Research on Tabular Deep Learning: Papers & Packages

A comprehensive toolkit and benchmark for tabular data learning, featuring 30+ deep methods, more than 10 classical methods, and 300 diverse tabular datasets.

[ICML 2023] The official implementation of the paper "TabDDPM: Modelling Tabular Data with Diffusion Models"

Mambular is a Python package that simplifies tabular deep learning by providing a suite of models for regression, classification, and distributional regression tasks. It includes models such as Mambular, TabM, FT-Transformer, TabulaRNN, TabTransformer, and tabular ResNets.

ML models + benchmark for tabular data classification and regression

TAT-QA (Tabular And Textual dataset for Question Answering) contains 16,552 questions associated with 2,757 hybrid contexts from real-world financial reports.

Connect to PBI/AAS Models via Python

Official PyTorch implementation of STUNT: Few-shot Tabular Learning with Self-generated Tasks from Unlabeled Tables (ICLR 2023 Spotlight).

augini: AI-Powered Tabular Data Assistant

Omnipy is a high level Python library for type-driven data wrangling and scalable workflow orchestration (under development)

Code accompanying AWS blog post "Build a Semantic Search Engine for Tabular Columns with Transformers and Amazon OpenSearch Service"

Implementation of the deep learning model with inference pipeline described in the paper "ConTextTab: A Semantics-Aware Tabular In-Context Learner".

Multi-modal data augmentation for machine learning

On Finetuning Tabular Foundation Models Paper Code

Implementation of the deep learning models with training and evaluation pipelines described in the paper "PORTAL: Scalable Tabular Foundation Models via Content-Specific Tokenization"

Preparatory scripts for BIDS tabular phenotypic data in large neuroimaging datasets.

Official Implementation of TMLR's paper: "TabCBM: Concept-based Interpretable Neural Networks for Tabular Data"

PyTorch Lightning implementation of PTaRL: Prototype-based Tabular Representation Learning via Space Calibration

[ICML 2024] BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Modern Hopfield Model

pretab is a flexible and extensible preprocessing library for tabular data, built on top of scikit-learn. It provides advanced transformations, spline and neural feature expansions, and seamless integration with embeddings – all designed for modern tabular ML workflows.