tabensemb.model.AutoGluon.predict

method

AutoGluon.predict(df: DataFrame, model_name: str, model: Any | None = None, derived_data: Dict | None = None, ignore_absence: bool = False, proba: bool = False, **kwargs) → ndarray

Make inferences on a new dataset using the selected model.

Parameters:

df:

A new tabular dataset.

model_name:

A selected name of a model, which is already trained.

model:

The model returned by _new_model(). If None, the model will be loaded from self.model.

derived_data:

Unstacked data derived from tabensemb.data.datamodule.DataModule.derive_unstacked(). If None, unstacked data will be re-derived.

ignore_absence:

Whether to ignore absent keys in derived_data. Use True only when the model does not use derived_data.

proba:

Return probabilities instead of predicted classes for classification models.

**kwargs:

Arguments of _predict().

Returns:

np.ndarray

Predicted target. Always 2d np.ndarray.