tabensemb.data.datamodule.DataModule._data_process

method

DataModule._data_process(warm_start: bool = False, skip_selector: bool = False, verbose: bool = True)

The main procedure to process data after splitting and imputation. Both scaled and unscaled data will be recorded. Note that processors will fit on training and validation datasets and transform the testing set by calling _data_preprocess() with different arguments.

Parameters:

warm_start

Whether to use fitted data processors to process the data.

skip_selector

True to skip feature selections.

verbose

Verbosity.