This collection is a *Curated set of Tabular Regression problems* containing datasets from a
multitude of domains.
All datasets from this benchmark suite satisfy the following conditions to the best of our
* Regression task on dense tabular data
* Independence of observations
* No time series or otherwise grouped observations that require special care, e.g. with respect to
* At least 500 observations
* The data is clearly described
* There datasets are not problematic from an ethical viewpoint
This regression benchmark suite has been partially funded by and developed as a part of the
Mathematical Research Data Initiative ([MaRDI](https://www.mardi4nfdi.de/about/mission)).