try categorical type for extensions

MODEL_REGISTRY.md

@@ -8,25 +8,25 @@
         "active": true
     },
     "sap": {
-        "version": "v0.15.0",
+        "version": "v0.16.0",
         "stage": {
-            "dev": "v0.15.0"
+            "dev": "v0.16.0"
         },
         "registered": true,
         "active": true
     },
     "heat": {
-        "version": "v0.6.0",
+        "version": "v0.7.0",
         "stage": {
-            "dev": "v0.6.0"
+            "dev": "v0.7.0"
         },
         "registered": true,
         "active": true
     },
     "carbon": {
-        "version": "v0.6.0",
+        "version": "v0.7.0",
         "stage": {
-            "dev": "v0.6.0"
+            "dev": "v0.7.0"
         },
         "registered": true,
         "active": true

modules/ml-pipeline/src/pipeline/configs/feature_processor_logic.py

@@ -45,6 +45,12 @@ def keep_non_zero_rdsap(df):
     return df
 
 
+def convert_extension_columns_to_categorical(df):
+    df["extension_count_starting"] = df["extension_count_starting"].astype("category")
+    df["extension_count_ending"] = df["extension_count_ending"].astype("category")
+    return df
+
+
 # def keep_ending_columns(df):
 #     ending_column_index = [ col_name.endswith("_ENDING") for col_name in list(df.columns)]
 #     keep_columns = df.columns[ending_column_index].to_list()
@@ -54,6 +60,7 @@ def keep_non_zero_rdsap(df):
 #     return df
 
 business_logic = {
+    "convert_extension_columns_to_categorical": convert_extension_columns_to_categorical,
     # "keep_non_zero_rdsap": keep_non_zero_rdsap,
     # "keep_flats": keep_flats,
     # "remove_minimum_habitable_room_size": remove_minimum_habitable_room_size,

modules/ml-pipeline/src/pipeline/configs/settings.yaml

@@ -36,7 +36,9 @@ default:
       drop_columns: [
         "heat_demand_change", "carbon_change", "rdsap_change", "heat_demand_ending", "carbon_ending", "days_to_starting", "days_to_ending",
         'number_habitable_rooms_starting', 'number_habitable_rooms_ending', 'number_heated_rooms_starting', 'number_heated_rooms_ending',
-        'number_habitable_rooms', 'number_heated_rooms', 'lighting_cost_starting', 'lighting_cost_ending', 'heating_cost_starting', 'heating_cost_ending', 'hot_water_cost_starting', 'hot_water_cost_ending',]
+        'number_habitable_rooms', 'number_heated_rooms', 'lighting_cost_starting',
+        'lighting_cost_ending', 'heating_cost_starting', 'heating_cost_ending', 'hot_water_cost_starting', 'hot_water_cost_ending',
+        'floor_thermal_transmittance', 'floor_thermal_transmittance_ending']
       retain_features: null
       # retain_features: ['uprn', 'sap_starting', 'hot_water_energy_eff_ending',
       #  'mainheat_energy_eff_ending', 'constituency', 'roof_energy_eff_ending',

modules/ml-pipeline/src/pipeline/dvc.lock

@@ -40,6 +40,8 @@ stages:
         - heating_cost_ending
         - hot_water_cost_starting
         - hot_water_cost_ending
+        - floor_thermal_transmittance
+        - floor_thermal_transmittance_ending
         default.feature_processor.feature_processor_config.retain_features:
         default.feature_processor.feature_processor_config.subsample_amount:
         default.feature_processor.feature_processor_config.subsample_seed: 0
@@ -55,8 +57,8 @@ stages:
     outs:
     - path: data/prepared_data/
       hash: md5
-      md5: cd82e70e3cfcd0421cddf9a2c4b3e476.dir
-      size: 49655735
+      md5: b4e5adebb1acfd9a84b785c6ae62ebec.dir
+      size: 47975005
       nfiles: 2
   build_model:
     cmd: python 2_build_model.py
@@ -67,8 +69,8 @@ stages:
       size: 4820
     - path: data/prepared_data
       hash: md5
-      md5: cd82e70e3cfcd0421cddf9a2c4b3e476.dir
-      size: 49655735
+      md5: b4e5adebb1acfd9a84b785c6ae62ebec.dir
+      size: 47975005
       nfiles: 2
     params:
       configs/build_model.yaml:
@@ -100,18 +102,18 @@ stages:
     outs:
     - path: data/fit_predictions/
       hash: md5
-      md5: a1c97fa17adae60ee97696c95cfebe7e.dir
-      size: 3618488
+      md5: 37d014de108113e6bf181d59c82cac9d.dir
+      size: 3617984
       nfiles: 1
     - path: data/model/
       hash: md5
-      md5: 7e6f38163a2cfe208ace26702ae28793.dir
-      size: 762167138
+      md5: 1978ae8a3960b419ab764fd47a3693ca.dir
+      size: 761166981
       nfiles: 35
     - path: metrics/fit_metrics.json
       hash: md5
-      md5: 8a6db7d72f9b4b8a5e411beaa3a2a413
-      size: 224
+      md5: f7ad23a19a3f53281e3fecd8b18568fe
+      size: 223
   generate_predictions:
     cmd: python 3_generate_predictions.py
     deps:
@@ -121,13 +123,13 @@ stages:
       size: 2464
     - path: data/model
       hash: md5
-      md5: 7e6f38163a2cfe208ace26702ae28793.dir
-      size: 762167138
+      md5: 1978ae8a3960b419ab764fd47a3693ca.dir
+      size: 761166981
       nfiles: 35
     - path: data/prepared_data
       hash: md5
-      md5: cd82e70e3cfcd0421cddf9a2c4b3e476.dir
-      size: 49655735
+      md5: b4e5adebb1acfd9a84b785c6ae62ebec.dir
+      size: 47975005
       nfiles: 2
     params:
       configs/settings.yaml:
@@ -139,8 +141,8 @@ stages:
     outs:
     - path: data/predictions/
       hash: md5
-      md5: 14729e943275748a19e86cf47d6561ee.dir
-      size: 508468
+      md5: e0472eb66c1876b95c9df7f71aff004d.dir
+      size: 508648
       nfiles: 1
   generate_metrics:
     cmd: python 4_generate_metrics.py
@@ -151,13 +153,13 @@ stages:
       size: 3484
     - path: data/predictions
       hash: md5
-      md5: 14729e943275748a19e86cf47d6561ee.dir
-      size: 508468
+      md5: e0472eb66c1876b95c9df7f71aff004d.dir
+      size: 508648
       nfiles: 1
     - path: data/prepared_data
       hash: md5
-      md5: cd82e70e3cfcd0421cddf9a2c4b3e476.dir
-      size: 49655735
+      md5: b4e5adebb1acfd9a84b785c6ae62ebec.dir
+      size: 47975005
       nfiles: 2
     params:
       configs/settings.yaml:
@@ -167,8 +169,8 @@ stages:
     outs:
     - path: metrics/metrics.json
       hash: md5
-      md5: e076cba5f5737fd6d95931931411fdd0
-      size: 225
+      md5: fc7b825ce93c38e3fa1b4124de195743
+      size: 224
   generate_scenerio_metrics:
     cmd: python 5_generate_scenarios.py
     deps:
@@ -188,9 +190,9 @@ stages:
     outs:
     - path: metrics/scenario_metrics.md
       hash: md5
-      md5: e210b90c92765061a7653842153f8054
+      md5: c6300902b312e95d42f5d6cf5b94efaf
       size: 356
     - path: metrics/scenario_table.md
       hash: md5
-      md5: 63cba3e3c1107ce8b91d33f3e4a82d05
+      md5: 61cc58bf790be5da91efe3aee9a78122
       size: 872