spacy-vectorizers

Scikit-learn compatible vectorizers built with spaCy NLP famework.

This repo contains customized scikit-learn compatible classes and vectorizers inspired by CountVectorizer, but with more accurate tokenization and lemmatization funcitonality with the help of spaCy NLP framework. Simple Keras-like punctuation removal support is also added.

Built on (prerequisites):

• Python 3.5.4
• scikit-learn 0.19.1
• spaCy 2.0.4

Usage:

Please refer to the Usage Examples & Tests Jupyter notebook or here.