Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Verwendung vortrainierter Modelle","local":"verwendung-vortrainierter-modelle","sections":[],"depth":1}"> | |
| <link href="/docs/course/pr_1069/de/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/entry/start.54e959b0.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/scheduler.49e4e380.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/singletons.84a526f1.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/index.9f446ad6.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/paths.6e45d2bb.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/entry/app.7f5691b0.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/index.fb15006d.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/nodes/0.a7b48091.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/each.e59479a4.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/nodes/21.24700c33.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/Tip.f590f2e1.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/CodeBlock.3f4fbe91.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/DocNotebookDropdown.ff7a6598.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/FrameworkSwitchCourse.f837e098.js"> | |
| <link rel="modulepreload" href="/docs/course/pr_1069/de/_app/immutable/chunks/getInferenceSnippets.233af260.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"Verwendung vortrainierter Modelle","local":"verwendung-vortrainierter-modelle","sections":[],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <div class="bg-white leading-none border border-gray-100 rounded-lg flex p-0.5 w-56 text-sm mb-4"><a class="flex justify-center flex-1 py-1.5 px-2.5 focus:outline-none !no-underline rounded-l bg-red-50 dark:bg-transparent text-red-600" href="?fw=pt"><svg class="mr-1.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><defs><clipPath id="a"><rect x="3.05" y="0.5" width="25.73" height="31" fill="none"></rect></clipPath></defs><g clip-path="url(#a)"><path d="M24.94,9.51a12.81,12.81,0,0,1,0,18.16,12.68,12.68,0,0,1-18,0,12.81,12.81,0,0,1,0-18.16l9-9V5l-.84.83-6,6a9.58,9.58,0,1,0,13.55,0ZM20.44,9a1.68,1.68,0,1,1,1.67-1.67A1.68,1.68,0,0,1,20.44,9Z" fill="#ee4c2c"></path></g></svg> Pytorch </a><a class="flex justify-center flex-1 py-1.5 px-2.5 focus:outline-none !no-underline rounded-r text-gray-500 filter grayscale" href="?fw=tf"><svg class="mr-1.5" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" aria-hidden="true" focusable="false" role="img" width="0.94em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 256 274"><path d="M145.726 42.065v42.07l72.861 42.07v-42.07l-72.86-42.07zM0 84.135v42.07l36.43 21.03V105.17L0 84.135zm109.291 21.035l-36.43 21.034v126.2l36.43 21.035v-84.135l36.435 21.035v-42.07l-36.435-21.034V105.17z" fill="#E55B2D"></path><path d="M145.726 42.065L36.43 105.17v42.065l72.861-42.065v42.065l36.435-21.03v-84.14zM255.022 63.1l-36.435 21.035v42.07l36.435-21.035V63.1zm-72.865 84.135l-36.43 21.035v42.07l36.43-21.036v-42.07zm-36.43 63.104l-36.436-21.035v84.135l36.435-21.035V210.34z" fill="#ED8E24"></path><path d="M145.726 0L0 84.135l36.43 21.035l109.296-63.105l72.861 42.07L255.022 63.1L145.726 0zm0 126.204l-36.435 21.03l36.435 21.036l36.43-21.035l-36.43-21.03z" fill="#F8BF3C"></path></svg> TensorFlow </a></div> <h1 class="relative group"><a id="verwendung-vortrainierter-modelle" 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="#verwendung-vortrainierter-modelle"><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>Verwendung vortrainierter Modelle</span></h1> <div class="flex space-x-1 absolute z-10 right-0 top-0"> <a href="https://colab.research.google.com/github/huggingface/notebooks/blob/master/course/chapter4/section2_pt.ipynb" target="_blank"><img alt="Open In Colab" class="!m-0" src="https://colab.research.google.com/assets/colab-badge.svg"></a> <a href="https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/master/course/chapter4/section2_pt.ipynb" target="_blank"><img alt="Open In Studio Lab" class="!m-0" src="https://studiolab.sagemaker.aws/studiolab.svg"></a></div> <p data-svelte-h="svelte-7y3kng">Der Model Hub erleichtert das Auswählen des passenden Modells, sodass es von downstream Libraries mit wenigen Codezeilen benutzt werden kann. Lass uns anschauen, wie genau man solche Modelle verwendet und wie man der Communinity zurück beitragen kann.</p> <p data-svelte-h="svelte-bbbe5r">Nehmen wir an, wir suchen nach einem französichbasierten Modell, das die “mask filling” Aufgabe kann.</p> <div class="flex justify-center" data-svelte-h="svelte-4k6hvy"><img src="https://huggingface.co/datasets/huggingface-course/documentation-images/resolve/main/en/chapter4/camembert.gif" alt="Selecting the Camembert model." width="80%"></div> <p data-svelte-h="svelte-qzga3h">Wir wählen den <code>camembert-base</code> Checkpoint aus, um es zu auszuprobieren. Das Kennzeichen <code>camembert-base</code> ist alles, was wir brauchen, um loszulegen! Wie in früheren Kapiteln gezeigt wurde, können wir das Modell mit der <code>pipeline()</code> Funktion instanziieren:</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> pipeline | |
| camembert_fill_mask = pipeline(<span class="hljs-string">"fill-mask"</span>, model=<span class="hljs-string">"camembert-base"</span>) | |
| results = camembert_fill_mask(<span class="hljs-string">"Le camembert est <mask> :)"</span>)<!-- HTML_TAG_END --></pre></div> <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-string">'sequence'</span>: <span class="hljs-string">'Le camembert est délicieux :)'</span>, <span class="hljs-string">'score'</span>: <span class="hljs-number">0.49091005325317383</span>, <span class="hljs-string">'token'</span>: <span class="hljs-number">7200</span>, <span class="hljs-string">'token_str'</span>: <span class="hljs-string">'délicieux'</span>}, | |
| {<span class="hljs-string">'sequence'</span>: <span class="hljs-string">'Le camembert est excellent :)'</span>, <span class="hljs-string">'score'</span>: <span class="hljs-number">0.1055697426199913</span>, <span class="hljs-string">'token'</span>: <span class="hljs-number">2183</span>, <span class="hljs-string">'token_str'</span>: <span class="hljs-string">'excellent'</span>}, | |
| {<span class="hljs-string">'sequence'</span>: <span class="hljs-string">'Le camembert est succulent :)'</span>, <span class="hljs-string">'score'</span>: <span class="hljs-number">0.03453313186764717</span>, <span class="hljs-string">'token'</span>: <span class="hljs-number">26202</span>, <span class="hljs-string">'token_str'</span>: <span class="hljs-string">'succulent'</span>}, | |
| {<span class="hljs-string">'sequence'</span>: <span class="hljs-string">'Le camembert est meilleur :)'</span>, <span class="hljs-string">'score'</span>: <span class="hljs-number">0.0330314114689827</span>, <span class="hljs-string">'token'</span>: <span class="hljs-number">528</span>, <span class="hljs-string">'token_str'</span>: <span class="hljs-string">'meilleur'</span>}, | |
| {<span class="hljs-string">'sequence'</span>: <span class="hljs-string">'Le camembert est parfait :)'</span>, <span class="hljs-string">'score'</span>: <span class="hljs-number">0.03007650189101696</span>, <span class="hljs-string">'token'</span>: <span class="hljs-number">1654</span>, <span class="hljs-string">'token_str'</span>: <span class="hljs-string">'parfait'</span>} | |
| ]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-q2shzx">So einfach kann man mit einer Pipeline ein Modell laden. Dabei muss man nur darauf achten, den passenden Checkpoint für die gewünschte Aufgabe zu selektieren. Zum Beispiel: Wir laden hier den <code>camembert-base</code> Checkpoint in die <code>fill-mask</code> Pipeline, was schon korrekt ist. Aber würden wir diesen Checkpoint in die <code>text-classification</code> Pipeline laden, wären die Ergebnisse völlig sinnlos, weil der “head” von <code>camembert-base</code> für diese Aufgabe einfach nicht passt! Wir empfehlen, den “Task Selector” auf der Hugging Face Hub Seite zu benutzen, um die richtigen Checkpoints auszuwählen:</p> <div class="flex justify-center" data-svelte-h="svelte-1cl7xbt"><img src="https://huggingface.co/datasets/huggingface-course/documentation-images/resolve/main/en/chapter4/tasks.png" alt="The task selector on the web interface." width="80%"></div> <p data-svelte-h="svelte-1epaz1v">Du kannst auch den Checkpoint mit der Modell-Architektur direkt instanziieren:</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> CamembertTokenizer, CamembertForMaskedLM | |
| tokenizer = CamembertTokenizer.from_pretrained(<span class="hljs-string">"camembert-base"</span>) | |
| model = CamembertForMaskedLM.from_pretrained(<span class="hljs-string">"camembert-base"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-j1ycsl">Dennoch empfehlen wir, dass man die <a href="https://huggingface.co/transformers/model_doc/auto?highlight=auto#auto-classes" rel="nofollow"><code>Auto*</code> classes</a> stattdessen benutzt, da diese architekturunabhängig sind. Das vorherige Code-Beispiel gilt nur für Checkpoints, die in die CamemBERT Architektur zu laden sind, aber mit den <code>Auto*</code> Klassen kann man Checkpoints ziemlich einfach tauschen:</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> AutoTokenizer, AutoModelForMaskedLM | |
| tokenizer = AutoTokenizer.from_pretrained(<span class="hljs-string">"camembert-base"</span>) | |
| model = AutoModelForMaskedLM.from_pretrained(<span class="hljs-string">"camembert-base"</span>)<!-- HTML_TAG_END --></pre></div> <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">Wenn du ein vortrainiertes Modell verwendest, prüf erstmal, wie genau das traininert wurde, mit welchen Datensätzen, sowie seine Einschränkungen und Biases. All diese Informationen sollten auf der Modellbeschreibungskarte stehen.</div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/course/blob/main/chapters/de/chapter4/2.mdx" 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_63aas9 = { | |
| assets: "/docs/course/pr_1069/de", | |
| base: "/docs/course/pr_1069/de", | |
| env: {} | |
| }; | |
| const element = document.currentScript.parentElement; | |
| const data = [null,null]; | |
| Promise.all([ | |
| import("/docs/course/pr_1069/de/_app/immutable/entry/start.54e959b0.js"), | |
| import("/docs/course/pr_1069/de/_app/immutable/entry/app.7f5691b0.js") | |
| ]).then(([kit, app]) => { | |
| kit.start(app, element, { | |
| node_ids: [0, 21], | |
| data, | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 18 kB
- Xet hash:
- 2b0b51b2d152188df10615f26dcda4bd9d484253d448acdedbaeea68ed64c583
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.