Single point where tabular data can be found for training and benchmarking Machine Learning (ML) models.
This repository consists of two parts
- Literature: Links to and summary of books, papers, blog posts etc. where tabular datasets are described.
- Data: The links to the actual datasets and util-functionality for convenient loading of theses datasets.
Tabular data is a domain where suprisingly simple ML models like decision tree based ones most of the times outperform complex deep learning models. Nevertheless, every now and then a new deep learning inspired idea comes up in research and often yields better rssults on the selected dataset in the paper but falls short performance-wise on other datasets not used in the paper. This reporsitory tries to give easy access to benchmarking datasets so upcoming, new ideas can be tested against a wealth of datasets.
WIP
WIP
WIP