For setup and examples, see Relational predictions and the TabPFN-Rel cookbook.
Installation
Use Python 3.11 or 3.12. Choose the hosted API or local inference:tabpfn-rel 0.0.6 and its RPI dependency, relarena-core 0.0.7. The context and query classes below are re-exported by tabpfn_rel; database and task specifications are available from relarena_core.userdb. The dependency is installed automatically. Both references are generated from source verified against the published wheels.
Reference topics
TabPFNRel
Select a backend, fit a context and predict a query.
Predictive context
Load a task from YAML, fit a model and work with cached features.
Prediction queries
Specify the entities and timestamp to predict.
Fitted predictor
Use the predictor returned by a context fit.
Database and task specifications
Define tables, keys and prediction tasks.
Model backends
Local and hosted model interfaces for RelArena.
Search spaces
Exported local and hosted search-space definitions.