From 51b704972058d3a14c12ac7cfd6dbaf04b03de26 Mon Sep 17 00:00:00 2001
From: Michael Duong <michael123ster@gmail.com>
Date: Tue, 3 Oct 2023 23:46:37 +0000
Subject: [PATCH] add optimised model

---
 modules/ml-pipeline/src/pipeline/2_build_model.py         | 4 +---
 modules/ml-pipeline/src/pipeline/configs/build_model.yaml | 6 +++---
 modules/ml-pipeline/src/pipeline/core/MLModels.py         | 8 ++++++--
 modules/ml-pipeline/src/pipeline/dvc.lock                 | 5 ++---
 modules/ml-pipeline/src/pipeline/metrics/.gitignore       | 2 --
 5 files changed, 12 insertions(+), 13 deletions(-)
 delete mode 100644 modules/ml-pipeline/src/pipeline/metrics/.gitignore

diff --git a/modules/ml-pipeline/src/pipeline/2_build_model.py b/modules/ml-pipeline/src/pipeline/2_build_model.py
index 44d72cd..cae5cfd 100644
--- a/modules/ml-pipeline/src/pipeline/2_build_model.py
+++ b/modules/ml-pipeline/src/pipeline/2_build_model.py
@@ -93,10 +93,8 @@ def build_model(
     logger.info("--- Training model ---")
     logger.info("----------------------")
 
-    model_train_data = train_data.drop(columns=identifier_columns)
-
     model.train_model(
-        data=model_train_data,
+        data=train_data.drop(columns=identifier_columns),
         target=target,
         model_hyperparameters=model_hyperparameters,
     )
diff --git a/modules/ml-pipeline/src/pipeline/configs/build_model.yaml b/modules/ml-pipeline/src/pipeline/configs/build_model.yaml
index a5eb62a..d296e6a 100644
--- a/modules/ml-pipeline/src/pipeline/configs/build_model.yaml
+++ b/modules/ml-pipeline/src/pipeline/configs/build_model.yaml
@@ -1,7 +1,7 @@
 default:
   build_model:
     model_type: AutogluonAutoML
-    model_save_filepath: ./data/model/autogluonmodel/
+    model_save_filepath: ./data/model/optimised/
     fit_metrics_filepath: ./metrics/fit_metrics.json
 
     SKLearnLinearRegression: null
@@ -10,9 +10,9 @@ default:
       kernel: "linear"
 
     AutogluonAutoML:
-      output_filepath: ./data/model/autogluonmodel/
+      output_filepath: ./data/model/allmodels/
       problem_type: regression
       eval_metric: mean_squared_error #mean_absolute_error
-      time_limit: 1000
+      time_limit: 4000
       presets: medium_quality
       excluded_model_types: ['KNN', 'RF']
diff --git a/modules/ml-pipeline/src/pipeline/core/MLModels.py b/modules/ml-pipeline/src/pipeline/core/MLModels.py
index 4d9a9e9..4cf8b08 100644
--- a/modules/ml-pipeline/src/pipeline/core/MLModels.py
+++ b/modules/ml-pipeline/src/pipeline/core/MLModels.py
@@ -165,8 +165,12 @@ class AutogluonAutoML:
         if self.model is None:
             raise KeyError("No model trained/ loaded - unable to save")
 
-        logger.info("In local development mode - no need for s3 client")
-        logger.info("Using AutoGluon Model - Model saving already occured")
+        logger.info(
+            "Using AutoGluon Model - Model saving is using optimised deployment mode"
+        )
+
+        logger.info("Saving optimised model")
+        self.model.clone_for_deployment(str(path))
 
         return str(path)
 
diff --git a/modules/ml-pipeline/src/pipeline/dvc.lock b/modules/ml-pipeline/src/pipeline/dvc.lock
index 1cc260c..51fcfbe 100644
--- a/modules/ml-pipeline/src/pipeline/dvc.lock
+++ b/modules/ml-pipeline/src/pipeline/dvc.lock
@@ -10,7 +10,6 @@ stages:
     params:
       configs/settings.yaml:
         default.feature_processor.feature_processor_config.drop_columns:
-        - UPRN
         - HEAT_DEMAND_CHANGE
         - CARBON_CHANGE
         - RDSAP_CHANGE
@@ -30,8 +29,8 @@ stages:
     outs:
     - path: data/prepared_data/
       hash: md5
-      md5: 951ad046d4fca2b977a314f9520e8235.dir
-      size: 28249626
+      md5: 26f4de635a4c10cd4a07df1ffb4bbb2c.dir
+      size: 33889000
       nfiles: 2
   build_model:
     cmd: python 2_build_model.py
diff --git a/modules/ml-pipeline/src/pipeline/metrics/.gitignore b/modules/ml-pipeline/src/pipeline/metrics/.gitignore
deleted file mode 100644
index e6fbc8d..0000000
--- a/modules/ml-pipeline/src/pipeline/metrics/.gitignore
+++ /dev/null
@@ -1,2 +0,0 @@
-/fit_metrics.json
-/metrics.json