Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"효과적이고 효율적인 Diffusion","local":"효과적이고-효율적인-diffusion","sections":[{"title":"속도","local":"속도","sections":[],"depth":2},{"title":"메모리","local":"메모리","sections":[],"depth":2},{"title":"품질","local":"품질","sections":[{"title":"더 나은 체크포인트","local":"더-나은-체크포인트","sections":[],"depth":3},{"title":"더 나은 파이프라인 구성 요소","local":"더-나은-파이프라인-구성-요소","sections":[],"depth":3},{"title":"더 나은 프롬프트 엔지니어링","local":"더-나은-프롬프트-엔지니어링","sections":[],"depth":3}],"depth":2},{"title":"다음 단계","local":"다음-단계","sections":[],"depth":2}],"depth":1}"> | |
| <link href="/docs/diffusers/pr_11739/ko/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/entry/start.3a826cf6.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/scheduler.23542ac5.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/singletons.c67d8ae0.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/index.7166da64.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/paths.f1d82941.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/entry/app.694c4502.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/preload-helper.84b08082.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/index.9b1f405b.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/nodes/0.556378bd.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/each.e59479a4.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/nodes/20.54bbd1c3.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.99afe1ab.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/CodeBlock.1c9c6858.js"> | |
| <link rel="modulepreload" href="/docs/diffusers/pr_11739/ko/_app/immutable/chunks/DocNotebookDropdown.68a629d2.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"효과적이고 효율적인 Diffusion","local":"효과적이고-효율적인-diffusion","sections":[{"title":"속도","local":"속도","sections":[],"depth":2},{"title":"메모리","local":"메모리","sections":[],"depth":2},{"title":"품질","local":"품질","sections":[{"title":"더 나은 체크포인트","local":"더-나은-체크포인트","sections":[],"depth":3},{"title":"더 나은 파이프라인 구성 요소","local":"더-나은-파이프라인-구성-요소","sections":[],"depth":3},{"title":"더 나은 프롬프트 엔지니어링","local":"더-나은-프롬프트-엔지니어링","sections":[],"depth":3}],"depth":2},{"title":"다음 단계","local":"다음-단계","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 max-sm:gap-0.5 h-6 max-sm:h-5 px-2 max-sm:px-1.5 text-[11px] max-sm:text-[9px] font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0"><svg class="w-3 h-3 max-sm:w-2.5 max-sm:h-2.5" 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></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-6 max-sm:h-5 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible w-3 h-3 max-sm:w-2.5 max-sm:h-2.5 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg></button></div> </div> <div class="flex space-x-1 " style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"> <div class="relative colab-dropdown "> <button class=" " type="button"> <img alt="Open In Colab" class="!m-0" src="https://colab.research.google.com/assets/colab-badge.svg"> </button> </div> <div class="relative colab-dropdown "> <button class=" " type="button"> <img alt="Open In Studio Lab" class="!m-0" src="https://studiolab.sagemaker.aws/studiolab.svg"> </button> </div></div> <h1 class="relative group"><a id="효과적이고-효율적인-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="#효과적이고-효율적인-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>효과적이고 효율적인 Diffusion</span></h1> <p data-svelte-h="svelte-18ayv9y">특정 스타일로 이미지를 생성하거나 원하는 내용을 포함하도록<code>DiffusionPipeline</code>을 설정하는 것은 까다로울 수 있습니다. 종종 만족스러운 이미지를 얻기까지 <code>DiffusionPipeline</code>을 여러 번 실행해야 하는 경우가 많습니다. 그러나 무에서 유를 창조하는 것은 특히 추론을 반복해서 실행하는 경우 계산 집약적인 프로세스입니다.</p> <p data-svelte-h="svelte-oyyvon">그렇기 때문에 파이프라인에서 <em>계산</em>(속도) 및 <em>메모리</em>(GPU RAM) 효율성을 극대화하여 추론 주기 사이의 시간을 단축하여 더 빠르게 반복할 수 있도록 하는 것이 중요합니다.</p> <p data-svelte-h="svelte-1enrctx">이 튜토리얼에서는 <code>DiffusionPipeline</code>을 사용하여 더 빠르고 효과적으로 생성하는 방법을 안내합니다.</p> <p data-svelte-h="svelte-1oj4hh8"><a href="https://huggingface.co/stable-diffusion-v1-5/stable-diffusion-v1-5" rel="nofollow"><code>stable-diffusion-v1-5/stable-diffusion-v1-5</code></a> 모델을 불러와서 시작합니다:</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> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline | |
| model_id = <span class="hljs-string">"stable-diffusion-v1-5/stable-diffusion-v1-5"</span> | |
| pipeline = DiffusionPipeline.from_pretrained(model_id)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-ssflff">예제 프롬프트는 “portrait of an old warrior chief” 이지만, 자유롭게 자신만의 프롬프트를 사용해도 됩니다:</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 -->prompt = <span class="hljs-string">"portrait photo of a old warrior chief"</span><!-- HTML_TAG_END --></pre></div> <h2 class="relative group"><a id="속도" 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="#속도"><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>속도</span></h2> <blockquote class="tip" data-svelte-h="svelte-1sothi9"><p>💡 GPU에 액세스할 수 없는 경우 다음과 같은 GPU 제공업체에서 무료로 사용할 수 있습니다!. <a href="https://colab.research.google.com/" rel="nofollow">Colab</a></p></blockquote> <p data-svelte-h="svelte-1rfznlg">추론 속도를 높이는 가장 간단한 방법 중 하나는 Pytorch 모듈을 사용할 때와 같은 방식으로 GPU에 파이프라인을 배치하는 것입니다:</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 -->pipeline = pipeline.to(<span class="hljs-string">"cuda"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-qirl6n">동일한 이미지를 사용하고 개선할 수 있는지 확인하려면 <a href="https://pytorch.org/docs/stable/generated/torch.Generator.html" rel="nofollow"><code>Generator</code></a>를 사용하고 <a href="./using-diffusers/reusing_seeds">재현성</a>에 대한 시드를 설정하세요:</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">import</span> torch | |
| generator = torch.Generator(<span class="hljs-string">"cuda"</span>).manual_seed(<span class="hljs-number">0</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-8nqpod">이제 이미지를 생성할 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 class=""><!-- HTML_TAG_START -->image = pipeline(prompt, generator=generator).images[<span class="hljs-number">0</span>] | |
| image<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-1of5nwm"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_1.png"></div> <p data-svelte-h="svelte-dt1sqn">이 프로세스는 T4 GPU에서 약 30초가 소요되었습니다(할당된 GPU가 T4보다 나은 경우 더 빠를 수 있음). 기본적으로 <code>DiffusionPipeline</code>은 50개의 추론 단계에 대해 전체 <code>float32</code> 정밀도로 추론을 실행합니다. <code>float16</code>과 같은 더 낮은 정밀도로 전환하거나 추론 단계를 더 적게 실행하여 속도를 높일 수 있습니다.</p> <p data-svelte-h="svelte-1jr1trd"><code>float16</code>으로 모델을 로드하고 이미지를 생성해 보겠습니다:</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">import</span> torch | |
| pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16) | |
| pipeline = pipeline.to(<span class="hljs-string">"cuda"</span>) | |
| generator = torch.Generator(<span class="hljs-string">"cuda"</span>).manual_seed(<span class="hljs-number">0</span>) | |
| image = pipeline(prompt, generator=generator).images[<span class="hljs-number">0</span>] | |
| image<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-1why3l7"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_2.png"></div> <p data-svelte-h="svelte-1gv5jsc">이번에는 이미지를 생성하는 데 약 11초밖에 걸리지 않아 이전보다 3배 가까이 빨라졌습니다!</p> <blockquote class="tip" data-svelte-h="svelte-1v23von"><p>💡 파이프라인은 항상 <code>float16</code>에서 실행할 것을 강력히 권장하며, 지금까지 출력 품질이 저하되는 경우는 거의 없었습니다.</p></blockquote> <p data-svelte-h="svelte-a5gi7u">또 다른 옵션은 추론 단계의 수를 줄이는 것입니다. 보다 효율적인 스케줄러를 선택하면 출력 품질 저하 없이 단계 수를 줄이는 데 도움이 될 수 있습니다. 현재 모델과 호환되는 스케줄러는 <code>compatibles</code> 메서드를 호출하여 <code>DiffusionPipeline</code>에서 찾을 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 class=""><!-- HTML_TAG_START -->pipeline.scheduler.compatibles | |
| [ | |
| diffusers.schedulers.scheduling_lms_discrete.LMSDiscreteScheduler, | |
| diffusers.schedulers.scheduling_unipc_multistep.UniPCMultistepScheduler, | |
| diffusers.schedulers.scheduling_k_dpm_2_discrete.KDPM2DiscreteScheduler, | |
| diffusers.schedulers.scheduling_deis_multistep.DEISMultistepScheduler, | |
| diffusers.schedulers.scheduling_euler_discrete.EulerDiscreteScheduler, | |
| diffusers.schedulers.scheduling_dpmsolver_multistep.DPMSolverMultistepScheduler, | |
| diffusers.schedulers.scheduling_ddpm.DDPMScheduler, | |
| diffusers.schedulers.scheduling_dpmsolver_singlestep.DPMSolverSinglestepScheduler, | |
| diffusers.schedulers.scheduling_k_dpm_2_ancestral_discrete.KDPM2AncestralDiscreteScheduler, | |
| diffusers.schedulers.scheduling_heun_discrete.HeunDiscreteScheduler, | |
| diffusers.schedulers.scheduling_pndm.PNDMScheduler, | |
| diffusers.schedulers.scheduling_euler_ancestral_discrete.EulerAncestralDiscreteScheduler, | |
| diffusers.schedulers.scheduling_ddim.DDIMScheduler, | |
| ]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-gjxj1i">Stable Diffusion 모델은 일반적으로 약 50개의 추론 단계가 필요한 <code>PNDMScheduler</code>를 기본으로 사용하지만, <code>DPMSolverMultistepScheduler</code>와 같이 성능이 더 뛰어난 스케줄러는 약 20개 또는 25개의 추론 단계만 필요로 합니다. 새 스케줄러를 로드하려면 <code>ConfigMixin.from_config()</code> 메서드를 사용합니다:</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> diffusers <span class="hljs-keyword">import</span> DPMSolverMultistepScheduler | |
| pipeline.scheduler = DPMSolverMultistepScheduler.from_config(pipeline.scheduler.config)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-f76kxz"><code>num_inference_steps</code>를 20으로 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 class=""><!-- HTML_TAG_START -->generator = torch.Generator(<span class="hljs-string">"cuda"</span>).manual_seed(<span class="hljs-number">0</span>) | |
| image = pipeline(prompt, generator=generator, num_inference_steps=<span class="hljs-number">20</span>).images[<span class="hljs-number">0</span>] | |
| image<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-19w49w4"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_3.png"></div> <p data-svelte-h="svelte-gcc2pb">추론시간을 4초로 단축할 수 있었습니다! ⚡️</p> <h2 class="relative group"><a id="메모리" 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="#메모리"><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>메모리</span></h2> <p data-svelte-h="svelte-nqgs99">파이프라인 성능 향상의 또 다른 핵심은 메모리 사용량을 줄이는 것인데, 초당 생성되는 이미지 수를 최대화하려고 하는 경우가 많기 때문에 간접적으로 더 빠른 속도를 의미합니다. 한 번에 생성할 수 있는 이미지 수를 확인하는 가장 쉬운 방법은 <code>OutOfMemoryError</code>(OOM)이 발생할 때까지 다양한 배치 크기를 시도해 보는 것입니다.</p> <p data-svelte-h="svelte-j2noup">프롬프트 목록과 <code>Generators</code>에서 이미지 배치를 생성하는 함수를 만듭니다. 좋은 결과를 생성하는 경우 재사용할 수 있도록 각 <code>Generator</code>에 시드를 할당해야 합니다.</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">def</span> <span class="hljs-title function_">get_inputs</span>(<span class="hljs-params">batch_size=<span class="hljs-number">1</span></span>): | |
| generator = [torch.Generator(<span class="hljs-string">"cuda"</span>).manual_seed(i) <span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(batch_size)] | |
| prompts = batch_size * [prompt] | |
| num_inference_steps = <span class="hljs-number">20</span> | |
| <span class="hljs-keyword">return</span> {<span class="hljs-string">"prompt"</span>: prompts, <span class="hljs-string">"generator"</span>: generator, <span class="hljs-string">"num_inference_steps"</span>: num_inference_steps}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-wyajyh">또한 각 이미지 배치를 보여주는 기능이 필요합니다:</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> PIL <span class="hljs-keyword">import</span> Image | |
| <span class="hljs-keyword">def</span> <span class="hljs-title function_">image_grid</span>(<span class="hljs-params">imgs, rows=<span class="hljs-number">2</span>, cols=<span class="hljs-number">2</span></span>): | |
| w, h = imgs[<span class="hljs-number">0</span>].size | |
| grid = Image.new(<span class="hljs-string">"RGB"</span>, size=(cols * w, rows * h)) | |
| <span class="hljs-keyword">for</span> i, img <span class="hljs-keyword">in</span> <span class="hljs-built_in">enumerate</span>(imgs): | |
| grid.paste(img, box=(i % cols * w, i // cols * h)) | |
| <span class="hljs-keyword">return</span> grid<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-ubrl9r"><code>batch_size=4</code>부터 시작해 얼마나 많은 메모리를 소비했는지 확인합니다:</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 -->images = pipeline(**get_inputs(batch_size=<span class="hljs-number">4</span>)).images | |
| image_grid(images)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-i1t4xo">RAM이 더 많은 GPU가 아니라면 위의 코드에서 <code>OOM</code> 오류가 반환되었을 것입니다! 대부분의 메모리는 cross-attention 레이어가 차지합니다. 이 작업을 배치로 실행하는 대신 순차적으로 실행하면 상당한 양의 메모리를 절약할 수 있습니다. 파이프라인을 구성하여 <code>enable_attention_slicing()</code> 함수를 사용하기만 하면 됩니다:</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 -->pipeline.enable_attention_slicing()<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-u177c0">이제 <code>batch_size</code>를 8로 늘려보세요!</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 -->images = pipeline(**get_inputs(batch_size=<span class="hljs-number">8</span>)).images | |
| image_grid(images, rows=<span class="hljs-number">2</span>, cols=<span class="hljs-number">4</span>)<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-vxa9bu"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_5.png"></div> <p data-svelte-h="svelte-1nhne3o">이전에는 4개의 이미지를 배치로 생성할 수도 없었지만, 이제는 이미지당 약 3.5초 만에 8개의 이미지를 배치로 생성할 수 있습니다! 이는 아마도 품질 저하 없이 T4 GPU에서 가장 빠른 속도일 것입니다.</p> <h2 class="relative group"><a id="품질" 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="#품질"><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>품질</span></h2> <p data-svelte-h="svelte-yzzx1p">지난 두 섹션에서는 <code>fp16</code>을 사용하여 파이프라인의 속도를 최적화하고, 더 성능이 좋은 스케줄러를 사용하여 추론 단계의 수를 줄이고, attention slicing을 활성화하여 메모리 소비를 줄이는 방법을 배웠습니다. 이제 생성된 이미지의 품질을 개선하는 방법에 대해 집중적으로 알아보겠습니다.</p> <h3 class="relative group"><a id="더-나은-체크포인트" 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="#더-나은-체크포인트"><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>더 나은 체크포인트</span></h3> <p data-svelte-h="svelte-97fik6">가장 확실한 단계는 더 나은 체크포인트를 사용하는 것입니다. Stable Diffusion 모델은 좋은 출발점이며, 공식 출시 이후 몇 가지 개선된 버전도 출시되었습니다. 하지만 최신 버전을 사용한다고 해서 자동으로 더 나은 결과를 얻을 수 있는 것은 아닙니다. 여전히 다양한 체크포인트를 직접 실험해보고, <a href="https://minimaxir.com/2022/11/stable-diffusion-negative-prompt/" rel="nofollow">negative prompts</a> 사용 등 약간의 조사를 통해 최상의 결과를 얻어야 합니다.</p> <p data-svelte-h="svelte-1pk1tvb">이 분야가 성장함에 따라 특정 스타일을 연출할 수 있도록 세밀하게 조정된 고품질 체크포인트가 점점 더 많아지고 있습니다. <a href="https://huggingface.co/models?library=diffusers&sort=downloads" rel="nofollow">Hub</a>와 <a href="https://huggingface.co/spaces/huggingface-projects/diffusers-gallery" rel="nofollow">Diffusers Gallery</a>를 둘러보고 관심 있는 것을 찾아보세요!</p> <h3 class="relative group"><a id="더-나은-파이프라인-구성-요소" 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="#더-나은-파이프라인-구성-요소"><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>더 나은 파이프라인 구성 요소</span></h3> <p data-svelte-h="svelte-3m63vd">현재 파이프라인 구성 요소를 최신 버전으로 교체해 볼 수도 있습니다. Stability AI의 최신 <a href="https://huggingface.co/stabilityai/stable-diffusion-2-1/tree/main/vae" rel="nofollow">autodecoder</a>를 파이프라인에 로드하고 몇 가지 이미지를 생성해 보겠습니다:</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> diffusers <span class="hljs-keyword">import</span> AutoencoderKL | |
| vae = AutoencoderKL.from_pretrained(<span class="hljs-string">"stabilityai/sd-vae-ft-mse"</span>, torch_dtype=torch.float16).to(<span class="hljs-string">"cuda"</span>) | |
| pipeline.vae = vae | |
| images = pipeline(**get_inputs(batch_size=<span class="hljs-number">8</span>)).images | |
| image_grid(images, rows=<span class="hljs-number">2</span>, cols=<span class="hljs-number">4</span>)<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-610rhb"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_6.png"></div> <h3 class="relative group"><a id="더-나은-프롬프트-엔지니어링" 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="#더-나은-프롬프트-엔지니어링"><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>더 나은 프롬프트 엔지니어링</span></h3> <p data-svelte-h="svelte-15owmr0">이미지를 생성하는 데 사용하는 텍스트 프롬프트는 <em>prompt engineering</em>이라고 할 정도로 매우 중요합니다. 프롬프트 엔지니어링 시 고려해야 할 몇 가지 사항은 다음과 같습니다:</p> <ul data-svelte-h="svelte-1y16yq3"><li>생성하려는 이미지 또는 유사한 이미지가 인터넷에 어떻게 저장되어 있는가?</li> <li>내가 원하는 스타일로 모델을 유도하기 위해 어떤 추가 세부 정보를 제공할 수 있는가?</li></ul> <p data-svelte-h="svelte-17wtiaq">이를 염두에 두고 색상과 더 높은 품질의 디테일을 포함하도록 프롬프트를 개선해 봅시다:</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 -->prompt += <span class="hljs-string">", tribal panther make up, blue on red, side profile, looking away, serious eyes"</span> | |
| prompt += <span class="hljs-string">" 50mm portrait photography, hard rim lighting photography--beta --ar 2:3 --beta --upbeta"</span><!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-8cwtf">새로운 프롬프트로 이미지 배치를 생성합니다:</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 -->images = pipeline(**get_inputs(batch_size=<span class="hljs-number">8</span>)).images | |
| image_grid(images, rows=<span class="hljs-number">2</span>, cols=<span class="hljs-number">4</span>)<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-n1o5lk"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_7.png"></div> <p data-svelte-h="svelte-1a7wb1e">꽤 인상적입니다! <code>1</code>의 시드를 가진 <code>Generator</code>에 해당하는 두 번째 이미지에 피사체의 나이에 대한 텍스트를 추가하여 조금 더 조정해 보겠습니다:</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 -->prompts = [ | |
| <span class="hljs-string">"portrait photo of the oldest warrior chief, tribal panther make up, blue on red, side profile, looking away, serious eyes 50mm portrait photography, hard rim lighting photography--beta --ar 2:3 --beta --upbeta"</span>, | |
| <span class="hljs-string">"portrait photo of a old warrior chief, tribal panther make up, blue on red, side profile, looking away, serious eyes 50mm portrait photography, hard rim lighting photography--beta --ar 2:3 --beta --upbeta"</span>, | |
| <span class="hljs-string">"portrait photo of a warrior chief, tribal panther make up, blue on red, side profile, looking away, serious eyes 50mm portrait photography, hard rim lighting photography--beta --ar 2:3 --beta --upbeta"</span>, | |
| <span class="hljs-string">"portrait photo of a young warrior chief, tribal panther make up, blue on red, side profile, looking away, serious eyes 50mm portrait photography, hard rim lighting photography--beta --ar 2:3 --beta --upbeta"</span>, | |
| ] | |
| generator = [torch.Generator(<span class="hljs-string">"cuda"</span>).manual_seed(<span class="hljs-number">1</span>) <span class="hljs-keyword">for</span> _ <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-built_in">len</span>(prompts))] | |
| images = pipeline(prompt=prompts, generator=generator, num_inference_steps=<span class="hljs-number">25</span>).images | |
| image_grid(images)<!-- HTML_TAG_END --></pre></div> <div class="flex justify-center" data-svelte-h="svelte-1lkw2bx"><img src="https://huggingface.co/datasets/diffusers/docs-images/resolve/main/stable_diffusion_101/sd_101_8.png"></div> <h2 class="relative group"><a id="다음-단계" 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="#다음-단계"><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>다음 단계</span></h2> <p data-svelte-h="svelte-10t9jtg">이 튜토리얼에서는 계산 및 메모리 효율을 높이고 생성된 출력의 품질을 개선하기 위해 <code>DiffusionPipeline</code>을 최적화하는 방법을 배웠습니다. 파이프라인을 더 빠르게 만드는 데 관심이 있다면 다음 리소스를 살펴보세요:</p> <ul data-svelte-h="svelte-1xpkcqm"><li><a href="./optimization/torch2.0">PyTorch 2.0</a> 및 <a href="https://pytorch.org/docs/stable/generated/torch.compile.html" rel="nofollow"><code>torch.compile</code></a>이 어떻게 추론 속도를 5~300% 향상시킬 수 있는지 알아보세요. A100 GPU에서는 추론 속도가 최대 50%까지 빨라질 수 있습니다!</li> <li>PyTorch 2를 사용할 수 없는 경우, <a href="./optimization/xformers">xFormers</a>를 설치하는 것이 좋습니다. 메모리 효율적인 어텐션 메커니즘은 PyTorch 1.13.1과 함께 사용하면 속도가 빨라지고 메모리 소비가 줄어듭니다.</li> <li>모델 오프로딩과 같은 다른 최적화 기법은 <a href="./optimization/fp16">이 가이드</a>에서 다루고 있습니다.</li></ul> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/diffusers/blob/main/docs/source/ko/stable_diffusion.md" target="_blank"><svg class="mr-1" 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="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg> <span data-svelte-h="svelte-zjs2n5"><span class="underline">Update</span> on GitHub</span></a> <p></p> | |
| <script> | |
| { | |
| __sveltekit_18rq40q = { | |
| assets: "/docs/diffusers/pr_11739/ko", | |
| base: "/docs/diffusers/pr_11739/ko", | |
| env: {} | |
| }; | |
| const element = document.currentScript.parentElement; | |
| const data = [null,null]; | |
| Promise.all([ | |
| import("/docs/diffusers/pr_11739/ko/_app/immutable/entry/start.3a826cf6.js"), | |
| import("/docs/diffusers/pr_11739/ko/_app/immutable/entry/app.694c4502.js") | |
| ]).then(([kit, app]) => { | |
| kit.start(app, element, { | |
| node_ids: [0, 20], | |
| data, | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 57.3 kB
- Xet hash:
- 8d7105ab4cc12d015fec7918967a08f3317095695799502d58a8bcd6f7bda0bb
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.