Skip to content

pyndl implements a Naive discriminative learning which is a learning and classification models based on the Rescorla-Wagner equations in python3.

License

Notifications You must be signed in to change notification settings

quantling/pyndl

Repository files navigation

Pyndl - Naive Discriminative Learning in Python

https://github.com/quantling/pyndl/actions/workflows/python-test.yml/badge.svg?branch=main https://codecov.io/gh/quantling/pyndl/branch/main/graph/badge.svg?token=2GWUXRA9PD

pyndl is an implementation of Naive Discriminative Learning in Python. It was created to analyse huge amounts of text file corpora. Especially, it allows to efficiently apply the Rescorla-Wagner learning rule to these corpora.

Installation

The easiest way to install pyndl is using pip:

pip install --user pyndl

For more information have a look at the Installation Guide.

Documentation

pyndl uses sphinx to create a documentation manual. The documentation is hosted on Read the Docs.

Getting involved

The pyndl project welcomes help in the following ways:

For more information on how to contribute to pyndl have a look at the development section.

Authors and Contributers

pyndl was mainly developed by Konstantin Sering, Marc Weitz, David-Elias Künstle, Elnaz Shafaei Bajestan and Lennart Schneider. For the full list of contributers have a look at Github's Contributor summary.

Currently, it is maintained by Konstantin Sering and Marc Weitz.

Funding

pyndl was partially funded by the Humboldt grant, the ERC advanced grant (no. 742545) and by the University of Tübingen.

Acknowledgements

This package is build as a python replacement for the R ndl2 package. Some ideas on how to build the API and how to efficiently run the Rescorla Wagner iterative learning on large text corpora are inspired by the way the ndl2 package solves this problems. The ndl2 package is available on Github here.

About

pyndl implements a Naive discriminative learning which is a learning and classification models based on the Rescorla-Wagner equations in python3.

Resources

License

Stars

Watchers

Forks

Packages

No packages published