medium10 pts · first answer countsmodel-evaluationcalibration
Accuracy after temperature scaling
You divide a neural network's logits by a learned temperature T = 1.8 before softmax to calibrate it. What happens to top-1 accuracy?
Discussion 0
Join free to post in the discussion.
Related questions
- Big gap between train and validation accuracyMachine Learning Engineer
- Why keep a separate test setMachine Learning Engineer
- How often a row is validated in k-foldMachine Learning Engineer
- Why stratify the foldsMachine Learning Engineer
- Specificity by handMachine Learning Engineer
- The x-axis of an ROC curveMachine Learning Engineer
- MAE of three errorsMachine Learning Engineer
- MAPE by handMachine Learning Engineer