This repository contains the implementation of the paper:

TabNAT: A Continuous-Discrete Joint Generative Framework for Tabular Data
Forty-Second International Conference on Machine Learning (ICML 2025)
Hengrui Zhang*, Liancheng Fang*, Qitian Wu, Philip S. Yu

Python version: 3.10

Create environment

conda create -n tabnat python=3.10
conda activate tabnat

Install pytorch

pip install torch torchvision torchaudio

or via conda

conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.7 -c pytorch -c nvidia

Install other dependencies

pip install -r requirements.txt

python download_and_process.py

bash run.sh