Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / transformers /pr_33913 /ko /gguf.html

rtrm

about 1 month ago

download

raw

16.8 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"GGUF와 Transformers의 상호작용","local":"gguf-and-interaction-with-transformers","sections":[{"title":"Transformers 내 지원","local":"support-within-transformers","sections":[{"title":"지원되는 양자화 유형","local":"supported-quantization-types","sections":[],"depth":3},{"title":"지원되는 모델 아키텍처","local":"supported-model-architectures","sections":[],"depth":3}],"depth":2},{"title":"사용 예시","local":"example-usage","sections":[],"depth":2}],"depth":1}">
	<link href="/docs/transformers/pr_33913/ko/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/entry/start.2d0ec41c.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/scheduler.bdbef820.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/singletons.665fa29a.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/index.8a885b74.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/paths.47fdf6cc.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/entry/app.e5c46936.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/index.33f81d56.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/nodes/0.7e28132d.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/each.e59479a4.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/nodes/20.0fa4cc38.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/CodeBlock.362b34a4.js">
	<link rel="modulepreload" href="/docs/transformers/pr_33913/ko/_app/immutable/chunks/EditOnGithub.a9246e21.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"GGUF와 Transformers의 상호작용","local":"gguf-and-interaction-with-transformers","sections":[{"title":"Transformers 내 지원","local":"support-within-transformers","sections":[{"title":"지원되는 양자화 유형","local":"supported-quantization-types","sections":[],"depth":3},{"title":"지원되는 모델 아키텍처","local":"supported-model-architectures","sections":[],"depth":3}],"depth":2},{"title":"사용 예시","local":"example-usage","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="gguf-and-interaction-with-transformers" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#gguf-and-interaction-with-transformers"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>GGUF와 Transformers의 상호작용</span></h1> <p data-svelte-h="svelte-11f2wai">GGUF 파일 형식은 <a href="https://github.com/ggerganov/ggml" rel="nofollow">GGML</a>과 그에 의존하는 다른 라이브러리, 예를 들어 매우 인기 있는 <a href="https://github.com/ggerganov/llama.cpp" rel="nofollow">llama.cpp</a>이나 <a href="https://github.com/ggerganov/whisper.cpp" rel="nofollow">whisper.cpp</a>에서 추론을 위한 모델을 저장하는데 사용됩니다.</p> <p data-svelte-h="svelte-19nl15r">이 파일 형식은 <a href="https://huggingface.co/docs/hub/en/gguf" rel="nofollow">Hugging Face Hub</a>에서 지원되며, 파일 내의 텐서와 메타데이터를 신속하게 검사할 수 있는 기능을 제공합니다.</p> <p data-svelte-h="svelte-2q169p">이 형식은 “단일 파일 형식(single-file-format)“으로 설계되었으며, 하나의 파일에 설정 속성, 토크나이저 어휘, 기타 속성뿐만 아니라 모델에서 로드되는 모든 텐서가 포함됩니다. 이 파일들은 파일의 양자화 유형에 따라 다른 형식으로 제공됩니다. 다양한 양자화 유형에 대한 간략한 설명은 <a href="https://huggingface.co/docs/hub/en/gguf#quantization-types" rel="nofollow">여기</a>에서 확인할 수 있습니다.</p> <h2 class="relative group"><a id="support-within-transformers" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#support-within-transformers"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>Transformers 내 지원</span></h2> <p data-svelte-h="svelte-rt99ms"><code>transformers</code> 내에서 <code>gguf</code> 파일을 로드할 수 있는 기능을 추가하여 GGUF 모델의 추가 학습/미세 조정을 제공한 후 <code>ggml</code> 생태계에서 다시 사용할 수 있도록 <code>gguf</code> 파일로 변환하는 기능을 제공합니다. 모델을 로드할 때 먼저 FP32로 역양자화한 후, PyTorch에서 사용할 수 있도록 가중치를 로드합니다.</p> <blockquote data-svelte-h="svelte-1itg8a0"><p>[!NOTE]
	지원은 아직 초기 단계에 있으며, 다양한 양자화 유형과 모델 아키텍처에 대해 이를 강화하기 위한 기여를 환영합니다.</p></blockquote> <p data-svelte-h="svelte-1hwmv9b">현재 지원되는 모델 아키텍처와 양자화 유형은 다음과 같습니다:</p> <h3 class="relative group"><a id="supported-quantization-types" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#supported-quantization-types"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>지원되는 양자화 유형</span></h3> <p data-svelte-h="svelte-1irch4b">초기에 지원되는 양자화 유형은 Hub에서 공유된 인기 있는 양자화 파일에 따라 결정되었습니다.</p> <ul data-svelte-h="svelte-1f29qgp"><li>F32</li> <li>F16</li> <li>BF16</li> <li>Q4_0</li> <li>Q4_1</li> <li>Q5_0</li> <li>Q5_1</li> <li>Q8_0</li> <li>Q2_K</li> <li>Q3_K</li> <li>Q4_K</li> <li>Q5_K</li> <li>Q6_K</li> <li>IQ1_S</li> <li>IQ1_M</li> <li>IQ2_XXS</li> <li>IQ2_XS</li> <li>IQ2_S</li> <li>IQ3_XXS</li> <li>IQ3_S</li> <li>IQ4_XS</li> <li>IQ4_NL</li></ul> <blockquote data-svelte-h="svelte-27l9s6"><p>[!NOTE]
	GGUF 역양자화를 지원하려면 <code>gguf>=0.10.0</code> 설치가 필요합니다.</p></blockquote> <h3 class="relative group"><a id="supported-model-architectures" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#supported-model-architectures"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>지원되는 모델 아키텍처</span></h3> <p data-svelte-h="svelte-knydm">현재 지원되는 모델 아키텍처는 Hub에서 매우 인기가 많은 아키텍처들로 제한되어 있습니다:</p> <ul data-svelte-h="svelte-gmjjmd"><li>LLaMa</li> <li>Mistral</li> <li>Qwen2</li> <li>Qwen2Moe</li> <li>Phi3</li> <li>Bloom</li></ul> <h2 class="relative group"><a id="example-usage" class="header-link block pr-1.5 text-lg no-hover:hidden with-hover:absolute with-hover:p-1.5 with-hover:opacity-0 with-hover:group-hover:opacity-100 with-hover:right-full" href="#example-usage"><span><svg class="" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 256"><path d="M167.594 88.393a8.001 8.001 0 0 1 0 11.314l-67.882 67.882a8 8 0 1 1-11.314-11.315l67.882-67.881a8.003 8.003 0 0 1 11.314 0zm-28.287 84.86l-28.284 28.284a40 40 0 0 1-56.567-56.567l28.284-28.284a8 8 0 0 0-11.315-11.315l-28.284 28.284a56 56 0 0 0 79.196 79.197l28.285-28.285a8 8 0 1 0-11.315-11.314zM212.852 43.14a56.002 56.002 0 0 0-79.196 0l-28.284 28.284a8 8 0 1 0 11.314 11.314l28.284-28.284a40 40 0 0 1 56.568 56.567l-28.285 28.285a8 8 0 0 0 11.315 11.314l28.284-28.284a56.065 56.065 0 0 0 0-79.196z" fill="currentColor"></path></svg></span></a> <span>사용 예시</span></h2> <p data-svelte-h="svelte-ha7vjv"><code>transformers</code>에서 <code>gguf</code> 파일을 로드하려면 <code>from_pretrained</code> 메소드에 <code>gguf_file</code> 인수를 지정해야 합니다. 동일한 파일에서 토크나이저와 모델을 로드하는 방법은 다음과 같습니다:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoTokenizer, AutoModelForCausalLM

	model_id = <span class="hljs-string">"TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF"</span>
	filename = <span class="hljs-string">"tinyllama-1.1b-chat-v1.0.Q6_K.gguf"</span>

	tokenizer = AutoTokenizer.from_pretrained(model_id, gguf_file=filename)
	model = AutoModelForCausalLM.from_pretrained(model_id, gguf_file=filename)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-55k75b">이제 PyTorch 생태계에서 모델의 양자화되지 않은 전체 버전에 접근할 수 있으며, 다른 여러 도구들과 결합하여 사용할 수 있습니다.</p> <p data-svelte-h="svelte-13eqcmr"><code>gguf</code> 파일로 다시 변환하려면 llama.cpp의 <a href="https://github.com/ggerganov/llama.cpp/blob/master/convert-hf-to-gguf.py" rel="nofollow"><code>convert-hf-to-gguf.py</code></a>를 사용하는 것을 권장합니다.</p> <p data-svelte-h="svelte-1ppx0pk">위의 스크립트를 완료하여 모델을 저장하고 다시 <code>gguf</code>로 내보내는 방법은 다음과 같습니다:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->tokenizer.save_pretrained(<span class="hljs-string">'directory'</span>)
	model.save_pretrained(<span class="hljs-string">'directory'</span>)

	!python ${path_to_llama_cpp}/convert-hf-to-gguf.py ${directory}<!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/ko/gguf.md" target="_blank"><span data-svelte-h="svelte-1kd6by1"><</span> <span data-svelte-h="svelte-x0xyl0">></span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a> <p></p>

	<script>
	{
	__sveltekit_1wfirp1 = {
	assets: "/docs/transformers/pr_33913/ko",
	base: "/docs/transformers/pr_33913/ko",
	env: {}
	};

	const element = document.currentScript.parentElement;

	const data = [null,null];

	Promise.all([
	import("/docs/transformers/pr_33913/ko/_app/immutable/entry/start.2d0ec41c.js"),
	import("/docs/transformers/pr_33913/ko/_app/immutable/entry/app.e5c46936.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 20],
	data,
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 16.8 kB
Xet hash:: e381cacb84878d5782c1f621e28efcd7724de0824c518fed28caa9fee73ee4ea

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.