diffusers/v0.18.2/ko/optimization/onnx.html

<meta charset="utf-8" /><meta http-equiv="content-security-policy" content=""><meta name="hf:doc:metadata" content="{&quot;local&quot;:&quot;onnx&quot;,&quot;sections&quot;:[{&quot;local&quot;:&quot;&quot;,&quot;title&quot;:&quot;설치&quot;},{&quot;local&quot;:&quot;stable-diffusion&quot;,&quot;title&quot;:&quot;Stable Diffusion 추론&quot;},{&quot;local&quot;:&quot;&quot;,&quot;title&quot;:&quot;알려진 이슈들&quot;}],&quot;title&quot;:&quot;추론을 위해 ONNX 런타임을 사용하는 방법&quot;}" data-svelte="svelte-1phssyn">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/assets/pages/__layout.svelte-hf-doc-builder.css">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/start-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/chunks/vendor-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/chunks/paths-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/pages/__layout.svelte-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/pages/optimization/onnx.mdx-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/chunks/IconCopyLink-hf-doc-builder.js">
	<link rel="modulepreload" href="/docs/diffusers/v0.18.2/ko/_app/chunks/CodeBlock-hf-doc-builder.js"> 






<h1 class="relative group"><a id="onnx" 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="#onnx"><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>추론을 위해 ONNX 런타임을 사용하는 방법
	</span></h1>

<p>🤗 Diffusers는 ONNX Runtime과 호환되는 Stable Diffusion 파이프라인을 제공합니다. 이를 통해 ONNX(CPU 포함)를 지원하고 PyTorch의 가속 버전을 사용할 수 없는 모든 하드웨어에서 Stable Diffusion을 실행할 수 있습니다.</p>
<h2 id="">설치</h2>
<p>다음 명령어로 ONNX Runtime를 지원하는 🤗 Optimum를 설치합니다:</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><!-- HTML_TAG_START -->pip <span class="hljs-keyword">install</span> optimum[<span class="hljs-string">&quot;onnxruntime&quot;</span>]<!-- HTML_TAG_END --></pre></div>
<h2 class="relative group"><a id="stable-diffusion" 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="#stable-diffusion"><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>Stable Diffusion 추론
	</span></h2>

<p>아래 코드는 ONNX 런타임을 사용하는 방법을 보여줍니다. <code>StableDiffusionPipeline</code> 대신 <code>OnnxStableDiffusionPipeline</code>을 사용해야 합니다.
PyTorch 모델을 불러오고 즉시 ONNX 형식으로 변환하려는 경우 <code>export=True</code>로 설정합니다.</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><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> optimum.onnxruntime <span class="hljs-keyword">import</span> ORTStableDiffusionPipeline

model_id = <span class="hljs-string">&quot;runwayml/stable-diffusion-v1-5&quot;</span>
pipe = ORTStableDiffusionPipeline.from_pretrained(model_id, export=<span class="hljs-literal">True</span>)
prompt = <span class="hljs-string">&quot;a photo of an astronaut riding a horse on mars&quot;</span>
images = pipe(prompt).images[<span class="hljs-number">0</span>]
pipe.save_pretrained(<span class="hljs-string">&quot;./onnx-stable-diffusion-v1-5&quot;</span>)<!-- HTML_TAG_END --></pre></div>
<p>파이프라인을 ONNX 형식으로 오프라인으로 내보내고 나중에 추론에 사용하려는 경우,
<a href="https://huggingface.co/docs/optimum/main/en/exporters/onnx/usage_guides/export_a_model#exporting-a-model-to-onnx-using-the-cli" rel="nofollow"><code>optimum-cli export</code></a> 명령어를 사용할 수 있습니다:</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><!-- HTML_TAG_START -->optimum-cli <span class="hljs-built_in">export</span> onnx --model runwayml/stable-diffusion-v1-5 sd_v15_onnx/<!-- HTML_TAG_END --></pre></div>
<p>그 다음 추론을 수행합니다:</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><!-- HTML_TAG_START --><span class="hljs-keyword">from</span> optimum.onnxruntime <span class="hljs-keyword">import</span> ORTStableDiffusionPipeline

model_id = <span class="hljs-string">&quot;sd_v15_onnx&quot;</span>
pipe = ORTStableDiffusionPipeline.from_pretrained(model_id)
prompt = <span class="hljs-string">&quot;a photo of an astronaut riding a horse on mars&quot;</span>
images = pipe(prompt).images[<span class="hljs-number">0</span>]<!-- HTML_TAG_END --></pre></div>
<p>Notice that we didn’t have to specify <code>export=True</code> above.</p>
<p><a href="https://huggingface.co/docs/optimum/" rel="nofollow">Optimum 문서</a>에서 더 많은 예시를 찾을 수 있습니다.</p>
<h2 id="">알려진 이슈들</h2>
<ul><li>여러 프롬프트를 배치로 생성하면 너무 많은 메모리가 사용되는 것 같습니다. 이를 조사하는 동안, 배치 대신 반복 방법이 필요할 수도 있습니다.</li></ul>


		<script type="module" data-hydrate="1ge1mui">
		import { start } from "/docs/diffusers/v0.18.2/ko/_app/start-hf-doc-builder.js";
		start({
			target: document.querySelector('[data-hydrate="1ge1mui"]').parentNode,
			paths: {"base":"/docs/diffusers/v0.18.2/ko","assets":"/docs/diffusers/v0.18.2/ko"},
			session: {},
			route: false,
			spa: false,
			trailing_slash: "never",
			hydrate: {
				status: 200,
				error: null,
				nodes: [
					import("/docs/diffusers/v0.18.2/ko/_app/pages/__layout.svelte-hf-doc-builder.js"),
						import("/docs/diffusers/v0.18.2/ko/_app/pages/optimization/onnx.mdx-hf-doc-builder.js")
				],
				params: {}
			}
		});
	</script>