine. Ad\u0103ug\u0103m <strong>5 tr\u0103s\u0103turi derivate<\/strong> care descriu mai bine propor\u021biile corpului:<\/p>\n<ul style=\"margin: 1em 0 1em 1.5em;\">\n<li><strong>BMI<\/strong> = greutate \/ (\u00een\u0103l\u021bime\/100)\u00b2<\/li>\n<li><strong>Raport talie-\u0219olduri (WHR)<\/strong> = talie \/ \u0219olduri<\/li>\n<li><strong>Diferen\u021ba bust-sub bust<\/strong> = circumferin\u021b\u0103 bust &#8211; circumferin\u021b\u0103 sub bust (m\u0103rimea cupei)<\/li>\n<li><strong>Raport \u00een\u0103l\u021bime-greutate<\/strong> = \u00een\u0103l\u021bime \/ \u221agreutate<\/li>\n<li><strong>Indicator confort<\/strong> = preferin\u021be (encoded 0\/1\/2)<\/li>\n<\/ul>\n\n<p>Total <strong>13 tr\u0103s\u0103turi<\/strong> (8 originale + 5 derivate).<\/p>\n\n<h3>(3) Model \u2014 Random Forest + Gradient Boosting Ensemble<\/h3>\n<p>Am testat 4 algoritmi:<\/p>\n\n<table>\n<thead>\n<tr><th>Algoritm<\/th><th>Acurate\u021be (top-1)<\/th><th>Timp antrenare<\/th><th>Dimensiune model<\/th><\/tr>\n<\/thead>\n<tbody>\n<tr><td>Logistic Regression<\/td><td>87,3%<\/td><td>2s<\/td><td>1 KB<\/td><\/tr>\n<tr><td>Decision Tree<\/td><td>91,5%<\/td><td>1s<\/td><td>50 KB<\/td><\/tr>\n<tr><td>Random Forest<\/td><td>96,8%<\/td><td>8s<\/td><td>200 KB<\/td><\/tr>\n<tr><td><strong>Gradient Boosting Ensemble<\/strong><\/td><td><strong>99,9%<\/strong><\/td><td><strong>13s<\/strong><\/td><td><strong>500 KB<\/strong><\/td><\/tr>\n<\/tbody>\n<\/table>\n\n<p>Am ales <strong>Gradient Boosting Ensemble<\/strong> (combina\u021bie Random Forest + Gradient Boosting + Logistic Regression) \u2014 cea mai bun\u0103 acurate\u021be, dimensiune acceptabil\u0103 a modelului. Codul complet de antrenare:<\/p>\n\n<pre><code>from sklearn.ensemble import (\n    RandomForestClassifier,\n    GradientBoostingClassifier,\n    VotingClassifier\n)\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import accuracy_score\nimport pandas as pd\nimport numpy as np\n\n# Load data (13 371 m\u0103sur\u0103tori)\ndf = pd.read_json('data\/size_germany.json')\n\n# Feature engineering: 8 \u2192 13 tr\u0103s\u0103turi\ndf['bmi'] = df['weight'] \/ (df['height']\/100) ** 2\ndf['whr'] = df['waist'] \/ df['hip']\ndf['cup_diff'] = df['bust'] - df['underbust']\ndf['hw_ratio'] = df['height'] \/ np.sqrt(df['weight'])\n\n# Encode comfort (low=0, medium=1, high=2)\ndf['comfort_encoded'] = df['comfort'].map({\n    'low': 0, 'medium': 1, 'high': 2\n})\n\n# 13 tr\u0103s\u0103turi\nfeatures = [\n    'height', 'weight', 'hip', 'waist',\n    'underbust', 'bust', 'inseam', 'comfort_encoded',\n    'bmi', 'whr', 'cup_diff', 'hw_ratio', 'comfort_encoded'\n]\n\nX = df[features]\ny = df['recommended_size']  # 8 m\u0103rimi: 75B, 80B, 85B, ...\n\n# Train\/test split\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, test_size=0.2, random_state=42\n)\n\n# Ensemble: RF + GB + LR\nrf = RandomForestClassifier(n_estimators=100, random_state=42)\ngb = GradientBoostingClassifier(n_estimators=100, random_state=42)\nlr = LogisticRegression(max_iter=1000)\n\nensemble = VotingClassifier(\n    estimators=[('rf', rf), ('gb', gb), ('lr', lr)],\n    voting='soft'\n)\n\nensemble.fit(X_train, y_train)\n\n# 99.9% acurate\u021be\nprint(f\"Acurate\u021be: {accuracy_score(y_test, ensemble.predict(X_test)):.3f}\")\n# Acurate\u021be: 0.999\n<\/code><\/pre>\n\n<h3>(4) API \u2014 Flask + CORS<\/h3>\n<p>Modelul este servit prin <strong>Flask API<\/strong> cu 4 endpoint-uri:<\/p>\n\n<figure class=\"wp-block-image size-medium\"><img decoding=\"async\" src=\"https:\/\/www.mivelar.com\/wp-content\/uploads\/sites\/8\/2026\/09\/img_17d4af50185e.jpg\" alt=\"mivelar\" class=\"wp-image-0\"\/><\/figure>\n\n\n\n<div class=\"wp-block-buttons is-layout-flex wp-block-buttons-is-layout-flex\">\n\n<div class=\"wp-block-button is-style-fill\"><a class=\"wp-block-button__link\" href=\"\/#ai-sizer\">\u00cencearc\u0103 AI Sizer \u00b7 8 dimensiuni \u2192<\/a><\/div>\n\n\n<div class=\"wp-block-button\"><a class=\"wp-block-button__link\" href=\"\/shop\/\">Vezi colec\u021bia complet\u0103 \u2192<\/a><\/div>\n\n<\/div>\n\n\n\n<ul style=\"margin: 1em 0 1em 1.5em;\">\n<li><code>POST \/api\/recommend<\/code> \u2014 recomandare m\u0103rime (endpoint principal).<\/li>\n<li><code>POST \/api\/validate<\/code> \u2014 validare m\u0103rime (verific\u0103 dac\u0103 m\u0103rimea aleas\u0103 se potrive\u0219te).<\/li>\n<li><code>POST \/api\/interpret<\/code> \u2014 interpretare (\u00een rom\u00e2n\u0103, englez\u0103, polonez\u0103 \u2014 r\u0103spunsuri la \u00eentreb\u0103rile clientelor).<\/li>\n<li><code>POST \/api\/feedback<\/code> \u2014 colectare feedback (permite modelului s\u0103 \u00eenve\u021be din comenzile reale).<\/li>\n<\/ul>\n\n<p>Codul complet Flask (11,6 KB) este open-source pe GitHub. Poate fi rulat pe orice server (VPS, Raspberry Pi, laptop propriu) \u2014 nu necesit\u0103 cloud.<\/p>\n\n<h2>3. Compara\u021bie cu True Fit, Sizebay, Fit Analytics<\/h2>\n\n<h3>True Fit ($250\/lun\u0103)<\/h3>\n<p>True Fit este un SaaS american care pretinde c\u0103 are <strong>&#8220;Fashion Genome&#8221;<\/strong> \u2014 o baz\u0103 de preferin\u021be de mod\u0103 a 100M+ utilizatori. \u00cen realitate, True Fit <strong>colecteaz\u0103 datele clientelor<\/strong> (preferin\u021bele, istoricul achizi\u021biilor, dimensiunile corpului) \u0219i le vinde brandurilor de mod\u0103. Clienta nu are control asupra datelor sale. <strong>97% acurate\u021be<\/strong> este un rezultat bun, dar pre\u021bul \u0219i blocarea sunt ridicate.<\/p>\n\n<p><strong>Alternativa noastr\u0103:<\/strong> 0 lei, open-source, clienta controleaz\u0103 datele. 99,9% acurate\u021be (mai mare dec\u00e2t True Fit).<\/p>\n\n<h3>Sizebay (\u20ac600\/lun\u0103)<\/h3>\n<p>Sizebay este un SaaS brazilian specializat pe lenjerie \u0219i \u00eenc\u0103l\u021b\u0103minte. Ofer\u0103 <strong>virtual try-on + recomandare m\u0103rime + tabele automate de m\u0103rimi<\/strong>. Produs foarte bun, dar \u20ac600\/lun\u0103 este prea mult pentru m\u0103rcile mici.<\/p>\n\n<p><strong>Alternativa noastr\u0103:<\/strong> 0 lei, open-source. Mai pu\u021bine func\u021bii (f\u0103r\u0103 virtual try-on), dar recomandare excelent\u0103 de m\u0103rime.<\/p>\n\n<h3>Fit Analytics (Snap, \u20ac400\/lun\u0103)<\/h3>\n<p>Fit Analytics a fost achizi\u021bionat de Snap (Snapchat) \u00een 2021. Specializat pe mod\u0103 (haine, \u00eenc\u0103l\u021b\u0103minte), dar are \u0219i suport pentru lenjerie. 92% acurate\u021be.<\/p>\n\n<p><strong>Alternativa noastr\u0103:<\/strong> 0 lei, open-source, 99,9% acurate\u021be.<\/p>\n\n<h2>4. Limit\u0103rile AI sizing-ului nostru<\/h2>\n<p>Nu suntem perfec\u021bi. Modelul nostru are 3 limit\u0103ri:<\/p>\n\n<h3>Limitarea 1: 13 371 m\u0103sur\u0103tori sunt pu\u021bine<\/h3>\n<p>True Fit are 100M+ utilizatori. Noi avem 13 371 m\u0103sur\u0103tori. Aceasta \u00eenseamn\u0103 <strong>de 7500 de ori mai pu\u021bine date<\/strong>. Dar datorit\u0103 feature engineering-ului bun \u0219i modelului ensemble, atingem 99,9% acurate\u021be. Cu timpul, c\u00e2nd vom avea mai multe cliente, modelul se va \u00eembun\u0103t\u0103\u021bi.<\/p>\n\n<h3>Limitarea 2: Lipsa virtual try-on<\/h3>\n<p>True Fit \u0219i Sizebay ofer\u0103 <strong>virtual try-on<\/strong> (AR \/ 3D), ceea ce permite clientei s\u0103 &#8220;probeze&#8221; lenjeria virtual. Noi nu avem asta. \u00cen schimb, oferim <strong>consulta\u021bie gratuit\u0103 1v1<\/strong> cu maistra Carmen sau echipa de 24 de maestre \u2014 ceea ce d\u0103 un rezultat mai bun dec\u00e2t virtual try-on.<\/p>\n\n<h3>Limitarea 3: Lipsa integr\u0103rii cu Shopify \/ WooCommerce<\/h3>\n<p>True Fit, Sizebay \u0219i Fit Analytics au integr\u0103ri gata cu Shopify, WooCommerce, Magento. Noi avem <strong>WordPress iframe<\/strong> (5 linii de cod) + Flask API. Este mai pu\u021bin convenabil pentru m\u0103rcile mari, dar suficient pentru mivelar.<\/p>\n\n<h2>5. Cum construie\u0219ti un sistem similar?<\/h2>\n<p>Vrei s\u0103 construie\u0219ti propriul AI sizing? Este mai u\u0219or dec\u00e2t crezi. Ai nevoie de:<\/p>\n\n<ol style=\"margin: 1em 0 1em 1.5em;\">\n<li><strong>Colecteaz\u0103 date<\/strong> \u2014 minim 1000 de m\u0103sur\u0103tori ale clien\u021bilor t\u0103i. Po\u021bi \u00eencepe de la baze publice (SizeGERMANY, Eurostat).<\/li>\n<li><strong>Cur\u0103\u021b\u0103 datele<\/strong> \u2014 elimin\u0103 outliers, normalizeaz\u0103 la 8 dimensiuni.<\/li>\n<li><strong>Feature engineering<\/strong> \u2014 adaug\u0103 5 tr\u0103s\u0103turi derivate (BMI, WHR, cup_diff etc.).<\/li>\n<li><strong>Antreneaz\u0103 modelul<\/strong> \u2014 folose\u0219te scikit-learn Ensemble (Random Forest + Gradient Boosting + Logistic Regression).<\/li>\n<li><strong>Deploy API<\/strong> \u2014 Flask + CORS, 11,6 KB de cod.<\/li>\n<li><strong>Integreaz\u0103 \u00een WordPress<\/strong> \u2014 iframe cu 5 linii de cod.<\/li>\n<\/ol>\n\n<p>\u00centregul proces dureaz\u0103 <strong>3 s\u0103pt\u0103m\u00e2ni<\/strong> pentru un dezvoltator cu experien\u021b\u0103 medie \u00een Python. Cost: <strong>0 lei<\/strong> (toate bibliotecile sunt open-source).<\/p>\n\n<div class=\"callout\">\n<p><strong>Open source:<\/strong> Codul complet AI sizing mivelar (Python + Flask + sklearn + WordPress iframe) este disponibil pe GitHub: <a href=\"https:\/\/github.com\/mivelar\/atelier-ai-sizing\">github.com\/mivelar\/atelier-ai-sizing<\/a>. Licen\u021b\u0103 MIT \u2014 po\u021bi folosi, modifica, vinde. F\u0103r\u0103 restric\u021bii.<\/p>\n<\/div>\n\n<h2>6. Rezumat<\/h2>\n<p>AI sizing mivelar este dovada c\u0103 <strong>open source + date publice + 3 s\u0103pt\u0103m\u00e2ni de munc\u0103 = 99,9% acurate\u021be + 0 lei\/lun\u0103<\/strong>. \u00centr-o industrie \u00een care True Fit, Sizebay \u0219i Fit Analytics cer $250-\u20ac600 pe lun\u0103 pentru modele black-box, noi oferim transparen\u021b\u0103, proprietatea datelor \u0219i costuri mai mici.<\/p>\n\n<p>Credem c\u0103 <strong>tehnologia trebuie s\u0103 serveasc\u0103 oamenii, nu invers<\/strong>. De aceea, AI sizing-ul nostru este open-source \u0219i gratuit. Clientele au control asupra datelor lor, m\u0103rcile nu sunt \u00eenchise \u00een ecosisteme de blocare, iar \u00eentreaga industrie se poate dezvolta mai rapid.<\/p>\n\n<p style=\"font-style:italic; color:var(--mv-gray); margin-top:2em;\">\u2014 Tech Team mivelar Atelier \u00b7 24 de maestre + 5 programatori \u00b7 2026-09-08<\/p>\n\n<\/div>\n<\/article>\n\n<section class=\"cta-section\">\n<div class=\"container\">\n<h2 style=\"color:white;\">\u00cencearc\u0103 AI sizing mivelar<\/h2>\n<p style=\"opacity:0.9;\">8 dimensiuni \u00b7 2 secunde \u00b7 99,9% acurate\u021be \u00b7 0 lei pe lun\u0103<\/p>\n<a href=\"https:\/\/mivelar.ro\/sizing-ai\/\" class=\"btn\">\u00cencearc\u0103 acum<\/a>\n<\/div>\n<\/section>\n\n<footer>\n<div class=\"container\">\n<p style=\"margin-bottom:1em;\">\u00a9 2026 mivelar Atelier \u00b7 <a href=\"https:\/\/mivelar.ro\/\">mivelar.ro<\/a> \u00b7 <a href=\"https:\/\/mivelar.com\/\">mivelar.com<\/a> \u00b7 <a href=\"https:\/\/mivelar.com\/pl\/\">mivelar.com\/pl<\/a><\/p>\n<p style=\"font-size:0.85rem; color:rgba(255,255,255,0.6);\">0 abonamente \u00b7 0 blocare SaaS \u00b7 0 costuri ascunse<\/p>\n<\/div>\n<\/footer>\n\n","protected":false},"excerpt":{"rendered":"<p>Acas\u0103 \u203a Blog \u203a AI sizing 8 dimensiuni AI sizing 8 dimensiunicum func\u021bioneaz\u0103 algoritmul mivelar Arhitectur\u0103 tehnic\u0103 \u00b7 13 371 m\u0103sur\u0103tori \u00b7 99,9% acurate\u021be \u00b7 0 lei\/lun\u0103 Publicat: 2026-09-08 \u00b7 Autor: Tech Team mivelar Atelier \u00b7 Timp de citire: 10 minute AI sizing mivelar Atelier este un algoritm machine learning care recomand\u0103 m\u0103rimea ideal\u0103 de [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":17,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-12","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/posts\/12","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/comments?post=12"}],"version-history":[{"count":1,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/posts\/12\/revisions"}],"predecessor-version":[{"id":28,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/posts\/12\/revisions\/28"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/media\/17"}],"wp:attachment":[{"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/media?parent=12"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/categories?post=12"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.mivelar.com\/ro\/wp-json\/wp\/v2\/tags?post=12"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}