Buckets:
| <meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Clasificación de audio","local":"clasificación-de-audio","sections":[{"title":"Carga el dataset MInDS-14","local":"carga-el-dataset-minds-14","sections":[],"depth":2},{"title":"Preprocesamiento","local":"preprocesamiento","sections":[],"depth":2},{"title":"Evaluación","local":"evaluación","sections":[],"depth":2},{"title":"Entrenamiento","local":"entrenamiento","sections":[],"depth":2},{"title":"Inference","local":"inference","sections":[],"depth":2}],"depth":1}"> | |
| <link href="/docs/transformers/main/es/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/entry/start.ef1076ce.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/scheduler.80ba32b1.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/singletons.45a91c7d.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/index.ee85231c.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/paths.258ac66e.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/entry/app.b1bc2a15.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/preload-helper.dd42f084.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/index.2df20f66.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/nodes/0.8f4d7807.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/each.e59479a4.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/nodes/33.2c78cf61.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/Tip.52dd0ccf.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/MermaidChart.svelte_svelte_type_style_lang.9fb840ae.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/Youtube.7aa0749e.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/CodeBlock.1f1e6bf6.js"> | |
| <link rel="modulepreload" href="/docs/transformers/main/es/_app/immutable/chunks/DocNotebookDropdown.ce9a79d4.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"Clasificación de audio","local":"clasificación-de-audio","sections":[{"title":"Carga el dataset MInDS-14","local":"carga-el-dataset-minds-14","sections":[],"depth":2},{"title":"Preprocesamiento","local":"preprocesamiento","sections":[],"depth":2},{"title":"Evaluación","local":"evaluación","sections":[],"depth":2},{"title":"Entrenamiento","local":"entrenamiento","sections":[],"depth":2},{"title":"Inference","local":"inference","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <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> </div> <div class="flex space-x-1 " style="float: right; margin-left: 10px; display: inline-flex; position: relative; z-index: 10;"> <div class="relative colab-dropdown "> <button class=" " type="button"> <img alt="Open In Colab" class="!m-0" src="https://colab.research.google.com/assets/colab-badge.svg"> </button> </div> <div class="relative colab-dropdown "> <button class=" " type="button"> <img alt="Open In Studio Lab" class="!m-0" src="https://studiolab.sagemaker.aws/studiolab.svg"> </button> </div></div> <h1 class="relative group"><a id="clasificación-de-audio" 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="#clasificación-de-audio"><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>Clasificación de audio</span></h1> <iframe class="w-full xl:w-4/6 h-80" src="https://www.youtube-nocookie.com/embed/KWwzcmG98Ds" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> <p data-svelte-h="svelte-1xmo7gs">Clasificación de audio - al igual que con texto — asigna una etiqueta de clase como salida desde las entradas de datos. La diferencia única es en vez de entrada de texto, tiene formas de onda de audio. Algunas aplicaciones prácticas de clasificación incluye identificar la intención del hablante, identificación del idioma, y la clasificación de animales por sus sonidos.</p> <p data-svelte-h="svelte-1j7e2nf">En esta guía te mostraremos como:</p> <ol data-svelte-h="svelte-1hi7ym6"><li>Hacer fine-tuning al modelo <a href="https://huggingface.co/facebook/wav2vec2-base" rel="nofollow">Wav2Vec2</a> en el dataset <a href="https://huggingface.co/datasets/PolyAI/minds14" rel="nofollow">MInDS-14</a> para clasificar la intención del hablante.</li> <li>Usar tu modelo ajustado para tareas de inferencia.</li></ol> <blockquote class="tip"><p data-svelte-h="svelte-b08hpg">Consulta la <a href="https://huggingface.co/tasks/audio-classification" rel="nofollow">página de la tarea</a> de clasificación de audio para acceder a más información sobre los modelos, datasets, y métricas asociados.</p></blockquote> <p data-svelte-h="svelte-dzv42z">Antes de comenzar, asegúrate de haber instalado todas las librerías necesarias:</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="language-bash "><!-- HTML_TAG_START -->pip install transformers datasets evaluate<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-4kxhez">Te aconsejamos iniciar sesión con tu cuenta de Hugging Face para que puedas subir tu modelo y compartirlo con la comunidad. Cuando se te solicite, ingresa tu token para iniciar sesión:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> huggingface_hub <span class="hljs-keyword">import</span> notebook_login | |
| <span class="hljs-meta">>>> </span>notebook_login()<!-- HTML_TAG_END --></pre></div> <h2 class="relative group"><a id="carga-el-dataset-minds-14" 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-minds-14"><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>Carga el dataset MInDS-14</span></h2> <p data-svelte-h="svelte-1sc23cw">Comencemos cargando el dataset MInDS-14 con la biblioteca de 🤗 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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset, Audio | |
| <span class="hljs-meta">>>> </span>minds = load_dataset(<span class="hljs-string">"PolyAI/minds14"</span>, name=<span class="hljs-string">"en-US"</span>, split=<span class="hljs-string">"train"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-19t7vnl">Divide el conjunto de <code>train</code> (entrenamiento) en un conjunto de entrenamiento y prueba mas pequeño con el método <code>train_test_split</code>. De esta forma, tendrás la oportunidad para experimentar y asegúrate de que todo funcióne antes de invertir más tiempo entrenando con el dataset entero.</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>minds = minds.train_test_split(test_size=<span class="hljs-number">0.2</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1br3wmv">Ahora échale un vistazo al 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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>minds | |
| DatasetDict({ | |
| train: Dataset({ | |
| features: [<span class="hljs-string">'path'</span>, <span class="hljs-string">'audio'</span>, <span class="hljs-string">'transcription'</span>, <span class="hljs-string">'english_transcription'</span>, <span class="hljs-string">'intent_class'</span>, <span class="hljs-string">'lang_id'</span>], | |
| num_rows: <span class="hljs-number">450</span> | |
| }) | |
| test: Dataset({ | |
| features: [<span class="hljs-string">'path'</span>, <span class="hljs-string">'audio'</span>, <span class="hljs-string">'transcription'</span>, <span class="hljs-string">'english_transcription'</span>, <span class="hljs-string">'intent_class'</span>, <span class="hljs-string">'lang_id'</span>], | |
| num_rows: <span class="hljs-number">113</span> | |
| }) | |
| })<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-ptkmu9">Aunque el dataset contiene mucha información útil, como los campos <code>land_id</code> (identificador del lenguaje) y <code>english_transcription</code> (transcripción al inglés), en esta guía nos enfocaremos en los campos <code>audio</code> y <code>intent_class</code> (clase de intención). Puedes quitar las otras columnas con cel método <code>remove_columns</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>minds = minds.remove_columns([<span class="hljs-string">"path"</span>, <span class="hljs-string">"transcription"</span>, <span class="hljs-string">"english_transcription"</span>, <span class="hljs-string">"lang_id"</span>])<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-ckoybq">Aquí está un ejemplo:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>minds[<span class="hljs-string">"train"</span>][<span class="hljs-number">0</span>] | |
| {<span class="hljs-string">'audio'</span>: {<span class="hljs-string">'array'</span>: array([ <span class="hljs-number">0.</span> , <span class="hljs-number">0.</span> , <span class="hljs-number">0.</span> , ..., -<span class="hljs-number">0.00048828</span>, | |
| -<span class="hljs-number">0.00024414</span>, -<span class="hljs-number">0.00024414</span>], dtype=float32), | |
| <span class="hljs-string">'path'</span>: <span class="hljs-string">'/root/.cache/huggingface/datasets/downloads/extracted/f14948e0e84be638dd7943ac36518a4cf3324e8b7aa331c5ab11541518e9368c/en-US~APP_ERROR/602b9a5fbb1e6d0fbce91f52.wav'</span>, | |
| <span class="hljs-string">'sampling_rate'</span>: <span class="hljs-number">8000</span>}, | |
| <span class="hljs-string">'intent_class'</span>: <span class="hljs-number">2</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1829lf7">Hay dos campos:</p> <ul data-svelte-h="svelte-1f92w94"><li><code>audio</code>: un <code>array</code> (arreglo) unidimensional de la señal de voz que se obtiene al cargar y volver a muestrear el archivo de audio.</li> <li><code>intent_class</code>: representa el identificador de la clase de la intención del hablante.</li></ul> <p data-svelte-h="svelte-uyw0r">Crea un diccionario que asigne el nombre de la etiqueta a un número entero y viceversa para facilitar la obtención del nombre de la etiqueta a partir de su identificador.</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>labels = minds[<span class="hljs-string">"train"</span>].features[<span class="hljs-string">"intent_class"</span>].names | |
| <span class="hljs-meta">>>> </span>label2id, id2label = <span class="hljs-built_in">dict</span>(), <span class="hljs-built_in">dict</span>() | |
| <span class="hljs-meta">>>> </span><span class="hljs-keyword">for</span> i, label <span class="hljs-keyword">in</span> <span class="hljs-built_in">enumerate</span>(labels): | |
| <span class="hljs-meta">... </span> label2id[label] = <span class="hljs-built_in">str</span>(i) | |
| <span class="hljs-meta">... </span> id2label[<span class="hljs-built_in">str</span>(i)] = label<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-romwjb">Ahora puedes convertir el identificador de la etiqueta a un nombre de etiqueta:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>id2label[<span class="hljs-built_in">str</span>(<span class="hljs-number">2</span>)] | |
| <span class="hljs-string">'app_error'</span><!-- HTML_TAG_END --></pre></div> <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 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>Preprocesamiento</span></h2> <p data-svelte-h="svelte-1sgkp2g">Seguidamente carga el feature extractor (función de extracción de características) de Wav2Vec para procesar la señal de audio:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoFeatureExtractor | |
| <span class="hljs-meta">>>> </span>feature_extractor = AutoFeatureExtractor.from_pretrained(<span class="hljs-string">"facebook/wav2vec2-base"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-qw88cp">El dataset MInDS-14 tiene una tasa de muestreo de 8kHz (puedes encontrar esta información en su <a href="https://huggingface.co/datasets/PolyAI/minds14" rel="nofollow">tarjeta de dataset</a>), lo que significa que tendrás que volver a muestrear el dataset a 16kHZ para poder usar el modelo Wav2Vec2 preentranado:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>minds = minds.cast_column(<span class="hljs-string">"audio"</span>, Audio(sampling_rate=<span class="hljs-number">16_000</span>)) | |
| <span class="hljs-meta">>>> </span>minds[<span class="hljs-string">"train"</span>][<span class="hljs-number">0</span>] | |
| {<span class="hljs-string">'audio'</span>: {<span class="hljs-string">'array'</span>: array([ <span class="hljs-number">2.2098757e-05</span>, <span class="hljs-number">4.6582241e-05</span>, -<span class="hljs-number">2.2803260e-05</span>, ..., | |
| -<span class="hljs-number">2.8419291e-04</span>, -<span class="hljs-number">2.3305941e-04</span>, -<span class="hljs-number">1.1425107e-04</span>], dtype=float32), | |
| <span class="hljs-string">'path'</span>: <span class="hljs-string">'/root/.cache/huggingface/datasets/downloads/extracted/f14948e0e84be638dd7943ac36518a4cf3324e8b7aa331c5ab11541518e9368c/en-US~APP_ERROR/602b9a5fbb1e6d0fbce91f52.wav'</span>, | |
| <span class="hljs-string">'sampling_rate'</span>: <span class="hljs-number">16000</span>}, | |
| <span class="hljs-string">'intent_class'</span>: <span class="hljs-number">2</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-vvxaqw">Ahora vamos a crear una función de preprocesamiento:</p> <ol data-svelte-h="svelte-edgbxy"><li>Invoque la columna <code>audio</code> para cargar, y si es necesario, volver a muestrear al archivo de audio.</li> <li>Comprueba si la frecuencia de muestreo del archivo de audio coincide con la frecuencia de muestreo de los datos de audio con los que se entrenó previamente el modelo. Puedes encontrar esta información en la <a href="https://huggingface.co/facebook/wav2vec2-base" rel="nofollow">tarjeta de modelo</a> de Wav2Vec2.</li> <li>Establece una longitud de entrada máxima para agrupar entradas más largas sin truncarlas.</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="language-py "><!-- HTML_TAG_START --><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> audio_arrays = [x[<span class="hljs-string">"array"</span>] <span class="hljs-keyword">for</span> x <span class="hljs-keyword">in</span> examples[<span class="hljs-string">"audio"</span>]] | |
| <span class="hljs-meta">... </span> inputs = feature_extractor( | |
| <span class="hljs-meta">... </span> audio_arrays, sampling_rate=feature_extractor.sampling_rate, max_length=<span class="hljs-number">16000</span>, truncation=<span class="hljs-literal">True</span> | |
| <span class="hljs-meta">... </span> ) | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> inputs<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-pp9odp">Para aplicar la función de preprocesamiento a todo el dataset, puedes usar la función <code>map</code> de 🤗 Datasets. Acelera la función <code>map</code> haciendo <code>batched=True</code> para procesar varios elementos del dataset a la vez. Quitas las columnas que no necesites con el método <code>[~datasets.Dataset.remove_columns]</code> y cambia el nombre de <code>intent_class</code> a <code>label</code>, como requiere el 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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>encoded_minds = minds.<span class="hljs-built_in">map</span>(preprocess_function, remove_columns=<span class="hljs-string">"audio"</span>, batched=<span class="hljs-literal">True</span>) | |
| <span class="hljs-meta">>>> </span>encoded_minds = encoded_minds.rename_column(<span class="hljs-string">"intent_class"</span>, <span class="hljs-string">"label"</span>)<!-- HTML_TAG_END --></pre></div> <h2 class="relative group"><a id="evaluación" 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="#evaluación"><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>Evaluación</span></h2> <p data-svelte-h="svelte-xg6zj4">A menudo es útil incluir una métrica durante el entrenamiento para evaluar el rendimiento de tu modelo. Puedes cargar un método de evaluación rapidamente con la biblioteca de 🤗 <a href="https://huggingface.co/docs/evaluate/index" rel="nofollow">Evaluate</a>. Para esta tarea, puedes usar la métrica de <a href="https://huggingface.co/spaces/evaluate-metric/accuracy" rel="nofollow">exactitud</a> (accuracy). Puedes ver la <a href="https://huggingface.co/docs/evaluate/a_quick_tour" rel="nofollow">guía rápida</a> de 🤗 Evaluate para aprender más de cómo cargar y computar una métrica:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">import</span> evaluate | |
| <span class="hljs-meta">>>> </span>accuracy = evaluate.load(<span class="hljs-string">"accuracy"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-s7nkfp">Ahora crea una función que le pase tus predicciones y etiquetas a <code>compute</code> para calcular la exactitud:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">import</span> numpy <span class="hljs-keyword">as</span> np | |
| <span class="hljs-meta">>>> </span><span class="hljs-keyword">def</span> <span class="hljs-title function_">compute_metrics</span>(<span class="hljs-params">eval_pred</span>): | |
| <span class="hljs-meta">... </span> predictions = np.argmax(eval_pred.predictions, axis=<span class="hljs-number">1</span>) | |
| <span class="hljs-meta">... </span> <span class="hljs-keyword">return</span> accuracy.compute(predictions=predictions, references=eval_pred.label_ids)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1sl6wg0">Ahora tu función <code>compute_metrics</code> (computar métricas) está lista y podrás usarla cuando estés preparando tu entrenamiento.</p> <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> <blockquote class="tip"><p data-svelte-h="svelte-1hoblby">¡Si no tienes experiencia haciéndo <em>fine-tuning</em> a un modelo con el <code>Trainer</code>, échale un vistazo al tutorial básico <a href="../training#train-with-pytorch-trainer">aquí</a>!</p></blockquote> <p data-svelte-h="svelte-1y4i0ou">¡Ya puedes empezar a entrenar tu modelo! Carga Wav2Vec2 con <code>AutoModelForAudioClassification</code> junto con el especifica el número de etiquetas, y pasa al modelo los <em>mappings</em> entre el número entero de etiqueta y la clase de etiqueta.</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForAudioClassification, TrainingArguments, Trainer | |
| <span class="hljs-meta">>>> </span>num_labels = <span class="hljs-built_in">len</span>(id2label) | |
| <span class="hljs-meta">>>> </span>model = AutoModelForAudioClassification.from_pretrained( | |
| <span class="hljs-meta">... </span> <span class="hljs-string">"facebook/wav2vec2-base"</span>, num_labels=num_labels, label2id=label2id, id2label=id2label | |
| <span class="hljs-meta">... </span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-yip3uv">Al llegar a este punto, solo quedan tres pasos:</p> <ol data-svelte-h="svelte-1c3fyoj"><li>Define tus hiperparámetros de entrenamiento en <code>TrainingArguments</code>. El único parámetro obligatorio es <code>output_dir</code> (carpeta de salida), el cual especifica dónde guardar tu modelo. Puedes subir este modelo al Hub haciendo <code>push_to_hub=True</code> (debes haber iniciado sesión en Hugging Face para subir tu modelo). Al final de cada época, el <code>Trainer</code> evaluará la exactitud y guardará el punto de control del entrenamiento.</li> <li>Pásale los argumentos del entrenamiento al <code>Trainer</code> junto con el modelo, el dataset, el tokenizer, el data collator y la función <code>compute_metrics</code>.</li> <li>Llama el método <code>train()</code> para hacerle fine-tuning a tu 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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>training_args = TrainingArguments( | |
| <span class="hljs-meta">... </span> output_dir=<span class="hljs-string">"my_awesome_mind_model"</span>, | |
| <span class="hljs-meta">... </span> eval_strategy=<span class="hljs-string">"epoch"</span>, | |
| <span class="hljs-meta">... </span> save_strategy=<span class="hljs-string">"epoch"</span>, | |
| <span class="hljs-meta">... </span> learning_rate=<span class="hljs-number">3e-5</span>, | |
| <span class="hljs-meta">... </span> per_device_train_batch_size=<span class="hljs-number">32</span>, | |
| <span class="hljs-meta">... </span> gradient_accumulation_steps=<span class="hljs-number">4</span>, | |
| <span class="hljs-meta">... </span> per_device_eval_batch_size=<span class="hljs-number">32</span>, | |
| <span class="hljs-meta">... </span> num_train_epochs=<span class="hljs-number">10</span>, | |
| <span class="hljs-meta">... </span> warmup_steps=<span class="hljs-number">0.1</span>, | |
| <span class="hljs-meta">... </span> logging_steps=<span class="hljs-number">10</span>, | |
| <span class="hljs-meta">... </span> load_best_model_at_end=<span class="hljs-literal">True</span>, | |
| <span class="hljs-meta">... </span> metric_for_best_model=<span class="hljs-string">"accuracy"</span>, | |
| <span class="hljs-meta">... </span> push_to_hub=<span class="hljs-literal">True</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=encoded_minds[<span class="hljs-string">"train"</span>], | |
| <span class="hljs-meta">... </span> eval_dataset=encoded_minds[<span class="hljs-string">"test"</span>], | |
| <span class="hljs-meta">... </span> processing_class=feature_extractor, | |
| <span class="hljs-meta">... </span> compute_metrics=compute_metrics, | |
| <span class="hljs-meta">... </span>) | |
| <span class="hljs-meta">>>> </span>trainer.train()<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-1hgs9nu">Una vez que el entrenamiento haya sido completado, comparte tu modelo en el Hub con el método <code>push_to_hub()</code> para que todo el mundo puede usar tu 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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>trainer.push_to_hub()<!-- HTML_TAG_END --></pre></div> <blockquote class="tip"><p data-svelte-h="svelte-n1zqug">Para ver un ejemplo más detallado de comó hacerle fine-tuning a un modelo para clasificación, échale un vistazo al correspondiente <a href="https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/audio_classification.ipynb" rel="nofollow">PyTorch notebook</a>.</p></blockquote> <h2 class="relative group"><a id="inference" 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="#inference"><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>Inference</span></h2> <p data-svelte-h="svelte-5ii2rt">¡Genial, ahora que le has hecho <em>fine-tuned</em> a un modelo, puedes usarlo para hacer inferencia!</p> <p data-svelte-h="svelte-kdy0mi">Carga el archivo de audio para hacer inferencia. Recuerda volver a muestrear la tasa de muestreo del archivo de audio para que sea la misma del modelo si es necesario.</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset, Audio | |
| <span class="hljs-meta">>>> </span>dataset = load_dataset(<span class="hljs-string">"PolyAI/minds14"</span>, name=<span class="hljs-string">"en-US"</span>, split=<span class="hljs-string">"train"</span>) | |
| <span class="hljs-meta">>>> </span>dataset = dataset.cast_column(<span class="hljs-string">"audio"</span>, Audio(sampling_rate=<span class="hljs-number">16000</span>)) | |
| <span class="hljs-meta">>>> </span>sampling_rate = dataset.features[<span class="hljs-string">"audio"</span>].sampling_rate | |
| <span class="hljs-meta">>>> </span>audio_file = dataset[<span class="hljs-number">0</span>][<span class="hljs-string">"audio"</span>][<span class="hljs-string">"path"</span>]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-136oep4">La manera más simple de probar tu modelo para hacer inferencia es usarlo en un <code>pipeline()</code>. Puedes instanciar un <code>pipeline</code> para clasificación de audio con tu modelo y pasarle tu archivo de audio:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> pipeline | |
| <span class="hljs-meta">>>> </span>classifier = pipeline(<span class="hljs-string">"audio-classification"</span>, model=<span class="hljs-string">"stevhliu/my_awesome_minds_model"</span>) | |
| <span class="hljs-meta">>>> </span>classifier(audio_file) | |
| [ | |
| {<span class="hljs-string">'score'</span>: <span class="hljs-number">0.09766869246959686</span>, <span class="hljs-string">'label'</span>: <span class="hljs-string">'cash_deposit'</span>}, | |
| {<span class="hljs-string">'score'</span>: <span class="hljs-number">0.07998877018690109</span>, <span class="hljs-string">'label'</span>: <span class="hljs-string">'app_error'</span>}, | |
| {<span class="hljs-string">'score'</span>: <span class="hljs-number">0.0781070664525032</span>, <span class="hljs-string">'label'</span>: <span class="hljs-string">'joint_account'</span>}, | |
| {<span class="hljs-string">'score'</span>: <span class="hljs-number">0.07667109370231628</span>, <span class="hljs-string">'label'</span>: <span class="hljs-string">'pay_bill'</span>}, | |
| {<span class="hljs-string">'score'</span>: <span class="hljs-number">0.0755252093076706</span>, <span class="hljs-string">'label'</span>: <span class="hljs-string">'balance'</span>} | |
| ]<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-o5jka0">También puedes replicar de forma manual los resultados del <code>pipeline</code> si lo deseas:</p> <p data-svelte-h="svelte-1sprg68">Carga el feature extractor para preprocesar el archivo de audio y devuelve el <code>input</code> como un tensor de PyTorch:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoFeatureExtractor | |
| <span class="hljs-meta">>>> </span>feature_extractor = AutoFeatureExtractor.from_pretrained(<span class="hljs-string">"stevhliu/my_awesome_minds_model"</span>) | |
| <span class="hljs-meta">>>> </span>inputs = feature_extractor(dataset[<span class="hljs-number">0</span>][<span class="hljs-string">"audio"</span>][<span class="hljs-string">"array"</span>], sampling_rate=sampling_rate, return_tensors=<span class="hljs-string">"pt"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-xfmo9s">Pásale tus entradas al modelo y devuelve los logits:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> transformers <span class="hljs-keyword">import</span> AutoModelForAudioClassification | |
| <span class="hljs-meta">>>> </span>model = AutoModelForAudioClassification.from_pretrained(<span class="hljs-string">"stevhliu/my_awesome_minds_model"</span>) | |
| <span class="hljs-meta">>>> </span><span class="hljs-keyword">with</span> torch.no_grad(): | |
| <span class="hljs-meta">... </span> logits = model(**inputs).logits<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-cf4cid">Obtén los identificadores de los clases con mayor probabilidad y usa el <em>mapping</em> <code>id2label</code> del modelo para convertirle a una etiqueta:</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="language-py "><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">import</span> torch | |
| <span class="hljs-meta">>>> </span>predicted_class_ids = torch.argmax(logits).item() | |
| <span class="hljs-meta">>>> </span>predicted_label = model.config.id2label[predicted_class_ids] | |
| <span class="hljs-meta">>>> </span>predicted_label | |
| <span class="hljs-string">'cash_deposit'</span><!-- HTML_TAG_END --></pre></div> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/transformers/blob/main/docs/source/es/tasks/audio_classification.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 data-svelte-h="svelte-zjs2n5"><span class="underline">Update</span> on GitHub</span></a> <p></p> | |
| <script> | |
| { | |
| __sveltekit_1buzbw1 = { | |
| assets: "/docs/transformers/main/es", | |
| base: "/docs/transformers/main/es", | |
| env: {} | |
| }; | |
| const element = document.currentScript.parentElement; | |
| const data = [null,null]; | |
| Promise.all([ | |
| import("/docs/transformers/main/es/_app/immutable/entry/start.ef1076ce.js"), | |
| import("/docs/transformers/main/es/_app/immutable/entry/app.b1bc2a15.js") | |
| ]).then(([kit, app]) => { | |
| kit.start(app, element, { | |
| node_ids: [0, 33], | |
| data, | |
| form: null, | |
| error: null | |
| }); | |
| }); | |
| } | |
| </script> | |
Xet Storage Details
- Size:
- 66.3 kB
- Xet hash:
- 6ebd1544ee06d27bafdae7dcec7a6ba7740f97aa2ba205ee8867831ed28c4460
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.