TabPFN

  1. Home
  2. Wiki
TabPFN
Developer(s)Noah Hollmann, Samuel Müller, Lennart Purucker, Arjun Krishnakumar, Max Körfer, Shi Bin Hoo, Robin Tibor Schirrmeister, Frank Hutter, Leo Grinsztajn, Klemens Flöge, Oscar Key & Sauraj Gambhir [1]
Initial releaseSeptember 16, 2023; 21 months ago (2023-09-16)[2][3]
Written inPython [3]
Operating systemLinux, macOS, Microsoft Windows[3]
TypeMachine learning
LicenseApache License 2.0
Websitegithub.com/PriorLabs/TabPFN

TabPFN (Tabular Prior-data Fitted Network) is a machine learning model that uses a transformer architecture for supervised classification and regression tasks on small to medium-sized tabular datasets, e.g., up to 10,000 samples.[1]

Overview

[edit]

First developed in 2022, TabPFN v2 was published in 2025 in Nature (journal) by Hollmann and co-authors.[1] The source code is published on GitHub under a modified Apache License and on PyPi.[4]

TabPFN v1 was introduced in a 2022 pre-print and presented at ICLR 2023.[2] Prior Labs, founded in 2024, aims to commercialize TabPFN.[5]

TabPFN supports classification, regression and generative tasks,[1] and its TabPFN-TS extension adds time series forecasting.[6]

Pre-training

[edit]

TabPFN addresses challenges in modeling tabular data[7][8] with Prior-Data Fitted Networks,[9] by using a transformer pre-trained on synthetic tabular datasets.[2][10]

It is pre-trained once on around 130 million synthetic datasets generated using Structural Causal Models or Bayesian Neural Networks, simulating real-world data characteristics like missing values or noise.[1] This enables TabPFN to process new datasets in a single forward pass, adapting to the input without retraining.[2] The model’s transformer encoder processes features and labels by alternating attention across rows and columns, capturing relationships within the data.[11] TabPFN v2, an updated version, handles numerical and categorical features, missing values, and supports tasks like regression and synthetic data generation.[1]

TabPFN's pre-training exclusively uses synthetically generated datasets, avoiding benchmark contamination and the costs of curating real-world data.[2] TabPFN v2 was pre-trained on approximately 130 million such datasets, each serving as a "meta-datapoint".[1]

The synthetic datasets are primarily drawn from a prior distribution embodying causal reasoning principles, using Structural Causal Models (SCMs) or Bayesian Neural Networks (BNNs). Random inputs are passed through these models to generate outputs, with a bias towards simpler causal structures. The process generates diverse datasets that simulate real-world imperfections like missing values, imbalanced data and noise. During pre-training, TabPFN predicts the masked target values of new data points given training data points and their known targets, effectively learning a generic learning algorithm that is executed by running a neural network forward pass.[1]

Since TabPFN is pre-trained, in contrast to other deep learning methods, it does not require costly hyperparameter optimization.[11]

Applications

[edit]

Applications for TabPFN have been investigated for domains such as Time Series Forecasting,[6] chemoproteomics,[12] insurance risk classification,[13] medical diagnostics,[14][15][16][17] metagenomics,[18] wildfire propagation modeling,[19] and others.

See also

[edit]

References

[edit]
  1. ^ a b c d e f g h Hollmann, N.; Müller, S.; Purucker, L. (2025). "Accurate predictions on small data with a tabular foundation model". Nature. 637 (8045): 319–326. Bibcode:2025Natur.637..319H. doi:10.1038/s41586-024-08328-6. PMC 11711098. PMID 39780007.
  2. ^ a b c d e Hollmann, Noah (2023). TabPFN: A transformer that solves small tabular classification problems in a second. International Conference on Learning Representations (ICLR).
  3. ^ a b c Python Package Index (PyPI) - tabpfn https://pypi.org/project/tabpfn/
  4. ^ PriorLabs/TabPFN, Prior Labs, 2025-06-22, retrieved 2025-06-23
  5. ^ Kahn, Jeremy (5 February 2025). "AI has struggled to analyze tables and spreadsheets. This German startup thinks its breakthrough is about to change that". Fortune.
  6. ^ a b "TabPFN Time Series". GitHub.
  7. ^ Shwartz-Ziv, Ravid; Armon, Amitai (2022). "Tabular data: Deep learning is not all you need". Information Fusion. 81: 84–90. arXiv:2106.03253. doi:10.1016/j.inffus.2021.11.011.
  8. ^ Grinsztajn, Léo; Oyallon, Edouard; Varoquaux, Gaël (2022). Why do tree-based models still outperform deep learning on typical tabular data?. Proceedings of the 36th International Conference on Neural Information Processing Systems (NIPS '22). pp. 507–520.
  9. ^ Müller, Samuel (2022). Transformers can do Bayesian inference. International Conference on Learning Representations (ICLR).
  10. ^ McCarter, Calvin (May 7, 2024). "What exactly has TabPFN learned to do? | ICLR Blogposts 2024". iclr-blogposts.github.io. Retrieved 2025-06-22.
  11. ^ a b McElfresh, Duncan C. (8 January 2025). "The AI tool that can interpret any spreadsheet instantly". Nature. 637 (8045): 274–275. Bibcode:2025Natur.637..274M. doi:10.1038/d41586-024-03852-x. PMID 39780000.
  12. ^ Offensperger, Fabian; Tin, Gary; Duran-Frigola, Miquel; Hahn, Elisa; Dobner, Sarah; Ende, Christopher W. am; Strohbach, Joseph W.; Rukavina, Andrea; Brennsteiner, Vincenth; Ogilvie, Kevin; Marella, Nara; Kladnik, Katharina; Ciuffa, Rodolfo; Majmudar, Jaimeen D.; Field, S. Denise; Bensimon, Ariel; Ferrari, Luca; Ferrada, Evandro; Ng, Amanda; Zhang, Zhechun; Degliesposti, Gianluca; Boeszoermenyi, Andras; Martens, Sascha; Stanton, Robert; Müller, André C.; Hannich, J. Thomas; Hepworth, David; Superti-Furga, Giulio; Kubicek, Stefan; Schenone, Monica; Winter, Georg E. (26 April 2024). "Large-scale chemoproteomics expedites ligand discovery and predicts ligand behavior in cells". Science. 384 (6694): eadk5864. Bibcode:2024Sci...384k5864O. doi:10.1126/science.adk5864. PMID 38662832.
  13. ^ Chu, Jasmin Z. K.; Than, Joel C. M.; Jo, Hudyjaya Siswoyo (2024). "Deep Learning for Cross-Selling Health Insurance Classification". 2024 International Conference on Green Energy, Computing and Sustainable Technology (GECOST). pp. 453–457. doi:10.1109/GECOST60902.2024.10475046. ISBN 979-8-3503-5790-5.
  14. ^ Alzakari, Sarah A.; Aldrees, Asma; Umer, Muhammad; Cascone, Lucia; Innab, Nisreen; Ashraf, Imran (December 2024). "Artificial intelligence-driven predictive framework for early detection of still birth". SLAS Technology. 29 (6): 100203. doi:10.1016/j.slast.2024.100203. PMID 39424101.
  15. ^ El-Melegy, Moumen; Mamdouh, Ahmed; Ali, Samia; Badawy, Mohamed; El-Ghar, Mohamed Abou; Alghamdi, Norah Saleh; El-Baz, Ayman (21 June 2024). "Prostate Cancer Diagnosis via Visual Representation of Tabular Data and Deep Transfer Learning". Bioengineering. 11 (7): 635. doi:10.3390/bioengineering11070635. PMC 11274351. PMID 39061717.
  16. ^ Karabacak, Mert; Schupper, Alexander; Carr, Matthew; Margetis, Konstantinos (August 2024). "A machine learning-based approach for individualized prediction of short-term outcomes after anterior cervical corpectomy". Asian Spine Journal. 18 (4): 541–549. doi:10.31616/asj.2024.0048. PMC 11366553. PMID 39113482.
  17. ^ Liu, Yanqing; Su, Zhenyi; Tavana, Omid; Gu, Wei (June 2024). "Understanding the complexity of p53 in a new era of tumor suppression". Cancer Cell. 42 (6): 946–967. doi:10.1016/j.ccell.2024.04.009. PMC 11190820. PMID 38729160.
  18. ^ Perciballi, Giulia; Granese, Federica; Fall, Ahmad; Zehraoui, Farida; Prifti, Edi; Zucker, Jean-Daniel (10 October 2024). Adapting TabPFN for Zero-Inflated Metagenomic Data. Table Representation Learning Workshop at NeurIPS 2024.
  19. ^ Khanmohammadi, Sadegh; Cruz, Miguel G.; Perrakis, Daniel D.B.; Alexander, Martin E.; Arashpour, Mehrdad (September 2024). "Using AutoML and generative AI to predict the type of wildfire propagation in Canadian conifer forests". Ecological Informatics. 82: 102711. doi:10.1016/j.ecoinf.2024.102711.