Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Respuesta a preguntas","local":"respuesta-a-preguntas","sections":[{"title":"Carga el dataset SQuAD","local":"carga-el-dataset-squad","sections":[],"depth":2},{"title":"Preprocesamiento","local":"preprocesamiento","sections":[],"depth":2},{"title":"Entrenamiento","local":"entrenamiento","sections":[],"depth":2}],"depth":1}"/> | |
| <link href="/docs/transformers/main/es/_app/immutable/entry/start.Cex4YXjh.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/D_b9q9jZ.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/Dl9aQtdI.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/entry/app.DNxQXwTy.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/ByWXHg74.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/NvhOjA8g.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/BWsCTqVr.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/DsnmJJEf.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/nodes/0.DUkXUlQz.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/DkDiSK5v.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/nodes/38.DLjHYNJ9.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/Cy51lpuW.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/71x9jLwp.js" rel="modulepreload"> | |
| <link href="/docs/transformers/main/es/_app/immutable/chunks/Cwgg6SDS.js" rel="modulepreload"> | |
| <!--1cfiqxb--><meta name="hf:doc:metadata" content="{"title":"Respuesta a preguntas","local":"respuesta-a-preguntas","sections":[{"title":"Carga el dataset SQuAD","local":"carga-el-dataset-squad","sections":[],"depth":2},{"title":"Preprocesamiento","local":"preprocesamiento","sections":[],"depth":2},{"title":"Entrenamiento","local":"entrenamiento","sections":[],"depth":2}],"depth":1}"/><!----> | |
| <link href="/docs/transformers/main/es/_app/immutable/assets/0.tn0RQdqM.css" rel="modulepreload"> <!--[--><!--[0--><!--[--><!--[0--><!--[--><p></p> <div class="items-center shrink-0 min-w-[100px] max-sm:min-w-[50px] justify-end ml-auto flex" style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"><div class="inline-flex rounded-md max-sm:rounded-sm"><button class="inline-flex items-center gap-1 h-7 max-sm:h-7 px-2 max-sm:px-1.5 text-sm font-medium text-gray-800 border border-r-0 rounded-l-md max-sm:rounded-l-sm border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-live="polite"><span class="inline-flex items-center justify-center rounded-md p-0.5 max-sm:p-0 hover:text-gray-800 dark:hover:text-gray-200"><svg class="sm:size-3.5 size-3" 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><!----></span> <span>Copy page</span></button> <button class="inline-flex items-center justify-center w-6 max-sm:w-5 h-7 max-sm:h-7 disabled:pointer-events-none text-sm text-gray-500 hover:text-gray-700 dark:hover:text-white rounded-r-md max-sm:rounded-r-sm border border-l transition border-gray-200 bg-white hover:shadow-inner dark:border-gray-850 dark:bg-gray-950 dark:text-gray-200 dark:hover:bg-gray-800" aria-haspopup="menu" aria-expanded="false" aria-label="Open copy menu"><svg class="transition-transform text-gray-400 overflow-visible sm:size-3.5 size-3 rotate-0" width="1em" height="1em" viewBox="0 0 12 7" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M1 1L6 6L11 1" stroke="currentColor"></path></svg><!----></button></div> <!--[-1--><!--]--></div><!----> <!--[0--><h1 class="relative group"><a id="respuesta-a-preguntas" 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="#respuesta-a-preguntas"><span><svg 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>Respuesta a preguntas</span></h1><!--]--><!----> <iframe class="w-full xl:w-4/6 h-80" src="https://www.youtube-nocookie.com/embed/ajPx5LwJD-I" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen=""></iframe><!----> <p>La respuesta a preguntas devuelve una respuesta a partir de una pregunta dada. Existen dos formas comunes de responder preguntas:</p> <ul><li>Extractiva: extraer la respuesta a partir del contexto dado.</li> <li>Abstractiva: generar una respuesta que responda correctamente la pregunta a partir del contexto dado.</li></ul> <p>Esta guía te mostrará como hacer fine-tuning de <a href="https://huggingface.co/distilbert/distilbert-base-uncased" rel="nofollow">DistilBERT</a> en el dataset <a href="https://huggingface.co/datasets/squad" rel="nofollow">SQuAD</a> para responder preguntas de forma extractiva.</p> <blockquote class="tip"><p>Revisa la <a href="https://huggingface.co/tasks/question-answering" rel="nofollow">página de la tarea</a> de responder preguntas para tener más información sobre otras formas de responder preguntas y los modelos, datasets y métricas asociadas.</p><!----></blockquote><!----> <!--[1--><h2 class="relative group"><a id="carga-el-dataset-squad" 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="#carga-el-dataset-squad"><span><svg 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>Carga el dataset SQuAD</span></h2><!--]--><!----> <p>Carga el dataset SQuAD 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 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="language-py "><!----><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset | |
| <span class="hljs-meta">>>> </span>squad = load_dataset(<span class="hljs-string">"squad"</span>)<!----></pre></div><!----> <p>Ahora, échale un vistazo a una muestra:</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 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="language-py "><!----><span class="hljs-meta">>>> </span>squad[<span class="hljs-string">"train"</span>][<span class="hljs-number">0</span>] | |
| {<span class="hljs-string">'answers'</span>: {<span class="hljs-string">'answer_start'</span>: [<span class="hljs-number">515</span>], <span class="hljs-string">'text'</span>: [<span class="hljs-string">'Saint Bernadette Soubirous'</span>]}, | |
| <span class="hljs-string">'context'</span>: <span class="hljs-string">'Architecturally, the school has a Catholic character. Atop the Main Building\'s gold dome is a golden statue of the Virgin Mary. Immediately in front of the Main Building and facing it, is a copper statue of Christ with arms upraised with the legend "Venite Ad Me Omnes". Next to the Main Building is the Basilica of the Sacred Heart. Immediately behind the basilica is the Grotto, a Marian place of prayer and reflection. It is a replica of the grotto at Lourdes, France where the Virgin Mary reputedly appeared to Saint Bernadette Soubirous in 1858. At the end of the main drive (and in a direct line that connects through 3 statues and the Gold Dome), is a simple, modern stone statue of Mary.'</span>, | |
| <span class="hljs-string">'id'</span>: <span class="hljs-string">'5733be284776f41900661182'</span>, | |
| <span class="hljs-string">'question'</span>: <span class="hljs-string">'To whom did the Virgin Mary allegedly appear in 1858 in Lourdes France?'</span>, | |
| <span class="hljs-string">'title'</span>: <span class="hljs-string">'University_of_Notre_Dame'</span> | |
| }<!----></pre></div><!----> <p>El campo <code>answers</code> es un diccionario que contiene la posición inicial de la respuesta y el <code>texto</code> de la respuesta.</p> <!--[1--><h2 class="relative group"><a id="preprocesamiento" 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="#preprocesamiento"><span><svg 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>Preprocesamiento</span></h2><!--]--><!----> <iframe class="w-full xl:w-4/6 h-80" src="https://www.youtube-nocookie.com/embed/qgaM0weJHpA" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen=""></iframe><!----> <p>Carga el tokenizer de DistilBERT para procesar los campos <code>question</code> (pregunta) y <code>context</code> (contexto):</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 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="language-py "><!----><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoTokenizer | |
| <span class="hljs-meta">>>> </span>tokenizer = AutoTokenizer.from_pretrained(<span class="hljs-string">"distilbert/distilbert-base-uncased"</span>)<!----></pre></div><!----> <p>Hay algunos pasos de preprocesamiento específicos para la tarea de respuesta a preguntas que debes tener en cuenta:</p> <ol><li>Algunos ejemplos en un dataset pueden tener un contexto que supera la longitud máxima de entrada de un modelo. Trunca solamente el contexto asignándole el valor <code>"only_second"</code> al parámetro <code>truncation</code>.</li> <li>A continuación, mapea las posiciones de inicio y fin de la respuesta al contexto original asignándole el valor <code>True</code> al parámetro <code>return_offsets_mapping</code>.</li> <li>Una vez tengas el mapeo, puedes encontrar los tokens de inicio y fin de la respuesta. Usa el método <a href="https://huggingface.co/docs/tokenizers/python/latest/api/reference.html#tokenizers.Encoding.sequence_ids" rel="nofollow"><code>sequence_ids</code></a> para encontrar qué parte de la lista de tokens desplazados corresponde a la pregunta y cuál corresponde al contexto.</li></ol> <p>A continuación puedes ver como se crea una función para truncar y mapear los tokens de inicio y fin de la respuesta al <code>context</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 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="language-py "><!----><span class="hljs-meta">>>> </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> questions = [q.strip() <span class="hljs-keyword">for</span> q <span class="hljs-keyword">in</span> examples[<span class="hljs-string">"question"</span>]] | |
| <span class="hljs-meta">... </span> inputs = tokenizer( | |
| <span class="hljs-meta">... </span> questions, | |
| <span class="hljs-meta">... </span> examples[<span class="hljs-string">"context"</span>], | |
| <span class="hljs-meta">... </span> max_length=<span class="hljs-number">384</span>, | |
| <span class="hljs-meta">... </span> truncation=<span class="hljs-string">"only_second"</span>, | |
| <span class="hljs-meta">... </span> return_offsets_mapping=<span class="hljs-literal">True</span>, | |
| <span class="hljs-meta">... </span> padding=<span class="hljs-string">"max_length"</span>, | |
| <span class="hljs-meta">... </span> ) | |
| <span class="hljs-meta">... </span> offset_mapping = inputs.pop(<span class="hljs-string">"offset_mapping"</span>) | |
| <span class="hljs-meta">... </span> answers = examples[<span class="hljs-string">"answers"</span>] | |
| <span class="hljs-meta">... </span> start_positions = [] | |
| <span class="hljs-meta">... </span> end_positions = [] | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">for</span> i, offset <span class="hljs-keyword">in</span> <span class="hljs-built_in">enumerate</span>(offset_mapping): | |
| <span class="hljs-meta">... </span> answer = answers[i] | |
| <span class="hljs-meta">... </span> start_char = answer[<span class="hljs-string">"answer_start"</span>][<span class="hljs-number">0</span>] | |
| <span class="hljs-meta">... </span> end_char = answer[<span class="hljs-string">"answer_start"</span>][<span class="hljs-number">0</span>] + <span class="hljs-built_in">len</span>(answer[<span class="hljs-string">"text"</span>][<span class="hljs-number">0</span>]) | |
| <span class="hljs-meta">... </span> sequence_ids = inputs.sequence_ids(i) | |
| <span class="hljs-meta">... </span> <span class="hljs-comment"># Encuentra el inicio y el fin del contexto</span> | |
| <span class="hljs-meta">... </span> idx = <span class="hljs-number">0</span> | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">while</span> sequence_ids[idx] != <span class="hljs-number">1</span>: | |
| <span class="hljs-meta">... </span> idx += <span class="hljs-number">1</span> | |
| <span class="hljs-meta">... </span> context_start = idx | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">while</span> sequence_ids[idx] == <span class="hljs-number">1</span>: | |
| <span class="hljs-meta">... </span> idx += <span class="hljs-number">1</span> | |
| <span class="hljs-meta">... </span> context_end = idx - <span class="hljs-number">1</span> | |
| <span class="hljs-meta">... </span> <span class="hljs-comment"># Si la respuesta entera no está dentro del contexto, etiquétala como (0, 0)</span> | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">if</span> offset[context_start][<span class="hljs-number">0</span>] > end_char <span class="hljs-keyword">or</span> offset[context_end][<span class="hljs-number">1</span>] < start_char: | |
| <span class="hljs-meta">... </span> start_positions.append(<span class="hljs-number">0</span>) | |
| <span class="hljs-meta">... </span> end_positions.append(<span class="hljs-number">0</span>) | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">else</span>: | |
| <span class="hljs-meta">... </span> <span class="hljs-comment"># De lo contrario, esta es la posición de los tokens de inicio y fin</span> | |
| <span class="hljs-meta">... </span> idx = context_start | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">while</span> idx <= context_end <span class="hljs-keyword">and</span> offset[idx][<span class="hljs-number">0</span>] <= start_char: | |
| <span class="hljs-meta">... </span> idx += <span class="hljs-number">1</span> | |
| <span class="hljs-meta">... </span> start_positions.append(idx - <span class="hljs-number">1</span>) | |
| <span class="hljs-meta">... </span> idx = context_end | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">while</span> idx >= context_start <span class="hljs-keyword">and</span> offset[idx][<span class="hljs-number">1</span>] >= end_char: | |
| <span class="hljs-meta">... </span> idx -= <span class="hljs-number">1</span> | |
| <span class="hljs-meta">... </span> end_positions.append(idx + <span class="hljs-number">1</span>) | |
| <span class="hljs-meta">... </span> inputs[<span class="hljs-string">"start_positions"</span>] = start_positions | |
| <span class="hljs-meta">... </span> inputs[<span class="hljs-string">"end_positions"</span>] = end_positions | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> inputs<!----></pre></div><!----> <p>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. | |
| Quita las columnas que no necesites:</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 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="language-py "><!----><span class="hljs-meta">>>> </span>tokenized_squad = squad.<span class="hljs-built_in">map</span>(preprocess_function, batched=<span class="hljs-literal">True</span>, remove_columns=squad[<span class="hljs-string">"train"</span>].column_names)<!----></pre></div><!----> <p>Usa el <code>DefaultDataCollator</code> para crear un lote de ejemplos. A diferencia de los otros collators de datos en 🤗 Transformers, el <code>DefaultDataCollator</code> no aplica ningún procesamiento adicional (como el rellenado).</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 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="language-py "><!----><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> DefaultDataCollator | |
| <span class="hljs-meta">>>> </span>data_collator = DefaultDataCollator()<!----></pre></div><!----> <!--[1--><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 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><!--]--><!----> <p>Carga el modelo DistilBERT con <code>AutoModelForQuestionAnswering</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 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="language-py "><!----><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForQuestionAnswering, TrainingArguments, Trainer | |
| <span class="hljs-meta">>>> </span>model = AutoModelForQuestionAnswering.from_pretrained(<span class="hljs-string">"distilbert/distilbert-base-uncased"</span>)<!----></pre></div><!----> <blockquote class="tip"><p>Para familiarizarte con el fine-tuning con <code>Trainer</code>, ¡mira el tutorial básico <a href="../training#finetune-with-trainer">aquí</a>!</p><!----></blockquote><!----> <p>En este punto, solo quedan tres pasos:</p> <ol><li>Definir tus hiperparámetros de entrenamiento en <code>TrainingArguments</code>.</li> <li>Pasarle los argumentos del entrenamiento al <code>Trainer</code> junto 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 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="language-py "><!----><span class="hljs-meta">>>> </span>training_args = TrainingArguments( | |
| <span class="hljs-meta">... </span> output_dir=<span class="hljs-string">"./results"</span>, | |
| <span class="hljs-meta">... </span> eval_strategy=<span class="hljs-string">"epoch"</span>, | |
| <span class="hljs-meta">... </span> learning_rate=<span class="hljs-number">2e-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">>>> </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_squad[<span class="hljs-string">"train"</span>], | |
| <span class="hljs-meta">... </span> eval_dataset=tokenized_squad[<span class="hljs-string">"validation"</span>], | |
| <span class="hljs-meta">... </span> processing_class=tokenizer, | |
| <span class="hljs-meta">... </span> data_collator=data_collator, | |
| <span class="hljs-meta">... </span>) | |
| <span class="hljs-meta">>>> </span>trainer.train()<!----></pre></div><!----> <blockquote class="tip"><p>Para un ejemplo con mayor profundidad de cómo hacer fine-tuning a un modelo para responder preguntas, échale un vistazo al <a href="https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb" rel="nofollow">cuaderno de PyTorch</a> o al <a href="https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb" rel="nofollow">cuaderno de TensorFlow</a> correspondiente.</p><!----></blockquote><!----> <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/question_answering.md" target="_blank"><svg class="mr-1" 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="M31,16l-7,7l-1.41-1.41L28.17,16l-5.58-5.59L24,9l7,7z"></path><path d="M1,16l7-7l1.41,1.41L3.83,16l5.58,5.59L8,23l-7-7z"></path><path d="M12.419,25.484L17.639,6.552l1.932,0.518L14.351,26.002z"></path></svg><!----> <span><span class="underline">Update</span> on GitHub</span></a><!----> <p></p><!--]--><!----><!--]--><!--]--><!--]--> <!--[-1--><!--]--><!--]--> | |
| <script> | |
| { | |
| __sveltekit_y4vo64 = { | |
| base: "/docs/transformers/main/es", | |
| assets: "/docs/transformers/main/es" | |
| }; | |
| const element = document.currentScript.parentElement; | |
| Promise.all([ | |
| import("/docs/transformers/main/es/_app/immutable/entry/start.Cex4YXjh.js"), | |
| import("/docs/transformers/main/es/_app/immutable/entry/app.DNxQXwTy.js") | |
| ]).then(([kit, app]) => { | |
| kit.start(app, element, { | |
| node_ids: [0, 38], | |
| data: [null,null], | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 34.8 kB
- Xet hash:
- eb3e748c0937c3fce7f15efaeba63e0ee766d4968e2405eb5bdab66f1186b438
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.