Merge branch 'master' into master

.cloud/.azure/test/test_aci_deploy.json

@@ -0,0 +1,8 @@
+{
+    "name": "TESTSERVICENAME",
+    "inference_source_directory": "tests/deploy/",
+    "inference_entry_script": "score.py",
+    "conda_file": "environment.yml",
+    "test_enabled": false,
+    "delete_service_after_deployment": true
+}

.cloud/.azure/test/test_aks_deploy.json

@@ -0,0 +1,8 @@
+{
+    "name": "TESTSERVICENAME",
+    "deployment_compute_target": "aks-intTest",
+    "inference_source_directory": "tests/deploy/",
+    "inference_entry_script": "score.py",
+    "conda_file": "environment.yml",
+    "delete_service_after_deployment": true
+}

.cloud/.azure/test/test_workspace.json

@@ -0,0 +1,4 @@
+{
+    "name": "gha_and_aml_workspace",
+    "resource_group": "gha_and_aml_rg"
+}

.github/workflows/integration.yml

@@ -1,5 +1,5 @@
 name: Integration Test
-on: [push]
+on: [push , pull_request]
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -8,9 +8,38 @@ jobs:
         id: checkout_repository
         uses: actions/checkout@master
       
-      - name: Self test
-        id: selftest
-        uses: azure/aml-deploy@master
+      # Connect to existing workspace
+      - name: Create Workspace
+        id: ws_create
+        uses: Azure/aml-workspace@v1
         with:
           azure_credentials: ${{ secrets.AZURE_CREDENTIALS }}
-        continue-on-error: true
+          parameters_file: "test/test_workspace.json"
+      
+      - name: Update Service Name Bash for ACI
+        run: |
+          sed -i -e 's/TESTSERVICENAME/'aci$RANDOM'/g' .cloud/.azure/test/test_aci_deploy.json
+      
+      # Deploy model in Azure Machine Learning to ACI
+      - name: Deploy model to ACI
+        id: aml_deploy_aci
+        uses: ./
+        with:
+            azure_credentials: ${{ secrets.AZURE_CREDENTIALS }}
+            model_name:  mymodel
+            model_version: 1
+            parameters_file: "test/test_aci_deploy.json"
+      
+      - name: Update Service Name Bash for AKS
+        run: |
+          sed -i -e 's/TESTSERVICENAME/'aks$RANDOM'/g' .cloud/.azure/test/test_aks_deploy.json
+      
+      # Deploy model in Azure Machine Learning to AKS
+      - name: Deploy model to AKS
+        id: aml_deploy_aks
+        uses: ./
+        with:
+            azure_credentials: ${{ secrets.AZURE_CREDENTIALS }}
+            model_name:  mymodel
+            model_version: 1
+            parameters_file: "test/test_aks_deploy.json"

README.md

@@ -3,7 +3,6 @@
 
 # GitHub Action for deploying Machine Learning Models to Azure
 
-
 ## Usage
 
 The Deploy Machine Learning Models to Azure action will deploy your model on [Azure Machine Learning](https://azure.microsoft.com/en-us/services/machine-learning/) using GitHub Actions.

code/main.py

@@ -40,7 +40,7 @@ def main():
         azure_credentials = json.loads(azure_credentials)
     except JSONDecodeError:
         print("::error::Please paste output of `az ad sp create-for-rbac --name <your-sp-name> --role contributor --scopes /subscriptions/<your-subscriptionId>/resourceGroups/<your-rg> --sdk-auth` as value of secret variable: AZURE_CREDENTIALS")
-        raise AMLConfigurationException(f"Incorrect or poorly formed output from azure credentials saved in AZURE_CREDENTIALS secret. See setup in https://github.com/Azure/aml-compute/blob/master/README.md")
+        raise AMLConfigurationException("Incorrect or poorly formed output from azure credentials saved in AZURE_CREDENTIALS secret. See setup in https://github.com/Azure/aml-compute/blob/master/README.md")
 
     # Checking provided parameters
     print("::debug::Checking provided parameters")

tests/deploy/environment.yml

@@ -0,0 +1,15 @@
+# Conda environment specification. Details about the Conda environment file format:
+# https://conda.io/docs/user-guide/tasks/manage-environments.html#create-env-file-manually
+
+name: deployment_env
+dependencies:
+  - python=3.6.2
+  - scikit-learn=0.22.2
+  - numpy=1.18.1
+  - pip:
+    - azureml-defaults==1.1.5
+    - azureml-monitoring==0.1.0a18
+    - inference-schema==1.0.2
+    - inference-schema[numpy-support]==1.0.2
+channels:
+  - conda-forge

tests/deploy/score.py

@@ -0,0 +1,40 @@
+import joblib
+import numpy as np
+
+from azureml.core import Model
+from azureml.monitoring import ModelDataCollector
+from inference_schema.schema_decorators import input_schema, output_schema
+from inference_schema.parameter_types.numpy_parameter_type import NumpyParameterType
+from inference_schema.parameter_types.standard_py_parameter_type import StandardPythonParameterType
+
+
+# The init() method is called once, when the web service starts up.
+# Typically you would deserialize the model file, as shown here using joblib,
+# and store it in a global variable so your run() method can access it later.
+def init():
+    global model
+    global inputs_dc, prediction_dc
+    # The AZUREML_MODEL_DIR environment variable indicates
+    # a directory containing the model file you registered.
+    model_path = Model.get_model_path(model_name="mymodel")
+    model = joblib.load(model_path)
+    inputs_dc = ModelDataCollector("sample-model", designation="inputs", feature_names=["feat1", "feat2", "feat3", "feat4"])
+    prediction_dc = ModelDataCollector("sample-model", designation="predictions", feature_names=["prediction"])
+
+
+# The run() method is called each time a request is made to the scoring API.
+# Shown here are the optional input_schema and output_schema decorators
+# from the inference-schema pip package. Using these decorators on your
+# run() method parses and validates the incoming payload against
+# the example input you provide here. This will also generate a Swagger
+# API document for your web service.
+@input_schema('data', NumpyParameterType(np.array([[0.1, 1.2, 2.3, 3.4]])))
+@output_schema(StandardPythonParameterType({'predict': [['Iris-virginica']]}))
+def run(data):
+    # Use the model object loaded by init().
+    result = model.predict(data)
+    inputs_dc.collect(data)  # this call is saving our input data into Azure Blob
+    prediction_dc.collect(result)  # this call is saving our input data into Azure Blob
+
+    # You can return any JSON-serializable object.
+    return {"predict": result.tolist()}