🤗 Datasets is a lightweight library providing two main features:
- one-line dataloaders for many public datasets: one liners to download and pre-process any of the major public datasets (in 467 languages and dialects!) provided on the HuggingFace Datasets Hub. With a simple command like
squad_dataset = load_dataset("squad")
, get any of these datasets ready to use in a dataloader for training/evaluating a ML model (Numpy/Pandas/PyTorch/TensorFlow/JAX), - efficient data pre-processing: simple, fast and reproducible data pre-processing for the above public datasets as well as your own local datasets in CSV/JSON/text. With simple commands like
tokenized_dataset = dataset.map(tokenize_example)
, efficiently prepare the dataset for inspection and ML model evaluation and training.
🎓 Documentation 🕹 Colab tutorial
🔎 Find a dataset in the Hub 🌟 Add a new dataset to the Hub
🤗 Datasets also provides access to +15 evaluation metrics and is designed to let the community easily add and share new datasets and evaluation metrics.
🤗 Datasets has many additional interesting features:
- Thrive on large datasets: 🤗 Datasets naturally frees the user from RAM memory limitation, all datasets are memory-mapped using an efficient zero-serialization cost backend (Apache Arrow).
- Smart caching: never wait for your data to process several times.
- Lightweight and fast with a transparent and pythonic API (multi-processing/caching/memory-mapping).
- Built-in interoperability with NumPy, pandas, PyTorch, Tensorflow 2 and JAX.
🤗 Datasets originated from a fork of the awesome TensorFlow Datasets and the HuggingFace team want to deeply thank the TensorFlow Datasets team for building this amazing library. More details on the differences between 🤗 Datasets and tfds
can be found in the section Main differences between 🤗 Datasets and tfds
.
🤗 Datasets can be installed from PyPi and has to be installed in a virtual environment (venv or conda for instance)
pip install datasets
🤗 Datasets can be installed using conda as follows:
conda install -c huggingface -c conda-forge datasets
Follow the installation pages of TensorFlow and PyTorch to see how to install them with conda.
For more details on installation, check the installation page in the documentation: https://huggingface.co/docs/datasets/installation.html
If you plan to use 🤗 Datasets with PyTorch (1.0+), TensorFlow (2.2+) or pandas, you should also install PyTorch, TensorFlow or pandas.
For more details on using the library with NumPy, pandas, PyTorch or TensorFlow, check the quick tour page in the documentation: https://huggingface.co/docs/datasets/quicktour.html
🤗 Datasets is made to be very simple to use. The main methods are:
datasets.list_datasets()
to list the available datasetsdatasets.load_dataset(dataset_name, **kwargs)
to instantiate a datasetdatasets.list_metrics()
to list the available metricsdatasets.load_metric(metric_name, **kwargs)
to instantiate a metric
Here is a quick example:
from datasets import list_datasets, load_dataset, list_metrics, load_metric
# Print all the available datasets
print(list_datasets())
# Load a dataset and print the first example in the training set
squad_dataset = load_dataset('squad')
print(squad_dataset['train'][0])
# List all the available metrics
print(list_metrics())
# Load a metric
squad_metric = load_metric('squad')
# Process the dataset - add a column with the length of the context texts
dataset_with_length = squad_dataset.map(lambda x: {"length": len(x["context"])})
# Process the dataset - tokenize the context texts (using a tokenizer from the 🤗 Transformers library)
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained('bert-base-cased')
tokenized_dataset = squad_dataset.map(lambda x: tokenizer(x['context']), batched=True)
For more details on using the library, check the quick tour page in the documentation: https://huggingface.co/docs/datasets/quicktour.html and the specific pages on:
- Loading a dataset https://huggingface.co/docs/datasets/loading_datasets.html
- What's in a Dataset: https://huggingface.co/docs/datasets/exploring.html
- Processing data with 🤗 Datasets: https://huggingface.co/docs/datasets/processing.html
- Writing your own dataset loading script: https://huggingface.co/docs/datasets/add_dataset.html
- etc.
Another introduction to 🤗 Datasets is the tutorial on Google Colab here:
We have a very detailed step-by-step guide to add a new dataset to the datasets already provided on the HuggingFace Datasets Hub.
You will find the step-by-step guide here to add a dataset to this repository.
You can also have your own repository for your dataset on the Hub under your or your organization's namespace and share it with the community. More information in the documentation section about dataset sharing.
If you are familiar with the great TensorFlow Datasets, here are the main differences between 🤗 Datasets and tfds
:
- the scripts in 🤗 Datasets are not provided within the library but are queried, downloaded/cached and dynamically loaded upon request
- 🤗 Datasets also provides evaluation metrics in a similar fashion to the datasets, i.e. as dynamically installed scripts with a unified API. This gives access to the pair of a benchmark dataset and a benchmark metric for instance for benchmarks like SQuAD or GLUE.
- the backend serialization of 🤗 Datasets is based on Apache Arrow instead of TF Records and leverage python dataclasses for info and features with some diverging features (we mostly don't do encoding and store the raw data as much as possible in the backend serialization cache).
- the user-facing dataset object of 🤗 Datasets is not a
tf.data.Dataset
but a built-in framework-agnostic dataset class with methods inspired by what we like intf.data
(like amap()
method). It basically wraps a memory-mapped Arrow table cache.
Similar to TensorFlow Datasets, 🤗 Datasets is a utility library that downloads and prepares public datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have license to use them. It is your responsibility to determine whether you have permission to use the dataset under the dataset's license.
If you're a dataset owner and wish to update any part of it (description, citation, etc.), or do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thanks for your contribution to the ML community!
If you want to cite this framework you can use this:
@software{quentin_lhoest_2021_5148649,
author = {Quentin Lhoest and
Albert Villanova del Moral and
Patrick von Platen and
Thomas Wolf and
Yacine Jernite and
Abhishek Thakur and
Lewis Tunstall and
Suraj Patil and
Mariama Drame and
Julien Chaumond and
Julien Plu and
Joe Davison and
Simon Brandeis and
Teven Le Scao and
Victor Sanh and
Kevin Canwen Xu and
Nicolas Patry and
Angelina McMillan-Major and
Philipp Schmid and
Sylvain Gugger and
Steven Liu and
Sylvain Lesage and
Lysandre Debut and
Théo Matussière and
Clément Delangue and
Stas Bekman},
title = {huggingface/datasets: 1.11.0},
month = jul,
year = 2021,
publisher = {Zenodo},
version = {1.11.0},
doi = {10.5281/zenodo.5148649},
url = {https://doi.org/10.5281/zenodo.5148649}
}