org.deeplearning4j.parallelism.trainer

Interface Trainer

All Superinterfaces:

Runnable

All Known Subinterfaces:

CommunicativeTrainer

All Known Implementing Classes:

DefaultTrainer, SymmetricTrainer

public interface Trainer
extends Runnable

A Trainer is an individual worker used in ParallelWrapper for handling training in multi core situations.

Author:

Adam Gibson

Method Summary

Modifier and Type	Method and Description
boolean	averagingRequired() This method returns TRUE if this Trainer implementation assumes periodic aver
void	feedDataSet(org.nd4j.linalg.dataset.api.DataSet dataSet, long etlTime) Train on a DataSet
void	feedMultiDataSet(org.nd4j.linalg.dataset.api.MultiDataSet dataSet, long etlTime) Train on a MultiDataSet
org.deeplearning4j.nn.api.Model	getModel() THe current model for the trainer
boolean	isRunning()
void	setUncaughtExceptionHandler(Thread.UncaughtExceptionHandler handler) Set the Thread.UncaughtExceptionHandler for this Trainer
void	shutdown() Shutdown this worker
void	start() Start this trainer
void	updateModel(org.deeplearning4j.nn.api.Model model) Update the current Model for the worker
void	waitTillRunning() Block the main thread till the trainer is up and running.

Methods inherited from interface java.lang.Runnable

run

Method Detail

feedMultiDataSet

void feedMultiDataSet(@NonNull
                      org.nd4j.linalg.dataset.api.MultiDataSet dataSet,
                      long etlTime)

Train on a MultiDataSet

Parameters:

dataSet - the data set to train on

feedDataSet

void feedDataSet(@NonNull
                 org.nd4j.linalg.dataset.api.DataSet dataSet,
                 long etlTime)

Train on a DataSet

Parameters:

dataSet - the data set to train on

getModel

org.deeplearning4j.nn.api.Model getModel()

THe current model for the trainer

Returns:

the current Model for the worker

updateModel

void updateModel(@NonNull
                 org.deeplearning4j.nn.api.Model model)

Update the current Model for the worker

Parameters:

model - the new model for this worker

isRunning

boolean isRunning()

shutdown

void shutdown()

Shutdown this worker

waitTillRunning

void waitTillRunning()

Block the main thread till the trainer is up and running.

setUncaughtExceptionHandler

void setUncaughtExceptionHandler(Thread.UncaughtExceptionHandler handler)

Set the Thread.UncaughtExceptionHandler for this Trainer

Parameters:

handler - the handler for uncaught errors

start

void start()

Start this trainer

averagingRequired

boolean averagingRequired()

This method returns TRUE if this Trainer implementation assumes periodic aver

Returns: