Uses of Class
org.deeplearning4j.parallelism.ParallelWrapper.Builder

Packages that use ParallelWrapper.Builder

Package	Description
org.deeplearning4j.parallelism

Uses of ParallelWrapper.Builder in org.deeplearning4j.parallelism

Methods in org.deeplearning4j.parallelism that return ParallelWrapper.Builder

Modifier and Type	Method	Description
ParallelWrapper.Builder	ParallelWrapper.Builder.averageUpdaters(boolean reallyAverage)	This method enables/disables updaters averaging.
ParallelWrapper.Builder	ParallelWrapper.Builder.averagingFrequency(int freq)	Model averaging frequency.
ParallelWrapper.Builder	ParallelWrapper.Builder.gradientsAccumulator(@NonNull org.deeplearning4j.optimize.solvers.accumulation.GradientsAccumulator accumulator)	This method allows you to specify GradientsAccumulator instance to be used in this ParallelWrapper instance PLEASE NOTE: This method is applicable only to gradients sharing mechanics.
ParallelWrapper.Builder	ParallelWrapper.Builder.modelParamsSupplier(org.nd4j.common.function.Supplier<org.nd4j.linalg.api.ndarray.INDArray> supplier)	This method attaches supplier that'll probably provide model params update PLEASE NOTE: This method is mostly used in Spark environment as part of fault tolerance logic
ParallelWrapper.Builder	ParallelWrapper.Builder.prefetchBuffer(int size)	Size of prefetch buffer that will be used for background data prefetching.
ParallelWrapper.Builder	ParallelWrapper.Builder.reportScoreAfterAveraging(boolean reallyReport)	This method enables/disables averaged model score reporting
ParallelWrapper.Builder	ParallelWrapper.Builder.residualPostProcessor(org.deeplearning4j.optimize.solvers.accumulation.encoding.ResidualPostProcessor residualPostProcessor)	Set the residual post processor algorithm.
ParallelWrapper.Builder	ParallelWrapper.Builder.temporaryMemory(@NonNull Long numBytes)	This method allows to define amount of temporary memory that will be used for gradients sharing.
ParallelWrapper.Builder	ParallelWrapper.Builder.thresholdAlgorithm(org.deeplearning4j.optimize.solvers.accumulation.encoding.ThresholdAlgorithm thresholdAlgorithm)	Set the threshold algorithm.
ParallelWrapper.Builder	ParallelWrapper.Builder.trainerContextArgs(Object... trainerContextArgs)	Transer context args are for calling a TrainerContext init method when ParallelWrapper starts training
ParallelWrapper.Builder	ParallelWrapper.Builder.trainerFactory(@NonNull TrainerContext trainerContext)	Specify a TrainerContext for the given ParallelWrapper instance.
ParallelWrapper.Builder	ParallelWrapper.Builder.trainingMode(@NonNull ParallelWrapper.TrainingMode mode)	This method allows you to specify training mode for this instance of PW. 1) AVERAGING - stands for parameters averaging.
ParallelWrapper.Builder	ParallelWrapper.Builder.updaterParamsSupplier(org.nd4j.common.function.Supplier<org.nd4j.linalg.api.ndarray.INDArray> supplier)	This method attaches supplier that'll probably provide updater params update PLEASE NOTE: This method is mostly used in Spark environment as part of fault tolerance logic
ParallelWrapper.Builder	ParallelWrapper.Builder.workers(int num)	This method allows to configure number of workers that'll be used for parallel training
ParallelWrapper.Builder	ParallelWrapper.Builder.workspaceMode(@NonNull org.deeplearning4j.nn.conf.WorkspaceMode mode)	This method allows to override model's WorkspaceMode configuration option