Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / transformers /main /ja /perf_infer_cpu.html

rtrm

29 days ago

download

raw

16 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Efficient Inference on CPU","local":"efficient-inference-on-cpu","sections":[{"title":"BetterTransformer for faster inference","local":"bettertransformer-for-faster-inference","sections":[],"depth":2},{"title":"PyTorch JITモード（TorchScript）","local":"pytorch-jitモードtorchscript","sections":[{"title":"JITモードでのIPEXグラフ最適化","local":"jitモードでのipexグラフ最適化","sections":[{"title":"IPEX installation:","local":"ipex-installation","sections":[],"depth":4}],"depth":3},{"title":"Usage of JIT-mode","local":"usage-of-jit-mode","sections":[],"depth":3}],"depth":2}],"depth":1}">
	<link href="/docs/transformers/main/ja/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/entry/start.1486e459.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/scheduler.9bc65507.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/singletons.eee55cbf.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/index.3b203c72.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/paths.59da1547.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/entry/app.d9ae818f.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/index.707bf1b6.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/nodes/0.c06aa070.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/each.e59479a4.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/nodes/112.46b0ff76.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/Tip.c2ecdbf4.js">
	<link rel="modulepreload" href="/docs/transformers/main/ja/_app/immutable/chunks/EditOnGithub.922df6ba.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"Efficient Inference on CPU","local":"efficient-inference-on-cpu","sections":[{"title":"BetterTransformer for faster inference","local":"bettertransformer-for-faster-inference","sections":[],"depth":2},{"title":"PyTorch JITモード（TorchScript）","local":"pytorch-jitモードtorchscript","sections":[{"title":"JITモードでのIPEXグラフ最適化","local":"jitモードでのipexグラフ最適化","sections":[{"title":"IPEX installation:","local":"ipex-installation","sections":[],"depth":4}],"depth":3},{"title":"Usage of JIT-mode","local":"usage-of-jit-mode","sections":[],"depth":3}],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="efficient-inference-on-cpu" 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="#efficient-inference-on-cpu"><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>Efficient Inference on CPU</span></h1> <p data-svelte-h="svelte-173ykpk">このガイドは、CPU上で大規模なモデルの効率的な推論に焦点を当てています。</p> <h2 class="relative group"><a id="bettertransformer-for-faster-inference" 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="#bettertransformer-for-faster-inference"><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>BetterTransformer for faster inference</span></h2> <p data-svelte-h="svelte-1gv96n8">最近、テキスト、画像、および音声モデルのCPU上での高速な推論のために<code>BetterTransformer</code>を統合しました。詳細については、この統合に関するドキュメンテーションを<a href="https://huggingface.co/docs/optimum/bettertransformer/overview" rel="nofollow">こちら</a>で確認してください。</p> <h2 class="relative group"><a id="pytorch-jitモードtorchscript" 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="#pytorch-jitモードtorchscript"><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>PyTorch JITモード（TorchScript）</span></h2> <p data-svelte-h="svelte-s6mz7v">TorchScriptは、PyTorchコードからシリアライズ可能で最適化可能なモデルを作成する方法です。任意のTorchScriptプログラムは、Python依存性のないプロセスで保存およびロードできます。
	デフォルトのイーガーモードと比較して、PyTorchのjitモードは通常、オペレーターフュージョンなどの最適化手法によりモデル推論のパフォーマンスが向上します。</p> <p data-svelte-h="svelte-1vr7y5u">TorchScriptの簡単な紹介については、<a href="https://pytorch.org/tutorials/beginner/Intro_to_TorchScript_tutorial.html#tracing-modules" rel="nofollow">PyTorch TorchScriptチュートリアル</a>を参照してください。</p> <h3 class="relative group"><a id="jitモードでのipexグラフ最適化" 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="#jitモードでのipexグラフ最適化"><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>JITモードでのIPEXグラフ最適化</span></h3> <p data-svelte-h="svelte-gbkgfl">Intel® Extension for PyTorchは、Transformersシリーズモデルのjitモードにさらなる最適化を提供します。Intel® Extension for PyTorchをjitモードで使用することを強くお勧めします。Transformersモデルからよく使用されるオペレーターパターンのいくつかは、既にIntel® Extension for PyTorchでjitモードのフュージョンに対応しています。これらのフュージョンパターン（Multi-head-attentionフュージョン、Concat Linear、Linear+Add、Linear+Gelu、Add+LayerNormフュージョンなど）は有効でパフォーマンスが良いです。フュージョンの利点は、ユーザーに透過的に提供されます。分析によれば、最も人気のある質問応答、テキスト分類、トークン分類のNLPタスクの約70％が、これらのフュージョンパターンを使用してFloat32精度とBFloat16混合精度の両方でパフォーマンスの利点を得ることができます。</p> <p data-svelte-h="svelte-qo6507"><a href="https://intel.github.io/intel-extension-for-pytorch/cpu/latest/tutorials/features/graph_optimization.html" rel="nofollow">IPEXグラフ最適化の詳細情報</a>を確認してください。</p> <h4 class="relative group"><a id="ipex-installation" 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="#ipex-installation"><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>IPEX installation:</span></h4> <p data-svelte-h="svelte-1u3jvxp">IPEXのリリースはPyTorchに従っています。<a href="https://intel.github.io/intel-extension-for-pytorch/" rel="nofollow">IPEXのインストール方法</a>を確認してください。</p> <h3 class="relative group"><a id="usage-of-jit-mode" 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="#usage-of-jit-mode"><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>Usage of JIT-mode</span></h3> <p data-svelte-h="svelte-15m7esz">Trainerで評価または予測のためにJITモードを有効にするには、ユーザーはTrainerコマンド引数に<code>jit_mode_eval</code>を追加する必要があります。</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-uj65sm">PyTorch >= 1.14.0の場合、jitモードはjit.traceでdict入力がサポートされているため、予測と評価に任意のモデルに利益をもたらす可能性があります。</p> <p data-svelte-h="svelte-1mjd070">PyTorch < 1.14.0の場合、jitモードはforwardパラメーターの順序がjit.traceのタプル入力の順序と一致するモデルに利益をもたらす可能性があります（質問応答モデルなど）。jit.traceがタプル入力の順序と一致しない場合、テキスト分類モデルなど、jit.traceは失敗し、これをフォールバックさせるために例外でキャッチしています。ログはユーザーに通知するために使用されます。</p></div> <p data-svelte-h="svelte-r9hc9d"><a href="https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering" rel="nofollow">Transformers質問応答の使用例</a>を参考にしてください。</p> <ul data-svelte-h="svelte-143w2h7"><li><p>Inference using jit mode on CPU:</p> <pre>python run_qa.py \
	--model_name_or_path csarron/bert-base-uncased-squad-v1 \
	--dataset_name squad \
	--do_eval \
	--max_seq_length 384 \
	--doc_stride 128 \
	--output_dir /tmp/ \
	--no_cuda \
	<b>--jit_mode_eval </b></pre></li> <li><p>Inference with IPEX using jit mode on CPU:</p> <pre>python run_qa.py \
	--model_name_or_path csarron/bert-base-uncased-squad-v1 \
	--dataset_name squad \
	--do_eval \
	--max_seq_length 384 \
	--doc_stride 128 \
	--output_dir /tmp/ \
	--no_cuda \
	<b>--use_ipex \</b>
	<b>--jit_mode_eval</b></pre></li></ul> <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/ja/perf_infer_cpu.md" target="_blank"><span data-svelte-h="svelte-1kd6by1"><</span> <span data-svelte-h="svelte-x0xyl0">></span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a> <p></p>

	<script>
	{
	__sveltekit_jement = {
	assets: "/docs/transformers/main/ja",
	base: "/docs/transformers/main/ja",
	env: {}
	};

	const element = document.currentScript.parentElement;

	const data = [null,null];

	Promise.all([
	import("/docs/transformers/main/ja/_app/immutable/entry/start.1486e459.js"),
	import("/docs/transformers/main/ja/_app/immutable/entry/app.d9ae818f.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 112],
	data,
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 16 kB
Xet hash:: df89b1ede598a9176e13fa7ae9fc8914c4d1477ab51ee50bbaf708eccc02e2a3

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