sakura.utils.data_splitter.DataSplitter.get_k_fold_cv_split

DataSplitter.get_k_fold_cv_split(base: ndarray) → dict

Obtain cross validation foldings directly from 1~k labels, 0 considered to be not selected

Parameters:

base (np.ndarray[base.dtype, np.integer]) – The predefined label vector to work with

Returns:

A dictionary with k folding keys, mapped to k dictionaries with “train” and “test” split codes

Return type:

dict[str, dict[str, np.ndarray]]