Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Model bileşenlerini device_map ile GPU'lar arasında dağıtın","local":"split-model-components-across-gpus-with-devicemap","sections":[{"title":"İsteği kodla","local":"encode-the-prompt","sections":[],"depth":2},{"title":"Parçalı dönüştürücü ile gürültü azaltma","local":"denoise-with-the-sharded-transformer","sections":[],"depth":2},{"title":"Latentleri çözümlen","local":"decode-the-latents","sections":[],"depth":2}],"depth":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/336.BxJyk7RE.js" rel="modulepreload"> | |
| <link href="/docs/diffusers/main/en/_app/immutable/chunks/DSPDl1gz.js" rel="modulepreload"> | |
| <!--lo1fle--><meta name="hf:doc:metadata" content="{"title":"Split model components across GPUs with device_map","local":"split-model-components-across-gpus-with-devicemap","sections":[{"title":"Encode the prompt","local":"encode-the-prompt","sections":[],"depth":2},{"title":"Denoise with the sharded transformer","local":"denoise-with-the-sharded-transformer","sections":[],"depth":2},{"title":"Decode the latents","local":"decode-the-latents","sections":[],"depth":2}],"depth":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="split-model-components-across-gpus-with-devicemap" 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="#split-model-components-across-gpus-with-devicemap"><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><!---->Model bileşenlerini device_map ile GPU'lar arasında dağıtın<!----></span></h1><!--]--><!----> <p><code>device_map</code> argümanı, model bileşenlerini ayrı GPU'lara yerleştirir. Bir model tek bir GPU'ya sığmıyorsa ve GPU'larınız hızlı bir bağlantıya sahip değilse bunu kullanın. Diğer stratejilerle karşılaştırmak için <a href="../training/distributed_inference">Distributed inference</a> bölümüne, <code>device_map</code> için daha fazla seçenek için ise <a href="./memory#device-placement">Device placement</a> bölümüne bakın.</p> <p>Aşağıdaki örnek, her bileşeni yalnızca ona ihtiyaç duyan aşamada yükler ve büyük olanları iki GPU arasında dağıtmak için <code>device_map</code> kullanır.</p> <!--[1--><h2 class="relative group"><a id="encode-the-prompt" 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="#encode-the-prompt"><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><!---->İsteği kodla<!----></span></h2><!--]--><!----> <p>Metin kodlayıcıları ayrı GPU'lara yerleştirmek için <code>device_map="balanced"</code> değerini ayarlayın. Her GPU'nun kullanabileceği bellek miktarını sınırlamak için <code>max_memory</code> parametresini kullanın. Yalnızca metin kodlayıcıların yüklenmesini sağlamak için <code>transformer</code> ve <code>vae</code> değerlerini <code>None</code> olarak ayarlayın.</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> Kopyalandı</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> FluxPipeline | |
| <span class="hljs-keyword">import</span> torch | |
| prompt = <span class="hljs-string">""" | |
| cinematic film still of a cat sipping a margarita in a pool in Palm Springs, California | |
| highly detailed, high budget hollywood movie, cinemascope, moody, epic, gorgeous, film grain | |
| """</span> | |
| pipeline = FluxPipeline.from_pretrained( | |
| <span class="hljs-string">"black-forest-labs/FLUX.1-dev"</span>, | |
| transformer=<span class="hljs-literal">None</span>, | |
| vae=<span class="hljs-literal">None</span>, | |
| device_map=<span class="hljs-string">"balanced"</span>, | |
| max_memory={<span class="hljs-number">0</span>: <span class="hljs-string">"16GB"</span>, <span class="hljs-number">1</span>: <span class="hljs-string">"16GB"</span>}, | |
| dtype=torch.bfloat16 | |
| ) | |
| <span class="hljs-keyword">with</span> torch.no_grad(): | |
| <span class="hljs-built_in">print</span>(<span class="hljs-string">"Encoding prompts."</span>) | |
| prompt_embeds, pooled_prompt_embeds, text_ids = pipeline.encode_prompt( | |
| prompt=prompt, prompt_2=<span class="hljs-literal">None</span>, max_sequence_length=<span class="hljs-number">512</span> | |
| )<!----></pre></div><!----> <p>Metin gömcelerine sahip olduğunuzda, transformer için yer açmak üzere metin kodlayıcıları GPU'dan kaldırın.</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> Kopyalandı</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">import</span> gc | |
| <span class="hljs-keyword">def</span> <span class="hljs-title function_">flush</span>(): | |
| gc.collect() | |
| torch.cuda.empty_cache() | |
| torch.cuda.reset_max_memory_allocated() | |
| torch.cuda.reset_peak_memory_stats() | |
| <span class="hljs-keyword">del</span> pipeline.text_encoder | |
| <span class="hljs-keyword">del</span> pipeline.text_encoder_2 | |
| <span class="hljs-keyword">del</span> pipeline.tokenizer | |
| <span class="hljs-keyword">del</span> pipeline.tokenizer_2 | |
| <span class="hljs-keyword">del</span> pipeline | |
| flush()<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="denoise-with-the-sharded-transformer" 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="#denoise-with-the-sharded-transformer"><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><!---->Parçalı dönüştürücü ile gürültü azaltma<!----></span></h2><!--]--><!----> <p><code>device_map="auto"</code> değerini ayarlayarak Accelerate'in dönüştürücüyü yerleştirmesini sağlayın. İlk GPU'yu doldurmadan diğerine geçmez ve yalnızca GPU'ların belleği tükendiğinde CPU veya diski kullanır. İki küçük GPU'da dönüştürücü, her ikisi arasında bölünmüş olarak konumlandırılır. CPU veya diskte bulunan ağırlıklar çıkarımı yavaşlatır.</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> Kopyalandı</div><!----></button><!----></div> <pre class="language-py "><!----><span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> AutoModel | |
| <span class="hljs-keyword">import</span> torch | |
| transformer = AutoModel.from_pretrained( | |
| <span class="hljs-string">"black-forest-labs/FLUX.1-dev"</span>, | |
| subfolder=<span class="hljs-string">"transformer"</span>, | |
| device_map=<span class="hljs-string">"auto"</span>, | |
| dtype=torch.bfloat16 | |
| )<!----></pre></div><!----> <blockquote class="tip"><p>Dönüştürücünün cihazlar arasında nasıl dağıtıldığını görmek için <code>transformer.hf_device_map</code> değerini kontrol edin. <code>device_map</code> ile yüklenen bir pipeline, her bileşenin nereye yerleştirildiğini gösteren bir <code>hf_device_map</code> özniteliğine de sahiptir.</p></blockquote> <p>Sadece dönüştürücüyle bir pipeline yükleyin ve <code>output_type="latent"</code> ayarlayın; böylece görseller yerine latentler döner.</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> Kopyalandı</div><!----></button><!----></div> <pre class="language-py "><!---->pipeline = FluxPipeline.from_pretrained( | |
| <span class="hljs-string">"black-forest-labs/FLUX.1-dev"</span>, | |
| text_encoder=<span class="hljs-literal">None</span>, | |
| text_encoder_2=<span class="hljs-literal">None</span>, | |
| tokenizer=<span class="hljs-literal">None</span>, | |
| tokenizer_2=<span class="hljs-literal">None</span>, | |
| vae=<span class="hljs-literal">None</span>, | |
| transformer=transformer, | |
| dtype=torch.bfloat16 | |
| ) | |
| <span class="hljs-built_in">print</span>(<span class="hljs-string">"Running denoising."</span>) | |
| height, width = <span class="hljs-number">768</span>, <span class="hljs-number">1360</span> | |
| latents = pipeline( | |
| prompt_embeds=prompt_embeds, | |
| pooled_prompt_embeds=pooled_prompt_embeds, | |
| num_inference_steps=<span class="hljs-number">50</span>, | |
| guidance_scale=<span class="hljs-number">3.5</span>, | |
| height=height, | |
| width=width, | |
| output_type=<span class="hljs-string">"latent"</span>, | |
| ).images<!----></pre></div><!----> <!--[1--><h2 class="relative group"><a id="decode-the-latents" 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="#decode-the-latents"><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><!---->Latentleri çözümlen<!----></span></h2><!--]--><!----> <p>Bellekten pipeline ve transformer'ı kaldırın ve latents'leri çözmek için bir VAE yükleyin. VAE, tek bir GPU'ya sığacak kadar küçüktür.</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> Kopyalandı</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> AutoencoderKL | |
| <span class="hljs-keyword">from</span> diffusers.image_processor <span class="hljs-keyword">import</span> VaeImageProcessor | |
| <span class="hljs-keyword">del</span> pipeline.transformer | |
| <span class="hljs-keyword">del</span> pipeline | |
| <span class="hljs-keyword">del</span> transformer | |
| flush() | |
| vae = AutoencoderKL.from_pretrained( | |
| <span class="hljs-string">"black-forest-labs/FLUX.1-dev"</span>, subfolder=<span class="hljs-string">"vae"</span>, dtype=torch.bfloat16 | |
| ).to(<span class="hljs-string">"cuda"</span>) <span class="hljs-comment"># or "mps", "xpu", "cpu"</span> | |
| vae_scale_factor = <span class="hljs-number">2</span> ** (<span class="hljs-built_in">len</span>(vae.config.block_out_channels) - <span class="hljs-number">1</span>) | |
| image_processor = VaeImageProcessor(vae_scale_factor=vae_scale_factor) | |
| <span class="hljs-keyword">with</span> torch.no_grad(): | |
| <span class="hljs-built_in">print</span>(<span class="hljs-string">"Running decoding."</span>) | |
| latents = FluxPipeline._unpack_latents(latents, height, width, vae_scale_factor) | |
| latents = (latents / vae.config.scaling_factor) + vae.config.shift_factor | |
| image = vae.decode(latents, return_dict=<span class="hljs-literal">False</span>)[<span class="hljs-number">0</span>] | |
| image = image_processor.postprocess(image, output_type=<span class="hljs-string">"pil"</span>) | |
| image[<span class="hljs-number">0</span>].save(<span class="hljs-string">"split_transformer.png"</span>)<!----></pre></div><!----> <p>Belirli bir aşamada ihtiyacınız olan modelleri seçici olarak yükleyip boşaltarak ve en büyük modelleri birden çok GPU arasında paylaştırarak, tüketici GPU'larında büyük modellerle çıkarım (inference) çalıştırabilirsiniz.</p> <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/optimization/device_map.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, 336], | |
| data: [null,null], | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 24.4 kB
- Xet hash:
- 3c1fa9b8625fc9e987609d7f0b4ef7cef4f85e31a1417ec51e930a29312e46d7
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.