diffusers/pr_12509/zh/optimization/cache.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;缓存&quot;,&quot;local&quot;:&quot;缓存&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;金字塔注意力广播&quot;,&quot;local&quot;:&quot;金字塔注意力广播&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;FasterCache&quot;,&quot;local&quot;:&quot;fastercache&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}">
		<link href="/docs/diffusers/pr_12509/zh/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/entry/start.636dab22.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/scheduler.e4ff9b64.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/singletons.348a6756.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/index.f9be34a7.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/paths.03439956.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/entry/app.c3d2b8c4.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/preload-helper.02e582cb.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/index.09f1bca0.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/nodes/0.5565e359.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/nodes/22.e57f731f.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.115c2aca.js">
		<link rel="modulepreload" href="/docs/diffusers/pr_12509/zh/_app/immutable/chunks/CodeBlock.41f8ac86.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;缓存&quot;,&quot;local&quot;:&quot;缓存&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;金字塔注意力广播&quot;,&quot;local&quot;:&quot;金字塔注意力广播&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;FasterCache&quot;,&quot;local&quot;:&quot;fastercache&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;: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>  <h1 class="relative group"><a id="缓存" 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="#缓存"><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>缓存</span></h1> <p data-svelte-h="svelte-1f1nqg4">缓存通过存储和重用不同层的中间输出（如注意力层和前馈层）来加速推理，而不是在每个推理步骤执行整个计算。它显著提高了生成速度，但以更多内存为代价，并且不需要额外的训练。</p> <p data-svelte-h="svelte-1onzyg1">本指南向您展示如何在 Diffusers 中使用支持的缓存方法。</p>  <h2 class="relative group"><a id="金字塔注意力广播" 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="#金字塔注意力广播"><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>金字塔注意力广播</span></h2> <p data-svelte-h="svelte-3i3btm"><a href="https://huggingface.co/papers/2408.12588" rel="nofollow">金字塔注意力广播 (PAB)</a> 基于这样一种观察：在生成过程的连续时间步之间，注意力输出差异不大。注意力差异在交叉注意力层中最小，并且通常在一个较长的时间步范围内被缓存。其次是时间注意力和空间注意力层。</p> <blockquote class="tip" data-svelte-h="svelte-r7ebaz"><p>并非所有视频模型都有三种类型的注意力（交叉、时间和空间）！</p></blockquote> <p data-svelte-h="svelte-1js4ud9">PAB 可以与其他技术（如序列并行性和无分类器引导并行性（数据并行性））结合，实现近乎实时的视频生成。</p> <p data-svelte-h="svelte-jg5pkc">设置并传递一个 <code>PyramidAttentionBroadcastConfig</code> 到管道的变换器以启用它。<code>spatial_attention_block_skip_range</code> 控制跳过空间注意力块中注意力计算的频率，<code>spatial_attention_timestep_skip_range</code> 是要跳过的时间步范围。注意选择一个合适的范围，因为较小的间隔可能导致推理速度变慢，而较大的间隔可能导致生成质量降低。</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
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> CogVideoXPipeline, PyramidAttentionBroadcastConfig

pipeline = CogVideoXPipeline.from_pretrained(<span class="hljs-string">&quot;THUDM/CogVideoX-5b&quot;</span>, torch_dtype=torch.bfloat16)
pipeline.to(<span class="hljs-string">&quot;cuda&quot;</span>)

config = PyramidAttentionBroadcastConfig(
    spatial_attention_block_skip_range=<span class="hljs-number">2</span>,
    spatial_attention_timestep_skip_range=(<span class="hljs-number">100</span>, <span class="hljs-number">800</span>),
    current_timestep_callback=<span class="hljs-keyword">lambda</span>: pipe.current_timestep,
)
pipeline.transformer.enable_cache(config)<!-- HTML_TAG_END --></pre></div>  <h2 class="relative group"><a id="fastercache" 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="#fastercache"><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>FasterCache</span></h2> <p data-svelte-h="svelte-1op45ku"><a href="https://huggingface.co/papers/2410.19355" rel="nofollow">FasterCache</a> 缓存并重用注意力特征，类似于 <a href="#pyramid-attention-broadcast">PAB</a>，因为每个连续时间步的输出差异很小。</p> <p data-svelte-h="svelte-1dt33dj">此方法在使用无分类器引导进行采样时（在大多数基础模型中常见），也可能选择跳过无条件分支预测，并且
如果连续时间步之间的预测潜在输出存在显著冗余，则从条件分支预测中估计它。</p> <p data-svelte-h="svelte-1emi3m1">设置并将 <code>FasterCacheConfig</code> 传递给管道的 transformer 以启用它。</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
<span class="hljs-keyword">from</span> diffusers <span class="hljs-keyword">import</span> CogVideoXPipeline, FasterCacheConfig

pipe line= CogVideoXPipeline.from_pretrained(<span class="hljs-string">&quot;THUDM/CogVideoX-5b&quot;</span>, torch_dtype=torch.bfloat16)
pipeline.to(<span class="hljs-string">&quot;cuda&quot;</span>)

config = FasterCacheConfig(
    spatial_attention_block_skip_range=<span class="hljs-number">2</span>,
    spatial_attention_timestep_skip_range=(-<span class="hljs-number">1</span>, <span class="hljs-number">681</span>),
    current_timestep_callback=<span class="hljs-keyword">lambda</span>: pipe.current_timestep,
    attention_weight_callback=<span class="hljs-keyword">lambda</span> _: <span class="hljs-number">0.3</span>,
    unconditional_batch_skip_range=<span class="hljs-number">5</span>,
    unconditional_batch_timestep_skip_range=(-<span class="hljs-number">1</span>, <span class="hljs-number">781</span>),
    tensor_format=<span class="hljs-string">&quot;BFCHW&quot;</span>,
)
pipeline.transformer.enable_cache(config)<!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/diffusers/blob/main/docs/source/zh/optimization/cache.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_1tglod5 = {
						assets: "/docs/diffusers/pr_12509/zh",
						base: "/docs/diffusers/pr_12509/zh",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/diffusers/pr_12509/zh/_app/immutable/entry/start.636dab22.js"),
						import("/docs/diffusers/pr_12509/zh/_app/immutable/entry/app.c3d2b8c4.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 22],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>