dlordinal: a Python package for deep ordinal classification

TL;DR

dlordinal is a comprehensive Python library built on PyTorch that unifies state-of-the-art deep ordinal classification methods, including loss functions, output layers, and evaluation metrics, to facilitate research and application in ordinal data problems.

Contribution

It introduces a unified, open-source Python package that consolidates recent deep ordinal classification techniques and evaluation metrics, streamlining research and development in this area.

Findings

Includes various loss functions and output layers tailored for ordinal data

Provides suitable evaluation metrics based on class distance

Facilitates implementation of deep ordinal classification models

Abstract

dlordinal is a new Python library that unifies many recent deep ordinal classification methodologies available in the literature. Developed using PyTorch as underlying framework, it implements the top performing state-of-the-art deep learning techniques for ordinal classification problems. Ordinal approaches are designed to leverage the ordering information present in the target variable. Specifically, it includes loss functions, various output layers, dropout techniques, soft labelling methodologies, and other classification strategies, all of which are appropriately designed to incorporate the ordinal information. Furthermore, as the performance metrics to assess novel proposals in ordinal classification depend on the distance between target and predicted classes in the ordinal scale, suitable ordinal evaluation metrics are also included. dlordinal is distributed under the BSD-3-Clause license and is available at https://github.com/ayrna/dlordinal.