torchtext

This repository consists of:

• torchtext.datasets: The raw text iterators for common NLP datasets
• torchtext.data: Some basic NLP building blocks (tokenizers, metrics, functionals etc.)
• torchtext.nn: NLP related modules
• examples: Example NLP workflows with PyTorch and torchtext library.

Note: the legacy code discussed in torchtext v0.7.0 release note has been retired to torchtext.legacy folder. Those legacy code will not be maintained by the development team, and we plan to fully remove them in the future release. See torchtext.legacy folder for more details.

Installation

We recommend Anaconda as Python package management system. Please refer to pytorch.org for the detail of PyTorch installation. The following is the corresponding torchtext versions and supported Python versions.

Version Compatibility

PyTorch version	torchtext version	Supported Python version
nightly build	master	3.6+
1.8	0.9	3.6+
1.7	0.8	3.6+
1.6	0.7	3.6+
1.5	0.6	3.5+
1.4	0.5	2.7, 3.5+
0.4 and below	0.2.3	2.7, 3.5+

Using conda:

conda install -c pytorch torchtext

Using pip:

pip install torchtext

Optional requirements

If you want to use English tokenizer from SpaCy, you need to install SpaCy and download its English model:

pip install spacy
python -m spacy download en_core_web_sm

Alternatively, you might want to use the Moses tokenizer port in SacreMoses (split from NLTK). You have to install SacreMoses:

pip install sacremoses

For torchtext 0.5 and below, sentencepiece:

conda install -c powerai sentencepiece

Building from source

To build torchtext from source, you need git, CMake and C++11 compiler such as g++.:

git clone https://github.com/pytorch/text torchtext
cd torchtext
git submodule update --init --recursive

# Linux
python setup.py clean install

# OSX
MACOSX_DEPLOYMENT_TARGET=10.9 CC=clang CXX=clang++ python setup.py clean install

# or ``python setup.py develop`` if you are making modifications.

Note

When building from source, make sure that you have the same C++ compiler as the one used to build PyTorch. A simple way is to build PyTorch from source and use the same environment to build torchtext. If you are using the nightly build of PyTorch, checkout the environment it was built with conda (here) and pip (here).

Documentation

Find the documentation here.

Datasets

The datasets module currently contains:

• Language modeling: WikiText2, WikiText103, PennTreebank, EnWik9
• Machine translation: IWSLT2016, IWSLT2017
• Sequence tagging (e.g. POS/NER): UDPOS, CoNLL2000Chunking
• Question answering: SQuAD1, SQuAD2
• Text classification: AG_NEWS, SogouNews, DBpedia, YelpReviewPolarity, YelpReviewFull, YahooAnswers, AmazonReviewPolarity, AmazonReviewFull, IMDB

For example, to access the raw text from the AG_NEWS dataset:

>>> from torchtext.datasets import AG_NEWS
>>> train_iter = AG_NEWS(split='train')
>>> next(train_iter)
>>> # Or iterate with for loop
>>> for (label, line) in train_iter:
>>>     print(label, line)
>>> # Or send to DataLoader
>>> from torch.utils.data import DataLoader
>>> train_iter = AG_NEWS(split='train')
>>> dataloader = DataLoader(train_iter, batch_size=8, shuffle=False)

A tutorial for the end-to-end text classification workflow can be found in PyTorch tutorial

[Prototype] Experimental Code

We have re-written several building blocks under torchtext.experimental:

• Transforms: some basic data processing building blocks
• Vocabulary: a vocabulary to numericalize tokens
• Vectors: the vectors to convert tokens into tensors.

These prototype building blocks in the experimental folder are available in the nightly release only. The nightly packages are accessible via Pip and Conda for Windows, Mac, and Linux. For example, Linux users can install the nightly wheels with the following command:

pip install --pre --upgrade torch torchtext -f https://download.pytorch.org/whl/nightly/cpu/torch_nightly.html

For more detailed instructions, please refer to Install PyTorch. It should be noted that the new building blocks are still under development, and the APIs have not been solidified.

[BC Breaking] Legacy

In v0.9.0 release, we move the following legacy code to torchtext.legacy. This is part of the work to revamp the torchtext library and the motivation has been discussed in Issue #664:

• torchtext.legacy.data.field
• torchtext.legacy.data.batch
• torchtext.legacy.data.example
• torchtext.legacy.data.iterator
• torchtext.legacy.data.pipeline
• torchtext.legacy.datasets

We have a migration tutorial to help users switch to the torchtext datasets in v0.9.0 release. For the users who still want the legacy components, they can add legacy to the import path.

Disclaimer on Datasets

This 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 the dataset. 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!