Using SpaCy

SpaCy is a Python library for Natural Language Processing (NLP) such as tokenization, named entity recognition with pre-trained models for several languages.

Documentation for SpaCy is available at

Installing SpaCy

In a code environment, you need to install the spacy package.

To add a specific pre-trained model, you can add the URL of the pip package for that model, as specified in the Installation via pip page of the SpaCy documentation.

For example for the English model, your code env’s Requested Packages could be:


See SpaCy’s Models page for a list of languages.

Using SpaCy models

In a python notebook or recipe (using the aforementioned code environment), you can then import spacy and use spacy.load with the model package name:

import spacy
nlp = spacy.load("en_core_web_sm")
doc = nlp(u"This is an example sentence.")