Buckets:

hf-doc-build
/

doc-dev

Files

xet

hf-doc-build/doc-dev / datasets /main /en /create_dataset.html

rtrm

28 days ago

download

raw

31.3 kB

	<meta charset="utf-8" /><meta name="hf:doc:metadata" content="{"title":"Create a dataset","local":"create-a-dataset","sections":[{"title":"File-based builders","local":"file-based-builders","sections":[],"depth":2},{"title":"Folder-based builders","local":"folder-based-builders","sections":[],"depth":2},{"title":"From Python dictionaries","local":"from-python-dictionaries","sections":[],"depth":2}],"depth":1}">
	<link href="/docs/datasets/main/en/_app/immutable/assets/0.e3b0c442.css" rel="modulepreload">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/entry/start.4d44eea4.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/scheduler.bdbef820.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/singletons.36b689ad.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/index.8a885b74.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/paths.27092e28.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/entry/app.d83067e8.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/index.c0aea24a.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/nodes/0.bfb01985.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/each.e59479a4.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/nodes/14.04ba0abc.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/CodeBlock.6ccca92e.js">
	<link rel="modulepreload" href="/docs/datasets/main/en/_app/immutable/chunks/EditOnGithub.725ee0c1.js"><!-- HEAD_svelte-u9bgzb_START --><meta name="hf:doc:metadata" content="{"title":"Create a dataset","local":"create-a-dataset","sections":[{"title":"File-based builders","local":"file-based-builders","sections":[],"depth":2},{"title":"Folder-based builders","local":"folder-based-builders","sections":[],"depth":2},{"title":"From Python dictionaries","local":"from-python-dictionaries","sections":[],"depth":2}],"depth":1}"><!-- HEAD_svelte-u9bgzb_END --> <p></p> <h1 class="relative group"><a id="create-a-dataset" 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="#create-a-dataset"><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>Create a dataset</span></h1> <p data-svelte-h="svelte-v9imqp">Sometimes, you may need to create a dataset if you’re working with your own data. Creating a dataset with 🤗 Datasets confers all the advantages of the library to your dataset: fast loading and processing, <a href="stream">stream enormous datasets</a>, <a href="https://huggingface.co/course/chapter5/4?fw=pt#the-magic-of-memory-mapping" rel="nofollow">memory-mapping</a>, and more. You can easily and rapidly create a dataset with 🤗 Datasets low-code approaches, reducing the time it takes to start training a model. In many cases, it is as easy as <a href="upload_dataset#upload-with-the-hub-ui">dragging and dropping</a> your data files into a dataset repository on the Hub.</p> <p data-svelte-h="svelte-1cv0g2h">In this tutorial, you’ll learn how to use 🤗 Datasets low-code methods for creating all types of datasets:</p> <ul data-svelte-h="svelte-1lzxsuu"><li>Folder-based builders for quickly creating an image or audio dataset</li> <li><code>from_</code> methods for creating datasets from local files</li></ul> <h2 class="relative group"><a id="file-based-builders" 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="#file-based-builders"><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>File-based builders</span></h2> <p data-svelte-h="svelte-109f0ey">🤗 Datasets supports many common formats such as <code>csv</code>, <code>json/jsonl</code>, <code>parquet</code>, <code>txt</code>.</p> <p data-svelte-h="svelte-1qx83so">For example it can read a dataset made up of one or several CSV files (in this case, pass your CSV files as a list):</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset
	<span class="hljs-meta">>>> </span>dataset = load_dataset(<span class="hljs-string">"csv"</span>, data_files=<span class="hljs-string">"my_file.csv"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-y2h2bt">To get the list of supported formats and code examples, follow this guide <a href="https://huggingface.co/docs/datasets/loading#local-and-remote-files" rel="nofollow">here</a>.</p> <h2 class="relative group"><a id="folder-based-builders" 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="#folder-based-builders"><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>Folder-based builders</span></h2> <p data-svelte-h="svelte-gu1tts">There are two folder-based builders, <code>ImageFolder</code> and <code>AudioFolder</code>. These are low-code methods for quickly creating an image or speech and audio dataset with several thousand examples. They are great for rapidly prototyping computer vision and speech models before scaling to a larger dataset. Folder-based builders takes your data and automatically generates the dataset’s features, splits, and labels. Under the hood:</p> <ul data-svelte-h="svelte-1qgcixv"><li><code>ImageFolder</code> uses the <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Image">Image</a> feature to decode an image file. Many image extension formats are supported, such as jpg and png, but other formats are also supported. You can check the complete <a href="https://github.com/huggingface/datasets/blob/b5672a956d5de864e6f5550e493527d962d6ae55/src/datasets/packaged_modules/imagefolder/imagefolder.py#L39" rel="nofollow">list</a> of supported image extensions.</li> <li><code>AudioFolder</code> uses the <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Audio">Audio</a> feature to decode an audio file. Audio extensions such as wav and mp3 are supported, and you can check the complete <a href="https://github.com/huggingface/datasets/blob/b5672a956d5de864e6f5550e493527d962d6ae55/src/datasets/packaged_modules/audiofolder/audiofolder.py#L39" rel="nofollow">list</a> of supported audio extensions.</li></ul> <p data-svelte-h="svelte-v88ugk">The dataset splits are generated from the repository structure, and the label names are automatically inferred from the directory name.</p> <p data-svelte-h="svelte-cn9wsj">For example, if your image dataset (it is the same for an audio dataset) is stored like this:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->pokemon<span class="hljs-regexp">/train/g</span>rass/bulbasaur.png
	pokemon<span class="hljs-regexp">/train/</span>fire/charmander.png
	pokemon<span class="hljs-regexp">/train/</span>water/squirtle.png

	pokemon<span class="hljs-regexp">/test/g</span>rass/ivysaur.png
	pokemon<span class="hljs-regexp">/test/</span>fire/charmeleon.png
	pokemon<span class="hljs-regexp">/test/</span>water/wartortle.png<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-5wcoi">Then this is how the folder-based builder generates an example:</p> <div class="flex justify-center" data-svelte-h="svelte-mid65z"><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/datasets/folder-based-builder.png"></div> <p data-svelte-h="svelte-1oxusca">Create the image dataset by specifying <code>imagefolder</code> in <a href="/docs/datasets/main/en/package_reference/loading_methods#datasets.load_dataset">load_dataset()</a>:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset

	<span class="hljs-meta">>>> </span>dataset = load_dataset(<span class="hljs-string">"imagefolder"</span>, data_dir=<span class="hljs-string">"/path/to/pokemon"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-13hsz50">An audio dataset is created in the same way, except you specify <code>audiofolder</code> in <a href="/docs/datasets/main/en/package_reference/loading_methods#datasets.load_dataset">load_dataset()</a> instead:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> load_dataset

	<span class="hljs-meta">>>> </span>dataset = load_dataset(<span class="hljs-string">"audiofolder"</span>, data_dir=<span class="hljs-string">"/path/to/folder"</span>)<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-ai9qzj">Any additional information about your dataset, such as text captions or transcriptions, can be included with a <code>metadata.csv</code> file in the folder containing your dataset. The metadata file needs to have a <code>file_name</code> column that links the image or audio file to its corresponding metadata:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START -->file_name, <span class="hljs-built_in">text</span>
	bulbasaur.png, There <span class="hljs-keyword">is</span> a plant seed <span class="hljs-keyword">on</span> <span class="hljs-keyword">its</span> <span class="hljs-keyword">back</span> right <span class="hljs-keyword">from</span> <span class="hljs-keyword">the</span> <span class="hljs-built_in">day</span> this Pokémon <span class="hljs-keyword">is</span> born.
	charmander.png, It has a preference <span class="hljs-keyword">for</span> hot things.
	squirtle.png, When <span class="hljs-keyword">it</span> retracts <span class="hljs-keyword">its</span> long neck <span class="hljs-keyword">into</span> <span class="hljs-keyword">its</span> shell, <span class="hljs-keyword">it</span> squirts out water <span class="hljs-keyword">with</span> vigorous force.<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-c9ukhj">To learn more about each of these folder-based builders, check out the and <a href="https://huggingface.co/docs/datasets/image_dataset#imagefolder"><span class="underline decoration-yellow-400 decoration-2 font-semibold">ImageFolder</span></a> or <a href="https://huggingface.co/docs/datasets/audio_dataset#audiofolder"><span class="underline decoration-pink-400 decoration-2 font-semibold">AudioFolder</span></a> guides.</p> <h2 class="relative group"><a id="from-python-dictionaries" 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="#from-python-dictionaries"><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>From Python dictionaries</span></h2> <p data-svelte-h="svelte-8iho7z">You can also create a dataset from data in Python dictionaries. There are two ways you can create a dataset using the <code>from_</code> methods:</p> <ul><li><p data-svelte-h="svelte-sk6mcz">The <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Dataset.from_generator">from_generator()</a> method is the most memory-efficient way to create a dataset from a <a href="https://wiki.python.org/moin/Generators" rel="nofollow">generator</a> due to a generators iterative behavior. This is especially useful when you’re working with a really large dataset that may not fit in memory, since the dataset is generated on disk progressively and then memory-mapped.</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> Dataset
	<span class="hljs-meta">>>> </span><span class="hljs-keyword">def</span> <span class="hljs-title function_">gen</span>():
	<span class="hljs-meta">... </span> <span class="hljs-keyword">yield</span> {<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"bulbasaur"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"grass"</span>}
	<span class="hljs-meta">... </span> <span class="hljs-keyword">yield</span> {<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"squirtle"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"water"</span>}
	<span class="hljs-meta">>>> </span>ds = Dataset.from_generator(gen)
	<span class="hljs-meta">>>> </span>ds[<span class="hljs-number">0</span>]
	{<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"bulbasaur"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"grass"</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-2gehho">A generator-based <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.IterableDataset">IterableDataset</a> needs to be iterated over with a <code>for</code> loop for example:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> IterableDataset
	<span class="hljs-meta">>>> </span>ds = IterableDataset.from_generator(gen)
	<span class="hljs-meta">>>> </span><span class="hljs-keyword">for</span> example <span class="hljs-keyword">in</span> ds:
	<span class="hljs-meta">... </span> <span class="hljs-built_in">print</span>(example)
	{<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"bulbasaur"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"grass"</span>}
	{<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"squirtle"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"water"</span>}<!-- HTML_TAG_END --></pre></div></li> <li><p data-svelte-h="svelte-b40znq">The <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Dataset.from_dict">from_dict()</a> method is a straightforward way to create a dataset from a dictionary:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span><span class="hljs-keyword">from</span> datasets <span class="hljs-keyword">import</span> Dataset
	<span class="hljs-meta">>>> </span>ds = Dataset.from_dict({<span class="hljs-string">"pokemon"</span>: [<span class="hljs-string">"bulbasaur"</span>, <span class="hljs-string">"squirtle"</span>], <span class="hljs-string">"type"</span>: [<span class="hljs-string">"grass"</span>, <span class="hljs-string">"water"</span>]})
	<span class="hljs-meta">>>> </span>ds[<span class="hljs-number">0</span>]
	{<span class="hljs-string">"pokemon"</span>: <span class="hljs-string">"bulbasaur"</span>, <span class="hljs-string">"type"</span>: <span class="hljs-string">"grass"</span>}<!-- HTML_TAG_END --></pre></div> <p data-svelte-h="svelte-v3dq16">To create an image or audio dataset, chain the <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Dataset.cast_column">cast_column()</a> method with <a href="/docs/datasets/main/en/package_reference/main_classes#datasets.Dataset.from_dict">from_dict()</a> and specify the column and feature type. For example, to create an audio dataset:</p> <div class="code-block relative"><div class="absolute top-2.5 right-4"><button class="inline-flex items-center relative text-sm focus:text-green-500 cursor-pointer focus:outline-none transition duration-200 ease-in-out opacity-0 mx-0.5 text-gray-600 " title="code excerpt" type="button"><svg class="" xmlns="http://www.w3.org/2000/svg" aria-hidden="true" fill="currentColor" focusable="false" role="img" width="1em" height="1em" preserveAspectRatio="xMidYMid meet" viewBox="0 0 32 32"><path d="M28,10V28H10V10H28m0-2H10a2,2,0,0,0-2,2V28a2,2,0,0,0,2,2H28a2,2,0,0,0,2-2V10a2,2,0,0,0-2-2Z" transform="translate(0)"></path><path d="M4,18H2V4A2,2,0,0,1,4,2H18V4H4Z" transform="translate(0)"></path><rect fill="none" width="32" height="32"></rect></svg> <div class="absolute pointer-events-none transition-opacity bg-black text-white py-1 px-2 leading-tight rounded font-normal shadow left-1/2 top-full transform -translate-x-1/2 translate-y-2 opacity-0"><div class="absolute bottom-full left-1/2 transform -translate-x-1/2 w-0 h-0 border-black border-4 border-t-0" style="border-left-color: transparent; border-right-color: transparent; "></div> Copied</div></button></div> <pre class=""><!-- HTML_TAG_START --><span class="hljs-meta">>>> </span>audio_dataset = Dataset.from_dict({<span class="hljs-string">"audio"</span>: [<span class="hljs-string">"path/to/audio_1"</span>, ..., <span class="hljs-string">"path/to/audio_n"</span>]}).cast_column(<span class="hljs-string">"audio"</span>, Audio())<!-- HTML_TAG_END --></pre></div></li></ul> <p data-svelte-h="svelte-uhu1ny">Now that you know how to create a dataset, consider sharing it on the Hub so the community can also benefit from your work! Go on to the next section to learn how to share your dataset.</p> <a class="!text-gray-400 !no-underline text-sm flex items-center not-prose mt-4" href="https://github.com/huggingface/datasets/blob/main/docs/source/create_dataset.mdx" target="_blank"><span data-svelte-h="svelte-1kd6by1"><</span> <span data-svelte-h="svelte-x0xyl0">></span> <span data-svelte-h="svelte-1dajgef"><span class="underline ml-1.5">Update</span> on GitHub</span></a> <p></p>

	<script>
	{
	__sveltekit_w3org2 = {
	assets: "/docs/datasets/main/en",
	base: "/docs/datasets/main/en",
	env: {}
	};

	const element = document.currentScript.parentElement;

	const data = [null,null];

	Promise.all([
	import("/docs/datasets/main/en/_app/immutable/entry/start.4d44eea4.js"),
	import("/docs/datasets/main/en/_app/immutable/entry/app.d83067e8.js")
	]).then(([kit, app]) => {
	kit.start(app, element, {
	node_ids: [0, 14],
	data,
	form: null,
	error: null
	});
	});
	}
	</script>

Xet Storage Details

Size:: 31.3 kB
Xet hash:: 5775f9dc6170834723d1f13eca918fda8de88e9191b5fdfd0dcde936b651db02

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