Glossary term
Glossary term
Foundations
The loss function used in binary logistic regression.
Click the icon to see the math.
See Logistic regression: Loss and regularization in Machine Learning Crash Course for more information.
Created for this library
An ad-tech team uses log loss as the primary training metric for its click model because it aligns with the calibrated probability the bidder requires.
A medical screening team reports log loss as a probability-calibration metric alongside accuracy and recall.
A churn modeling team uses log loss during training because the production system uses calibrated probabilities to prioritize outreach.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License