<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Huggingface on [ MECANIK DEV ]</title><link>https://mecanik.dev/ko/tags/huggingface/</link><description>Recent content in Huggingface on [ MECANIK DEV ]</description><generator>Hugo</generator><language>ko</language><copyright>저작권 © 2020-{year} [MECANIK DEV]. 모든 권리 보유.</copyright><lastBuildDate>Thu, 07 Aug 2025 20:00:00 +0100</lastBuildDate><atom:link href="https://mecanik.dev/ko/tags/huggingface/index.xml" rel="self" type="application/rss+xml"/><item><title>Tiny BPE Trainer – 빠르고 가벼운 C++ 기반 BPE 트레이너</title><link>https://mecanik.dev/ko/posts/tiny-bpe-trainer-a-fast-and-lightweight-bpe-trainer-in-c++/</link><pubDate>Thu, 07 Aug 2025 20:00:00 +0100</pubDate><guid>https://mecanik.dev/ko/posts/tiny-bpe-trainer-a-fast-and-lightweight-bpe-trainer-in-c++/</guid><description>&lt;h2 id="tiny-bpe-trainer-소개">Tiny BPE Trainer 소개&lt;/h2>
&lt;p>GPT부터 RoBERTa까지 오늘날 대부분의 최신 NLP 모델은 **바이트 쌍 인코딩(BPE)**을 사용한 하위 단어 토큰화에 의존합니다. 하지만 순수 C++로 자신만의 어휘를 학습시키고 싶다면 어떻게 해야 할까요?&lt;/p></description></item></channel></rss>