Data for machine learning · Question 57 of 60
The data is so scarce that no holdout test set is possible; only the average of the cross-validation is reported. How is this value to be assessed?
1 answer is correct. Hard
Show answer
Answer
- A It is optimistically biased and not an unbiased estimate of performance
Explanation
If data is so scarce that no separate test set can be held back, the cross-validation average rates performance too favorably. It is then not suitable as an unbiased estimate.