Evaluation metrics and uncertainty · MLD-1216
Temperature scaling
Calibrating model confidence by fitting a scalar temperature on held-out logits.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Reduce overconfidence without changing class ranking.
Original caveat
One scalar cannot correct every form of miscalibration.
Context and aliases
Level: Intermediate. Stages: Evaluation · Research.
Tags: ML · DL · SLM · LLM.
Aliases: None recorded.
Related concepts
Original references
- scikit-learn: cross-validation ↗ · Reference link
- Google Machine Learning Glossary ↗ · Consulted
- scikit-learn: evaluation metrics ↗ · Reference link
- Original reference URL ↗