Model/docs/sap-spec
Khalim Conn-Kowlessar a1c9d2a14d Record post-P5 parity-probe baseline (2026-05-19)
100-cert probe, seed=7, sap_score window 5..99. MAE 4.29
(vs 8.41 on 2026-05-18 with the older 20..95 window — the
delta blends calculator improvements with sample-window
change, so this is logged as the post-P5 reference, not as
"P5 reduced MAE".)

P5 itself was pure trace exposure; the calculator's SAP
output should be numerically unchanged. The headline finding
from this run is primary-energy over-prediction: PE MAE
44.40 kWh/m², bias +39.66 — now the dominant signal with
SAP residuals halved. Each end-use PE contribution surfaces
on SapResult.intermediate per P5.12, so the next session
can localise the bias without re-instrumenting.
2026-05-19 16:19:01 +00:00
..
CALCULATOR_DESIGN_SKETCH.md ADR-0009: pivot to deterministic SAP 10.3 calculator (Accepted) 2026-05-17 21:27:21 +00:00
HANDOVER_FRESH_REVIEW.md docs: handover for fresh-context SAP calculator review 2026-05-18 19:03:40 +00:00
HANDOVER_SYSTEMATIC_REVIEW.md docs: ADR-0010 retargets calculator to SAP 10.2; rewrite handover 2026-05-19 09:54:24 +00:00
PARITY_FINDINGS.md Record post-P5 parity-probe baseline (2026-05-19) 2026-05-19 16:19:01 +00:00
rdsap-10-specification-2025-06-10.pdf ADR-0009: pivot to deterministic SAP 10.3 calculator (Accepted) 2026-05-17 21:27:21 +00:00
sap-10-2-full-specification-2025-03-14.pdf slice S-B9: SAP 10.2/10.3 Table 12 spec-correct prices + Table 12a fix 2026-05-18 15:14:11 +00:00
sap-10-3-full-specification-2026-01-13.pdf ADR-0009: pivot to deterministic SAP 10.3 calculator (Accepted) 2026-05-17 21:27:21 +00:00
SPEC_COVERAGE.md docs: SAP 10.2 / RdSAP 10 spec coverage map 2026-05-18 16:55:17 +00:00