Hugging Face's logo

Buckets:
hf-doc-build
/
doc-dev

Files
xet
hf-doc-build/doc-dev / transformers /pr_33913 /es /tasks /multiple_choice.html
rtrm's picture
rtrm
download
raw
58.9 kB
<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Selección múltiple&quot;,&quot;local&quot;:&quot;selección-múltiple&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Cargar el dataset SWAG&quot;,&quot;local&quot;:&quot;cargar-el-dataset-swag&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Preprocesmaiento&quot;,&quot;local&quot;:&quot;preprocesmaiento&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Entrenamiento&quot;,&quot;local&quot;:&quot;entrenamiento&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}">
 <link href="/docs/transformers/pr_33913/es/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/entry/start.41996def.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/scheduler.36a0863c.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/singletons.be078126.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/index.733708bb.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/paths.9d726baf.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/entry/app.9ed505df.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/index.f891bdb2.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/nodes/0.cfca1b02.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/each.e59479a4.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/nodes/37.60ac4ff2.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/Tip.a8272f7f.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/CodeBlock.3ec784ea.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/Markdown.7b58822e.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/globals.7f7f1b26.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/stores.300cf1d0.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/IconTensorflow.8e6f27b1.js">
 <link rel="modulepreload" href="/docs/transformers/pr_33913/es/_app/immutable/chunks/EditOnGithub.a58e27a9.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{&quot;title&quot;:&quot;Selección múltiple&quot;,&quot;local&quot;:&quot;selección-múltiple&quot;,&quot;sections&quot;:[{&quot;title&quot;:&quot;Cargar el dataset SWAG&quot;,&quot;local&quot;:&quot;cargar-el-dataset-swag&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Preprocesmaiento&quot;,&quot;local&quot;:&quot;preprocesmaiento&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2},{&quot;title&quot;:&quot;Entrenamiento&quot;,&quot;local&quot;:&quot;entrenamiento&quot;,&quot;sections&quot;:[],&quot;depth&quot;:2}],&quot;depth&quot;:1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="selección-múltiple" 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="#selección-múltiple"><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>Selección múltiple</span></h1> <p data-svelte-h="svelte-50nup6">La tarea de selección múltiple es parecida a la de responder preguntas, con la excepción de que se dan varias opciones de respuesta junto con el contexto. El modelo se entrena para escoger la respuesta correcta
entre varias opciones a partir del contexto dado.</p> <p data-svelte-h="svelte-1tgwhyx">Esta guía te mostrará como hacerle fine-tuning a <a href="https://huggingface.co/google-bert/bert-base-uncased" rel="nofollow">BERT</a> en la configuración <code>regular</code> del dataset <a href="https://huggingface.co/datasets/swag" rel="nofollow">SWAG</a>, de forma
que seleccione la mejor respuesta a partir de varias opciones y algún contexto.</p> <h2 class="relative group"><a id="cargar-el-dataset-swag" 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="#cargar-el-dataset-swag"><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>Cargar el dataset SWAG</span></h2> <p data-svelte-h="svelte-vnn33k">Carga el dataset SWAG con la biblioteca 🤗 Datasets:</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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset
<span class="hljs-meta">&gt;&gt;&gt; </span>swag = load_dataset(<span class="hljs-string">&quot;swag&quot;</span>, <span class="hljs-string">&quot;regular&quot;</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-x1ncwr">Ahora, échale un vistazo a un ejemplo del dataset:</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">&gt;&gt;&gt; </span>swag[<span class="hljs-string">&quot;train&quot;</span>][<span class="hljs-number">0</span>]
{<span class="hljs-string">&#x27;ending0&#x27;</span>: <span class="hljs-string">&#x27;passes by walking down the street playing their instruments.&#x27;</span>,
 <span class="hljs-string">&#x27;ending1&#x27;</span>: <span class="hljs-string">&#x27;has heard approaching them.&#x27;</span>,
 <span class="hljs-string">&#x27;ending2&#x27;</span>: <span class="hljs-string">&quot;arrives and they&#x27;re outside dancing and asleep.&quot;</span>,
 <span class="hljs-string">&#x27;ending3&#x27;</span>: <span class="hljs-string">&#x27;turns the lead singer watches the performance.&#x27;</span>,
 <span class="hljs-string">&#x27;fold-ind&#x27;</span>: <span class="hljs-string">&#x27;3416&#x27;</span>,
 <span class="hljs-string">&#x27;gold-source&#x27;</span>: <span class="hljs-string">&#x27;gold&#x27;</span>,
 <span class="hljs-string">&#x27;label&#x27;</span>: <span class="hljs-number">0</span>,
 <span class="hljs-string">&#x27;sent1&#x27;</span>: <span class="hljs-string">&#x27;Members of the procession walk down the street holding small horn brass instruments.&#x27;</span>,
 <span class="hljs-string">&#x27;sent2&#x27;</span>: <span class="hljs-string">&#x27;A drum line&#x27;</span>,
 <span class="hljs-string">&#x27;startphrase&#x27;</span>: <span class="hljs-string">&#x27;Members of the procession walk down the street holding small horn brass instruments. A drum line&#x27;</span>,
 <span class="hljs-string">&#x27;video-id&#x27;</span>: <span class="hljs-string">&#x27;anetv_jkn6uvmqwh4&#x27;</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1ri0vdd">Los campos <code>sent1</code> y <code>sent2</code> muestran cómo comienza una oración, y cada campo <code>ending</code> indica cómo podría terminar. Dado el comienzo de la oración, el modelo debe escoger el final de oración correcto indicado por el campo <code>label</code>.</p> <h2 class="relative group"><a id="preprocesmaiento" 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="#preprocesmaiento"><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>Preprocesmaiento</span></h2> <p data-svelte-h="svelte-1iq2bgx">Carga el tokenizer de BERT para procesar el comienzo de cada oración y los cuatro finales posibles:</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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoTokenizer
<span class="hljs-meta">&gt;&gt;&gt; </span>tokenizer = AutoTokenizer.from_pretrained(<span class="hljs-string">&quot;google-bert/bert-base-uncased&quot;</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1jkglaf">La función de preprocesmaiento debe hacer lo siguiente:</p> <ol data-svelte-h="svelte-12xaauq"><li>Hacer cuatro copias del campo <code>sent1</code> de forma que se pueda combinar cada una con el campo <code>sent2</code> para recrear la forma en que empieza la oración.</li> <li>Combinar <code>sent2</code> con cada uno de los cuatro finales de oración posibles.</li> <li>Aplanar las dos listas para que puedas tokenizarlas, y luego des-aplanarlas para que cada ejemplo tenga los campos <code>input_ids</code>, <code>attention_mask</code> y <code>labels</code> correspondientes.</li></ol> <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">&gt;&gt;&gt; </span>ending_names = [<span class="hljs-string">&quot;ending0&quot;</span>, <span class="hljs-string">&quot;ending1&quot;</span>, <span class="hljs-string">&quot;ending2&quot;</span>, <span class="hljs-string">&quot;ending3&quot;</span>]
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">def</span> <span class="hljs-title function_">preprocess_function</span>(<span class="hljs-params">examples</span>):
<span class="hljs-meta">... </span> first_sentences = [[context] * <span class="hljs-number">4</span> <span class="hljs-keyword">for</span> context <span class="hljs-keyword">in</span> examples[<span class="hljs-string">&quot;sent1&quot;</span>]]
<span class="hljs-meta">... </span> question_headers = examples[<span class="hljs-string">&quot;sent2&quot;</span>]
<span class="hljs-meta">... </span> second_sentences = [
<span class="hljs-meta">... </span> [<span class="hljs-string">f&quot;<span class="hljs-subst">{header}</span> <span class="hljs-subst">{examples[end][i]}</span>&quot;</span> <span class="hljs-keyword">for</span> end <span class="hljs-keyword">in</span> ending_names] <span class="hljs-keyword">for</span> i, header <span class="hljs-keyword">in</span> <span class="hljs-built_in">enumerate</span>(question_headers)
<span class="hljs-meta">... </span> ]
<span class="hljs-meta">... </span> first_sentences = <span class="hljs-built_in">sum</span>(first_sentences, [])
<span class="hljs-meta">... </span> second_sentences = <span class="hljs-built_in">sum</span>(second_sentences, [])
<span class="hljs-meta">... </span> tokenized_examples = tokenizer(first_sentences, second_sentences, truncation=<span class="hljs-literal">True</span>)
<span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> {k: [v[i : i + <span class="hljs-number">4</span>] <span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-number">0</span>, <span class="hljs-built_in">len</span>(v), <span class="hljs-number">4</span>)] <span class="hljs-keyword">for</span> k, v <span class="hljs-keyword">in</span> tokenized_examples.items()}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-f9xmqh">Usa la función <code>map</code> de 🤗 Datasets para aplicarle la función de preprocesamiento al dataset entero. Puedes acelerar la función <code>map</code> haciendo <code>batched=True</code> para procesar varios elementos del dataset a la vez.</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 -->tokenized_swag = swag.<span class="hljs-built_in">map</span>(preprocess_function, batched=<span class="hljs-literal">True</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1vi4idm">🤗 Transformers no tiene un collator de datos para la tarea de selección múltiple, así que tendrías que crear uno. Puedes adaptar el <code>DataCollatorWithPadding</code> para crear un lote de ejemplos para selección múltiple. Este también
le <em>añadirá relleno de manera dinámica</em> a tu texto y a las etiquetas para que tengan la longitud del elemento más largo en su lote, de forma que tengan una longitud uniforme. Aunque es posible rellenar el texto en la función <code>tokenizer</code> haciendo
<code>padding=True</code>, el rellenado dinámico es más eficiente.</p> <p data-svelte-h="svelte-1010f04">El <code>DataCollatorForMultipleChoice</code> aplanará todas las entradas del modelo, les aplicará relleno y luego des-aplanará los resultados:</p> <div class="space-y-10 py-6 2xl:py-8 2xl:-mx-4"> <div class="border border-gray-200 rounded-xl px-4 relative"><div class="flex h-[22px] mt-[-12.5px] justify-between leading-none"><div class="flex px-1 items-center space-x-1 bg-white dark:bg-gray-950"><svg class="" 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> <span>Pytorch</span></div> <div class="cursor-pointer flex items-center justify-center space-x-1 text-sm px-2 bg-white dark:bg-gray-950 hover:underline leading-none"><svg class="" width="0.9em" height="0.9em" viewBox="0 0 10 9" fill="currentColor" xmlns="http://www.w3.org/2000/svg"><path d="M1.39125 1.9725L0.0883333 0.669997L0.677917 0.0804138L8.9275 8.33041L8.33792 8.91958L6.95875 7.54041C6.22592 8.00523 5.37572 8.25138 4.50792 8.25C2.26125 8.25 0.392083 6.63333 0 4.5C0.179179 3.52946 0.667345 2.64287 1.39167 1.9725H1.39125ZM5.65667 6.23833L5.04667 5.62833C4.81335 5.73996 4.55116 5.77647 4.29622 5.73282C4.04129 5.68918 3.80617 5.56752 3.62328 5.38463C3.44039 5.20175 3.31874 4.96663 3.27509 4.71169C3.23144 4.45676 3.26795 4.19456 3.37958 3.96125L2.76958 3.35125C2.50447 3.75187 2.38595 4.2318 2.4341 4.70978C2.48225 5.18777 2.6941 5.63442 3.0338 5.97411C3.37349 6.31381 3.82015 6.52567 4.29813 6.57382C4.77611 6.62197 5.25605 6.50345 5.65667 6.23833ZM2.83042 1.06666C3.35 0.862497 3.91625 0.749997 4.50792 0.749997C6.75458 0.749997 8.62375 2.36666 9.01583 4.5C8.88816 5.19404 8.60119 5.84899 8.1775 6.41333L6.56917 4.805C6.61694 4.48317 6.58868 4.15463 6.48664 3.84569C6.3846 3.53675 6.21162 3.256 5.98156 3.02594C5.7515 2.79588 5.47075 2.6229 5.16181 2.52086C4.85287 2.41882 4.52433 2.39056 4.2025 2.43833L2.83042 1.06708V1.06666Z" fill="currentColor"></path></svg> <span>Hide Pytorch content</span></div></div> <div class="framework-content"> <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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> dataclasses <span class="hljs-keyword">import</span> dataclass
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers.tokenization_utils_base <span class="hljs-keyword">import</span> PreTrainedTokenizerBase, PaddingStrategy
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> typing <span class="hljs-keyword">import</span> <span class="hljs-type">Optional</span>, <span class="hljs-type">Union</span>
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> torch
<span class="hljs-meta">&gt;&gt;&gt; </span>@dataclass
<span class="hljs-meta">... </span><span class="hljs-keyword">class</span> <span class="hljs-title class_">DataCollatorForMultipleChoice</span>:
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;&quot;&quot;
<span class="hljs-meta">... </span> Collator de datos que le añadirá relleno de forma automática a las entradas recibidas para
<span class="hljs-meta">... </span> una tarea de selección múltiple.
<span class="hljs-meta">... </span> &quot;&quot;&quot;</span>
<span class="hljs-meta">... </span> tokenizer: PreTrainedTokenizerBase
<span class="hljs-meta">... </span> padding: <span class="hljs-type">Union</span>[<span class="hljs-built_in">bool</span>, <span class="hljs-built_in">str</span>, PaddingStrategy] = <span class="hljs-literal">True</span>
<span class="hljs-meta">... </span> max_length: <span class="hljs-type">Optional</span>[<span class="hljs-built_in">int</span>] = <span class="hljs-literal">None</span>
<span class="hljs-meta">... </span> pad_to_multiple_of: <span class="hljs-type">Optional</span>[<span class="hljs-built_in">int</span>] = <span class="hljs-literal">None</span>
<span class="hljs-meta">... </span> <span class="hljs-keyword">def</span> <span class="hljs-title function_">__call__</span>(<span class="hljs-params">self, features</span>):
<span class="hljs-meta">... </span> label_name = <span class="hljs-string">&quot;label&quot;</span> <span class="hljs-keyword">if</span> <span class="hljs-string">&quot;label&quot;</span> <span class="hljs-keyword">in</span> features[<span class="hljs-number">0</span>].keys() <span class="hljs-keyword">else</span> <span class="hljs-string">&quot;labels&quot;</span>
<span class="hljs-meta">... </span> labels = [feature.pop(label_name) <span class="hljs-keyword">for</span> feature <span class="hljs-keyword">in</span> features]
<span class="hljs-meta">... </span> batch_size = <span class="hljs-built_in">len</span>(features)
<span class="hljs-meta">... </span> num_choices = <span class="hljs-built_in">len</span>(features[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;input_ids&quot;</span>])
<span class="hljs-meta">... </span> flattened_features = [
<span class="hljs-meta">... </span> [{k: v[i] <span class="hljs-keyword">for</span> k, v <span class="hljs-keyword">in</span> feature.items()} <span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(num_choices)] <span class="hljs-keyword">for</span> feature <span class="hljs-keyword">in</span> features
<span class="hljs-meta">... </span> ]
<span class="hljs-meta">... </span> flattened_features = <span class="hljs-built_in">sum</span>(flattened_features, [])
<span class="hljs-meta">... </span> batch = self.tokenizer.pad(
<span class="hljs-meta">... </span> flattened_features,
<span class="hljs-meta">... </span> padding=self.padding,
<span class="hljs-meta">... </span> max_length=self.max_length,
<span class="hljs-meta">... </span> pad_to_multiple_of=self.pad_to_multiple_of,
<span class="hljs-meta">... </span> return_tensors=<span class="hljs-string">&quot;pt&quot;</span>,
<span class="hljs-meta">... </span> )
<span class="hljs-meta">... </span> batch = {k: v.view(batch_size, num_choices, -<span class="hljs-number">1</span>) <span class="hljs-keyword">for</span> k, v <span class="hljs-keyword">in</span> batch.items()}
<span class="hljs-meta">... </span> batch[<span class="hljs-string">&quot;labels&quot;</span>] = torch.tensor(labels, dtype=torch.int64)
<span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> batch<!-- HTML_TAG_END --></pre></div></div></div> <div class="border border-gray-200 rounded-xl px-4 relative"><div class="flex h-[22px] mt-[-12.5px] justify-between leading-none"><div class="flex px-1 items-center space-x-1 bg-white dark:bg-gray-950"><svg class="" 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> <span>TensorFlow</span></div> <div class="cursor-pointer flex items-center justify-center space-x-1 text-sm px-2 bg-white dark:bg-gray-950 hover:underline leading-none"><svg class="" width="0.9em" height="0.9em" viewBox="0 0 10 9" fill="currentColor" xmlns="http://www.w3.org/2000/svg"><path d="M1.39125 1.9725L0.0883333 0.669997L0.677917 0.0804138L8.9275 8.33041L8.33792 8.91958L6.95875 7.54041C6.22592 8.00523 5.37572 8.25138 4.50792 8.25C2.26125 8.25 0.392083 6.63333 0 4.5C0.179179 3.52946 0.667345 2.64287 1.39167 1.9725H1.39125ZM5.65667 6.23833L5.04667 5.62833C4.81335 5.73996 4.55116 5.77647 4.29622 5.73282C4.04129 5.68918 3.80617 5.56752 3.62328 5.38463C3.44039 5.20175 3.31874 4.96663 3.27509 4.71169C3.23144 4.45676 3.26795 4.19456 3.37958 3.96125L2.76958 3.35125C2.50447 3.75187 2.38595 4.2318 2.4341 4.70978C2.48225 5.18777 2.6941 5.63442 3.0338 5.97411C3.37349 6.31381 3.82015 6.52567 4.29813 6.57382C4.77611 6.62197 5.25605 6.50345 5.65667 6.23833ZM2.83042 1.06666C3.35 0.862497 3.91625 0.749997 4.50792 0.749997C6.75458 0.749997 8.62375 2.36666 9.01583 4.5C8.88816 5.19404 8.60119 5.84899 8.1775 6.41333L6.56917 4.805C6.61694 4.48317 6.58868 4.15463 6.48664 3.84569C6.3846 3.53675 6.21162 3.256 5.98156 3.02594C5.7515 2.79588 5.47075 2.6229 5.16181 2.52086C4.85287 2.41882 4.52433 2.39056 4.2025 2.43833L2.83042 1.06708V1.06666Z" fill="currentColor"></path></svg> <span>Hide TensorFlow content</span></div></div> <div class="framework-content"> <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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> dataclasses <span class="hljs-keyword">import</span> dataclass
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers.tokenization_utils_base <span class="hljs-keyword">import</span> PreTrainedTokenizerBase, PaddingStrategy
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> typing <span class="hljs-keyword">import</span> <span class="hljs-type">Optional</span>, <span class="hljs-type">Union</span>
<span class="hljs-meta">&gt;&gt;&gt; </span><span class="hljs-keyword">import</span> tensorflow <span class="hljs-keyword">as</span> tf
<span class="hljs-meta">&gt;&gt;&gt; </span>@dataclass
<span class="hljs-meta">... </span><span class="hljs-keyword">class</span> <span class="hljs-title class_">DataCollatorForMultipleChoice</span>:
<span class="hljs-meta">... </span> <span class="hljs-string">&quot;&quot;&quot;
<span class="hljs-meta">... </span> Data collator that will dynamically pad the inputs for multiple choice received.
<span class="hljs-meta">... </span> &quot;&quot;&quot;</span>
<span class="hljs-meta">... </span> tokenizer: PreTrainedTokenizerBase
<span class="hljs-meta">... </span> padding: <span class="hljs-type">Union</span>[<span class="hljs-built_in">bool</span>, <span class="hljs-built_in">str</span>, PaddingStrategy] = <span class="hljs-literal">True</span>
<span class="hljs-meta">... </span> max_length: <span class="hljs-type">Optional</span>[<span class="hljs-built_in">int</span>] = <span class="hljs-literal">None</span>
<span class="hljs-meta">... </span> pad_to_multiple_of: <span class="hljs-type">Optional</span>[<span class="hljs-built_in">int</span>] = <span class="hljs-literal">None</span>
<span class="hljs-meta">... </span> <span class="hljs-keyword">def</span> <span class="hljs-title function_">__call__</span>(<span class="hljs-params">self, features</span>):
<span class="hljs-meta">... </span> label_name = <span class="hljs-string">&quot;label&quot;</span> <span class="hljs-keyword">if</span> <span class="hljs-string">&quot;label&quot;</span> <span class="hljs-keyword">in</span> features[<span class="hljs-number">0</span>].keys() <span class="hljs-keyword">else</span> <span class="hljs-string">&quot;labels&quot;</span>
<span class="hljs-meta">... </span> labels = [feature.pop(label_name) <span class="hljs-keyword">for</span> feature <span class="hljs-keyword">in</span> features]
<span class="hljs-meta">... </span> batch_size = <span class="hljs-built_in">len</span>(features)
<span class="hljs-meta">... </span> num_choices = <span class="hljs-built_in">len</span>(features[<span class="hljs-number">0</span>][<span class="hljs-string">&quot;input_ids&quot;</span>])
<span class="hljs-meta">... </span> flattened_features = [
<span class="hljs-meta">... </span> [{k: v[i] <span class="hljs-keyword">for</span> k, v <span class="hljs-keyword">in</span> feature.items()} <span class="hljs-keyword">for</span> i <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(num_choices)] <span class="hljs-keyword">for</span> feature <span class="hljs-keyword">in</span> features
<span class="hljs-meta">... </span> ]
<span class="hljs-meta">... </span> flattened_features = <span class="hljs-built_in">sum</span>(flattened_features, [])
<span class="hljs-meta">... </span> batch = self.tokenizer.pad(
<span class="hljs-meta">... </span> flattened_features,
<span class="hljs-meta">... </span> padding=self.padding,
<span class="hljs-meta">... </span> max_length=self.max_length,
<span class="hljs-meta">... </span> pad_to_multiple_of=self.pad_to_multiple_of,
<span class="hljs-meta">... </span> return_tensors=<span class="hljs-string">&quot;tf&quot;</span>,
<span class="hljs-meta">... </span> )
<span class="hljs-meta">... </span> batch = {k: tf.reshape(v, (batch_size, num_choices, -<span class="hljs-number">1</span>)) <span class="hljs-keyword">for</span> k, v <span class="hljs-keyword">in</span> batch.items()}
<span class="hljs-meta">... </span> batch[<span class="hljs-string">&quot;labels&quot;</span>] = tf.convert_to_tensor(labels, dtype=tf.int64)
<span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> batch<!-- HTML_TAG_END --></pre></div> </div></div> </div> <h2 class="relative group"><a id="entrenamiento" 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="#entrenamiento"><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>Entrenamiento</span></h2> <div class="space-y-10 py-6 2xl:py-8 2xl:-mx-4"> <div class="border border-gray-200 rounded-xl px-4 relative"><div class="flex h-[22px] mt-[-12.5px] justify-between leading-none"><div class="flex px-1 items-center space-x-1 bg-white dark:bg-gray-950"><svg class="" 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> <span>Pytorch</span></div> <div class="cursor-pointer flex items-center justify-center space-x-1 text-sm px-2 bg-white dark:bg-gray-950 hover:underline leading-none"><svg class="" width="0.9em" height="0.9em" viewBox="0 0 10 9" fill="currentColor" xmlns="http://www.w3.org/2000/svg"><path d="M1.39125 1.9725L0.0883333 0.669997L0.677917 0.0804138L8.9275 8.33041L8.33792 8.91958L6.95875 7.54041C6.22592 8.00523 5.37572 8.25138 4.50792 8.25C2.26125 8.25 0.392083 6.63333 0 4.5C0.179179 3.52946 0.667345 2.64287 1.39167 1.9725H1.39125ZM5.65667 6.23833L5.04667 5.62833C4.81335 5.73996 4.55116 5.77647 4.29622 5.73282C4.04129 5.68918 3.80617 5.56752 3.62328 5.38463C3.44039 5.20175 3.31874 4.96663 3.27509 4.71169C3.23144 4.45676 3.26795 4.19456 3.37958 3.96125L2.76958 3.35125C2.50447 3.75187 2.38595 4.2318 2.4341 4.70978C2.48225 5.18777 2.6941 5.63442 3.0338 5.97411C3.37349 6.31381 3.82015 6.52567 4.29813 6.57382C4.77611 6.62197 5.25605 6.50345 5.65667 6.23833ZM2.83042 1.06666C3.35 0.862497 3.91625 0.749997 4.50792 0.749997C6.75458 0.749997 8.62375 2.36666 9.01583 4.5C8.88816 5.19404 8.60119 5.84899 8.1775 6.41333L6.56917 4.805C6.61694 4.48317 6.58868 4.15463 6.48664 3.84569C6.3846 3.53675 6.21162 3.256 5.98156 3.02594C5.7515 2.79588 5.47075 2.6229 5.16181 2.52086C4.85287 2.41882 4.52433 2.39056 4.2025 2.43833L2.83042 1.06708V1.06666Z" fill="currentColor"></path></svg> <span>Hide Pytorch content</span></div></div> <div class="framework-content"> <p data-svelte-h="svelte-6bdz67">Carga el modelo BERT con <code>AutoModelForMultipleChoice</code>:</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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForMultipleChoice, TrainingArguments, Trainer
<span class="hljs-meta">&gt;&gt;&gt; </span>model = AutoModelForMultipleChoice.from_pretrained(<span class="hljs-string">&quot;google-bert/bert-base-uncased&quot;</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"><p data-svelte-h="svelte-1sco78m">Para familiarizarte con el fine-tuning con <code>Trainer</code>, ¡mira el tutorial básico <a href="../training#finetune-with-trainer">aquí</a>!</p></div> <p data-svelte-h="svelte-bd5x35">En este punto, solo quedan tres pasos:</p> <ol data-svelte-h="svelte-k4ouyy"><li>Definir tus hiperparámetros de entrenamiento en <code>TrainingArguments</code>.</li> <li>Pasarle los argumentos del entrenamiento al <code>Trainer</code> jnto con el modelo, el dataset, el tokenizer y el collator de datos.</li> <li>Invocar el método <code>train()</code> para realizar el fine-tuning del modelo.</li></ol> <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">&gt;&gt;&gt; </span>training_args = TrainingArguments(
<span class="hljs-meta">... </span> output_dir=<span class="hljs-string">&quot;./results&quot;</span>,
<span class="hljs-meta">... </span> eval_strategy=<span class="hljs-string">&quot;epoch&quot;</span>,
<span class="hljs-meta">... </span> learning_rate=<span class="hljs-number">5e-5</span>,
<span class="hljs-meta">... </span> per_device_train_batch_size=<span class="hljs-number">16</span>,
<span class="hljs-meta">... </span> per_device_eval_batch_size=<span class="hljs-number">16</span>,
<span class="hljs-meta">... </span> num_train_epochs=<span class="hljs-number">3</span>,
<span class="hljs-meta">... </span> weight_decay=<span class="hljs-number">0.01</span>,
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>trainer = Trainer(
<span class="hljs-meta">... </span> model=model,
<span class="hljs-meta">... </span> args=training_args,
<span class="hljs-meta">... </span> train_dataset=tokenized_swag[<span class="hljs-string">&quot;train&quot;</span>],
<span class="hljs-meta">... </span> eval_dataset=tokenized_swag[<span class="hljs-string">&quot;validation&quot;</span>],
<span class="hljs-meta">... </span> processing_class=tokenizer,
<span class="hljs-meta">... </span> data_collator=DataCollatorForMultipleChoice(tokenizer=tokenizer),
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>trainer.train()<!-- HTML_TAG_END --></pre></div></div></div> <div class="border border-gray-200 rounded-xl px-4 relative"><div class="flex h-[22px] mt-[-12.5px] justify-between leading-none"><div class="flex px-1 items-center space-x-1 bg-white dark:bg-gray-950"><svg class="" 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> <span>TensorFlow</span></div> <div class="cursor-pointer flex items-center justify-center space-x-1 text-sm px-2 bg-white dark:bg-gray-950 hover:underline leading-none"><svg class="" width="0.9em" height="0.9em" viewBox="0 0 10 9" fill="currentColor" xmlns="http://www.w3.org/2000/svg"><path d="M1.39125 1.9725L0.0883333 0.669997L0.677917 0.0804138L8.9275 8.33041L8.33792 8.91958L6.95875 7.54041C6.22592 8.00523 5.37572 8.25138 4.50792 8.25C2.26125 8.25 0.392083 6.63333 0 4.5C0.179179 3.52946 0.667345 2.64287 1.39167 1.9725H1.39125ZM5.65667 6.23833L5.04667 5.62833C4.81335 5.73996 4.55116 5.77647 4.29622 5.73282C4.04129 5.68918 3.80617 5.56752 3.62328 5.38463C3.44039 5.20175 3.31874 4.96663 3.27509 4.71169C3.23144 4.45676 3.26795 4.19456 3.37958 3.96125L2.76958 3.35125C2.50447 3.75187 2.38595 4.2318 2.4341 4.70978C2.48225 5.18777 2.6941 5.63442 3.0338 5.97411C3.37349 6.31381 3.82015 6.52567 4.29813 6.57382C4.77611 6.62197 5.25605 6.50345 5.65667 6.23833ZM2.83042 1.06666C3.35 0.862497 3.91625 0.749997 4.50792 0.749997C6.75458 0.749997 8.62375 2.36666 9.01583 4.5C8.88816 5.19404 8.60119 5.84899 8.1775 6.41333L6.56917 4.805C6.61694 4.48317 6.58868 4.15463 6.48664 3.84569C6.3846 3.53675 6.21162 3.256 5.98156 3.02594C5.7515 2.79588 5.47075 2.6229 5.16181 2.52086C4.85287 2.41882 4.52433 2.39056 4.2025 2.43833L2.83042 1.06708V1.06666Z" fill="currentColor"></path></svg> <span>Hide TensorFlow content</span></div></div> <div class="framework-content"> <p data-svelte-h="svelte-1dn6k4v">Para realizar el fine-tuning de un modelo en TensorFlow, primero convierte tus datasets al formato <code>tf.data.Dataset</code> con el método <code>prepare_tf_dataset()</code>.</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">&gt;&gt;&gt; </span>data_collator = DataCollatorForMultipleChoice(tokenizer=tokenizer)
<span class="hljs-meta">&gt;&gt;&gt; </span>tf_train_set = model.prepare_tf_dataset(
<span class="hljs-meta">... </span> tokenized_swag[<span class="hljs-string">&quot;train&quot;</span>],
<span class="hljs-meta">... </span> shuffle=<span class="hljs-literal">True</span>,
<span class="hljs-meta">... </span> batch_size=batch_size,
<span class="hljs-meta">... </span> collate_fn=data_collator,
<span class="hljs-meta">... </span>)
<span class="hljs-meta">&gt;&gt;&gt; </span>tf_validation_set = model.prepare_tf_dataset(
<span class="hljs-meta">... </span> tokenized_swag[<span class="hljs-string">&quot;validation&quot;</span>],
<span class="hljs-meta">... </span> shuffle=<span class="hljs-literal">False</span>,
<span class="hljs-meta">... </span> batch_size=batch_size,
<span class="hljs-meta">... </span> collate_fn=data_collator,
<span class="hljs-meta">... </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"><p data-svelte-h="svelte-66s4ry">Para familiarizarte con el fine-tuning con Keras, ¡mira el tutorial básico <a href="training#finetune-with-keras">aquí</a>!</p></div> <p data-svelte-h="svelte-81yq28">Prepara una función de optimización, un programa para la tasa de aprendizaje y algunos hiperparámetros de entrenamiento:</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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> create_optimizer
<span class="hljs-meta">&gt;&gt;&gt; </span>batch_size = <span class="hljs-number">16</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>num_train_epochs = <span class="hljs-number">2</span>
<span class="hljs-meta">&gt;&gt;&gt; </span>total_train_steps = (<span class="hljs-built_in">len</span>(tokenized_swag[<span class="hljs-string">&quot;train&quot;</span>]) // batch_size) * num_train_epochs
<span class="hljs-meta">&gt;&gt;&gt; </span>optimizer, schedule = create_optimizer(init_lr=<span class="hljs-number">5e-5</span>, num_warmup_steps=<span class="hljs-number">0</span>, num_train_steps=total_train_steps)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1v9x5sp">Carga el modelo BERT con <code>TFAutoModelForMultipleChoice</code>:</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">&gt;&gt;&gt; </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> TFAutoModelForMultipleChoice
<span class="hljs-meta">&gt;&gt;&gt; </span>model = TFAutoModelForMultipleChoice.from_pretrained(<span class="hljs-string">&quot;google-bert/bert-base-uncased&quot;</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-jy7mgr">Configura el modelo para entrenarlo con <a href="https://keras.io/api/models/model_training_apis/#compile-method" rel="nofollow"><code>compile</code></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">&gt;&gt;&gt; </span>model.<span class="hljs-built_in">compile</span>(optimizer=optimizer)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1dz4xz1">Invoca el método <a href="https://keras.io/api/models/model_training_apis/#fit-method" rel="nofollow"><code>fit</code></a> para realizar el fine-tuning del modelo:</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">&gt;&gt;&gt; </span>model.fit(x=tf_train_set, validation_data=tf_validation_set, epochs=<span class="hljs-number">2</span>)<!-- HTML_TAG_END --></pre></div> </div></div> </div> <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/es/tasks/multiple_choice.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_yx9cql = {
 assets: "/docs/transformers/pr_33913/es",
 base: "/docs/transformers/pr_33913/es",
 env: {}
 };
 const element = document.currentScript.parentElement;
 const data = [null,null];
 Promise.all([
 import("/docs/transformers/pr_33913/es/_app/immutable/entry/start.41996def.js"),
 import("/docs/transformers/pr_33913/es/_app/immutable/entry/app.9ed505df.js")
 ]).then(([kit, app]) => {
 kit.start(app, element, {
 node_ids: [0, 37],
 data,
 form: null,
 error: null
 });
 });
 }
 </script>

Xet Storage Details

Size:
58.9 kB
·
Xet hash:
7093114ebb45ea12f541d3296042e51edf931cb8cef5af10731a5420c618b948

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