Merge pull request #200 from Hestia-Homes/main

Fixed docker image, testing new pushing and caching v4 in github actions file

.github/workflows/deploy_sap_model_lambda.yml

@@ -22,10 +22,6 @@ jobs:
           npm install -g serverless
           npm install -g serverless-domain-manager
 
-      - name: Install dependencies
-        run: |
-          python -m pip install --upgrade pip
-
       - name: AWS credentials for dev
         if: github.ref == 'refs/heads/dev'
         uses: aws-actions/configure-aws-credentials@v1
@@ -54,28 +50,15 @@ jobs:
       - name: Setup Docker
         uses: docker/setup-buildx-action@v1
 
-      - name: Cache Docker layers
-        uses: actions/cache@v2
+      # Building and pushing Docker image with caching
+      - name: Build and push Docker image
+        uses: docker/build-push-action@v4
         with:
-          path: /tmp/.buildx-cache
-          key: ${{ runner.os }}-buildx-${{ github.sha }}
-          restore-keys: |
-            ${{ runner.os }}-buildx-
-
-      - name: Build Docker Image
-        run: |
-          cd model_data/simulation_system
-          docker buildx create --use
-          docker buildx build --load --cache-from=type=local,src=/tmp/.buildx-cache --cache-to=type=local,dest=/tmp/.buildx-cache --build-arg BUILDKIT_INLINE_CACHE=1 --tag lambda-sap-prediction-image:${{ github.sha }} -f Dockerfiles/Dockerfile.prediction.lambda .
-
-      - name: Login to ECR
-        run: |
-          aws ecr get-login-password --region eu-west-2 | docker login --username AWS --password-stdin ${{ steps.set_ecr_credentials.outputs.ecr_uri }}
-
-      - name: Tag and Push Docker Image to ECR
-        run: |
-          docker tag lambda-sap-prediction-image:${{ github.sha }} ${{ steps.set_ecr_credentials.outputs.ecr_uri }}:${{ github.sha }}
-          docker push ${{ steps.set_ecr_credentials.outputs.ecr_uri }}:${{ github.sha }}
+          context: ./model_data/simulation_system
+          push: true
+          tags: ${{ steps.set_ecr_credentials.outputs.ecr_uri }}:${{ github.sha }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
 
       - name: Deploy to AWS Lambda via Serverless
         env:

model_data/simulation_system/Dockerfiles/Dockerfile.prediction.lambda

@@ -2,14 +2,14 @@ FROM public.ecr.aws/lambda/python:3.10
 
 # Set the working directory
 WORKDIR ${LAMBDA_TASK_ROOT}/simulation_system
+ENV PYTHONPATH "${PYTHONPATH}:${LAMBDA_TASK_ROOT}/simulation_system"
 
 # Install necessary build tools - required to test locally
 RUN yum install -y gcc python3-devel
 
 # Install python packages
-COPY requirements/predictions/predictions.txt ./simulation_system/requirements.txt
-COPY requirements/predictions/predictions-dev.txt ./simulation_system/requirements.txt
-RUN pip install --no-cache-dir -r ./simulation_system/requirements.txt
+COPY requirements/predictions/predictions.txt ./requirements.txt
+RUN pip install --no-cache-dir -r ./requirements.txt
 
 # Copy the project code to the working directory
 COPY ./core ./core
@@ -19,4 +19,4 @@ COPY ./handlers/predictions_app.py ./predictions_app.py
 COPY ./__init__.py ./__init__.py
 
 # Run off a lambda trigger
-CMD [ "predictions_app.handler" ]
+CMD [ "simulation_system.predictions_app.handler" ]

model_data/simulation_system/handlers/predictions_app.py

@@ -5,12 +5,10 @@ from predictions import prediction
 RUNTIME_ENVIRONMENT = os.environ.get("RUNTIME_ENVIRONMENT", "dev")
 
 
-def handler():
+def handler(event, context):
     """
     Take in event and trigger the prediction pipeline
     """
-    print("HIHIHI")
-    event = {}
 
     # Assuming a file in a bucket landing for now?
     # Assuming we have a model to use
@@ -19,7 +17,6 @@ def handler():
     # key = urllib.parse.unquote_plus(
     #     event["Records"][0]["s3"]["bucket"]["key"], encoding="utf-8"
     # )
-
     payload = event["body"]
     data_path = payload["file_location"]
     property_id = payload["property_id"]
@@ -32,7 +29,7 @@ def handler():
     # model_path = os.environ.get("MODEL_PATH", "http://minio:9000/data/model_directory/")
     model_path = os.environ.get(
         "MODEL_PATH",
-        "s3://retrofit-model-directory-{RUNTIME_ENVIRONMENT}/RDSAP_CHANGE/autogluon/rdsap_change-medium_quality-30"
+        f"s3://retrofit-model-directory-{RUNTIME_ENVIRONMENT}/RDSAP_CHANGE/autogluon/rdsap_change-medium_quality-30"
         "-2023-08-30_11-43-41/deployment/",
     )
 

model_data/simulation_system/predictions.py

@@ -23,6 +23,7 @@ from core.Settings import (
 TIMESTAMP = datetime.now().strftime(TIMESTAMP_FORMAT)
 RUNTIME_ENVIRONMENT = os.environ.get("RUNTIME_ENVIRONMENT", "dev")
 
+
 # FOR TESTING
 # For now just loading data first and then passing into function (i.e. as if we receive json data and convert to
 # DataFrame)
@@ -76,7 +77,7 @@ def prediction(
             exit(1)
     elif RUNTIME_ENVIRONMENT == "dev":
         registry_path = (
-            "s3://retrofit-model-directory-dev/RDSAP_CHANGE/model_registry.csv"
+            "s3://retrofit-model-directory-dev/model_directory/RDSAP_CHANGE/model_registry.csv"
         )
     else:
         raise NotImplemented("TO be implemented")
@@ -108,7 +109,7 @@ def prediction(
     if data_path and data is None:
         logger.info("Loading data from provided path")
         dataloader = dataloader_factory(runtime_environment=RUNTIME_ENVIRONMENT)
-        data = dataloader.load(filepath=data_path, index_col="UPRN")
+        data = dataloader.load(filepath=data_path, index_col="id")
 
         if data is None:
             raise ValueError("No data loaded")
@@ -134,7 +135,7 @@ def prediction(
 
     logger.info("--- Generating Predictions ---")
     prediction = model.generate_predictions(data=data)
-    return pd.concat([data["recommendation_id"], prediction], axis=1)
+    return pd.concat([data["id"], prediction], axis=1)
 
     # Save prediction some where?
     # prediction.to_csv("s3?")
@@ -167,7 +168,8 @@ if __name__ == "__main__":
     args = ingest_arguments()
 
     # Data can be passed in as JSON string: python3 predictions.py --data '{"TOTAL_FLOOR_AREA": 1}'
-    # Data path can be passed as so: python3 predictions.py --data-path ./model_build_data/change_data/rdsap_full/test_data.parquet
+    # Data path can be passed as so: python3 predictions.py --data-path
+    # ./model_build_data/change_data/rdsap_full/test_data.parquet
     prediction(
         target_column=args.target_column,
         model_path=args.model_path,