A B C D E F G H I L M N O P Q R S T U V W Z
All Classes All Packages
All Classes All Packages
All Classes All Packages
A
- accumulator - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- accumulator - Variable in class org.deeplearning4j.parallelism.trainer.SymmetricTrainer
- acquireModel() - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- addInput(INDArray...) - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- addInput(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- addInput(INDArray...) - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- addInput(INDArray[], INDArray[]) - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- addInput(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- addModelHolder(Integer, InplaceParallelInference.ModelHolder) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- addObserver(Observer) - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- averageUpdaters - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- averageUpdaters - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- averageUpdaters(boolean) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method enables/disables updaters averaging.
- AVERAGING - org.deeplearning4j.parallelism.ParallelWrapper.TrainingMode
-
Averaging every X epochs will be applied
- averagingFrequency - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- averagingFrequency - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- averagingFrequency - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- averagingFrequency(int) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Model averaging frequency.
- averagingRequired() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- averagingRequired() - Method in class org.deeplearning4j.parallelism.trainer.SymmetricTrainer
- averagingRequired() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
This method returns TRUE if this Trainer implementation assumes periodic aver
B
- BasicInferenceObservable - Class in org.deeplearning4j.parallelism.inference.observers
- BasicInferenceObservable(INDArray...) - Constructor for class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- BasicInferenceObservable(INDArray[], INDArray[]) - Constructor for class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- BasicInferenceObserver - Class in org.deeplearning4j.parallelism.inference.observers
- BasicInferenceObserver() - Constructor for class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObserver
- BATCHED - org.deeplearning4j.parallelism.inference.InferenceMode
-
input will be included into the batch if computation device is busy, and executed immediately otherwise
- BatchedInferenceObservable - Class in org.deeplearning4j.parallelism.inference.observers
- BatchedInferenceObservable() - Constructor for class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- batchLimit - Variable in class org.deeplearning4j.parallelism.ParallelInference
- batchLimit(int) - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method defines, how many input samples can be batched within given time frame.
- broadcastGradients(SharedGradient) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
This method will propagate gradients across all workers
- build() - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method builds new ParallelInference instance
- build() - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method returns ParallelWrapper instance
- Builder(Model) - Constructor for class org.deeplearning4j.parallelism.ParallelInference.Builder
- Builder(T) - Constructor for class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Build ParallelWrapper for MultiLayerNetwork
C
- checkOutputException() - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- cloneListener(TrainingListener) - Static method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- close() - Method in class org.deeplearning4j.parallelism.ParallelWrapper
- CommunicativeTrainer - Interface in org.deeplearning4j.parallelism.trainer
- configureListeners(String, Collection<TrainingListener>, Collection<TrainingListener>) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- create() - Method in interface org.deeplearning4j.parallelism.main.DataSetIteratorProviderFactory
-
Create an
DataSetIterator
- create() - Method in interface org.deeplearning4j.parallelism.main.MultiDataSetProviderFactory
-
Create an
MultiDataSetIterator
- create(String, int, Model, int, boolean, ParallelWrapper, WorkspaceMode, int) - Method in class org.deeplearning4j.parallelism.factory.DefaultTrainerContext
-
Create a
Trainer
based on the given parameters - create(String, int, Model, int, boolean, ParallelWrapper, WorkspaceMode, int) - Method in class org.deeplearning4j.parallelism.factory.SymmetricTrainerContext
-
Create a
Trainer
based on the given parameters - create(String, int, Model, int, boolean, ParallelWrapper, WorkspaceMode, int) - Method in interface org.deeplearning4j.parallelism.factory.TrainerContext
-
Create a
Trainer
based on the given parameters - CUSTOM - org.deeplearning4j.parallelism.ParallelWrapper.TrainingMode
-
This option assumes use of GradientsAccumulator with any MessageHandler
D
- DataSetIteratorProviderFactory - Interface in org.deeplearning4j.parallelism.main
- debug - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- DEFAULT_BATCH_LIMIT - Static variable in class org.deeplearning4j.parallelism.ParallelInference
- DEFAULT_INFERENCE_MODE - Static variable in class org.deeplearning4j.parallelism.ParallelInference
- DEFAULT_NUM_WORKERS - Static variable in class org.deeplearning4j.parallelism.ParallelInference
- DEFAULT_QUEUE_LIMIT - Static variable in class org.deeplearning4j.parallelism.ParallelInference
- DefaultTrainer - Class in org.deeplearning4j.parallelism.trainer
- DefaultTrainer() - Constructor for class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- DefaultTrainer.DefaultTrainerBuilder - Class in org.deeplearning4j.parallelism.trainer
- DefaultTrainerBuilder() - Constructor for class org.deeplearning4j.parallelism.trainer.DefaultTrainer.DefaultTrainerBuilder
- DefaultTrainerContext - Class in org.deeplearning4j.parallelism.factory
- DefaultTrainerContext() - Constructor for class org.deeplearning4j.parallelism.factory.DefaultTrainerContext
E
- EarlyStoppingParallelTrainer<T extends org.deeplearning4j.nn.api.Model> - Class in org.deeplearning4j.parallelism
- EarlyStoppingParallelTrainer(EarlyStoppingConfiguration<T>, T, DataSetIterator, MultiDataSetIterator, int, int, int) - Constructor for class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- EarlyStoppingParallelTrainer(EarlyStoppingConfiguration<T>, T, DataSetIterator, MultiDataSetIterator, EarlyStoppingListener<T>, int, int, int) - Constructor for class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- EarlyStoppingParallelTrainer(EarlyStoppingConfiguration<T>, T, DataSetIterator, MultiDataSetIterator, EarlyStoppingListener<T>, int, int, int, boolean, boolean) - Constructor for class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- encoderMemory - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- enqueueGradient(SharedGradient) - Method in interface org.deeplearning4j.parallelism.trainer.CommunicativeTrainer
- enqueueGradient(SharedGradient) - Method in class org.deeplearning4j.parallelism.trainer.SymmetricTrainer
-
Deprecated.
- esConfig - Variable in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- exception - Variable in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- exception - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- exceptionEncountered - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- executorService - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
F
- feedDataSet(DataSet, long) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Train on a
DataSet
- feedDataSet(DataSet, long) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- feedMultiDataSet(MultiDataSet, long) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- feedMultiDataSet(MultiDataSet, long) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Train on a
MultiDataSet
- FIFO - org.deeplearning4j.parallelism.inference.LoadBalanceMode
-
in this mode we'll be picking free node for next request, blocking if we don't have free nodes at the moment
- finalizeRound(Model, Model...) - Method in class org.deeplearning4j.parallelism.factory.DefaultTrainerContext
- finalizeRound(Model, Model...) - Method in class org.deeplearning4j.parallelism.factory.SymmetricTrainerContext
- finalizeRound(Model, Model...) - Method in interface org.deeplearning4j.parallelism.factory.TrainerContext
-
This method is called at averagingFrequency
- finalizeTraining(Model, Model...) - Method in class org.deeplearning4j.parallelism.factory.DefaultTrainerContext
- finalizeTraining(Model, Model...) - Method in class org.deeplearning4j.parallelism.factory.SymmetricTrainerContext
- finalizeTraining(Model, Model...) - Method in interface org.deeplearning4j.parallelism.factory.TrainerContext
-
This method is called
- fit() - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- fit(DataSetIterator) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
This method takes DataSetIterator, and starts training over it by scheduling DataSets to different executors
- fit(MultiDataSetIterator) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
- fit(DataSet) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- fit(MultiDataSet) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
G
- getCounter() - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- getCurrentModelsFromWorkers() - Method in class org.deeplearning4j.parallelism.InplaceParallelInference
- getCurrentModelsFromWorkers() - Method in class org.deeplearning4j.parallelism.ParallelInference
-
This method returns Models used in workers at this moment PLEASE NOTE: This method is NOT thread safe, and should NOT be used anywhere but tests
- getInputBatches() - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
-
Get input batches - and their associated input mask arrays, if any
Note that usually the returned list will be of size 1 - however, in the batched case, not all inputs can actually be batched (variable size inputs to fully convolutional net, for example). - getInputBatches() - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- getInputBatches() - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- getModel() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- getModel() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
THe current model for the trainer
- getModelForThisThread() - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- getModelForThread(long) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- getOutput() - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- getOutput() - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- getOutput() - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- getOutputs() - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
-
PLEASE NOTE: This method is for tests only
- getTermination() - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- getUuid() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
- getWorkerCounter(int) - Method in class org.deeplearning4j.parallelism.ParallelInference
- gradientsAccumulator - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- gradientsAccumulator(GradientsAccumulator) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
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.
H
- holders - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference
I
- incrementIteration() - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- inferenceMode - Variable in class org.deeplearning4j.parallelism.ParallelInference
- inferenceMode(InferenceMode) - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method allows you to define mode that'll be used during inference.
- InferenceMode - Enum in org.deeplearning4j.parallelism.inference
- InferenceObservable - Interface in org.deeplearning4j.parallelism.inference
- init() - Method in class org.deeplearning4j.parallelism.InplaceParallelInference
- init() - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- init() - Method in class org.deeplearning4j.parallelism.ParallelInference
- init() - Method in class org.deeplearning4j.parallelism.ParallelWrapper
- init(Model, Object...) - Method in class org.deeplearning4j.parallelism.factory.DefaultTrainerContext
-
Initialize the context
- init(Model, Object...) - Method in class org.deeplearning4j.parallelism.factory.SymmetricTrainerContext
-
Initialize the context
- init(Model, Object...) - Method in interface org.deeplearning4j.parallelism.factory.TrainerContext
-
Initialize the context
- INPLACE - org.deeplearning4j.parallelism.inference.InferenceMode
-
Inference will applied in the calling thread instead of workers.
- InplaceParallelInference - Class in org.deeplearning4j.parallelism
- InplaceParallelInference() - Constructor for class org.deeplearning4j.parallelism.InplaceParallelInference
- InplaceParallelInference.ModelHolder - Class in org.deeplearning4j.parallelism
- InplaceParallelInference.ModelSelector - Class in org.deeplearning4j.parallelism
- isCG - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- isLocked() - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- isMLN - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- isMQ - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- isMQ - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- isRunning() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- isRunning() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
- isStopped - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- iterationsCounter - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
L
- lastEtlTime - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- legacyAveraging - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- legacyAveraging - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- listeners - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- loadBalanceMode - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- loadBalanceMode - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- loadBalanceMode - Variable in class org.deeplearning4j.parallelism.ParallelInference.Builder
- loadBalanceMode - Variable in class org.deeplearning4j.parallelism.ParallelInference
- loadBalanceMode(LoadBalanceMode) - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method allows you to specify load balance mode
- LoadBalanceMode - Enum in org.deeplearning4j.parallelism.inference
- locker - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference
M
- main(String[]) - Static method in class org.deeplearning4j.parallelism.main.ParallelWrapperMain
- map - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- model - Variable in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- model - Variable in class org.deeplearning4j.parallelism.ParallelInference
- model - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- model - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- ModelHolder() - Constructor for class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- modelLock - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- modelLock - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- modelParamsSupplier - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- modelParamsSupplier - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- modelParamsSupplier(Supplier<INDArray>) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
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
- ModelSelector() - Constructor for class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- ModelSelector(LoadBalanceMode) - Constructor for class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- MultiDataSetProviderFactory - Interface in org.deeplearning4j.parallelism.main
N
- nanos - Variable in class org.deeplearning4j.parallelism.ParallelInference
- nullDataSet - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- nullMode - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
O
- ObservablesProvider(long, int, BlockingQueue<InferenceObservable>) - Constructor for class org.deeplearning4j.parallelism.ParallelInference.ObservablesProvider
- onRootModel - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- org.deeplearning4j.parallelism - package org.deeplearning4j.parallelism
- org.deeplearning4j.parallelism.factory - package org.deeplearning4j.parallelism.factory
- org.deeplearning4j.parallelism.inference - package org.deeplearning4j.parallelism.inference
- org.deeplearning4j.parallelism.inference.observers - package org.deeplearning4j.parallelism.inference.observers
- org.deeplearning4j.parallelism.main - package org.deeplearning4j.parallelism.main
- org.deeplearning4j.parallelism.trainer - package org.deeplearning4j.parallelism.trainer
- originalModel - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- output(double[]) - Method in class org.deeplearning4j.parallelism.ParallelInference
- output(float[]) - Method in class org.deeplearning4j.parallelism.ParallelInference
- output(ModelAdapter<T>, INDArray...) - Method in class org.deeplearning4j.parallelism.ParallelInference
-
This method does forward pass and returns output provided by OutputAdapter
- output(ModelAdapter<T>, INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.ParallelInference
-
This method does forward pass and returns output provided by OutputAdapter
- output(ModelAdapter<T>, INDArray[], INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference
-
This method does forward pass and returns output provided by OutputAdapter
- output(INDArray) - Method in class org.deeplearning4j.parallelism.ParallelInference
- output(INDArray...) - Method in class org.deeplearning4j.parallelism.ParallelInference
-
Generate predictions/output from the netwonk
- output(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- output(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelSelector
- output(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference
- output(INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.ParallelInference
-
Generate predictions/outputs from the network, optionally using input masks for predictions
- output(INDArray, INDArray) - Method in class org.deeplearning4j.parallelism.ParallelInference
- output(DataSet) - Method in class org.deeplearning4j.parallelism.ParallelInference
P
- ParallelInference - Class in org.deeplearning4j.parallelism
- ParallelInference() - Constructor for class org.deeplearning4j.parallelism.ParallelInference
- ParallelInference.Builder - Class in org.deeplearning4j.parallelism
- ParallelInference.ObservablesProvider - Class in org.deeplearning4j.parallelism
- parallelWrapper - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- ParallelWrapper - Class in org.deeplearning4j.parallelism
- ParallelWrapper(Model, int, int) - Constructor for class org.deeplearning4j.parallelism.ParallelWrapper
- ParallelWrapper.Builder<T extends org.deeplearning4j.nn.api.Model> - Class in org.deeplearning4j.parallelism
- ParallelWrapper.TrainingMode - Enum in org.deeplearning4j.parallelism
- ParallelWrapperMain - Class in org.deeplearning4j.parallelism.main
- ParallelWrapperMain() - Constructor for class org.deeplearning4j.parallelism.main.ParallelWrapperMain
- position - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- postInit() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
-
This method does post-initialization configuration of Model.
- postInit() - Method in class org.deeplearning4j.parallelism.trainer.SymmetricTrainer
- prefetchBuffer(int) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Size of prefetch buffer that will be used for background data prefetching.
- prefetchSize - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- prefetchSize - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- pretrain() - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
Q
- queue - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- queue - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- queueLimit - Variable in class org.deeplearning4j.parallelism.ParallelInference
- queueLimit(int) - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method defines buffer queue size.
- queueMDS - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
R
- releaseModel(Model) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- replicas - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- replicatedModel - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- reportScore - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- reportScore - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- reportScoreAfterAveraging(boolean) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method enables/disables averaged model score reporting
- reset() - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- residualPostProcessor - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- residualPostProcessor(ResidualPostProcessor) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Set the residual post processor algorithm.
- rootDevice - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- ROUND_ROBIN - org.deeplearning4j.parallelism.inference.LoadBalanceMode
-
In this mode, `n+1 % nodes` node will be used for next request
- run() - Method in class org.deeplearning4j.parallelism.main.ParallelWrapperMain
- run() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- runMain(String...) - Method in class org.deeplearning4j.parallelism.main.ParallelWrapperMain
- running - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
S
- selector - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference
- SEQUENTIAL - org.deeplearning4j.parallelism.inference.InferenceMode
-
input will be passed into the model as is
- setCounter(int) - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- setInput(Observer, INDArray) - Method in class org.deeplearning4j.parallelism.ParallelInference.ObservablesProvider
- setInput(Observer, INDArray...) - Method in class org.deeplearning4j.parallelism.ParallelInference.ObservablesProvider
- setInput(Observer, INDArray[], INDArray[]) - Method in class org.deeplearning4j.parallelism.ParallelInference.ObservablesProvider
- setLatestScore(double) - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- setListener(EarlyStoppingListener<T>) - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- setListeners(Collection<TrainingListener>) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
This method allows you to specify trainingListeners for this model.
- setListeners(TrainingListener...) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
This method allows you to specify trainingListeners for this model.
- setListeners(StatsStorageRouter, Collection<? extends TrainingListener>) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
Set the listeners, along with a StatsStorageRouter that the results will be shuffled to (in the case of any listeners that implement the
RoutingIterationListener
interface) - setListeners(StatsStorageRouter, TrainingListener...) - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
Set the listeners, along with a StatsStorageRouter that the results will be shuffled to (in the case of any listeners that implement the
RoutingIterationListener
interface) - setOutputBatches(List<INDArray[]>) - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- setOutputBatches(List<INDArray[]>) - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- setOutputBatches(List<INDArray[]>) - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- setOutputException(Exception) - Method in interface org.deeplearning4j.parallelism.inference.InferenceObservable
- setOutputException(Exception) - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObservable
- setPosition(int) - Method in class org.deeplearning4j.parallelism.inference.observers.BatchedInferenceObservable
- setTermination(boolean) - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- setTerminationReason(IterationTerminationCondition) - Method in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- setUncaughtExceptionHandler(Thread.UncaughtExceptionHandler) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Set the
Thread.UncaughtExceptionHandler
for thisTrainer
- setupIfNeccessary() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- SHARED_GRADIENTS - org.deeplearning4j.parallelism.ParallelWrapper.TrainingMode
-
Models within ParallelWrapper instance will share gradients updates
- shouldStop - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- shouldUpdate - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- shutdown() - Method in class org.deeplearning4j.parallelism.ParallelInference
-
This method gracefully shuts down ParallelInference instance
- shutdown() - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
This method causes all threads used for parallel training to stop
- shutdown() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- shutdown() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Shutdown this worker
- sourceModel - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- start() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Start this trainer
- stop() - Method in class org.deeplearning4j.parallelism.main.ParallelWrapperMain
-
Stop the ParallelWrapper main.
- stopFit - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- stopFit() - Method in class org.deeplearning4j.parallelism.ParallelWrapper
-
Will stop a fit operation from continuing to iterate.
- storageRouter - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- SymmetricTrainer - Class in org.deeplearning4j.parallelism.trainer
- SymmetricTrainer(Model, String, int, WorkspaceMode, ParallelWrapper, boolean) - Constructor for class org.deeplearning4j.parallelism.trainer.SymmetricTrainer
- SymmetricTrainerContext - Class in org.deeplearning4j.parallelism.factory
- SymmetricTrainerContext() - Constructor for class org.deeplearning4j.parallelism.factory.SymmetricTrainerContext
T
- targetDeviceId - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- temporaryMemory(Long) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method allows to define amount of temporary memory that will be used for gradients sharing.
- terminationReason - Variable in class org.deeplearning4j.parallelism.EarlyStoppingParallelTrainer
- threadId - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- thresholdAlgorithm - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- thresholdAlgorithm(ThresholdAlgorithm) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Set the threshold algorithm.
- thrownException - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- Trainer - Interface in org.deeplearning4j.parallelism.trainer
- trainerContext - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- trainerContext - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- TrainerContext - Interface in org.deeplearning4j.parallelism.factory
- trainerContextArgs - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- trainerContextArgs - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- trainerContextArgs(Object...) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Transer context args are for calling a
TrainerContext
init method whenParallelWrapper
starts training - trainerFactory(TrainerContext) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
Specify a
TrainerContext
for the givenParallelWrapper
instance. - trainingMode - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- trainingMode(ParallelWrapper.TrainingMode) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method allows you to specify training mode for this instance of PW.
1) AVERAGING - stands for parameters averaging.
U
- update(Observable, Object) - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObserver
- updateModel(Model) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- updateModel(Model) - Method in class org.deeplearning4j.parallelism.InplaceParallelInference
- updateModel(Model) - Method in class org.deeplearning4j.parallelism.ParallelInference
-
This method allows to update Model used for inference in runtime, without queue reset
- updateModel(Model) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- updateModel(Model) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Update the current
Model
for the worker - updateModelParams(INDArray) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- updateModelParams(INDArray) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
This method updates replicated model params
- updaterParamsSupplier - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- updaterParamsSupplier - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- updaterParamsSupplier(Supplier<INDArray>) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
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
- updateUpdaterParams(INDArray) - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- updateUpdaterParams(INDArray) - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
This method updates updater params of the replicated model
- useMDS - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- uuid - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- uuid - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
V
- valueOf(String) - Static method in enum org.deeplearning4j.parallelism.inference.InferenceMode
-
Returns the enum constant of this type with the specified name.
- valueOf(String) - Static method in enum org.deeplearning4j.parallelism.inference.LoadBalanceMode
-
Returns the enum constant of this type with the specified name.
- valueOf(String) - Static method in enum org.deeplearning4j.parallelism.ParallelWrapper.TrainingMode
-
Returns the enum constant of this type with the specified name.
- values() - Static method in enum org.deeplearning4j.parallelism.inference.InferenceMode
-
Returns an array containing the constants of this enum type, in the order they are declared.
- values() - Static method in enum org.deeplearning4j.parallelism.inference.LoadBalanceMode
-
Returns an array containing the constants of this enum type, in the order they are declared.
- values() - Static method in enum org.deeplearning4j.parallelism.ParallelWrapper.TrainingMode
-
Returns an array containing the constants of this enum type, in the order they are declared.
W
- waitTillDone() - Method in class org.deeplearning4j.parallelism.inference.observers.BasicInferenceObserver
-
FOR DEBUGGING ONLY, TO BE REMOVED BEFORE MERGE
- waitTillRunning() - Method in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- waitTillRunning() - Method in interface org.deeplearning4j.parallelism.trainer.Trainer
-
Block the main thread till the trainer is up and running.
- wasAveraged - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- workerCounter - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- workers - Variable in class org.deeplearning4j.parallelism.InplaceParallelInference.ModelHolder
- workers - Variable in class org.deeplearning4j.parallelism.ParallelInference
- workers - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- workers - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- workers(int) - Method in class org.deeplearning4j.parallelism.ParallelInference.Builder
-
This method defines, how many model copies will be used for inference.
- workers(int) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method allows to configure number of workers that'll be used for parallel training
- workspaceMode - Variable in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
- workspaceMode - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
- workspaceMode - Variable in class org.deeplearning4j.parallelism.trainer.DefaultTrainer
- workspaceMode(WorkspaceMode) - Method in class org.deeplearning4j.parallelism.ParallelWrapper.Builder
-
This method allows to override model's WorkspaceMode configuration option
Z
- zoo - Variable in class org.deeplearning4j.parallelism.ParallelWrapper
All Classes All Packages