diff --git a/modules/ml-pipeline/src/pipeline/configs/settings.yaml b/modules/ml-pipeline/src/pipeline/configs/settings.yaml
index e0c99a6..7b0f8e8 100644
--- a/modules/ml-pipeline/src/pipeline/configs/settings.yaml
+++ b/modules/ml-pipeline/src/pipeline/configs/settings.yaml
@@ -36,14 +36,78 @@ default:
       target: heating_kwh
       identifier_columns: ["uprn"]
       # drop_columns: ["heat_demand_change", "carbon_change", "rdsap_change", "heat_demand_ending", "carbon_ending", "days_to_starting", "days_to_ending"]
-      drop_columns: []
+      drop_columns: ["hot_water_kwh"]
       # [
       #   "sap_ending", "heat_demand_change", "carbon_change", "rdsap_change", "heat_demand_ending", "carbon_ending", 
       #   "heating_cost_ending", "hot_water_cost_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']
-      retain_features: null
+      retain_features: ['uprn', 'heating-cost-current',
+ 'co2-emissions-current',
+ 'hot-water-cost-current',
+ 'total-floor-area',
+ 'secondheat-description',
+ 'environment-impact-current',
+ 'floor-description',
+ 'mainheat-energy-eff',
+ 'current-energy-efficiency',
+ 'mainheat-env-eff',
+ 'walls-energy-eff',
+ 'roof-energy-eff',
+ 'property-type',
+ 'mainheat-description',
+ 'hot-water-env-eff',
+ 'mechanical-ventilation',
+ 'floor-level',
+ 'built-form',
+ 'walls-description',
+ 'mainheatcont-description',
+ 'roof-description',
+ 'energy-consumption-current',
+ 'construction-age-band',
+ 'hotwater-description',
+ 'lodgement-datetime',
+ 'main-fuel',
+ 'hot-water-energy-eff',
+ 'co2-emiss-curr-per-floor-area',
+ 'windows-energy-eff',
+ 'current-energy-rating',
+ 'lodgement-year',
+ 'extension-count',
+ 'number-open-fireplaces',
+ 'number-heated-rooms',
+ 'lodgement-date',
+ 'number-habitable-rooms',
+ 'windows-description',
+ 'local-authority',
+ 'photo-supply',
+ 'heat-loss-corridor',
+ 'posttown',
+ 'address',
+ 'flat-top-storey',
+ 'unheated-corridor-length',
+ 'fixed-lighting-outlets-count',
+ 'inspection-date',
+ 'tenure',
+ 'county',
+ 'constituency-label',
+ 'multi-glaze-proportion',
+ 'solar-water-heating-flag',
+ 'address2',
+ 'energy-tariff',
+ 'floor-height',
+ 'constituency',
+ 'uprn-source',
+ 'transaction-type',
+ 'floor-energy-eff',
+ 'postcode',
+ 'lodgement-month',
+ 'lighting-cost-current',
+ 'glazed-area',
+ 'address1',
+ 'floor-env-eff',
+ 'main-heating-controls']
       # retain_features: ['uprn', 'sap_starting', 'hot_water_energy_eff_ending',
       #  'mainheat_energy_eff_ending', 'constituency', 'roof_energy_eff_ending',
       #  'walls_energy_eff_ending', 'secondheat_description_ending',
diff --git a/modules/ml-pipeline/src/pipeline/dvc.lock b/modules/ml-pipeline/src/pipeline/dvc.lock
index 70615b1..14cd48c 100644
--- a/modules/ml-pipeline/src/pipeline/dvc.lock
+++ b/modules/ml-pipeline/src/pipeline/dvc.lock
@@ -20,8 +20,75 @@ stages:
       size: 4298
     params:
       configs/settings.yaml:
-        default.feature_processor.feature_processor_config.drop_columns: []
+        default.feature_processor.feature_processor_config.drop_columns:
+        - hot_water_kwh
         default.feature_processor.feature_processor_config.retain_features:
+        - uprn
+        - heating-cost-current
+        - co2-emissions-current
+        - hot-water-cost-current
+        - total-floor-area
+        - secondheat-description
+        - environment-impact-current
+        - floor-description
+        - mainheat-energy-eff
+        - current-energy-efficiency
+        - mainheat-env-eff
+        - walls-energy-eff
+        - roof-energy-eff
+        - property-type
+        - mainheat-description
+        - hot-water-env-eff
+        - mechanical-ventilation
+        - floor-level
+        - built-form
+        - walls-description
+        - mainheatcont-description
+        - roof-description
+        - energy-consumption-current
+        - construction-age-band
+        - hotwater-description
+        - lodgement-datetime
+        - main-fuel
+        - hot-water-energy-eff
+        - co2-emiss-curr-per-floor-area
+        - windows-energy-eff
+        - current-energy-rating
+        - lodgement-year
+        - extension-count
+        - number-open-fireplaces
+        - number-heated-rooms
+        - lodgement-date
+        - number-habitable-rooms
+        - windows-description
+        - local-authority
+        - photo-supply
+        - heat-loss-corridor
+        - posttown
+        - address
+        - flat-top-storey
+        - unheated-corridor-length
+        - fixed-lighting-outlets-count
+        - inspection-date
+        - tenure
+        - county
+        - constituency-label
+        - multi-glaze-proportion
+        - solar-water-heating-flag
+        - address2
+        - energy-tariff
+        - floor-height
+        - constituency
+        - uprn-source
+        - transaction-type
+        - floor-energy-eff
+        - postcode
+        - lodgement-month
+        - lighting-cost-current
+        - glazed-area
+        - address1
+        - floor-env-eff
+        - main-heating-controls
         default.feature_processor.feature_processor_config.subsample_amount:
         default.feature_processor.feature_processor_config.subsample_seed: 0
         default.feature_processor.feature_processor_config.target: heating_kwh
@@ -36,8 +103,8 @@ stages:
     outs:
     - path: data/prepared_data/
       hash: md5
-      md5: c162c4b80826f42b33cce656446460f3.dir
-      size: 23784411
+      md5: d1ca07d66c3e28c133d0561423e6d2c8.dir
+      size: 14503223
       nfiles: 2
   build_model:
     cmd: python 2_build_model.py
@@ -48,8 +115,8 @@ stages:
       size: 4820
     - path: data/prepared_data
       hash: md5
-      md5: c162c4b80826f42b33cce656446460f3.dir
-      size: 23784411
+      md5: d1ca07d66c3e28c133d0561423e6d2c8.dir
+      size: 14503223
       nfiles: 2
     params:
       configs/build_model.yaml:
@@ -81,18 +148,18 @@ stages:
     outs:
     - path: data/fit_predictions/
       hash: md5
-      md5: c6925db5c6b2ff0f95e97aed727462a1.dir
-      size: 726994
+      md5: e08a232adc7f805d5d97ed7e93d667b3.dir
+      size: 726970
       nfiles: 1
     - path: data/model/
       hash: md5
-      md5: f8cd16b81139a2ed1f40009204b5bb67.dir
-      size: 557447134
-      nfiles: 36
+      md5: 3daab783532ba88d40eb905ff65b0f1c.dir
+      size: 400927883
+      nfiles: 37
     - path: metrics/fit_metrics.json
       hash: md5
-      md5: 35cc792ba9bda3755928561c512aed3c
-      size: 214
+      md5: 9d6a478739e42b2f5f8500de585e9cf9
+      size: 212
   generate_predictions:
     cmd: python 3_generate_predictions.py
     deps:
@@ -102,13 +169,13 @@ stages:
       size: 2464
     - path: data/model
       hash: md5
-      md5: f8cd16b81139a2ed1f40009204b5bb67.dir
-      size: 557447134
-      nfiles: 36
+      md5: 3daab783532ba88d40eb905ff65b0f1c.dir
+      size: 400927883
+      nfiles: 37
     - path: data/prepared_data
       hash: md5
-      md5: c162c4b80826f42b33cce656446460f3.dir
-      size: 23784411
+      md5: d1ca07d66c3e28c133d0561423e6d2c8.dir
+      size: 14503223
       nfiles: 2
     params:
       configs/settings.yaml:
@@ -120,7 +187,7 @@ stages:
     outs:
     - path: data/predictions/
       hash: md5
-      md5: 4edc571b115ec0b8be1d2689e97ff36f.dir
+      md5: 98a3db098cf2ad9bf786fb77b0ce643f.dir
       size: 77479
       nfiles: 1
   generate_metrics:
@@ -132,13 +199,13 @@ stages:
       size: 3484
     - path: data/predictions
       hash: md5
-      md5: 4edc571b115ec0b8be1d2689e97ff36f.dir
+      md5: 98a3db098cf2ad9bf786fb77b0ce643f.dir
       size: 77479
       nfiles: 1
     - path: data/prepared_data
       hash: md5
-      md5: c162c4b80826f42b33cce656446460f3.dir
-      size: 23784411
+      md5: d1ca07d66c3e28c133d0561423e6d2c8.dir
+      size: 14503223
       nfiles: 2
     params:
       configs/settings.yaml:
@@ -148,8 +215,8 @@ stages:
     outs:
     - path: metrics/metrics.json
       hash: md5
-      md5: 32811206191a4b2f24234d1f94b80b70
-      size: 222
+      md5: 41ccaa41fd34009602d0df571e6453e9
+      size: 219
   generate_scenerio_metrics:
     cmd: python 5_generate_scenarios.py
     deps: