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
What is target leakage?
B is correct. A feature that's only populated, or only meaningful, after the outcome you're predicting has already happened is encoding the answer, not predicting it.
Question 02
Why is a random (non-chronological) train/test split sometimes the wrong choice?
B is correct. Random shuffling can interleave future and past rows, letting a model implicitly learn from data that wouldn't have existed yet at prediction time.
Question 03
Why does this lesson recommend pinning library versions and fixing random seeds?
B is correct. Reproducibility is about being able to verify a result, not about improving it: the same seed and the same library versions should always produce the same output.