Merge pull request #1 from Hestia-Homes/master

Master

.gitignore

@@ -158,3 +158,82 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+
+# Covers JetBrains IDEs: IntelliJ, RubyMine, PhpStorm, AppCode, PyCharm, CLion, Android Studio, WebStorm and Rider
+# Reference: https://intellij-support.jetbrains.com/hc/en-us/articles/206544839
+
+# User-specific stuff
+.idea/**/workspace.xml
+.idea/**/tasks.xml
+.idea/**/usage.statistics.xml
+.idea/**/dictionaries
+.idea/**/shelf
+
+# AWS User-specific
+.idea/**/aws.xml
+
+# Generated files
+.idea/**/contentModel.xml
+
+# Sensitive or high-churn files
+.idea/**/dataSources/
+.idea/**/dataSources.ids
+.idea/**/dataSources.local.xml
+.idea/**/sqlDataSources.xml
+.idea/**/dynamic.xml
+.idea/**/uiDesigner.xml
+.idea/**/dbnavigator.xml
+
+# Gradle
+.idea/**/gradle.xml
+.idea/**/libraries
+
+# Gradle and Maven with auto-import
+# When using Gradle or Maven with auto-import, you should exclude module files,
+# since they will be recreated, and may cause churn.  Uncomment if using
+# auto-import.
+.idea/artifacts
+.idea/compiler.xml
+.idea/jarRepositories.xml
+.idea/modules.xml
+.idea/*.iml
+.idea/modules
+*.iml
+*.ipr
+
+# CMake
+cmake-build-*/
+
+# Mongo Explorer plugin
+.idea/**/mongoSettings.xml
+
+# File-based project format
+*.iws
+
+# IntelliJ
+out/
+
+# mpeltonen/sbt-idea plugin
+.idea_modules/
+
+# JIRA plugin
+atlassian-ide-plugin.xml
+
+# Cursive Clojure plugin
+.idea/replstate.xml
+
+# SonarLint plugin
+.idea/sonarlint/
+
+# Crashlytics plugin (for Android Studio and IntelliJ)
+com_crashlytics_export_strings.xml
+crashlytics.properties
+crashlytics-build.properties
+fabric.properties
+
+# Editor-based Rest Client
+.idea/httpRequests
+
+# Android studio 3.1+ serialized cache file
+.idea/caches/build_file_checksums.ser
+

.idea/.gitignore

@@ -0,0 +1,3 @@
+# Default ignored files
+/shelf/
+/workspace.xml

.idea/Model.iml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$" />
+    <orderEntry type="jdk" jdkName="Python 3.10 (hestia-data)" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+</module>

.idea/codeStyles/codeStyleConfig.xml

@@ -0,0 +1,5 @@
+<component name="ProjectCodeStyleConfiguration">
+  <state>
+    <option name="PREFERRED_PROJECT_CODE_STYLE" value="Default" />
+  </state>
+</component>

.idea/inspectionProfiles/Project_Default.xml

@@ -0,0 +1,77 @@
+<component name="InspectionProjectProfileManager">
+  <profile version="1.0">
+    <option name="myName" value="Project Default" />
+    <inspection_tool class="GrazieInspection" enabled="false" level="GRAMMAR_ERROR" enabled_by_default="false" />
+    <inspection_tool class="LanguageDetectionInspection" enabled="false" level="WARNING" enabled_by_default="false" />
+    <inspection_tool class="PyCompatibilityInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ourVersions">
+        <value>
+          <list size="1">
+            <item index="0" class="java.lang.String" itemvalue="3.10" />
+          </list>
+        </value>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="PyPackageRequirementsInspection" enabled="true" level="WARNING" enabled_by_default="true">
+      <option name="ignoredPackages">
+        <value>
+          <list size="48">
+            <item index="0" class="java.lang.String" itemvalue="click" />
+            <item index="1" class="java.lang.String" itemvalue="lxml" />
+            <item index="2" class="java.lang.String" itemvalue="requests" />
+            <item index="3" class="java.lang.String" itemvalue="schema" />
+            <item index="4" class="java.lang.String" itemvalue="levenshtein" />
+            <item index="5" class="java.lang.String" itemvalue="decorator" />
+            <item index="6" class="java.lang.String" itemvalue="environs" />
+            <item index="7" class="java.lang.String" itemvalue="scikit-learn" />
+            <item index="8" class="java.lang.String" itemvalue="tabulate" />
+            <item index="9" class="java.lang.String" itemvalue="python-dateutil" />
+            <item index="10" class="java.lang.String" itemvalue="python-dotenv" />
+            <item index="11" class="java.lang.String" itemvalue="marshmallow" />
+            <item index="12" class="java.lang.String" itemvalue="pycodestyle" />
+            <item index="13" class="java.lang.String" itemvalue="cycler" />
+            <item index="14" class="java.lang.String" itemvalue="numpy" />
+            <item index="15" class="java.lang.String" itemvalue="currency-symbols" />
+            <item index="16" class="java.lang.String" itemvalue="flake8" />
+            <item index="17" class="java.lang.String" itemvalue="patsy" />
+            <item index="18" class="java.lang.String" itemvalue="yfinance" />
+            <item index="19" class="java.lang.String" itemvalue="contextlib2" />
+            <item index="20" class="java.lang.String" itemvalue="pytest-cov" />
+            <item index="21" class="java.lang.String" itemvalue="certifi" />
+            <item index="22" class="java.lang.String" itemvalue="gevent" />
+            <item index="23" class="java.lang.String" itemvalue="pycountry" />
+            <item index="24" class="java.lang.String" itemvalue="xgboost" />
+            <item index="25" class="java.lang.String" itemvalue="coverage" />
+            <item index="26" class="java.lang.String" itemvalue="scipy" />
+            <item index="27" class="java.lang.String" itemvalue="botocore" />
+            <item index="28" class="java.lang.String" itemvalue="pytest" />
+            <item index="29" class="java.lang.String" itemvalue="prospector" />
+            <item index="30" class="java.lang.String" itemvalue="pandas" />
+            <item index="31" class="java.lang.String" itemvalue="pyyaml" />
+            <item index="32" class="java.lang.String" itemvalue="tqdm" />
+            <item index="33" class="java.lang.String" itemvalue="boto3" />
+            <item index="34" class="java.lang.String" itemvalue="s3transfer" />
+            <item index="35" class="java.lang.String" itemvalue="scikit-optimize" />
+            <item index="36" class="java.lang.String" itemvalue="pyaml" />
+            <item index="37" class="java.lang.String" itemvalue="future" />
+            <item index="38" class="java.lang.String" itemvalue="pylint" />
+            <item index="39" class="java.lang.String" itemvalue="matplotlib" />
+            <item index="40" class="java.lang.String" itemvalue="statsmodels" />
+            <item index="41" class="java.lang.String" itemvalue="isort" />
+            <item index="42" class="java.lang.String" itemvalue="pytz" />
+            <item index="43" class="java.lang.String" itemvalue="idna" />
+            <item index="44" class="java.lang.String" itemvalue="scikeras" />
+            <item index="45" class="java.lang.String" itemvalue="nose" />
+            <item index="46" class="java.lang.String" itemvalue="sphinx" />
+            <item index="47" class="java.lang.String" itemvalue="pydantic" />
+          </list>
+        </value>
+      </option>
+    </inspection_tool>
+    <inspection_tool class="SpellCheckingInspection" enabled="false" level="TYPO" enabled_by_default="false">
+      <option name="processCode" value="true" />
+      <option name="processLiterals" value="true" />
+      <option name="processComments" value="true" />
+    </inspection_tool>
+  </profile>
+</component>

.idea/inspectionProfiles/profiles_settings.xml

@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>

.idea/misc.xml

@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.10 (hestia-data)" project-jdk-type="Python SDK" />
+  <component name="PythonCompatibilityInspectionAdvertiser">
+    <option name="version" value="3" />
+  </component>
+</project>

.idea/modules.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/Model.iml" filepath="$PROJECT_DIR$/.idea/Model.iml" />
+    </modules>
+  </component>
+</project>

.idea/toolchains.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="NimToolchainService">
+    <option name="rootPaths">
+      <list />
+    </option>
+  </component>
+</project>

.idea/vcs.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
+  </component>
+</project>

README.md

@@ -0,0 +1,15 @@
+# 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
+
+# Folders
+
+### epc_data/
+This folder contains related to the reading and preparation of
+epc data

epc_data/EpcClean.py

@@ -0,0 +1,256 @@
+import re
+from typing import List, Dict, Any, Union, Tuple, Optional
+from collections import Counter
+
+
+class EpcClean:
+    """
+    Container for methods which we utilise for cleaning EPC data
+    """
+
+    U_VALUE_REGEX = re.compile(r"(\d+\.\d+)")
+    UNIT_REGEX = re.compile(r"(w/m-¦k)")
+
+    CLEANING_FIELDS: List[str] = [
+        "roof-description",
+        "floor-description",
+        "walls-description",
+        "mainheat-description"
+    ]
+
+    def __init__(self, data: List[Dict[str, Any]]) -> None:
+        """
+        EpcClean constructor.
+
+        :param data: List of dictionaries containing EPC data.
+        """
+        self.data: List[Dict[str, Any]] = data
+        self.unique_vals: Dict[str, Any] = {}
+        self.cleaned: Dict[str, List[Any]] = {}
+
+    def clean(self) -> None:
+        """
+        Cleans the EPC data, mapping text fields to property attributes.
+        """
+        self._init_empty_cleaned_obj()
+
+        for field in self.CLEANING_FIELDS:
+            self.unique_vals[field] = Counter([v[field] for v in self.data])
+
+        for description in self.unique_vals["roof-description"].keys():
+            self.cleaned["roof-description"].append(
+                {
+                    "original_description": description,
+                    **self.clean_roof(description)
+                }
+            )
+
+    def _init_empty_cleaned_obj(self) -> None:
+        """
+        Initializes an empty object for cleaned data.
+        """
+        self.cleaned = {field: [] for field in self.CLEANING_FIELDS}
+
+    @staticmethod
+    def _search_split_roof_description(desc: str) -> str:
+        """
+        Searches roof descriptions and looks for key words, determining a description about the roof's insulation.
+
+        :param desc: Description to be searched.
+        :return: Result of the search.
+        """
+        if desc == "insulated":
+            return "average"
+        if desc == "limited":
+            return "below average"
+        raise NotImplementedError("Handle me")
+
+    def _find_insulation_thickness(
+        self, description_lower: str, is_pitched: bool, is_roof_room: bool, is_flat: bool
+    ) -> Union[int, str, None]:
+        """
+        Finds insulation thickness in the description.
+
+        :param description_lower: Lowercase description.
+        :param is_pitched: Whether the roof is pitched.
+        :param is_roof_room: Whether there is a room in the roof.
+        :param is_flat: Whether the roof is flat.
+        :return: Insulation thickness if found, else None.
+        """
+        if "no insulation" in description_lower:
+            return 0
+
+        if is_pitched:
+            try:
+                thickness = description_lower.split("pitched,")[-1].split("mm")[0].strip()
+                if "+" in thickness:
+                    return thickness
+                try:
+                    return int(thickness)
+                except ValueError as int_error:
+                    raise ValueError(int_error)
+            except ValueError as _:
+                if "invalid input" in description_lower:
+                    return None
+                desc = description_lower.split("pitched,")[-1].strip().split(" ")[0]
+                return self._search_split_roof_description(desc)
+
+        if is_roof_room:
+            desc_split_lookup = {
+                "ceiling insulated": "average",
+                "thatched": "average",
+            }
+            # Just search for specific phrases
+            desc_split = description_lower.split("roof room(s),")[-1].strip()
+            res = desc_split_lookup.get(desc_split)
+            if res:
+                return res
+
+            desc = desc_split.split(" ")[0]
+            return self._search_split_roof_description(desc)
+
+        if is_flat:
+            # Just search for specific phrases
+            desc = description_lower.split("flat,")[-1].lstrip().split(" ")[0]
+            return self._search_split_roof_description(desc)
+
+        return None
+
+    def _extract_thermal_transmittence(self, description_lower: str) -> Tuple[Union[float, None], Union[str, None]]:
+        """
+        Extracts thermal transmittance from the description.
+
+        :param description_lower: Lowercase description.
+        :return: Tuple containing U-value and unit.
+        """
+        # Find U-value
+        u_value = re.search(self.U_VALUE_REGEX, description_lower)
+        if u_value is not None:
+            u_value = float(u_value.group(1))
+        else:
+            u_value = None
+
+        # Find unit
+        unit = re.search(self.UNIT_REGEX, description_lower)
+        if unit is not None:
+            unit = unit.group(1)
+        else:
+            unit = None
+
+        return u_value, unit
+
+    @staticmethod
+    def _make_clean_roof_output(
+        is_valid: bool,
+        at_rafters: bool,
+        is_pitched: bool,
+        is_roof_room: bool,
+        has_loft: bool,
+        insulation_thickness: str | int | None,
+        has_dwelling_above: bool,
+        assumed: bool,
+        is_flat: bool,
+        is_thatched: bool,
+        thermal_transmittence: Optional[float],
+        thermal_transmittence_unit: Optional[str]
+    ) -> Dict[str, Union[bool, str, None]]:
+        """
+        Utility function to ensure all the keys are present in the output.
+
+        :param is_valid: True if the roof descrption is valid, False otherwise
+        :param at_rafters: True if the insulation is at the rafters, False otherwise
+        :param is_pitched: True if the roof is pitched, False otherwise
+        :param is_roof_room: True if there is a room in the roof, False otherwise
+        :param has_loft: True if there is a loft, False otherwise
+        :param insulation_thickness: The thickness of the insulation
+        :param has_dwelling_above: True if there is a dwelling above, False otherwise
+        :param assumed: True if the roof type was assumed based on property age, False otherwise
+        :param is_flat: True if the roof is flat, False otherwise
+        :param is_thatched: True if the roof is thatched, False otherwise
+        :param thermal_transmittence: The thermal transmittence value of the roof, if known
+        :param thermal_transmittence_unit: The unit of thermal transmittence, if known
+        :return: A dictionary containing all the information about the roof.
+        """
+
+        return {
+            "is_valid": is_valid,
+            "at_rafters": at_rafters,
+            "is_pitched": is_pitched,
+            "is_roof_room": is_roof_room,
+            "has_loft": has_loft,
+            "insulation_thickness": insulation_thickness,
+            "has_dwelling_above": has_dwelling_above,
+            "assumed": assumed,
+            "is_flat": is_flat,
+            "is_thatched": is_thatched,
+            "thermal_transmittence": thermal_transmittence,
+            "thermal_transmittence_unit": thermal_transmittence_unit
+        }
+
+    def clean_roof(self, description: str) -> Dict[str, Union[str, bool, int, None]]:
+        """
+        We aim to extract features about the roof, so we can characterise it. We will check:
+        - If the roof is pitched
+        - If there is a room roof
+        - if there is a loft
+        - If it has insulation
+        - if so, what degree of insulation
+
+        :param description: Description of the roof.
+        :return: Dictionary of attributes of the roof.
+        """
+        description_lower = description.lower().strip()
+
+        if "another dwelling above" in description_lower or "other premises above" in description_lower:
+            return self._make_clean_roof_output(
+                is_valid="invalid" not in description_lower,
+                at_rafters="at rafters" in description_lower,
+                is_pitched=False,
+                is_roof_room=False,
+                has_loft=False,
+                insulation_thickness=0,
+                has_dwelling_above=True,
+                assumed="assumed" in description_lower,
+                is_flat="flat" in description_lower,
+                is_thatched=False,
+                thermal_transmittence=None,
+                thermal_transmittence_unit=None
+            )
+
+        is_pitched = "pitched" in description_lower
+        is_roof_room = "roof room" in description_lower
+        has_loft = "loft" in description_lower
+        is_flat = "flat" in description_lower
+        is_thatched = "thatched" in description_lower
+        at_rafters = "at rafters" in description_lower
+
+        thermal_transmittence, thermal_transmittence_unit, insulation_thickness = None, None, None
+        if "insulation" in description_lower or "insulated" in description_lower:
+            insulation_thickness = self._find_insulation_thickness(description_lower, is_pitched, is_roof_room, is_flat)
+        elif "thermal transmittance" in description_lower:
+            thermal_transmittence, thermal_transmittence_unit = self._extract_thermal_transmittence(description_lower)
+        elif is_thatched:
+            # Search for these features:
+            thermal_transmittence, thermal_transmittence_unit = self._extract_thermal_transmittence(description_lower)
+            insulation_thickness = self._find_insulation_thickness(
+                description_lower, is_pitched, is_roof_room, is_flat
+            )
+        elif description_lower == "pitched":
+            thermal_transmittence, thermal_transmittence_unit, insulation_thickness = None, None, None
+        else:
+            raise NotImplementedError("Not handled this")
+
+        return self._make_clean_roof_output(
+            is_valid="invalid" not in description_lower,
+            at_rafters=at_rafters,
+            is_pitched=is_pitched,
+            is_roof_room=is_roof_room,
+            has_loft=has_loft,
+            insulation_thickness=insulation_thickness,
+            has_dwelling_above=False,
+            assumed="assumed" in description_lower,
+            is_flat=is_flat,
+            is_thatched=is_thatched,
+            thermal_transmittence=thermal_transmittence,
+            thermal_transmittence_unit=thermal_transmittence_unit
+        )

epc_data/Property.py

@@ -0,0 +1,31 @@
+from epc_api.client import EpcClient
+from epc_data.config import EPC_AUTH_TOKEN
+
+
+class Property:
+
+    def __init__(self, postcode, address1, epc_client=None, data=None):
+        self.postcode = postcode
+        self.address1 = address1
+        self.data = data
+
+        if epc_client:
+            self.epc_client = epc_client
+        else:
+            self.epc_client = EpcClient(auth_token=EPC_AUTH_TOKEN)
+
+    def search_address_epc(self):
+        """
+        This method searches for an address in the EPC database and returns the first result
+        :return: property data
+        """
+        if self.data:
+            return
+
+        # This will fail if a property does not have an EPC - this has been documented as a case to handle
+        response = self.epc_client.domestic.search(params={"address": self.address1, "postcode": self.postcode})
+
+        if len(response["rows"]) > 1:
+            raise Exception("More than one result found for this address - investigate me")
+
+        self.data = response["rows"][0]

epc_data/README.md

@@ -0,0 +1,49 @@
+# Environment setup
+
+We're using conda to manage environments to circumvent the
+issues with Mac M1. This documentation will also cover Pycharm setup.
+
+We're working in python 3.10 so
+
+```commandline
+conda create -n hestia-data python=3.10
+```
+
+Then activate the environment
+
+```commandline
+conda activate hestia-data
+```
+
+To set up with Pycharm, run
+
+```commandline
+which python
+```    
+
+and grab the path to the python executable. Then in Pycharm, go to
+Settings > Project > Python Interpreter and click the gear icon
+to add a new interpreter. Select Conda and either paste the path to the python executable
+and click OK, or select the conda environment from the dropdown.
+
+You may need to restart Pycharm for the new interpreter to be recognised.
+
+To install project dependencies navigate to /epc_data and run
+
+```commandline
+pip install -r requirements.txt
+```
+
+### Running Tests
+
+If you are not in a virtual environment, activate it with
+
+```commandline
+conda activate envName
+```
+
+Then run
+
+```commandline
+python -m pytest
+```

epc_data/app.py

@@ -0,0 +1,42 @@
+from tqdm import tqdm
+
+from epc_data.temp_inputs import input_data
+from epc_data.Property import Property
+from epc_data.config import EPC_AUTH_TOKEN
+from epc_api.client import EpcClient
+from epc_data.downloader import pagenated_epc_download
+from epc_data.EpcClean import EpcClean
+
+
+def handler():
+    # To begin with, the input data is a list of dictionaries, however we would read this file in
+
+    epc_client = EpcClient(auth_token=EPC_AUTH_TOKEN)
+
+    input_properties = [
+        Property(postcode=config['postcode'], address1=config['address1'], epc_client=epc_client)
+        for config in input_data
+    ]
+
+    for p in input_properties:
+        p.search_address_epc()
+
+    local_authorities = {p.data['local-authority'] for p in input_properties}
+
+    data = []
+    for la in tqdm(local_authorities):
+        data.extend(
+            pagenated_epc_download(
+                client=epc_client,
+                params={"local-authority": la},
+                page_size=5000,
+                n_pages=10,
+            )
+        )
+
+    cleaner = EpcClean(data)
+
+    cleaner.clean()
+
+    import pandas as pd
+    df = pd.DataFrame(cleaner.cleaned["roof-description"])

epc_data/config.py

@@ -0,0 +1,6 @@
+import os
+from dotenv import load_dotenv
+
+load_dotenv(dotenv_path='epc_data/.env')
+
+EPC_AUTH_TOKEN = os.environ.get('EPC_AUTH_TOKEN')

epc_data/downloader.py

@@ -0,0 +1,26 @@
+import time
+
+
+def pagenated_epc_download(client, params, page_size, n_pages, verbose=0, slowdown=0.1):
+    offset_from = 0
+    n_completed = 0
+    results = []
+    complete = False
+    while not complete:
+        if verbose:
+            print("Pulling for page %s" % str(int(offset_from / page_size) + 1))
+        time.sleep(slowdown)
+        search_resp = client.domestic.search(params=params, offset_from=offset_from, size=page_size)
+
+        # Note: We can only make 10k queries for a single set of search queries.
+        # It might make sense to download data via zip for machine learning since we don't need this
+        # data to be perfectly up to date
+        if search_resp is None:
+            break
+        results.extend(search_resp["rows"])
+        if n_completed == n_pages:
+            complete = True
+        else:
+            offset_from += page_size
+
+    return results

epc_data/requirements.txt

@@ -0,0 +1,6 @@
+epc-api-python
+python-dotenv
+tqdm
+pandas
+mypy
+pytest

epc_data/temp_inputs.py

@@ -0,0 +1,11 @@
+# Temporary input data
+input_data = [
+    {
+        "address1": "28 Distillery Wharf",
+        "postcode": "w6 9bf"
+    },
+    {
+        "address1": "Flat 14 Godley V C House",
+        "postcode": "E2 0LP"
+    }
+]

epc_data/tests/test_EpcClean.py

@@ -0,0 +1,94 @@
+import pytest
+import pickle
+from epc_data.EpcClean import EpcClean
+from pathlib import Path
+from epc_data.tests.test_data.EpcClean_test_roof_cases import clean_roof_test_cases
+
+# For local testing
+if __file__ == "<input>":
+    input_data_path = Path("./epc_data/tests/test_data/EpcClean_inputs.obj")
+else:
+    current_file_path = Path(__file__)
+    input_data_path = current_file_path.parent / 'test_data' / 'EpcClean_inputs.obj'
+
+
+class TestEpcClean:
+
+    @staticmethod
+    def load_data(path):
+        with open(path, "rb") as file:
+            return pickle.load(file)
+
+    @pytest.fixture(autouse=True)
+    def setup_class(self):
+        self.cleaner = EpcClean(self.load_data(input_data_path))
+
+    def test_clean(self):
+        self.cleaner.clean()
+        assert len(self.cleaner.cleaned["roof-description"]) == len(self.cleaner.unique_vals["roof-description"])
+
+    def test__init_empty_cleaned_obj(self):
+        self.cleaner._init_empty_cleaned_obj()
+        assert all([len(values) == 0 for values in self.cleaner.cleaned.values()])
+
+    def test__search_split_roof_description(self):
+        assert self.cleaner._search_split_roof_description("insulated") == "average"
+        assert self.cleaner._search_split_roof_description("limited") == "below average"
+        with pytest.raises(NotImplementedError):
+            self.cleaner._search_split_roof_description("unknown")
+
+    def test__find_insulation_thickness(self):
+        assert self.cleaner._find_insulation_thickness("no insulation", False, False, False) == 0
+
+    def test__extract_thermal_transmittence(self):
+        description = "U-value of 2.3 w/m-¦k"
+        assert self.cleaner._extract_thermal_transmittence(description) == (2.3, "w/m-¦k")
+
+    def test_clean_roof(self):
+        result = self.cleaner.clean_roof('Pitched, 270 mm loft insulation')
+
+        # change the expected output based on your requirement
+        expected_output = {
+            "is_valid": True,
+            "at_rafters": False,
+            "is_pitched": True,
+            "is_roof_room": False,
+            "has_loft": True,
+            "insulation_thickness": 270,
+            "has_dwelling_above": False,
+            "assumed": False,
+            "is_flat": False,
+            "is_thatched": False,
+            "thermal_transmittence": None,
+            "thermal_transmittence_unit": None
+        }
+
+        assert result == expected_output
+
+        for test_case in clean_roof_test_cases:
+            result = self.cleaner.clean_roof(test_case['original_description'])
+            # Ensure the output ordering is correct
+            expected_result = {key: test_case[key] for key in result.keys()}
+            expected_result["desc"] = test_case["original_description"]
+            result["desc"] = test_case["original_description"]
+            assert result == expected_result
+
+    def test_clean_roof_with_dwelling_above(self):
+        result = self.cleaner.clean_roof('(another dwelling above)')
+
+        expected_output = {
+            "is_valid": True,
+            "at_rafters": False,
+            "is_pitched": False,
+            "is_roof_room": False,
+            "has_loft": False,
+            "insulation_thickness": 0,
+            "has_dwelling_above": True,
+            "assumed": False,
+            "is_flat": False,
+            "is_thatched": False,
+            "thermal_transmittence": None,
+            "thermal_transmittence_unit": None,
+        }
+
+        assert result == expected_output

epc_data/tests/test_data/EpcClean_test_roof_cases.py

@@ -0,0 +1,261 @@
+clean_roof_test_cases = [
+    {'original_description': 'Flat, limited insulation (assumed)', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 'below average', 'has_dwelling_above': False, 'assumed': True,
+     'is_flat': True, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': '(another dwelling above)', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 0, 'has_dwelling_above': True, 'assumed': False, 'is_flat': False, 'is_thatched': False,
+     'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, insulated (assumed)', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': True,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Flat, insulated (assumed)', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': True, 'is_flat': True,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 200 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 200, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 50 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 50, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, no insulation (assumed)', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 0, 'has_dwelling_above': False, 'assumed': True, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 150 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 150, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Flat, no insulation (assumed)', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 0, 'has_dwelling_above': False, 'assumed': True, 'is_flat': True,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Roof room(s), insulated (assumed)', 'is_pitched': False, 'is_roof_room': True,
+     'has_loft': False, 'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': True,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, 350 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 350, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 300 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 300, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.14 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.14, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, no insulation', 'is_pitched': True, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 0, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False, 'is_thatched': False,
+     'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, 100 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 100, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': '(other premises above)', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 0, 'has_dwelling_above': True, 'assumed': False, 'is_flat': False, 'is_thatched': False,
+     'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, 270 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 270, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, limited insulation (assumed)', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 'below average', 'has_dwelling_above': False, 'assumed': True,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, 250 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 250, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, insulated at rafters', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': False,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': True},
+    {'original_description': 'Pitched', 'is_pitched': True, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Roof room(s), insulated', 'is_pitched': False, 'is_roof_room': True, 'has_loft': False,
+     'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 75 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 75, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Roof room(s), no insulation (assumed)', 'is_pitched': False, 'is_roof_room': True,
+     'has_loft': False, 'insulation_thickness': 0, 'has_dwelling_above': False, 'assumed': True, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Flat, insulated', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': False, 'is_flat': True,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.1 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.1, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, *** INVALID INPUT Code : 57 *** loft insulation', 'is_pitched': True,
+     'is_roof_room': False, 'has_loft': True, 'insulation_thickness': None, 'has_dwelling_above': False,
+     'assumed': False, 'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None,
+     'thermal_transmittence_unit': None, 'is_valid': False, 'at_rafters': False},
+    {'original_description': 'Pitched, 12 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 12, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.10 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.1, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.13 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.13, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.11 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.11, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Roof room(s), ceiling insulated', 'is_pitched': False, 'is_roof_room': True,
+     'has_loft': False, 'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': False,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.12 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.12, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.20 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.2, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.16 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.16, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.15 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.15, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 0 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 0, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.08 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.08, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Flat, limited insulation', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 'below average', 'has_dwelling_above': False, 'assumed': False, 'is_flat': True,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Roof room(s), limited insulation (assumed)', 'is_pitched': False, 'is_roof_room': True,
+     'has_loft': False, 'insulation_thickness': 'below average', 'has_dwelling_above': False, 'assumed': True,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Pitched, insulated', 'is_pitched': True, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': 'average', 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.18 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.18, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, limited insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': 'below average', 'has_dwelling_above': False, 'assumed': False,
+     'is_flat': False, 'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None,
+     'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.19 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.19, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.17 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.17, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.48 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.48, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.42 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.42, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.58 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.58, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.45 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.45, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 1.00 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 1.0, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.2 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.2, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 400+ mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': '400+', 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 25 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 25, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.49 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.49, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.37 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.37, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Thatched', 'is_pitched': False, 'is_roof_room': False, 'has_loft': False,
+     'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False, 'is_thatched': True,
+     'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True, 'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.09 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.09, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.21 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.21, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.24 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.24, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Pitched, 400 mm loft insulation', 'is_pitched': True, 'is_roof_room': False,
+     'has_loft': True, 'insulation_thickness': 400, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.23 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.23, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.28 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.28, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.4 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.4, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Thatched, with additional insulation', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': True, 'thermal_transmittence': None, 'thermal_transmittence_unit': None, 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.26 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.26, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False},
+    {'original_description': 'Average thermal transmittance 0.22 W/m-¦K', 'is_pitched': False, 'is_roof_room': False,
+     'has_loft': False, 'insulation_thickness': None, 'has_dwelling_above': False, 'assumed': False, 'is_flat': False,
+     'is_thatched': False, 'thermal_transmittence': 0.22, 'thermal_transmittence_unit': 'w/m-¦k', 'is_valid': True,
+     'at_rafters': False}]