Module config_gen 

Ports TabFMClassifier/TabFMRegressor’s _generate_ensemble() — builds the n_estimators member configs (feature permutation, classification class-shift offset, categorical-value permutation, row-subsample pattern, normalization method). RNG consumption order matches the source exactly (see the plan doc / module comments below) so results are bit-identical to the real wrapper for the same random_state.

Structs

EnsembleConfigParams

MemberConfig

Enums

NormMethod

Functions

generate_ensemble

The full _generate_ensemble() port. See module docs for the exact RNG call order this must preserve: (1) feature permutations from the independent FeatureShuffler stream; then, from the main stream: (2) class-shift base offsets, (3) categorical permutations per-member (only if enabled), (4) row-subsample patterns per-member (only if enabled), (5) one shuffle() over the zipped per-member tuples; norm-method assignment happens last, by position, consuming no RNG.