junte/Model
1
0
Fork 0
mirror of https://github.com/Hestia-Homes/Model.git synced 2026-06-08 11:17:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

S0380.181: tighten heat-systems corpus residual tolerances to 1e-4 (all metrics)

Browse source 
The corpus residual-pin tolerances had drifted looser than the comment
above them claimed ("pin at 1e-4 relative to lodged precision"): SAP was
1e-3, cost ±£0.01, CO2 ±0.1 kg, PE ±0.1 kWh. A ±0.1 kg CO2 band could
silently mask a ~0.09 kg drift on a variant we report as EXACT.

The worksheet pins are extracted from the P960 PDF text, which prints
4 d.p., so the hard residual floor is ~5e-5 (half a unit in the last
printed digit) regardless of cascade precision. 1e-4 sits just above
that floor. All 41 variants hold at uniform 1e-4 on continuous SAP,
cost, CO2 AND PE — confirming the 37 EXACT variants are genuinely exact
to PDF print-rounding and the looser bands were masking nothing.

Aligns the guard with [[feedback-zero-error-strict]] /
[[feedback-continuous-sap-tolerance]] (basically zero error across all
four metrics). Test-only change; no cascade behaviour touched.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Khalim Conn-Kowlessar 2026-06-02 18:11:28 +00:00 • committed by Jun-te Kim
parent fe527e3a23
commit 3bbb9aa1a2
1 changed files with 12 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
backend/documents_parser/tests/test_heating_systems_corpus.py
View file

@ -68,13 +68,18 @@ _CORPUS_ROOT = (
# Per-pin absolute tolerances. Worksheet `SAP value` lodges 4 d.p.,
# (255) total fuel cost 4 d.p., (272) total CO2 4 d.p., (286) Total
# Primary energy kWh/year 4 d.p. — pin at 1e-4 relative to lodged
# precision so any drift outside cascade float noise fires.
_SAP_RESID_ABS_TOLERANCE = 0.001
_COST_RESID_ABS_TOLERANCE_GBP = 0.01
_CO2_RESID_ABS_TOLERANCE_KG = 0.1
_PE_RESID_ABS_TOLERANCE_KWH = 0.1
# (255)/(355) total fuel cost 4 d.p., (272)/(383) total CO2 4 d.p.,
# (286)/(483) Total Primary energy kWh/year 4 d.p. — so the hard floor
# on any residual is ~5e-5 (half a unit in the last printed digit),
# independent of cascade precision. Pin at 1e-4 on EVERY metric (per
# [[feedback-zero-error-strict]] / [[feedback-continuous-sap-tolerance]]
# — basically zero error across continuous SAP, cost, CO2 and PE) so
# any drift beyond PDF print-rounding fires loudly. All 41 variants hold
# at this tolerance; closures re-pin the smaller residual, never widen.
_SAP_RESID_ABS_TOLERANCE = 0.0001
_COST_RESID_ABS_TOLERANCE_GBP = 0.0001
_CO2_RESID_ABS_TOLERANCE_KG = 0.0001
_PE_RESID_ABS_TOLERANCE_KWH = 0.0001
@dataclass(frozen=True)