C2W1_Assignment Exercise 7

Course Q&A Machine Learning Engineering for Production(MLOps) Machine Learning Data Lifecycle in Production
1

According to the grader, I’ve completed Exercise 7 correctly. However, I don’t have any anomalies, although the instructions say I should.

You should see that `metformin-rosiglitazone`, `metformin-pioglitazone`, `payer_code` and medical_specialty` features have an anomaly (i.e. Unexpected string values) which is less than 1%.

Should I have created a new schema for the serving data?

2

By the end of exercise 7, you should have 3 anomalies for serving stats based on the updated schema which relaxes constraints for payer_code and medical_specialty. They are metformin-pioglitazone, metformin-rosiglitazone, readmitted.

3

Hi Balaji. Yes, that’s what the instructions say. But that’s not what I got. And the grader marked my assignment as correct.

4

Please click my name and message your notebook as an attachment.

5

Please use statistics parameter in calculate_and_display_anomalies correctly.

1 Like
6

Thanks for the hint. I figured it out.

I changed the call as you suggested, but the error persisted. It turns out that the error was in the function definition. I had hard coded eval_stats in the body of calculate_and_display_anomalies

It was odd that I was still able to pass the assignment with this error.

2 Likes