Documentation Index
Fetch the complete documentation index at: /llms.txt
Use this file to discover all available pages before exploring further.
Persist a fitted TabPFN estimator to path.
save_fitted_tabpfn_modelpath.
This stores the initialization parameters and the fitted state, but crucially
omits the large foundation model weights for efficiency.
save_fitted_tabpfn_model(
estimator: BaseEstimator,
path: Path | str,
) -> None
| Parameter | Type | Default | Description |
|---|---|---|---|
estimator | Base | Required | — |
path | Path | str | Required | — |
| Type | Description |
|---|---|
None | — |
load_fitted_tabpfn_modelsave_fitted_tabpfn_model.
load_fitted_tabpfn_model(
path: Path | str,
*,
device: DevicesSpecification = "auto",
) -> BaseEstimator
Type aliases
DevicesSpecification = torch.device | str | Sequence[torch.device | str] | Literal["auto"]
| Parameter | Type | Default | Description |
|---|---|---|---|
path | Path | str | Required | The .tabpfn_fit archive to load. |
device | Devices | "auto" | The device(s) to load onto. The archive does not record where the model was fitted, so the default resolves by availability like the constructors’ does; pass a device to pin it. |
| Type | Description |
|---|---|
Base | — |
ClassifierModelSpecsClassifierModelSpecs(
model: Architecture,
architecture_config: ArchitectureConfig,
inference_config: InferenceConfig,
)
| Name | Type | Default | Description |
|---|---|---|---|
model | — | model | — |
architecture_ | — | architecture_config | — |
inference_ | — | inference_config | — |
norm_ | — | None | — |
RegressorModelSpecsRegressorModelSpecs(
model: Architecture,
architecture_config: ArchitectureConfig,
inference_config: InferenceConfig,
norm_criterion: FullSupportBarDistribution,
)
| Name | Type | Default | Description |
|---|---|---|---|
model | — | model | — |
architecture_ | — | architecture_config | — |
inference_ | — | inference_config | — |
norm_ | — | norm_criterion | — |
Was this page helpful?