transformers/pr_28250/zh/perf_train_special.html

<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;在 Apple Silicon 芯片上进行 PyTorch 训练&quot;,&quot;local&quot;:&quot;在-apple-silicon-芯片上进行-pytorch-训练&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}">
		<link href="/docs/transformers/pr_28250/zh/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/entry/start.359aefb2.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/scheduler.9991993c.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/singletons.0d57c151.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/index.02cfeb18.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/paths.03bfa24f.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/entry/app.6c335e9f.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/index.ed60ef0f.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/nodes/0.2e1efaaf.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/each.e59479a4.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/nodes/54.17719f57.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/Tip.8eaeb7b5.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/CodeBlock.a73b7ee1.js">
		<link rel="modulepreload" href="/docs/transformers/pr_28250/zh/_app/immutable/chunks/EditOnGithub.ba269039.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;在 Apple Silicon 芯片上进行 PyTorch 训练&quot;,&quot;local&quot;:&quot;在-apple-silicon-芯片上进行-pytorch-训练&quot;,&quot;sections&quot;:[],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END -->      <p></p>   <h1 class="relative group"><a id="在-apple-silicon-芯片上进行-pytorch-训练" 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="#在-apple-silicon-芯片上进行-pytorch-训练"><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>在 Apple Silicon 芯片上进行 PyTorch 训练</span></h1> <p data-svelte-h="svelte-1ve9m0l">之前，在 Mac 上训练模型仅限于使用 CPU 训练。不过随着PyTorch v1.12的发布，您可以通过在 Apple Silicon 芯片的 GPU 上训练模型来显著提高性能和训练速度。这是通过将 Apple 的 Metal 性能着色器 (Metal Performance Shaders, MPS) 作为后端集成到PyTorch中实现的。<a href="https://pytorch.org/docs/stable/notes/mps.html" rel="nofollow">MPS后端</a> 将 PyTorch 操作视为自定义的 Metal 着色器来实现，并将对应模块部署到<code>mps</code>设备上。</p>  <div class="course-tip course-tip-orange bg-gradient-to-br dark:bg-gradient-to-r before:border-orange-500 dark:before:border-orange-800 from-orange-50 dark:from-gray-900 to-white dark:to-gray-950 border border-orange-50 text-orange-700 dark:text-gray-400"><p data-svelte-h="svelte-oq75au">某些 PyTorch 操作目前还未在 MPS 上实现，可能会抛出错误提示。可以通过设置环境变量<code>PYTORCH_ENABLE_MPS_FALLBACK=1</code>来使用CPU内核以避免这种情况发生（您仍然会看到一个<code>UserWarning</code>）。</p> <br> <p data-svelte-h="svelte-1x8iya3">如果您遇到任何其他错误，请在<a href="https://github.com/pytorch/pytorch/issues" rel="nofollow">PyTorch库</a>中创建一个 issue，因为<code>Trainer</code>类中只集成了 MPS 后端.</p></div> <p data-svelte-h="svelte-xvtsug">配置好<code>mps</code>设备后，您可以：</p> <ul data-svelte-h="svelte-5fu39o"><li>在本地训练更大的网络或更大的批量大小</li> <li>降低数据获取延迟，因为 GPU 的统一内存架构允许直接访问整个内存存储</li> <li>降低成本，因为您不需要再在云端 GPU 上训练或增加额外的本地 GPU</li></ul> <p data-svelte-h="svelte-7whwsu">在确保已安装PyTorch后就可以开始使用了。 MPS 加速支持macOS 12.3及以上版本。</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 -->pip install torch torchvision torchaudio<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1etdm8m"><code>TrainingArguments</code>类默认使用<code>mps</code>设备(如果可用)因此无需显式设置设备。例如，您可以直接运行<a href="https://github.com/huggingface/transformers/blob/main/examples/pytorch/text-classification/run_glue.py" rel="nofollow">run_glue.py</a>脚本，在无需进行任何修改的情况下自动启用 MPS 后端。</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 -->export TASK_NAME=mrpc

python examples/pytorch/text-classification/run_glue.py \
  --model_name_or_path google-bert/bert-base-cased \
  --task_name $TASK_NAME \
<span class="hljs-deletion">- --use_mps_device \</span>
  --do_train \
  --do_eval \
  --max_seq_length 128 \
  --per_device_train_batch_size 32 \
  --learning_rate 2e-5 \
  --num_train_epochs 3 \
  --output_dir /tmp/$TASK_NAME/ \
  --overwrite_output_dir<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1m152z5">用于<a href="https://pytorch.org/docs/stable/distributed.html#backends" rel="nofollow">分布式设置</a>的后端(如<code>gloo</code>和<code>nccl</code>)不支持<code>mps</code>设备，这也意味着使用 MPS 后端时只能在单个 GPU 上进行训练。</p> <p data-svelte-h="svelte-qu3rh1">您可以在<a href="https://pytorch.org/blog/introducing-accelerated-pytorch-training-on-mac/" rel="nofollow">Introducing Accelerated PyTorch Training on Mac</a>博客文章中了解有关 MPS 后端的更多信息。</p> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/zh/perf_train_special.md" target="_blank"><span data-svelte-h="svelte-1kd6by1">&lt;</span> <span data-svelte-h="svelte-x0xyl0">&gt;</span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a>  <p></p> 
			
			<script>
				{
					__sveltekit_qheazb = {
						assets: "/docs/transformers/pr_28250/zh",
						base: "/docs/transformers/pr_28250/zh",
						env: {}
					};

					const element = document.currentScript.parentElement;

					const data = [null,null];

					Promise.all([
						import("/docs/transformers/pr_28250/zh/_app/immutable/entry/start.359aefb2.js"),
						import("/docs/transformers/pr_28250/zh/_app/immutable/entry/app.6c335e9f.js")
					]).then(([kit, app]) => {
						kit.start(app, element, {
							node_ids: [0, 54],
							data,
							form: null,
							error: null
						});
					});
				}
			</script>