Tools · Question 9 of 12
What can Amazon SageMaker Clarify check before a model is trained?
2 answers are correct. Medium
Show answer
Answer
- B Whether the labels differ strongly between groups
- D Whether groups are unequally represented in the data
Explanation
Before training, Clarify calculates bias metrics on the data, such as an imbalance in group sizes or differences in the share of positive labels. After training it checks predictions per group. Latency, permissions and encryption are checked by other services.