tabula-8b-eval-suite

Evaluation suite used in our paper ["Large Scale Transfer Learning for Tabular Data via Language Modeling."](https://arxiv.org/abs/2406.12031)

This suite includes our preprocessed versions of benchmark datasets except the AutoML Multimodal Benchmark, which can be accessed by following the installation instructions in their repo [here](https://github.com/sxjscience/automl_multimodal_benchmark?tab=readme-ov-file#install-the-benchmark-suite).

We recommend using [`rtfm`](https://github.com/mlfoundations/rtfm) when evaluating models with these datasets.
See the `rtfm` repo for more information on using this data for evaluation.

本论文《基于语言建模的表格数据大规模迁移学习（Large Scale Transfer Learning for Tabular Data via Language Modeling）》（https://arxiv.org/abs/2406.12031）所采用的评测套件。

该套件包含了我们预处理后的各类基准数据集，仅AutoML多模态基准（AutoML Multimodal Benchmark）除外，该数据集可通过其官方仓库[此处](https://github.com/sxjscience/automl_multimodal_benchmark?tab=readme-ov-file#install-the-benchmark-suite)给出的安装说明获取。

我们推荐在使用此类数据集开展模型评测时采用[`rtfm`](https://github.com/mlfoundations/rtfm)工具。

有关使用该数据集进行评测的更多细节，请参阅`rtfm`的官方仓库文档。