StackedML
Practice
Labs
Questions
Models
Pricing
Sign in
Questions
/
Model Evaluation & Experimentation
/
Metrics (Classification)
/
Calibration
← Previous
Next →
173.
Definition of Calibration
easy
What does it mean for a classifier to be well-calibrated?
A
It means predicted probabilities match observed outcome frequencies — samples predicted at 0.7 are positive 70% of the time
B
It means the model achieves equal precision and recall at the optimal classification threshold for the given evaluation setting
C
It means the model produces consistent predictions across different random seeds and initializations
D
It means the model's predicted probabilities sum to 1.0 across all classes for every input sample
Sign in to verify your answer
← Back to Questions