471. Log-Scale Grid Search
medium
A practitioner uses grid search on a log-scale for learning rate: {0.001, 0.01, 0.1}. Why is a log-scale more appropriate than a linear scale?
A practitioner uses grid search on a log-scale for learning rate: {0.001, 0.01, 0.1}. Why is a log-scale more appropriate than a linear scale?