ML Made Simple

Exercise Part 5: Deploy a Model as a Service in Azure ML

Now that you’ve trained some models, it’s showtime! Let’s deploy the best one as a service that client applications can use. Think of it as setting up a lemonade stand, but for predictions!

---

Step 1: Deployment Options

• ACI vs. AKS: You can deploy as an Azure Container Instances (ACI) for testing or an Azure Kubernetes Service (AKS) cluster for production. We’ll use ACI for this exercise. It’s like choosing a bike for a leisurely ride instead of a race car.

---

Step 2: Deploying Your Model

1. Automated ML Page: In Azure Machine Learning studio, go to your Automated ML experiment’s Details tab.
2. Deploy the Best Model:
   • Service Name: predict-rentals
   • Description: Predict cycle rentals
   • Compute Type: Azure Container Instance
   • Enable Authentication: Selected
3. Deployment Process: After clicking Deploy, wait for the status to change from Running to Successful. You might need to hit Refresh now and then.

---

Step 3: Getting Endpoint Details

• Endpoints Page: In Azure Machine Learning studio, go to the Endpoints page.
• Consume Tab: Find the predict-rentals real-time endpoint and note the REST endpoint and Primary Key. These are like the secret codes to access your service.

---

Step 4: Testing the Deployed Service

1. Open a New Notebook:
   • In a new Azure Machine Learning studio tab, go to Notebooks (under Author).
   • Create a new file named Test-Bikes.ipynb in your user folder.
   • Make sure your compute instance is running.
2. Setup the Test Environment:
   • Collapse the file explorer to focus on your notebook.
   • Copy the required code block. Remember to select all, including endpoints and brackets.
3. Add the Test Code:
   • Paste the provided Python code into the notebook.
   • This code will use your deployed service to predict cycle rentals for a hypothetical five-day period.
4. Insert Endpoint and Key:
   • Go back to the Consume page and copy the REST endpoint and Primary Key.
   • Paste them into the notebook, replacing YOUR_ENDPOINT and YOUR_KEY.
5. Run the Code:
   • Save the notebook.
   • Click the run button next to the code cell.
   • Check the predicted number of rentals for each day.

Python Code:

endpoint = 'YOUR_ENDPOINT' #Replace with your endpoint 



key = 'YOUR_KEY' #Replace with your key 



 



import json 



import requests 



 



#An array of features based on five-day weather forecast 



x = [[1,1,2022,1,0,6,0,2,0.344167,0.363625,0.805833,0.160446], 



    [2,1,2022,1,0,0,0,2,0.363478,0.353739,0.696087,0.248539], 



    [3,1,2022,1,0,1,1,1,0.196364,0.189405,0.437273,0.248309], 



    [4,1,2022,1,0,2,1,1,0.2,0.212122,0.590435,0.160296], 



    [5,1,2022,1,0,3,1,1,0.226957,0.22927,0.436957,0.1869]] 



 



#Convert the array to JSON format 



input_json = json.dumps({"data": x}) 



 



#Set the content type and authentication for the request 



headers = {"Content-Type":"application/json", 



        "Authorization":"Bearer " + key} 





#Send the request 



response = requests.post(endpoint, input_json, headers=headers) 





#If we got a valid response, display the predictions 



if response.status_code == 200: 



    y = json.loads(response.json()) 



    print("Predictions:") 



    for i in range(len(x)): 



        print (" Day: {}. Predicted rentals: {}".format(i+1, max(0, round(y["result"][i])))) 



else: 



    print(response) 

---

Important Note:

• Ensure you copy the entire code block accurately to avoid errors.

---

And there you have it! You’ve successfully deployed a model as a service and tested it. You’re now a bona fide Azure ML service provider!