Views
No views yet
pip install tabpfn1@article{hollmann2025tabpfn,
2 title={Accurate predictions on small data with a tabular foundation model},
3 author={Hollmann, Noah and M{\"u}ller, Samuel and Purucker, Lennart and
4 Krishnakumar, Arjun and K{\"o}rfer, Max and Hoo, Shi Bin and
5 Schirrmeister, Robin Tibor and Hutter, Frank},
6 journal={Nature},
7 year={2025},
8 month={01},
9 day={09},
10 doi={10.1038/s41586-024-08328-6},
11 publisher={Springer Nature},
12 url={https://www.nature.com/articles/s41586-024-08328-6},
13}tabpfn-v2-classifier-finetuned-*.ckpt are content-identical aliases of the corresponding base checkpoints (e.g. tabpfn-v2-classifier-finetuned-gn2p4bpt-xp6f0iqb.ckpt is identical to tabpfn-v2-classifier-gn2p4bpt.ckpt; tabpfn-v2-classifier-finetuned-zk73skhh.ckpt is identical to tabpfn-v2-classifier-v2_default.ckpt). The finetuned-* filenames are retained as aliases so that pinned filenames in user code continue to resolve.