Contents

• Overview
• Annotating text
• API Reference
• Rule-based IE with Odin
• API Reference
• A walkthrough example
• Running the NLP server
• OpenIE for concept recognition
• Jupyter notebook visualizations
• Other ways of initializing the server
• Keeping the server running
• FAQ
• History

What is it?

py-processors is a Python wrapper for the CLU Lab’s processors NLP library. py-processors relies on processors-server.

Though compatible* with Python 2.x, this library was developed with 3.x in mind.

Requirements

The server component can be run either via docker or directly with java.

Option 1

• docker and the myedibleenso/processors-server container

Option 2

• Java 8
• processor-server (v3.1.0)
  • this dependency will be retrieved automatically during installation
• At least 2GB of RAM free for the server (I recommend 3GB+)

Installation

py-processors can be installed via pip. The library also has a jupyter extras module which adds widgets/visualizations to juypter notebooks.

basic installation

pip install py-processors

basic + jupyter notebook widgets

pip install py-processors[jupyter]

bleeding edge

pip install git+https://github.com/myedibleenso/py-processors.git

How to use it?

See the walkthrough example