This python library helps you with augmenting nlp for your machine learning projects. Visit this introduction to understand about Data Augmentation in NLP. Augmenter is the basic element of augmentation while Flow is a pipeline to orchestra multi augmenter together.

• Data Augmentation library for Text
• Does your NLP model able to prevent adversarial attack?
• How does Data Noising Help to Improve your NLP Model?
• Data Augmentation library for Speech Recognition
• Data Augmentation library for Audio
• Unsupervied Data Augmentation

• Example of Augmentation for Textual Inputs
• Example of Augmentation for Spectrogram Inputs
• Example of Augmentation for Audio Inputs
• Example of Orchestra Multiple Augmenters
• How to train TF-IDF model
• How to create custom augmentation
• API Documentation

The library supports python 3.5+ in linux and window platform.

To install the library:

pip install nlpaug numpy matplotlib python-dotenv

or install the latest version (include BETA features) from github directly

pip install git+https://github.com/makcedward/nlpaug.git numpy matplotlib python-dotenv

If you use ContextualWordEmbsAug or ContextualWordEmbsForSentenceAug, install the following dependencies as well

pip install torch>=1.2.0 transformers>=2.0.0

If you use AntonymAug, SynonymAug, install the following dependencies as well

pip install nltk

If you use WordEmbsAug (word2vec, glove or fasttext), downloading pre-trained model first

from nlpaug.util.file.download import DownloadUtil
DownloadUtil.download_word2vec(dest_dir='.') # Download word2vec model
DownloadUtil.download_glove(model_name='glove.6B', dest_dir='.') # Download GloVe model
DownloadUtil.download_fasttext(model_name='wiki-news-300d-1M', dest_dir='.') # Download fasttext model

If you use any one of audio augmenter, install the following dependencies as well

pip install librosa

0.0.9 Sep 30, 2019

• Added Swap Mode (adjacent, middle and random) for RandomAug (character level)
• Added SynonymAug (WordNet/ PPDB) and AntonymAug (WordNet)
• WordNetAug is deprecated. Uses SynonymAug instead
• Introduce parameter n. Returning more than 1 augmented data. Changing output format from text (or numpy) to list of text (or numpy) if n > 1
• Introduce parameter temperature in ContextualWordEmbsAug and ContextualWordEmbsForSentenceAug to control the randomness
• aug_n parameter is deprecated. This parameter will be replaced by top_k parameter
• Fixed tokenization issue #48
• Upgraded transformers dependency (or pytorch_transformer) to 2.0.0
• Upgraded PyTorch dependency to 1.2.0
• Added SplitAug

See changelog for more details.

This library uses data (e.g. capturing from internet), research (e.g. following augmenter idea), model (e.g. using pre-trained model) See data source for more details.