🔗 The TOC-UCO repository can be downloaded from this website. The following bash command can also be executed:
wget https://www.uco.es/grupos/ayrna/datasets/TOC-UCO.zip📕 In the tutorial notebook you can find the code on how to load the train-test partitions of each TOC-UCO dataset.
📘 In the baseline experiments script, the code for executing different methodologies in the TOC-UCO repository is provided.
💻 More about ordinal classification can be found in the dlordinal package, where several ordinal techniques are implemented, together with the most popular ordinal performance metrics.
📚 If you enjoyed this repository, we would appretiate a citation for the following work: If you enjoyed this framework, we would appretiate a citation for the following work:
@misc{ayllóngavilán2025tocucocomprehensiverepositorytabular,
title={TOC-UCO: a comprehensive repository of tabular ordinal classification datasets},
author={Rafael Ayllón-Gavilán and David Guijo-Rubio and Antonio Manuel Gómez-Orellana and David Guijo-Rubio and Francisco Bérchez-Moreno and Víctor Manuel Vargas-Yun and Pedro A. Gutiérrez},
year={2025},
eprint={2507.17348},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2507.17348},
}