Elamur Groupantsgparse-test-predict-paths-bounded-check
The groupant beside gparse/tests/predict_paths.rs in the Mova repository: what the code computes, what it never does, how to go back from its outputs, its invariants, cost and checks.
Facts
Implementation material. A Mova code model describes this code at one source revision: partial models, finite checks and their stated scope. It is not a reviewed general exposition, and its labels are shown as written, never raised.
Components
| component | label | where |
|---|---|---|
| approximant | partial | approximant |
| ast | partial | ../syntax.json |
| cost | symbolic source derived | invariant |
| descriptant | present | descriptant |
| invariant | partial | invariant |
| tests | unexecuted source derived scope | tests |
Obernants
The typed relations around the forward model: the behaviour complement (what the code never does), the reverse reconstruction (every input compatible with an output) and specialised views.
| id | type | label | where |
|---|---|---|---|
behavior_complement | behavior_complement | partial source derived model | behavior_complement |
reverse_reconstruction | reverse_reconstruction | partial source derived model | reverse_reconstruction |
input_boundary | input_boundary | partial source derived model | input_boundary |
repeat_protocol | repeat_protocol | partial source derived model | repeat_protocol |
timing_observation | timing_observation | partial source derived model | timing_observation |
unexecuted_effect | unexecuted_effect | partial source derived model | unexecuted_effect |
Model text
descriptant
Source-bound description
Original ignored unscored_matches_scored_and_repeated_batches loads GP_TEST_MODEL and GP_TEST_CORPUS, filters fitting sentences, compares scored/plain outputs through six alternating rounds and checks fallback/empty cases.
Epoch 2026-10-31
Prospective model was recorded before edits. Added tests use bounded synthetic fixtures; production algorithms unchanged. Coverage is partial, not proof. No trained corpus, live teacher or throughput experiment was executed.
approximant
Forward approximant
Original ignored unscored_matches_scored_and_repeated_batches loads GP_TEST_MODEL and GP_TEST_CORPUS, filters fitting sentences, compares scored/plain outputs through six alternating rounds and checks fallback/empty cases.
Domain: valid source-specific dimensions and resources stated in invariant; no automatic production substitution. Numerical observations are finite checks, not an end-to-end relative bound. Executable Rust surrogate: unsupported gap; existing exact tests are not surrogates. Future metric-specific caps may be 1/2/5/10%, never above 10%.
invariant
Invariant and cost
Requires CUDA and nonempty fitting corpus; every repeated output equals the initial scored reference.
Cost
Model I/O plus at least 14 predictions of bounded selected corpus; unexecuted in this epoch.
Allocation, finite arithmetic, platform dimensions and external failures are not universally certified.
models
Typed model catalog
behavior_complement
A changed tuple fails equality; no statement about latency gain follows merely from timing output.
reverse_reconstruction
A passing trace admits any model/corpus producing the same categorical repeated trees; six durations do not uniquely determine timing medians.
input_boundary
GP_TEST_SENTENCES defaults1024; filter requires 0<len<max_len and then take(limit).
repeat_protocol
Six rounds alternate ordering; equality is exact tuple comparison per run.
timing_observation
Median averages sorted duration positions2,3; no explicit device synchronize appears in this probe.
unexecuted_effect
External model/corpus loading and benchmark remain excluded; new CPU test below this scope supplies separate bounded evidence.
tests
Checks and limits
This exact fragment remains source-derived and unexecuted in this epoch. Parent/sibling selected evidence, not runtime coverage of this fragment: synthetic_cpu_prediction_preserves_slots_and_fallback
Selected commands run twice debug and twice release; exact receipts and source-to-test mapping retained in wave30 gparse external run directory. Tests do not certify allocation failure, all NaN/overflow inputs, trained-model accuracy or corpus timing. Compiled mutants cover selected boundaries; per-source coverage is recorded in the external map, not presumed.
This page in Markdown → https://md.groupants.elamur.ai/free/en/mova/gparse/tests/predict_paths.rs.groupant/bounded-check.groupant/