Skip to main content

save_fitted_tabpfn_model

View source
Persist a fitted TabPFN estimator to path. This stores the initialization parameters and the fitted state, but crucially omits the large foundation model weights for efficiency.
Parameters
Returns

load_fitted_tabpfn_model

View source
Load a fitted TabPFN estimator saved with save_fitted_tabpfn_model.
Parameters
Returns

An in-memory model and the metadata needed by either TabPFN estimator. Pass this object (or a list of objects) as model_path. The estimator selects the task, so the same specs can back both tasks of a multitask model. The model is used by reference, not copied; create an isolated evaluation model before wrapping live training weights.
Parameters