Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / accelerate /main /en /basic_tutorials /tpu.html

rtrm

29 days ago

download

raw

9.86 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"TPU training","local":"tpu-training","sections":[{"title":"Compilation","local":"compilation","sections":[],"depth":2},{"title":"Weight tying","local":"weight-tying","sections":[],"depth":2}],"depth":1}">
	<link href="/docs/accelerate/main/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/entry/start.2ea03080.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/scheduler.defa9a21.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/singletons.aff0b9fc.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/index.beade68d.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/paths.2c85d1a6.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/entry/app.e6812672.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/index.fe795e71.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/nodes/0.39c84d5d.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/each.e59479a4.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/nodes/8.dcd34024.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/CodeBlock.42404125.js">
	<link rel="modulepreload" href="/docs/accelerate/main/en/_app/immutable/chunks/EditOnGithub.0f575778.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"TPU training","local":"tpu-training","sections":[{"title":"Compilation","local":"compilation","sections":[],"depth":2},{"title":"Weight tying","local":"weight-tying","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="tpu-training" 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="#tpu-training"><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>TPU training</span></h1> <p data-svelte-h="svelte-1efoa6g">A <a href="https://cloud.google.com/tpu/docs/intro-to-tpu" rel="nofollow">TPU (Tensor Processing Unit)</a> is a type of hardware specifically designed for training models efficiently. Accelerate supports TPU training, but there are a few things you should be aware of, namely graph compilation. This tutorial briefly discusses compilation, and for more details, take a look at the <a href="../concept_guides/training_tpu">Training on TPUs with Accelerate</a> guide.</p> <h2 class="relative group"><a id="compilation" 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="#compilation"><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>Compilation</span></h2> <p data-svelte-h="svelte-76he8q">A TPU creates a graph of all the operations in the training step such as the forward pass, backward pass and optimizer step. This is why the first training step always takes a while because building and compiling this graph takes time. But once compilation is complete, it is cached and all subsequent steps are much faster.</p> <p data-svelte-h="svelte-1ga0n4t">The key is to avoid compiling your code again or else training is super slow. This means all your operations must be exactly the same:</p> <ul data-svelte-h="svelte-fxztf"><li>all tensors in your batches must have the same length (for example, no dynamic padding for NLP tasks)</li> <li>your code must be static (for example, no layers with for loops that have different lengths depending on the input such as a LSTM)</li></ul> <h2 class="relative group"><a id="weight-tying" 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="#weight-tying"><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>Weight tying</span></h2> <p data-svelte-h="svelte-1u75nqb">A common language model design is to tie the weights of the embedding and softmax layers. However, moving the model to a TPU (either yourself or passing it to the <a href="/docs/accelerate/main/en/package_reference/accelerator#accelerate.Accelerator.prepare">prepare()</a> method) breaks the weight tying and you’ll need to retie the weights.</p> <p data-svelte-h="svelte-w62ir7">To add special behavior (like weight tying) in your script for TPUs, set <code>distributed_type</code> to <code>DistributedType.TPU</code> first. Then you can use the <a href="https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel.tie_weights" rel="nofollow">tie_weights</a> method to tie the weights.</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">if</span> accelerator.distributed_type == DistributedType.TPU:
	model.tie_weights()<!-- 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/accelerate/blob/main/docs/source/basic_tutorials/tpu.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_1fyccrg = {
	assets: "/docs/accelerate/main/en",
	base: "/docs/accelerate/main/en",
	env: {}
	};

	const element = document.currentScript.parentElement;

	const data = [null,null];

	Promise.all([
	import("/docs/accelerate/main/en/_app/immutable/entry/start.2ea03080.js"),
	import("/docs/accelerate/main/en/_app/immutable/entry/app.e6812672.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 8],
	data,
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 9.86 kB
Xet hash:: 4bc064e68423c9bcce22933c11c51a4fdededd285bd6e7b21fbce63149ce6c40

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