Struct MemberConfig 

pub struct MemberConfig {
    pub feature_permutation: Vec<usize>,
    pub class_shift: usize,
    pub cat_permutation: Option<Vec<Vec<usize>>>,
    pub row_subsample: Option<Vec<usize>>,
    pub norm_method: NormMethod,
}

Fields

feature_permutation: Vec<usize>

Permutation of 0..n_features (or a subsampled+permuted subset if n_features > max_num_features).

class_shift: usize

Amount added (mod n_classes) to every training label fed to the model for this member. Always 0 for regression or when class-shift is disabled/inapplicable.

cat_permutation: Option<Vec<Vec<usize>>>

cat_col_index -> (original_code -> permuted_code), present only when permute_categorical=true (default: off, so this is None for every member).

row_subsample: Option<Vec<usize>>

Row indices to bag from 0..n_train, present only when row-subsampling is active (default: off, so this is None for every member — use all rows).

norm_method: NormMethod