2.0.0 16-11-2023

• now supports gensim >= 4

• now support fasttext models

• enhanced code to allow user to select top n words (synonyms/most similar words)

• added punctuation insertion

1.3.4 05-11-2020

• Fixed minor issues

1.3.3 21-10-2020

• Added support for Fasttext augmentation
• Added example notebook for Fasttext augmentation

1.3.2 10-06-2020

• minor updates

1.3.1 29-05-2020

• fix minor issues

1.3 29-05-2020

• added mixup augmentation algorithm for NLP

1.2 23-05-2020

• Added support for EDA algorithm
• Added examples using Jupyter notebook

1.1, 16-07-2019

Updated ReadMe and icons.

• Added licence icon.
• Release icon.
• Wheel icon.
• Python version icon.

Added pre-print paper citation.

Initial release, 16-07-2019

TextAugment is a Python 3 library for augmenting text for natural language processing applications. TextAugment stands on the giant shoulders of NLTK, Gensim, and TextBlob and plays nicely with them.

Requirements

• Python 3
  The following software packages are dependencies and will be installed automatically.

$ pip install numpy nltk gensim textblob googletrans 

The following code downloads wordnet, tokenizer, and part-of-speech tagger model.

nltk.download('wordnet')
nltk.download('punkt')
nltk.download('averaged_perceptron_tagger')

Install from pip [Recommended]

$ pip install textaugment

How to use

>>> from textaugment import Word2vec
>>> t = Word2vec(model='path/to/gensim/model'or 'gensim model itself')
>>> t.augment('The stories are good')
The films are good

Citation

@article{marivate2019improving,
  title={Improving short text classification through global augmentation methods},
  author={Marivate, Vukosi and Sefara, Tshephisho},
  journal={arXiv preprint arXiv:1907.03752},
  year={2019}
}

Built with ❤ on Python