Replacing a predictive model

3 Tasks

10 mins

Pega Customer Decision Hub 8.6

Pega Customer Decision Hub 8.8 Pega Customer Decision Hub 8.7

Visible to: All users

Beginner Pega Customer Decision Hub 8.6 English

Scenario

U+ Bank uses Pega Customer Decision Hub™ to personalize the credit card offers that a customer is presented on its website. To customers that are likely to leave the bank soon, the bank makes a proactive retention offer instead of a credit card offer. The bank has recorded historical churn data for its customer base. As a data scientist, you used the historical data to create a churn model using an external machine learning service, and you replace the active model with the new one.

Use the following credentials to log in to the exercise system:

Role	User name	Password
Data scientist	DataScientist	rules

Your assignment consists of the following tasks:

Task 1: Create a validation data set

Use ValidationData to create a data set for model validation.

Task 2: Place the new model in shadow mode

Place the ChurnH20 model in shadow mode in the Predict churn propensity prediction.

Task 3: Promote the new model to the active state

As a data scientist, replace the active model with the candidate model in the prediction.

You must initiate your own Pega instance to complete this Challenge.

Initialization may take up to 5 minutes so please be patient.

• Log In

Challenge Walkthrough

Detailed Tasks

1 Create a validation data set

1. Log in as Data scientist with user name DataScientist and password rules.
2. In the navigation pane on the left, click Intelligence > Prediction Studio > Data > Data sets.
3. In the upper right, click New.
4. Configure the new data set with the following information:
   1. Name: ValidationData
   2. Type: File
   3. Apply to: PegaCRM-Data-Customer
   4. Development branch: No branch
   5. Add to ruleset: PegaCRM-Artifacts
   6. Ruleset version: 01-01-99
      
5. Click Create.
6. In the Data source section, select Embedded file.
7. Download the ValidationData data set and extract the .csv file.
8. Click Upload file.
9. Click Choose File.
10. Select the ValidationData.csv file.
11. In the upper right, click Save.

2 Place the new model in shadow mode

1. Download the ChurnH20 model and extract the ChurnH2O.mojo file.
2. In the navigation pane on the left, click Predictions.
3. Click the Predict churn propensity tile.
   
4. In the Models tab, click the More icon of the Churn model.
   
5. Click Replace model.
6. Ensure that Model is selected, and then click Next.
7. In the Upload tab, click Choose File.
8. Select the ChurnH20.mojo file.
9. Click Next.
10. In the Dataset field, select ValidationData.
11. In the Outcome column field, select Outcome.
    
12. Click Next.
13. In the Model name field, enter ChurnH20.
14. Click Replace.
    Note: Keep monitoring the Status. During the process of replacing the model the status changes to CONFIGURATION IN PROGRESS, VALIDATION IN PROGRESS, and READY FOR REVIEW..

15. When the status of the H20 model changes to Ready for review, click H20 (M-1101).
    
16. Inspect the model comparison.
    
17. In the upper right, click Evaluate.
18. Confirm that Approve new candidate model and start shadowing (recommended) is selected.
19. In the Reason field, add the appropriate information ('The new model outperforms the active model') .
20. Click Save.
    

3 Promote the new model to the active state

1. In the Models tab, click the More icon of the ChurnH20 model, and select Promote model.
2. Click Promote model to confirm that you want to promote the model.
3. Confirm that the ChurnH20 model is now the active model.
   

---

This Challenge is to practice what you learned in the following Module:

• MLOps v1

---