Module outer_subsample 

Outer-score row subsampling and the model-layer row-measure bridge.

The canonical definitions of OuterScoreSubsample, RowSet, and WeightedOuterRow are neutral low-layer primitives that live in the gam-problem crate so the CustomFamily trait layer (and the model families above it) can depend on them downward without duplication. This module is the single model-layer authority that translates those primitives into custom-family fit options.

Structs

OuterScoreSubsample

Stratified row index subsample shared across outer-loop evaluations.

WeightedOuterRow

Enums

RowSet

Row selection for an outer-loop evaluation: either the full data (All) or a Horvitz–Thompson WeightedOuterRow subsample.

Constants

ARROW_ROW_CHUNK

Deterministic row-block tiling constant for the parallel reduction paths.

Functions

arrow_row_chunk_count

Number of ARROW_ROW_CHUNK-sized tiles covering n_rows.