Java example source code file (LearningRatePolicy.java)

This example Java source code file (LearningRatePolicy.java) is included in the alvinalexander.com "Java Source Code Warehouse" project. The intent of this project is to help you "Learn Java by Example" ^TM.

Learn more about this Java project at its project page.

Java - Java tags/keywords

exponential, inverse, learningratepolicy, none, poly, schedule, score, sigmoid, step

The LearningRatePolicy.java Java example source code

package org.deeplearning4j.nn.conf;

/**
 * Learning Rate Policy
 *
 * How to decay learning rate during training.
 *
 * <p>None = do not apply decay policy aka fixed in Caffe 

 * <p>Exponential = applies decay rate to the power of the # batches  

 * <p>Inverse = divide learning rate by negative (1 + decay rate * # batches)^power 

 * <p>Poly = polynomial decay that hits 0 when iterations are complete 

 * <p>Sigmoid = sigmoid decay rate 

 * <p>Step = decay rate to the power of the floor (nearest integer) of # of batches by # of steps 

 * <p>Schedule = rate to use at a specific iteration 

 * <p>Score = apply decay when score stops improving 

 */

// TODO provide options using epochs in addition to iterations

public enum LearningRatePolicy {
    None,
    Exponential,
    Inverse,
    Poly,
    Sigmoid,
    Step,
    Schedule,
    Score
}