spacy-ngram

spacy-ngram is a flexible SpaCy pipeline component for adding document- or sentence-level ngrams to your NLP pipeline. It extracts ngrams from lemmas (default) or tokens, handling stop words, punctuation, and digits automatically.

Table of Contents

• About the Project
• Getting Started
  • Prerequisites
  • Installation
• Usage
  • Quick Start
  • Custom Configuration
• Roadmap
• Contributing
• License
• Contact

About the Project

This component provides an easy way to enrich your Doc or Span objects with n-character or n-word sequences ( ngrams). It's designed to be lightweight and highly configurable.

Getting Started

Prerequisites

• Python 3.10+
• SpaCy 3.5.0+

Installation

1. Install from PyPI:

pip install spacy-ngram

2. Download a SpaCy model:

python -m spacy download en_core_web_sm

Usage

Quick Start

By default, the component adds unigrams and bigrams at the document level, filtering out stop words, punctuation, and digits.

import spacy
# the component is registered automatically on import
import spacy_ngram

nlp = spacy.load('en_core_web_sm')
nlp.add_pipe('spacy-ngram')

text = 'Quark soup is an interacting localized assembly of quarks and gluons.'
doc = nlp(text)

print(doc._.ngram_1)
# ['quark', 'soup', 'interact', 'localize', 'assembly', 'quark', 'gluon']

print(doc._.ngram_2)
# ['quark_soup', 'soup_interact', 'interact_localize', 'localize_assembly', 'assembly_quark', 'quark_gluon']

Custom Configuration

You can customize the extension name, ngram sizes, and whether to include Boundary-of-Sentence (<BOS>) / End-of-Sentence (<EOS>) tags.

nlp.add_pipe('spacy-ngram', config={
    'extension_name': 'my_ngrams',
    'ngrams': (2, 3),  # Extract bi- and trigrams
    'sentence_level': True,  # Process each sentence individually
    'doc_level': True,  # Also process the entire document
    'include_bos': True,  # Include <BOS> tags
    'include_eos': True,  # Include <EOS> tags
})

doc = nlp("This is a test. This is only a test.")

# access document-level trigrams
print(doc._.my_ngrams_3)

# access sentence-level bigrams
for sent in doc.sents:
    print(sent._.my_ngrams_2)

Pipeline Parameters

Versions

Uses SEMVER. See Releases.

Roadmap

See the open issues for a list of proposed features (and known issues).

Contributing

Contributions are welcome!

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Please use the issue tracker.