ML development risks · Question 41 of 60
A team suspects that a framework's metric calculation delivers wrong F1 values. Which mitigations fit this risk?
2 answers are correct. Hard
Show answer
Answer
- B Cross-checking the results, for example against manually calculated benchmarks
- D A review of the framework's evaluation code
Explanation
Here the framework calculates the evaluation incorrectly. What helps is looking at the evaluation code in a review and cross-checking the results, for example with a comparison measurement calculated by hand.