089e6ac9da
SAP 10.2 Table 4d (PDF p.170) "Heating type and responsiveness ...
depending on heat emitter":
Heat emitter R
------------------------------------------------- -----
Systems with radiators 1.0
Underfloor (wet) — pipes in insulated timber floor 1.0
Underfloor (wet) — pipes in screed above insulation 0.75
Underfloor (wet) — pipes in concrete slab 0.25
Warm air via fan coil units 1.0
Pre-S0380.89 the cascade `_responsiveness` had:
if isinstance(emitter, int) and emitter == 2:
return 0.25
return 1.0
But the Elmhurst cert-side enum (`_ELMHURST_HEAT_EMITTER_TO_SAP10` at
`datatypes/epc/domain/mapper.py:3646`) maps:
1 = Radiators
2 = Underfloor (in screed) ← spec R=0.75, NOT 0.25
3 = Underfloor (timber floor)
4 = Warm air
5 = Fan coils
The cascade silently treated screed UFH (Elmhurst code 2) as
concrete-slab UFH (R=0.25 — Table 4d's most thermally massive UFH
variant). The bug halved the actual spec responsiveness — for a screed
UFH cert, off-period temperature reduction was computed with R=0.25
instead of R=0.75, materially under-counting MIT drop and over-counting
SH demand.
The bug is latent on cohort + golden because `_first_main_heating`
picks main[0] and almost all certs lodge radiators (emitter=1) there.
Corpus audit (full JSON sweep): emitter=2 appears on 2 records and
in both cases on a secondary main slot (e.g. golden cert
0240-0200-5706-2365-8010 main[1]) — never on the selected first main.
The fix preempts the next cert that lodges screed UFH on main[0].
Fix:
- New `_RESPONSIVENESS_BY_EMITTER_CODE` dispatch dict reflecting
Table 4d per the Elmhurst cert-side enum (1: 1.0, 2: 0.75, 3: 1.0,
4: 1.0, 5: 1.0). "Concrete slab UFH" (Table 4d R=0.25) has no
cert-side enum entry — that variant would need a new mapper code
before the cascade can dispatch it.
- `_responsiveness` flipped to strict-raise per [[reference-
unmapped-sap-code]]: absent lodging (None / 0 / "") returns modal
R=1.0 default; lodging present but unmapped raises
`UnmappedSapCode("heat_emitter_type", value)`.
Tests (4 new, AAA-structure):
- `test_heat_emitter_code_2_underfloor_in_screed_routes_to_responsiveness_0p75_per_table_4d`
pins the bug fix: emitter=2 → R=0.75 (was 0.25)
- `test_heat_emitter_code_dispatch_table_4d_full_coverage`
pins all 5 Elmhurst emitter codes to spec-correct R
- `test_responsiveness_raises_unmapped_sap_code_on_unknown_emitter`
pins the strict-raise contract (hypothetical code 99 raises)
- `test_responsiveness_default_1p0_when_emitter_lodging_absent`
pins the absent-lodging contract (None / 0 / "" → 1.0)
Test baseline: 568 pass (was 564 + 4 new) + 9 expected
`test_sap_result_pin[000565-*]` fails unchanged. Cohort + golden
unaffected (all use emitter=1 on main[0]).
Pyright net-zero per touched file (one `pyright: ignore` added on the
absent-lodging test where `main_heating_control=None` is passed to a
dataclass declaring `Union[int, str]` — runtime data exhibits None
on certs lacking space-heating controls, so the test covers a real
codepath the type system doesn't model).
Per the user-requested "we keep debugging silent fallbacks" mandate,
this is the second slice (after S0380.88) in the strict-raise series.
Next candidates per [[reference-unmapped-sap-code]]: PV pitch +
overshading, meter→tariff, heat-network DLF, secondary-heating
fraction by category.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
|
||
|---|---|---|
| .devcontainer | ||
| .github/workflows | ||
| .idea | ||
| .vscode | ||
| applications | ||
| asset_list | ||
| backend | ||
| backlog | ||
| datatypes | ||
| deployment/terraform | ||
| docs/adr | ||
| domain | ||
| epr_data_exports | ||
| etl | ||
| infrastructure | ||
| model_data/requirements | ||
| orchestration | ||
| recommendations | ||
| repositories | ||
| scripts | ||
| sfr/principal_pitch | ||
| survey_report | ||
| tests | ||
| utilities | ||
| utils | ||
| .coveragerc | ||
| .dockerignore | ||
| .gitignore | ||
| __init__.py | ||
| ara_backend_design.md | ||
| BaseUtility.py | ||
| CLAUDE.md | ||
| conftest.py | ||
| CONTEXT.md | ||
| devcontainer.sh | ||
| Dockerfile.test | ||
| Dockerfile.test.dockerignore | ||
| Makefile | ||
| MEMORY.md | ||
| package-lock.json | ||
| package.json | ||
| pyproject.toml | ||
| pyrightconfig.json | ||
| pytest.ini | ||
| README.md | ||
| run_lambda_local.sh | ||
| serverless.yml | ||
| test.requirements.txt | ||
| tox.ini | ||
| UBIQUITOUS_LANGUAGE.md | ||
Model Repository
This repository contains the code pertaining to the development of the data science and machine learning products being utilised by Hestia.
The different folders in this repository relate to services that can be used independently, or can be imported and used as part of a larger application
Getting Started
Prerequisites
Dev Container Setup
This repo uses a Docker Compose-based dev container. The model-backend service joins a shared-dev Docker network so it can communicate with other local services (e.g. a frontend container) running on your machine.
VS Code users: The initializeCommand in devcontainer.json creates the shared-dev network automatically before the container starts. No manual step required — just open the repo and select Reopen in Container.
Non-VS Code / CI workflows: Run the following once before starting the container:
make dev-setup
This is idempotent and safe to re-run if the network already exists.
Folders
backend/
This folder contains the code for the fastapi backend service, which provides an interface to much of the functionality in this repository, for the frontend
model_data/
This folder contains related to the reading and preparation of assessment model data, including pulling out epc attributes
Testing
All tests can be run, against the configuration in pytest.ini running
pytest
This will run the complete panel of tests and report on coverage in the locations specified by the pytest.ini file.
To run tests in a specific service, e.g. inside of model_data, simply run
pytest --cov-config=model_data/.coveragerc --cov=model_data
This will produce the test results and coverage reports