org.ea.javacnn.trainers

Class Trainer

java.lang.Object

org.ea.javacnn.trainers.Trainer

Direct Known Subclasses:

AdaDeltaTrainer, AdaGradTrainer, AdamTrainer, NesterovTrainer, SGDTrainer, WindowGradTrainer

public abstract class Trainer
extends java.lang.Object

Trainers take the generated output of activations and gradients in order to modify the weights in the network to make a better prediction the next time the network runs with a data block.

Constructor Summary

Constructors

Constructor and Description
Trainer(JavaCNN net, int batch_size, float l2_decay)

Method Summary

Modifier and Type	Method and Description
void	initTrainData(BackPropResult bpr)
TrainResult	train(DataBlock x, int y)
abstract void	update(int i, int j, double gij, double[] p)

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Trainer

public Trainer(JavaCNN net,
               int batch_size,
               float l2_decay)

Method Detail

train

public TrainResult train(DataBlock x,
                         int y)

update

public abstract void update(int i,
                            int j,
                            double gij,
                            double[] p)

initTrainData

public void initTrainData(BackPropResult bpr)