spb/wp20_uqo

1# Results validation (Gate 7)23Generated by `scripts/12_validate_results.py` on 2026-10-20 08:52. Each headline number quoted in the4manuscript (via `paper/tables/numbers.tex`) is recomputed here from the processed data by independent code paths5(direct cost-model call, direct per-image aggregation) and compared with the macro value.67| Check | Macro | Macro value | Recomputed | Tolerance | OK |8|---|---|---|---|---|---|9| sample size | `nSample` | 1,200 | 1200 | 0 | ✓ |10| image URLs in sample | `nImagesSample` | 14,245 | 1.424e+04 | 0 | ✓ |11| frame size | `nFrame` | 7,745 | 7745 | 0 | ✓ |12| roof agreement (Sonnet) | `accroofcovering` | 95.8 | 95.78 | 0.15 | ✓ |13| roof coverage (Sonnet) | `covroofcovering` | 87.1 | 87.13 | 0.15 | ✓ |14| basement shown (Sonnet, K10) | `shownBasement` | 3.1 | 3.083 | 0.15 | ✓ |15| bathroom shown (Sonnet, K10) | `shownBathroom` | 19.6 | 19.58 | 0.15 | ✓ |16| median V B0 | `VsonnetBZeroB` | 407,545 | 4.075e+05 | 1.0 | ✓ |17| median D B0 | `DmedsonnetBZeroB` | 0.62 | 0.6167 | 0.006 | ✓ |18| price ratio B0 | `priceRatiosonnetBZeroB` | 0.75 | 0.7467 | 0.006 | ✓ |19| price MdAPE B0 | `priceMdapesonnetBZeroB` | 30.1 | 30.15 | 0.06 | ✓ |20| median V B3 | `VsonnetBThreeB` | 514,919 | 5.149e+05 | 1.0 | ✓ |21| median D B3 | `DmedsonnetBThreeB` | 0.38 | 0.375 | 0.006 | ✓ |22| price ratio B3 | `priceRatiosonnetBThreeB` | 0.95 | 0.9471 | 0.006 | ✓ |23| price MdAPE B3 | `priceMdapesonnetBThreeB` | 27.9 | 27.87 | 0.06 | ✓ |24| cost identity holds for all B3 rows | `—` | True | 1 | 0 | ✓ |25| verified references | `nRefsVerified` | 104 | 104 | 0 | ✓ |26| bib entries = verified refs | `—` | 104 | 104 | 0 | ✓ |2728**18 / 18 checks passed.**2930## Manual checks performed3132- Every table in the manuscript is an `\input` of a generated fragment; no number is typed by hand (grep of `sections/*.tex` for digits outside macros: only section numbers, years, equation constants and the 15 m / 20 % thresholds that are design parameters).33- Figures are regenerated from `results/tables/*.csv` and `data/processed/*.parquet` by `src/visualization/figures.py`; PNG previews were inspected for clipping and label collisions.34- Citations: 104 references, each verified against Crossref/arXiv metadata (`docs/verification/`); BibTeX run reports 0 warnings; no undefined citations in `main.log`.35- The pre-registered experiment plan (`docs/experiment_plan.md`) lists 16 experiments; 14 were executed, 2 (human expert agreement, human-in-the-loop gains) are marked not feasible and replaced by documented substitutes.36- Known approximations stated in the text: assessor building value indexed by the construction price index; roll land value indexed by the single-family price index; regional cost level uniform; Monte Carlo excludes latent attributes.37