---
title: "gparse-test-inference-softmax-comparison — Mova code model"
description: "Mova code model of gparse/tests/inference_softmax.rs: components with their labels, obernants and the model text."
url: "https://groupants.elamur.ai/free/en/mova/gparse/tests/inference_softmax.rs.groupant/comparison.groupant/"
---
Elamur: [Mova](https://md.mova.elamur.ai/free/en/) · [Dictionary](https://md.dictionary.elamur.ai/free/en/) · [Research](https://md.research.elamur.ai/free/en/) · **Groupants**

This page as HTML: <https://groupants.elamur.ai/free/en/mova/gparse/tests/inference_softmax.rs.groupant/comparison.groupant/>

# gparse-test-inference-softmax-comparison

The groupant beside gparse/tests/inference_softmax.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

- **kind:** groupant_fragment
- **status:** partial source bound model
- **source:** [`gparse/tests/inference_softmax.rs`](https://github.com/ElamurAI/mova-free/blob/main/gparse/tests/inference_softmax.rs)
- **source SHA-256:** `45f0dbb793a0b01b…`
- **covers:** compare checks outer sentence count, then each paired token count, counts unequal sentence vectors and unequal tuples, returns (sentences,tokens). (bytes 269–684)
- **part of:** [gparse-test-inference-softmax](https://md.groupants.elamur.ai/free/en/mova/gparse/tests/inference_softmax.rs.groupant/)
- **repository folder:** [`gparse/tests/inference_softmax.rs.groupant/comparison.groupant`](https://github.com/ElamurAI/mova-free/tree/main/gparse/tests/inference_softmax.rs.groupant/comparison.groupant)
- **repository commit:** `48f2e6f74146`

> **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](#approximant) |
| **ast** | partial | `../syntax.json` |
| **cost** | symbolic source derived | [invariant](#invariant) |
| **descriptant** | present | [descriptant](#descriptant) |
| **invariant** | partial | [invariant](#invariant) |
| **tests** | partial bounded checks | [tests](#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](#behavior-complement) |
| `reverse_reconstruction` | reverse_reconstruction | partial source derived model | [reverse_reconstruction](#reverse-reconstruction) |
| `shape_precedence` | shape_precedence | partial source derived model | [shape_precedence](#shape-precedence) |
| `tuple_metric` | tuple_metric | partial source derived model | [tuple_metric](#tuple-metric) |
| `sentence_metric` | sentence_metric | partial source derived model | [sentence_metric](#sentence-metric) |
| `effect_boundary` | effect_boundary | partial source derived model | [effect_boundary](#effect-boundary) |

## Model text


### descriptant

#### Source-bound description

compare checks outer sentence count, then each paired token count, counts unequal sentence vectors and unequal tuples, returns (sentences,tokens).

##### 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

compare checks outer sentence count, then each paired token count, counts unequal sentence vectors and unequal tuples, returns (sentences,tokens).

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

Aligned identical tuple sequences return (0,0); one tag change returns (1,1).

##### Cost

O(paired tuples plus compared string bytes), O(1) auxiliary counters.

Allocation, finite arithmetic, platform dimensions and external failures are not universally certified.

### models

## Typed model catalog

### behavior_complement

Shape mismatches cannot silently count as zero: outer or inner assert panics.

### reverse_reconstruction

Counts admit every placement and kind of tuple mismatch consistent with the two totals.

### shape_precedence

Outer length assertion occurs before loop; inner length assertion occurs before counting that sentence.

### tuple_metric

Head, relation, XPOS or UPOS difference each contributes one token, not one per field.

### sentence_metric

Multiple changed tokens in one sentence add one to sentence counter.

### effect_boundary

Pure borrowed comparisons; no CUDA or environment changes.

### tests

#### Checks and limits

Selected parent checks relevant to this fragment (not exhaustive): compare_counts_tuple_changes_and_shape_errors

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.