ML development risks · Question 43 of 60
A team finds that some records appear in both the training and the test dataset. Which risk is present, and which mitigation fits?
1 answer is correct. Hard
Show answer
Answer
- C Defective data allocation to the three datasets; data allocation review
Explanation
If data ends up in the wrong pot, for example test data in training, that is an ML development risk. A review of the data split uncovers it. The test data must stay separate from training and validation data.