Expand description
Engine orchestration for bounded exact n-ary induction.
The n-ary counterpart of the binary crate::induce_exact pipeline:
enumerate every canonical pattern (crate::nary::enumerate_patterns),
flatten the batch into the device layout
(crate::nary_layout::flatten_patterns), score it through the
device-resident launcher
(CudaKernelProvider::ilp_exact_nary_score_device), and reduce the
per-pattern coverage counts deterministically
(crate::reduce::reduce_nary).
Transfer budget: relation and example values never visit the host — ingestion is device-to-device column copies inside the launcher; row counts come from cached host-side metadata (a pure struct read); the only device-to-host transfers are the two per-pattern count arrays. The reduction then runs on those counts entirely host-side.
Structs§
- Induce
Exact Nary Request - Inputs to one
induce_exact_narycall. - Nary
Induction Config - Bounds and selection size for one
induce_exact_narycall. - Nary
Induction Result - Result of one
induce_exact_narycall. - Scored
Nary Candidate - One kept pattern with full metadata and diagnostics.
Functions§
- induce_
exact_ nary - Run exact n-ary induction against one request.