Contains projects associated to the development of modeling products
Find a file
Khalim Conn-Kowlessar c7419ca45a Slice S0380.142: §4 (61)m/(59)m cascade — cylinder presence gates combi=0 + primary loss applies for PCDB Table 322 boilers
SAP 10.2 §4 line 7702 (PDF p.137):

  Combi loss for each month from Table 3a, 3b or 3c (enter '0' if
  not a combi boiler)

SAP 10.2 Table 3 (PDF p.160) zero-loss list for primary circuit loss:

  Electric immersion heater
  Combi boiler (including when it is part of a combined heat pump and
  boiler package and provides all the hot water)
  CPSU (including electric CPSU)
  Boiler and thermal store within a single casing
  Separate boiler and thermal store connected by no more than 1.5 m
  of insulated pipework
  Direct-acting electric boiler
  Heat pump (...) with hot water vessel integral to package

Combi boilers are defined by Table 3's zero-loss list entry: they
provide instantaneous DHW with no storage vessel. A cert that lodges
a hot-water cylinder therefore has a non-combi heat generator —
the cylinder bypasses any instantaneous-DHW capability and the
boiler acts as a regular boiler for the DHW circuit.

Two compounding gaps for PCDB Table 322 (gas/oil boiler) records
with a lodged cylinder:

(a) (61)m combi loss: pre-slice the cascade routed every PCDB record
    through `pcdb_combi_loss_override` regardless of cylinder
    presence. For PCDB regular boilers (subsidiary_type=0, store_
    type=0, separate_dhw_tests=0) this dispatched to Table 3a row 1
    "Instantaneous without keep-hot" — 600 kWh/yr. Cert pcdb 1
    (Potterton KOA PCDB 716 + 110 L cylinder) exposed this: worksheet
    (61)m = 0 ; cascade was lodging 600 kWh/yr keep-hot loss on a
    regular oil boiler.

(b) (59)m primary loss: `_primary_loss_applies` gated on
    `main_heating_category in {1, 2}`. The Elmhurst path leaves
    `main_heating_category=None`, so the gate returned False even
    when the cert lodged a PCDB Table 322 (gas/oil boiler) record +
    a cylinder. Worksheet (59)m sum ~1177 kWh ; cascade was zero.

Fix:

- `_water_heating_worksheet_and_gains` now zeroes combi_loss_override
  whenever `epc.has_hot_water_cylinder` is True (top-level gate
  preceding the `pcdb_combi_loss_override` dispatch). Preserves the
  existing non-cylinder fallback for HP / no-PCDB / community-heat
  certs that lack a main_heating_category lodgement.

- `_primary_loss_applies` extends the Elmhurst-path fallback: when
  `main_heating_index_number` resolves to a PCDB Table 322 record,
  return True (the cert is implicitly a boiler — Table 3 row 1 covers
  any "heat generator (e.g. boiler) connected to a hot water storage
  vessel via insulated or uninsulated pipes").

Corpus impact:

- pcdb 1 (Potterton KOA + cylinder, the only PCDB Table 322 + cylinder
  combination in the corpus): SAP +3.40 → +2.86; cost -£75.68 →
  -£63.22; CO2 -397.02 → -328.74; PE -1601.74 → -1257.97.

- Golden cert 0390-2954-3640-2196-4175 (Firebird oil combi PCDF 9005
  + cylinder): PE -26.37 → -28.50; CO2 -2.55 → -2.75. Combi-loss
  removal (-600 kWh/yr) exceeded the primary-loss gain (~5-10 kWh
  given the cert's insulated pipework + thermostat lodging), so the
  net (62) shifted down. Direction is more spec-correct: the spec
  treats a combi feeding a cylinder as a regular boiler for DHW,
  matching the (61)m=0 + (59)m>0 worksheet behaviour.

Extended handover suite: 885 pass, 0 fail.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-01 16:28:48 +00:00
.devcontainer pr review, move domain and orhcestration 2026-06-01 14:00:31 +00:00
.github/workflows added landlord description overrides 2026-05-29 12:12:54 +00:00
.idea scaffolding for ml pipeline 2026-05-16 14:15:56 +00:00
.vscode added utils to allow easier subtask management 2026-03-02 15:15:39 +00:00
applications move landlord overrides 2026-06-01 15:46:46 +00:00
asset_list Remove EPC and asset_list changes unrelated to SAL handler 2026-05-22 15:36:46 +00:00
backend Slice S0380.142: §4 (61)m/(59)m cascade — cylinder presence gates combi=0 + primary loss applies for PCDB Table 322 boilers 2026-06-01 16:28:48 +00:00
backlog implemented onboarding 2026-04-21 20:23:33 +00:00
datatypes Slice S0380.133: derive solid-fuel main fuel from §14.0 EES Code 2026-06-01 16:28:48 +00:00
deployment/terraform added landlord description overrides 2026-05-29 12:12:54 +00:00
docs/adr move landlord overrides 2026-06-01 15:46:46 +00:00
domain Slice S0380.142: §4 (61)m/(59)m cascade — cylinder presence gates combi=0 + primary loss applies for PCDB Table 322 boilers 2026-06-01 16:28:48 +00:00
epr_data_exports allowing carbon and energy otimisation by removing slack 2025-07-31 19:13:16 +01:00
etl booking status 2026-06-01 16:28:47 +00:00
infrastructure GoogleSolarApiClient raises BuildingInsightsNotFoundError on 404 entity-not-found 🟩 2026-06-01 16:28:47 +00:00
model_data/requirements its working the way khalim wanted wiht postcode and then search that 2026-01-22 15:17:13 +00:00
orchestration moved classifier data transformation to an easy one 2026-06-01 14:53:34 +00:00
recommendations save 2026-05-07 15:55:44 +00:00
repositories move landlord overrides 2026-06-01 15:46:46 +00:00
scripts Slice S0380.39: bulk-fetch 38 cohort-2 EPC API JSONs for cross-mapper parity 2026-06-01 16:28:47 +00:00
sfr/principal_pitch added added historic epc data class with shape 2026-05-08 12:03:35 +00:00
survey_report quidos site notes extraction 2025-02-18 19:49:29 +00:00
tests GoogleSolarApi translates BuildingInsightsNotFoundError to sentinel dict 🟩 2026-06-01 16:28:47 +00:00
utilities get rid of comments 2026-05-20 13:21:11 +00:00
utils rename files in sharepoint to desired structure 2026-05-20 16:26:07 +00:00
.coveragerc fixed unit tests 2023-10-05 16:04:12 +01:00
.dockerignore deployment from infrastructure 2026-05-19 12:55:30 +00:00
.gitignore fixed merge conflicts from main 2026-05-26 11:21:09 +00:00
__init__.py added checking for directory before creation and made some minor style changes 2023-08-25 15:21:17 +01:00
ara_backend_design.md refactor: lift-and-shift packages/domain/src/domain/ml → domain/sap10_ml 2026-05-26 13:01:35 +00:00
BaseUtility.py fixed missing task and subtask for single remote assessments 2025-11-27 17:50:26 +00:00
CLAUDE.md fixed merge conflicts from main 2026-05-26 11:21:09 +00:00
conftest.py working on integrating new EPC api into address2UPRN 2026-04-27 11:32:44 +00:00
CONTEXT.md refactor: lift-and-shift packages/domain/src/domain/ml → domain/sap10_ml 2026-05-26 13:01:35 +00:00
devcontainer.sh add dev container 2026-04-17 14:50:57 +00:00
Dockerfile.test fix: address 22 project-wide test failures from previous sweep 2026-05-26 13:34:51 +00:00
Dockerfile.test.dockerignore deployment from infrastructure 2026-05-19 12:55:30 +00:00
Makefile adding to dev container to create shared network on start up 2026-04-25 15:03:07 +00:00
MEMORY.md memory 2026-04-02 10:24:31 +00:00
package-lock.json restructuring openUrpn code 2023-07-20 11:41:43 +01:00
package.json restructuring openUrpn code 2023-07-20 11:41:43 +01:00
pyproject.toml slice 14a: ml_training_data pkg + sample.py (CSV filter + random sample) 2026-05-16 17:39:43 +00:00
pyrightconfig.json slice 14d: build_features wires bulk reader -> mapper -> EpcMlTransform 2026-05-16 18:38:41 +00:00
pytest.ini refactor: lift-and-shift packages/domain/src/domain/ml → domain/sap10_ml 2026-05-26 13:01:35 +00:00
README.md adding to dev container to create shared network on start up 2026-04-25 15:03:07 +00:00
run_lambda_local.sh debugging local lambda run and updating the sap point checking condition 2023-09-13 18:47:12 +01:00
serverless.yml added logic to add to serverless 2026-04-22 12:39:44 +00:00
test.requirements.txt tests wrong environemnt 2026-05-29 16:17:06 +00:00
tox.ini removing playright install for integration test 2026-04-30 20:08:13 +00:00
UBIQUITOUS_LANGUAGE.md postcode_splitter: pure domain (UserAddress, sanitise_postcode, postcode_batching) 2026-05-19 16:45:47 +00:00

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