Check your understanding of this module before moving on.
?
0/3
Checkpoint quiz 3 questions
A quick check on this module's key ideas before you move on
Question 01
Why is "the code ran without errors" not sufficient evidence that a result is correct?
B is correct. Several of the most damaging ML bugs, leakage especially, never raise an exception. They just quietly produce a number that looks fine but isn't trustworthy.
Question 02
According to the recommended debugging order, what should you check before considering a model or algorithm change?
B is correct. Most real debugging sessions end with a data or preprocessing issue, not a model issue. Checking the cheaper, more common causes first saves time.
Question 03
Why can a model with a good test-set score still fail in production?
B is correct. A test-set score is a statement about the past. If the real-world data distribution shifts after that evaluation, the model can degrade without any single test ever having been wrong.