Hugging Face's logo

Buckets:
hf-doc-build
/
doc

Files
xet
hf-doc-build/doc / diffusers /main /es /using-diffusers /custom_pipeline_overview.html
mishig's picture
mishig
download
raw
34.7 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Pipelines y componentes de la comunidad&quot;,&quot;local&quot;:&quot;community-pipelines-and-components&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Carga de un pipeline de la comunidad&quot;,&quot;local&quot;:&quot;loading-a-community-pipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Componentes de la comunidad&quot;,&quot;local&quot;:&quot;community-components&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Próximos pasos&quot;,&quot;local&quot;:&quot;next-steps&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/>
<link href="/docs/diffusers/main/en/_app/immutable/entry/start.0S0_iEoG.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/I9eooN77.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/Bupkjr9f.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/entry/app.CPdOxCbw.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/CyvZ3FS5.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/l0QnnOED.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/DnVIJfC5.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/nodes/0.DteTvmQZ.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/CoSrdQax.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/nodes/392.Cp1y09dz.js" rel="modulepreload">
<link href="/docs/diffusers/main/en/_app/immutable/chunks/DSPDl1gz.js" rel="modulepreload">
<!--1nactrq--><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Community pipelines and components&quot;,&quot;local&quot;:&quot;community-pipelines-and-components&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Loading a community pipeline&quot;,&quot;local&quot;:&quot;loading-a-community-pipeline&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Community components&quot;,&quot;local&quot;:&quot;community-components&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Next steps&quot;,&quot;local&quot;:&quot;next-steps&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"/><!---->
<link href="/docs/diffusers/main/en/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: inline-end; margin-inline-start: 10px; display: inline-flex; position: relative; z-index: 10;"><div dir="ltr" class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm 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 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" 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 dir="auto">Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 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 sm:size-3.5 size-3 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> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="community-pipelines-and-components" 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="#community-pipelines-and-components"><span><svg 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><!---->Pipelines y componentes de la comunidad<!----></span></h1><!--]--><!----> <p>Los pipelines de la comunidad son clases <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline">DiffusionPipeline</a> que difieren de la implementación del artículo original. Proporcionan funcionalidad adicional o extienden la implementación del pipeline original.</p> <blockquote class="tip"><p>Echa un vistazo a los pipelines de la comunidad en <a href="https://github.com/huggingface/diffusers/tree/main/examples/community" rel="nofollow">diffusers/examples/community</a>, que incluyen ejemplos de inferencia y entrenamiento sobre cómo utilizarlos.</p></blockquote> <p>Los pipelines de la comunidad se almacenan en el Hub o en el repositorio de GitHub de Diffusers. Los pipelines del Hub son completamente personalizables (scheduler, modelos, código del pipeline, etc.), mientras que los pipelines de GitHub están limitados únicamente al código del pipeline personalizado. Compare los dos tipos de pipeline en la tabla siguiente.</p> <table><thead><tr><th></th><th>GitHub</th><th>Hub</th></tr></thead><tbody><tr><td>Uso</td><td><code>custom_pipeline="&lt;filename_without_.py>"</code></td><td>ID del repositorio de Hub o <code>trust_remote_code=True</code> cuando <code>_class_name</code> apunta a código personalizado</td></tr><tr><td>Proceso de revisión</td><td>Abre una Pull Request en GitHub y sométela a un proceso de revisión por parte del equipo de Diffusers antes de fusionarla. Esta opción es más lenta.</td><td>Sube directamente a un repositorio de Hub sin revisión. Esta es la opción más rápida.</td></tr><tr><td>Visibilidad</td><td>Incluido en el repositorio y la documentación oficiales de Diffusers.</td><td>Incluido en tu perfil de Hub y depende de tu propio uso y promoción para ganar visibilidad.</td></tr></tbody></table> <!--[1--><h2 class="relative group"><a id="loading-a-community-pipeline" 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="#loading-a-community-pipeline"><span><svg 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><!---->Carga de un pipeline de la comunidad<!----></span></h2><!--]--><!----> <p>Cargue cualquier tipo de pipeline de la comunidad pasando <code>custom_pipeline</code> a <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained">from_pretrained()</a>.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
<span class="hljs-string">&quot;stabilityai/stable-diffusion-3-medium-diffusers&quot;</span>,
custom_pipeline=<span class="hljs-string">&quot;pipeline_stable_diffusion_3_instruct_pix2pix&quot;</span>,
dtype=torch.float16,
device_map=<span class="hljs-string">&quot;cuda&quot;</span> <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
)<!----></pre></div><!----> <p>Agrega el argumento <code>custom_revision</code> a <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained">from_pretrained()</a> para cargar un pipeline de la comunidad desde una versión específica (por ejemplo, <code>v0.30.0</code> o <code>main</code>). De forma predeterminada, los pipelines de la comunidad se cargan desde la última versión estable de Diffusers.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
<span class="hljs-string">&quot;stabilityai/stable-diffusion-3-medium-diffusers&quot;</span>,
custom_pipeline=<span class="hljs-string">&quot;pipeline_stable_diffusion_3_instruct_pix2pix&quot;</span>,
custom_revision=<span class="hljs-string">&quot;main&quot;</span>,
dtype=torch.float16,
device_map=<span class="hljs-string">&quot;cuda&quot;</span> <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
)<!----></pre></div><!----> <blockquote class="warning"><p>Mientras el Hugging Face Hub <a href="https://huggingface.co/docs/hub/security-malware" rel="nofollow">analiza</a> los archivos, debes inspeccionar el código de la pipeline del Hub y asegurarte de que sea seguro.</p></blockquote> <p>Hay varias formas de cargar un pipeline de la comunidad.</p> <ul><li><p>Pase una ruta a <code>custom_pipeline</code> para cargar un pipeline de la comunidad local. El directorio debe contener un archivo <code>pipeline.py</code> que contenga la clase del pipeline.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
<span class="hljs-string">&quot;stabilityai/stable-diffusion-3-medium-diffusers&quot;</span>,
custom_pipeline=<span class="hljs-string">&quot;path/to/pipeline_directory&quot;</span>,
trust_remote_code=<span class="hljs-literal">True</span>,
dtype=torch.float16,
device_map=<span class="hljs-string">&quot;cuda&quot;</span> <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
)<!----></pre></div><!----></li> <li><p>El argumento <code>custom_pipeline</code> también es compatible con <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pipe">from_pipe()</a>, lo cual resulta útil para <a href="./loading#reusing-models-in-multiple-pipelines">reutilizar pipelines</a> sin utilizar memoria adicional. Limita el uso de memoria a solo el pipeline más grande cargado.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
pipeline_sd = DiffusionPipeline.from_pretrained(
<span class="hljs-string">&quot;emilianJR/CyberRealistic_V3&quot;</span>, dtype=torch.float16, device_map=<span class="hljs-string">&quot;cuda&quot;</span> <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span>
)
pipeline_lpw = DiffusionPipeline.from_pipe(
pipeline_sd, custom_pipeline=<span class="hljs-string">&quot;lpw_stable_diffusion&quot;</span>
)<!----></pre></div><!----> <p>El método <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pipe">from_pipe()</a> resulta especialmente útil para cargar pipelines de la comunidad, ya que muchos de ellos no tienen pesos preentrenados. Los pipelines de la comunidad generalmente añaden una funcionalidad sobre un pipeline existente.</p></li></ul> <!--[1--><h2 class="relative group"><a id="community-components" 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="#community-components"><span><svg 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><!---->Componentes de la comunidad<!----></span></h2><!--]--><!----> <p>Los componentes de la comunidad permiten crear pipelines con transformadores, UNets, VAEs y programadores personalizados no admitidos por Diffusers. Estos componentes requieren sus propios módulos de Python.</p> <p>Esta sección muestra cómo los usuarios pueden utilizar componentes de la comunidad para construir un pipeline de la comunidad, utilizando <a href="https://huggingface.co/showlab/show-1-base" rel="nofollow">showlab/show-1-base</a> como ejemplo.</p> <ol><li>Cargue los componentes necesarios, el programador y el procesador de imágenes. El codificador de texto generalmente se importa desde <a href="https://huggingface.co/docs/transformers/index" rel="nofollow">Transformers</a>.</li></ol> <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 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> Copiado</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> T5Tokenizer, T5EncoderModel, CLIPImageProcessor
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DPMSolverMultistepScheduler
pipeline_id = <span class="hljs-string">&quot;showlab/show-1-base&quot;</span>
tokenizer = T5Tokenizer.from_pretrained(pipeline_id, subfolder=<span class="hljs-string">&quot;tokenizer&quot;</span>)
text_encoder = T5EncoderModel.from_pretrained(pipeline_id, subfolder=<span class="hljs-string">&quot;text_encoder&quot;</span>)
scheduler = DPMSolverMultistepScheduler.from_pretrained(pipeline_id, subfolder=<span class="hljs-string">&quot;scheduler&quot;</span>)
feature_extractor = CLIPImageProcessor.from_pretrained(pipeline_id, subfolder=<span class="hljs-string">&quot;feature_extractor&quot;</span>)<!----></pre></div><!----> <blockquote class="warning"><p>En los pasos 2 y 3, la implementación personalizada de <a href="https://github.com/showlab/Show-1/blob/main/showone/models/unet_3d_condition.py" rel="nofollow">UNet</a> y <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/blob/main/unet/showone_unet_3d_condition.py" rel="nofollow">pipeline</a> debe coincidir con el formato mostrado en sus archivos para que este ejemplo funcione.</p></blockquote> <ol start="2"><li>Carga un <a href="https://github.com/showlab/Show-1/blob/main/showone/models/unet_3d_condition.py" rel="nofollow">UNet personalizado</a> que ya está implementado en <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/blob/main/unet/showone_unet_3d_condition.py" rel="nofollow">showone_unet_3d_condition.py</a>. El nombre de la clase <a href="/docs/diffusers/main/es/api/models/unet3d-cond#diffusers.UNet3DConditionModel">UNet3DConditionModel</a> se cambia a la implementación personalizada, <code>ShowOneUNet3DConditionModel</code>, porque <a href="/docs/diffusers/main/es/api/models/unet3d-cond#diffusers.UNet3DConditionModel">UNet3DConditionModel</a> ya existe en Diffusers. Todos los componentes necesarios para la clase <code>ShowOneUNet3DConditionModel</code> deben colocarse en <code>showone_unet_3d_condition.py</code>.</li></ol> <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 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> Copiado</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">from</span> showone_unet_3d_condition <span class="hljs-keyword">import</span> ShowOneUNet3DConditionModel
unet = ShowOneUNet3DConditionModel.from_pretrained(pipeline_id, subfolder=<span class="hljs-string">&quot;unet&quot;</span>)<!----></pre></div><!----> <ol start="3"><li>Cargue el código de la pipeline personalizada (ya implementado en <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/blob/main/pipeline_t2v_base_pixel.py" rel="nofollow">pipeline_t2v_base_pixel.py</a>). Este script contiene una clase personalizada <code>TextToVideoIFPipeline</code> para generar videos a partir de texto. Al igual que con la UNet personalizada, cualquier código necesario para <code>TextToVideoIFPipeline</code> debe colocarse en <code>pipeline_t2v_base_pixel.py</code>.</li></ol> <p>Inicializa <code>TextToVideoIFPipeline</code> con <code>ShowOneUNet3DConditionModel</code>.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> pipeline_t2v_base_pixel <span class="hljs-keyword">import</span> TextToVideoIFPipeline
pipeline = TextToVideoIFPipeline(
unet=unet,
text_encoder=text_encoder,
tokenizer=tokenizer,
scheduler=scheduler,
feature_extractor=feature_extractor,
)
pipeline = pipeline.to(<span class="hljs-string">&quot;cuda&quot;</span>) <span class="hljs-comment"># or &quot;mps&quot;, &quot;xpu&quot;, &quot;cpu&quot;</span><!----></pre></div><!----> <ol start="4"><li>Sube el pipeline al Hub para compartirlo con la comunidad.</li></ol> <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 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> Copiado</div><!----></button><!----></div> <pre class="language-python "><!---->pipeline.push_to_hub(<span class="hljs-string">&quot;custom-t2v-pipeline&quot;</span>)<!----></pre></div><!----> <p>Después de que el pipeline se haya subido correctamente, realiza los siguientes cambios.</p> <ul><li>Cambia el atributo <code>_class_name</code> en <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/blob/main/model_index.json#L2" rel="nofollow">model_index.json</a> a <code>["pipeline_t2v_base_pixel", "TextToVideoIFPipeline"]</code>.</li> <li>Sube <code>showone_unet_3d_condition.py</code> a la subcarpeta <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/blob/main/unet/showone_unet_3d_condition.py" rel="nofollow">unet</a>.</li> <li>Sube <code>pipeline_t2v_base_pixel.py</code> al <a href="https://huggingface.co/sayakpaul/show-1-base-with-code/tree/main" rel="nofollow">repositorio</a> de la pipeline.</li></ul> <p>Para ejecutar la inferencia, añade el argumento <code>trust_remote_code</code> al inicializar el pipeline para manejar toda la “magia” detrás de escena.</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 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> Copiado</div><!----></button><!----></div> <pre class="language-python "><!----><span class="hljs-keyword">import</span> torch
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> DiffusionPipeline
pipeline = DiffusionPipeline.from_pretrained(
<span class="hljs-string">&quot;&lt;change-username&gt;/&lt;change-id&gt;&quot;</span>, trust_remote_code=<span class="hljs-literal">True</span>, dtype=torch.float16
)<!----></pre></div><!----> <blockquote class="warning"><p>Como precaución adicional con <code>trust_remote_code=True</code>, pase un hash de commit al argumento <code>revision</code> en <a href="/docs/diffusers/main/es/api/pipelines/overview#diffusers.DiffusionPipeline.from_pretrained">from_pretrained()</a> para asegurarse de que el código no se haya actualizado con nuevo código malicioso (a menos que confíe por completo en los propietarios del modelo).</p></blockquote> <!--[1--><h2 class="relative group"><a id="next-steps" 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="#next-steps"><span><svg 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><!---->Próximos pasos<!----></span></h2><!--]--><!----> <ul><li>Echa un vistazo a la Issue <a href="https://github.com/huggingface/diffusers/issues/841" rel="nofollow">#841</a> para obtener más contexto sobre por qué estamos añadiendo pipelines de la comunidad para ayudar a todos a compartir su trabajo fácilmente sin que se vean frenados.</li> <li>Echa un vistazo al repositorio <a href="https://huggingface.co/stabilityai/japanese-stable-diffusion-xl/" rel="nofollow">stabilityai/japanese-stable-diffusion-xl</a> para un ejemplo adicional de un pipeline de la comunidad que también utiliza la función <code>trust_remote_code</code>.</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/en/using-diffusers/custom_pipeline_overview.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><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]-->
<script>
{
__sveltekit_1qsdbwf = {
base: "/docs/diffusers/main/en",
assets: "/docs/diffusers/main/en"
};
const element = document.currentScript.parentElement;
Promise.all([
import("/docs/diffusers/main/en/_app/immutable/entry/start.0S0_iEoG.js"),
import("/docs/diffusers/main/en/_app/immutable/entry/app.CPdOxCbw.js")
]).then(([kit, app]) => {
kit.start(app, element, {
node_ids: [0, 392],
data: [null,null],
form: null,
error: null
});
});
}
</script>

Xet Storage Details

Size:
34.7 kB
·
Xet hash:
b63e77f299a3ac00a212a5d171d6f2f6867590df9d1f6a9d29bb430baeb14ad8

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