Contents

tabben: A Benchmark for Tabular Data

tabben: A Benchmark for Tabular Data

This is the documentation for the tabben benchmark for tabular data, its datasets, and python API (tabben python package).

Getting Started

To install the package, you can use your favorite python package manager (we recommend Python 3.8). Using pip for example,

pip install tabben

Then you’re off to the races! Check out how to use the package, walk-throughs with different machine learning frameworks, or details about the datasets.

Interoperability with other packages

Depending on what you want to use tabben with, you may need to install other packages to interoperate, e.g. scikit-learn, PyTorch, autogluon.

Citing

If you find this benchmark useful or use it in your research, please consider citing:

next

Datasets