# yandex-research/rtdl

Research on Tabular Deep Learning: Papers & Packages

Repository: https://github.com/yandex-research/rtdl
Canonical: https://ross.abutalabs.com/products/rtdl
Language: Python
License: Apache-2.0
License Family: permissive
Topics: python, machine-learning, deep-learning, pytorch, research, ai, artificial-intelligence, tabular, neural-network, papers, tabular-data
Last push: 2026-04-17T09:43:50+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 77, release rhythm 48, longevity 100
- inputs: {"age_days": 1910, "days_push": 138, "days_rel": 138, "gap_med": 520, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1158, forks 124 (observed 2026-08-28T04:03:48.520669+00:00)

## What it is
RTDL is a curated collection of research papers and accompanying Python/PyTorch packages on deep learning for tabular data from Yandex Research. It serves as a hub linking papers (TabM, TabR, TabDDPM, etc.) to their code implementations and successor packages.

## Use cases
- find state-of-the-art deep learning models for tabular data
- reproduce tabular deep learning research papers
- benchmark neural networks against gradient boosting on tabular datasets
- get PyTorch implementations of FT-Transformer and MLP models for tabular data
- learn about numerical feature embeddings in neural networks
- generate synthetic tabular data with diffusion models
- follow new research on tabular deep learning

## When to choose
- you are doing research or building models on tabular data with neural networks
- you want paper-linked reference implementations of tabular DL methods
- you need benchmarks for tabular deep learning

## When to avoid
- you just need a production gradient boosting library like XGBoost or LightGBM
- you expect a single maintained pip package - the original rtdl package is deprecated in favor of separate packages
- you work with images, text, or other non-tabular data

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning
- domain: machine-learning, data-science, artificial-intelligence
- platform: python
- tags: tabular-data, pytorch, research-papers, neural-networks, benchmarks

## Member repositories
- yandex-research/rtdl (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:48.520669+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:31:59.507515+00:00, confidence not recorded.
  - readme: https://github.com/yandex-research/rtdl (fetched 2026-08-28T04:03:48.520669+00:00, sha e3a0f48cc91b)
- Data as of 2026-08-30T08:39:29.467469+00:00.