Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / transformers /main /zh /big_models.html

rtrm

27 days ago

download

raw

25.5 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"实例化大型模型","local":"实例化大型模型","sections":[{"title":"分片checkpoints","local":"分片checkpoints","sections":[],"depth":2},{"title":"低内存加载","local":"低内存加载","sections":[],"depth":2}],"depth":1}">
	<link href="/docs/transformers/main/zh/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/entry/start.a61b9c50.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/scheduler.9991993c.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/singletons.2822fe91.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/index.02cfeb18.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/paths.d66588b4.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/entry/app.99775688.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/index.7fc9a5e7.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/nodes/0.f4c5a5c1.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/each.e59479a4.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/nodes/5.b02dcc4e.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/Tip.9de92fc6.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/CodeBlock.e11cba92.js">
	<link rel="modulepreload" href="/docs/transformers/main/zh/_app/immutable/chunks/EditOnGithub.84ab7f0e.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"实例化大型模型","local":"实例化大型模型","sections":[{"title":"分片checkpoints","local":"分片checkpoints","sections":[],"depth":2},{"title":"低内存加载","local":"低内存加载","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="实例化大型模型" 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="#实例化大型模型"><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>实例化大型模型</span></h1> <p data-svelte-h="svelte-f980nr">当你想使用一个非常大的预训练模型时，一个挑战是尽量减少对内存的使用。通常从PyTorch开始的工作流程如下：</p> <ol data-svelte-h="svelte-qi8gup"><li>用随机权重创建你的模型。</li> <li>加载你的预训练权重。</li> <li>将这些预训练权重放入你的随机模型中。</li></ol> <p data-svelte-h="svelte-3m51fj">步骤1和2都需要完整版本的模型在内存中，这在大多数情况下不是问题，但如果你的模型开始达到几个GB的大小，这两个副本可能会让你超出内存的限制。更糟糕的是，如果你使用<code>torch.distributed</code>来启动分布式训练，每个进程都会加载预训练模型并将这两个副本存储在内存中。</p> <div class="course-tip bg-gradient-to-br dark:bg-gradient-to-r before:border-green-500 dark:before:border-green-800 from-green-50 dark:from-gray-900 to-white dark:to-gray-950 border border-green-50 text-green-700 dark:text-gray-400"><p data-svelte-h="svelte-1s8s2m3">请注意，随机创建的模型使用“空”张量进行初始化，这些张量占用内存空间但不填充它（因此随机值是给定时间内该内存块中的任何内容）。在第3步之后，对未初始化的权重执行适合模型/参数种类的随机初始化（例如正态分布），以尽可能提高速度！</p></div> <p data-svelte-h="svelte-1fsody1">在本指南中，我们将探讨 Transformers 提供的解决方案来处理这个问题。请注意，这是一个积极开发的领域，因此这里解释的API在将来可能会略有变化。</p> <h2 class="relative group"><a id="分片checkpoints" 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="#分片checkpoints"><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>分片checkpoints</span></h2> <p data-svelte-h="svelte-p12rar">自4.18.0版本起，占用空间超过10GB的模型检查点将自动分成较小的片段。在使用<code>model.save_pretrained(save_dir)</code>时，您最终会得到几个部分<code>checkpoints</code>（每个的大小都小于10GB）以及一个索引，该索引将参数名称映射到存储它们的文件。</p> <p data-svelte-h="svelte-fnq70n">您可以使用<code>max_shard_size</code>参数来控制分片之前的最大大小。为了示例的目的，我们将使用具有较小分片大小的普通大小的模型：让我们以传统的BERT模型为例。</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">from</span> transformers <span class="hljs-keyword">import</span> AutoModel

	model = AutoModel.from_pretrained(<span class="hljs-string">"google-bert/bert-base-cased"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1vlh5wl">如果您使用 <a href="%E6%A8%A1%E5%9E%8B%E9%A2%84%E8%AE%AD%E7%BB%83%E4%BF%9D%E5%AD%98"><code>PreTrainedModel.save_pretrained</code></a> 进行保存，您将得到一个新的文件夹，其中包含两个文件：模型的配置和权重：</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-meta">>>> </span><span class="hljs-keyword">import</span> os
	<span class="hljs-meta">>>> </span><span class="hljs-keyword">import</span> tempfile

	<span class="hljs-meta">>>> </span><span class="hljs-keyword">with</span> tempfile.TemporaryDirectory() <span class="hljs-keyword">as</span> tmp_dir:
	<span class="hljs-meta">... </span> model.save_pretrained(tmp_dir)
	<span class="hljs-meta">... </span> <span class="hljs-built_in">print</span>(<span class="hljs-built_in">sorted</span>(os.listdir(tmp_dir)))
	[<span class="hljs-string">'config.json'</span>, <span class="hljs-string">'pytorch_model.bin'</span>]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-15702pw">现在让我们使用最大分片大小为200MB：</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-meta">>>> </span><span class="hljs-keyword">with</span> tempfile.TemporaryDirectory() <span class="hljs-keyword">as</span> tmp_dir:
	<span class="hljs-meta">... </span> model.save_pretrained(tmp_dir, max_shard_size=<span class="hljs-string">"200MB"</span>)
	<span class="hljs-meta">... </span> <span class="hljs-built_in">print</span>(<span class="hljs-built_in">sorted</span>(os.listdir(tmp_dir)))
	[<span class="hljs-string">'config.json'</span>, <span class="hljs-string">'pytorch_model-00001-of-00003.bin'</span>, <span class="hljs-string">'pytorch_model-00002-of-00003.bin'</span>, <span class="hljs-string">'pytorch_model-00003-of-00003.bin'</span>, <span class="hljs-string">'pytorch_model.bin.index.json'</span>]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-yzeurb">在模型配置文件最上方，我们可以看到三个不同的权重文件，以及一个<code>index.json</code>索引文件。这样的<code>checkpoint</code>可以使用<a href="/docs/transformers/main/zh/main_classes/model#transformers.PreTrainedModel.from_pretrained">from_pretrained()</a>方法完全重新加载：</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-meta">>>> </span><span class="hljs-keyword">with</span> tempfile.TemporaryDirectory() <span class="hljs-keyword">as</span> tmp_dir:
	<span class="hljs-meta">... </span> model.save_pretrained(tmp_dir, max_shard_size=<span class="hljs-string">"200MB"</span>)
	<span class="hljs-meta">... </span> new_model = AutoModel.from_pretrained(tmp_dir)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-bnpz0s">对于大型模型来说，这样做的主要优点是在上述工作流程的步骤2中，每个<code>checkpoint</code>的分片在前一个分片之后加载，从而将内存中的内存使用限制在模型大小加上最大分片的大小。</p> <p data-svelte-h="svelte-18it1xo">在后台，索引文件用于确定<code>checkpoint</code>中包含哪些键以及相应的权重存储在哪里。我们可以像加载任何json一样加载该索引，并获得一个字典：</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-meta">>>> </span><span class="hljs-keyword">import</span> json

	<span class="hljs-meta">>>> </span><span class="hljs-keyword">with</span> tempfile.TemporaryDirectory() <span class="hljs-keyword">as</span> tmp_dir:
	<span class="hljs-meta">... </span> model.save_pretrained(tmp_dir, max_shard_size=<span class="hljs-string">"200MB"</span>)
	<span class="hljs-meta">... </span> <span class="hljs-keyword">with</span> <span class="hljs-built_in">open</span>(os.path.join(tmp_dir, <span class="hljs-string">"pytorch_model.bin.index.json"</span>), <span class="hljs-string">"r"</span>) <span class="hljs-keyword">as</span> f:
	<span class="hljs-meta">... </span> index = json.load(f)

	<span class="hljs-meta">>>> </span><span class="hljs-built_in">print</span>(index.keys())
	dict_keys([<span class="hljs-string">'metadata'</span>, <span class="hljs-string">'weight_map'</span>])<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-26dikn">目前元数据仅包括模型的总大小。我们计划在将来添加其他信息：</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-meta">>>> </span>index[<span class="hljs-string">"metadata"</span>]
	{<span class="hljs-string">'total_size'</span>: <span class="hljs-number">433245184</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1lig6ro">权重映射是该索引的主要部分，它将每个参数的名称（通常在PyTorch模型的<code>state_dict</code>中找到）映射到存储该参数的文件：</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-meta">>>> </span>index[<span class="hljs-string">"weight_map"</span>]
	{<span class="hljs-string">'embeddings.LayerNorm.bias'</span>: <span class="hljs-string">'pytorch_model-00001-of-00003.bin'</span>,
	<span class="hljs-string">'embeddings.LayerNorm.weight'</span>: <span class="hljs-string">'pytorch_model-00001-of-00003.bin'</span>,
	...<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-8dbmah">如果您想直接在模型内部加载这样的分片<code>checkpoint</code>，而不使用 [<code>PreTrainedModel.from_pretrained</code>](就像您会为完整<code>checkpoint</code>执行 <code>model.load_state_dict()</code> 一样)，您应该使用 <a href="/docs/transformers/main/zh/main_classes/model#transformers.modeling_utils.load_sharded_checkpoint">modeling_utils.load_sharded_checkpoint()</a>：</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-meta">>>> </span><span class="hljs-keyword">from</span> transformers.modeling_utils <span class="hljs-keyword">import</span> load_sharded_checkpoint

	<span class="hljs-meta">>>> </span><span class="hljs-keyword">with</span> tempfile.TemporaryDirectory() <span class="hljs-keyword">as</span> tmp_dir:
	<span class="hljs-meta">... </span> model.save_pretrained(tmp_dir, max_shard_size=<span class="hljs-string">"200MB"</span>)
	<span class="hljs-meta">... </span> load_sharded_checkpoint(model, tmp_dir)<!-- HTML_TAG_END --></pre></div> <h2 class="relative group"><a id="低内存加载" 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="#低内存加载"><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>低内存加载</span></h2> <p data-svelte-h="svelte-1qqd7tm">分片<code>checkpoints</code>在上述工作流的第2步中降低了内存使用，但为了在低内存环境中使用该模型，我们建议使用基于 Accelerate 库的工具。</p> <p data-svelte-h="svelte-kz6dhx">请阅读以下指南以获取更多信息：<a href="./main_classes/model#large-model-loading">使用 Accelerate 进行大模型加载</a></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/big_models.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_173uja2 = {
	assets: "/docs/transformers/main/zh",
	base: "/docs/transformers/main/zh",
	env: {}
	};

	const element = document.currentScript.parentElement;

	const data = [null,null];

	Promise.all([
	import("/docs/transformers/main/zh/_app/immutable/entry/start.a61b9c50.js"),
	import("/docs/transformers/main/zh/_app/immutable/entry/app.99775688.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 5],
	data,
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 25.5 kB
Xet hash:: b1158467599f20559f730386945a36d44a74b89d726b37d14b9155b9aa5af4d1

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