Fig.1

Concept

TabPFN

TabPFN (Tabular Prior-Data Fitted Network) is a transformer pre-trained once on millions of synthetic tabular datasets that then classifies new datasets with no gradient training on your data at all. Think of it like a foundation model for small tables, except instead of…

The rest of “TabPFN” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.

Log in to unlock

← Back to the library

TabPFN, explained · Fig. 1