Forte is a toolkit for building Natural Language Processing pipelines. This repository tries to wrap the fantastic collections of NLP libraries built by the community.

This project is part of the CASL Open Source family.

Get Started

• First, install the library along with the desired tools. Let's take AllenNLP as an example:

To install from PyPI:

pip install forte.allennlp

To install the latest from source

git clone https://github.com/asyml/forte-wrappers.git
cd forte-wrappers
pip install src/allennlp

Libraries and Tools Supported

• NLTK (Processors)
  • From PyPI: pip install forte.nltk
  • From source: pip install src/nltk
  • Features:
    • POS Tagger
    • Sentence Segmenter
    • Tokenizer
    • Lemmatizer
    • NER
• spaCy (Processors)
  • From PyPI: pip install forte.spacy
  • From source: pip install src/spacy
  • Features:
    • Tokenizer, Lemmatizer and POS Tagging
    • NER
• AllenNLP (Processors)
  • From PyPI: pip install forte.allennlp
  • From source: pip install src/allennlp
  • Features:
    • Tokenizer, POS Tagging
    • Semantic Role Labeling
    • Dependency Parsing
• Stanza (Processors)
  • From PyPI: pip install forte.stanza
  • From source: pip install src/stanza
  • Features:
    • Tokenization, POS Tagging, Lemmatizer
    • Dependency Parsing
• HuggingFace Models
  • From PyPI: pip install forte.huggingface
  • From source: pip install src/huggingface
  • Features:
    • BioBERT NER (Processors)
    • Zero Shot Classifier(Processors)
    • Question Answering(Processors)
• Vader Sentiment (Processors)
  • From PyPI: pip install forte.vader
  • From source: pip install src/vader
  • Features:
    • Sentiment Analysis
• Elastic Search (Processors)
  • From PyPI: pip install forte.elastic
  • From source: pip install src/elastic
  • Features:
    • Elastic Indexer
    • Elastic Search
• Faiss (Processors)
  • From PyPI: pip install forte.faiss
  • From source: pip install src/faiss
  • Features:
    • Faiss Indexer
• GPT2 (Processors)
  • From PyPI: pip install forte.gpt2
  • From source: pip install src/gpt2
  • Features:
    • GPT2 Text Generation
• Tweepy (Processors)
  • From PyPI: pip install forte.tweepy
  • From source: pip install src/tweepy
  • Features:
    • TwitterAPI Search

Contributing

If you are interested in making enhancement to this repository, Forte or other ASYML/CASL projects, please first go over our Code of Conduct and Contribution Guideline

License

This project is licensed by Apache License 2.0. The project provides wrappers to other open-sourced projects. To use them in your project, please check the license of the corresponding project.

Companies and Universities Supporting Forte