Skip navigation links
A B C D E F G H I J K L M N O P Q R S T U V W X Z 

A

abs() - Method in interface ai.djl.ndarray.NDArray
Returns the absolute value of this NDArray element-wise.
AbstractAccuracy - Class in ai.djl.training.evaluator
Accuracy is an Evaluator that computes the accuracy score.
AbstractAccuracy() - Constructor for class ai.djl.training.evaluator.AbstractAccuracy
Creates an accuracy evaluator that computes accuracy across axis 1 along the 0th index.
AbstractAccuracy(String, int) - Constructor for class ai.djl.training.evaluator.AbstractAccuracy
Creates an accuracy evaluator that computes accuracy across axis 1 along given index.
AbstractAccuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.AbstractAccuracy
Creates an accuracy evaluator.
AbstractBlock - Class in ai.djl.nn
AbstractBlock is an abstract implementation of Block.
AbstractBlock() - Constructor for class ai.djl.nn.AbstractBlock
 
AbstractCompositeLoss - Class in ai.djl.training.loss
AbstractCompositeLoss is a Loss class that can combine other Losses together to make a larger loss.
AbstractCompositeLoss(String) - Constructor for class ai.djl.training.loss.AbstractCompositeLoss
Constructs a composite loss with the given name.
Accuracy - Class in ai.djl.training.evaluator
Accuracy is the AbstractAccuracy with multiple classes.
Accuracy() - Constructor for class ai.djl.training.evaluator.Accuracy
Creates a multiclass accuracy evaluator that computes accuracy across axis 1 along the 0th index.
Accuracy(String, int) - Constructor for class ai.djl.training.evaluator.Accuracy
Creates a multiclass accuracy evaluator that computes accuracy across axis 1 along given index.
Accuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.Accuracy
Creates a multiclass accuracy evaluator.
accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.AbstractAccuracy
A helper for classes extending AbstractAccuracy.
accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.Accuracy
A helper for classes extending AbstractAccuracy.
accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.BinaryAccuracy
A helper for classes extending AbstractAccuracy.
accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.SingleShotDetectionAccuracy
 
accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.TopKAccuracy
 
acos() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse trigonometric cosine of this NDArray element-wise.
acosh() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse hyperbolic cosine of this NDArray element-wise.
ACTION_RECOGNITION - Static variable in interface ai.djl.Application.CV
 
Activation - Class in ai.djl.nn
Utility class that provides activation functions and blocks.
activation - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
Adam - Class in ai.djl.training.optimizer
Adam is a generalization of the AdaGrad Optimizer.
Adam(Adam.Builder) - Constructor for class ai.djl.training.optimizer.Adam
Creates a new instance of Adam optimizer.
adam() - Static method in class ai.djl.training.optimizer.Optimizer
Returns a new instance of Adam.Builder that can build an Adam optimizer.
Adam.Builder - Class in ai.djl.training.optimizer
The Builder to construct an Adam object.
add(List<String>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Adds the given sentence to the Vocabulary.
add(Number) - Method in interface ai.djl.ndarray.NDArray
Adds a number to this NDArray element-wise.
add(NDArray) - Method in interface ai.djl.ndarray.NDArray
Adds other NDArrays to this NDArray element-wise.
add(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Adds a number to the NDArray element-wise.
add(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Adds a NDArray to a number element-wise.
add(NDArray...) - Static method in class ai.djl.ndarray.NDArrays
Adds a NDArray to a NDArray element-wise.
add(Block) - Method in class ai.djl.nn.ParallelBlock
Adds the given Block to the block, which is one parallel branch.
add(Function<NDList, NDList>) - Method in class ai.djl.nn.ParallelBlock
Adds a LambdaBlock, that applies the given function, to the list of parallel branches.
add(Block) - Method in class ai.djl.nn.SequentialBlock
Adds the given Block to the block to be executed in order.
add(Function<NDList, NDList>) - Method in class ai.djl.nn.SequentialBlock
Adds a LambdaBlock that applies the given function to the sequence of blocks.
add(Hyperparameter<?>) - Method in class ai.djl.training.hyperparameter.param.HpSet
Adds a hyperparameter to the set.
add(Transform) - Method in class ai.djl.translate.Pipeline
Adds the given Transform to the list of transforms to be applied on the input when the transform method is called on this object.
add(int, Transform) - Method in class ai.djl.translate.Pipeline
Adds the given Transform to the list of transforms to be applied on the NDArray at the given index in the input NDList.
add(String, Transform) - Method in class ai.djl.translate.Pipeline
Adds the given Transform to the list of transforms to be applied on the NDArray with the given key as name in the input NDList.
addAccumulator(String) - Method in class ai.djl.training.evaluator.AbstractAccuracy
 
addAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError
 
addAccumulator(String) - Method in class ai.djl.training.evaluator.Evaluator
Adds an accumulator for the results of the evaluation with the given key.
addAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss
 
addAccumulator(String) - Method in class ai.djl.training.loss.Loss
Adds an accumulator for the results of the evaluation with the given key.
addAll(List<List<String>>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Adds the given list of sentences to the Vocabulary.
addAll(NDList) - Method in class ai.djl.ndarray.NDList
Appends all of the NDArrays in the specified NDList to the end of this NDList, in the order that they are returned by the specified NDList's iterator.
addAll(Shape) - Method in class ai.djl.ndarray.types.Shape
Joins a this shape with specified other shape.
addAll(Block...) - Method in class ai.djl.nn.ParallelBlock
Adds an array of blocks, each of which is a parallel branch.
addAll(Collection<Block>) - Method in class ai.djl.nn.ParallelBlock
Adds a Collection of blocks, each of which is a parallel branch.
addAll(Block...) - Method in class ai.djl.nn.SequentialBlock
Adds an array of blocks to be executed in sequence, in order.
addAll(Collection<Block>) - Method in class ai.djl.nn.SequentialBlock
Adds a Collection of blocks to be executed in sequence, in order.
addBooleanIndex(NDArray) - Method in class ai.djl.ndarray.index.NDIndex
Updates the NDIndex by appending a boolean NDArray.
addEvaluator(Evaluator) - Method in class ai.djl.training.DefaultTrainingConfig
Adds an Evaluator that needs to be computed during training.
addi(Number) - Method in interface ai.djl.ndarray.NDArray
Adds a number to this NDArray element-wise in place.
addi(NDArray) - Method in interface ai.djl.ndarray.NDArray
Adds other NDArrays to this NDArray element-wise in place.
addi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Adds a number to the NDArray element-wise in place.
addi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Adds a NDArray to a number element-wise in place.
addi(NDArray...) - Static method in class ai.djl.ndarray.NDArrays
Adds all of the NDArrays together element-wise in place.
addIndices(String) - Method in class ai.djl.ndarray.index.NDIndex
Updates the NDIndex by appending indices to the array.
addIndices(long...) - Method in class ai.djl.ndarray.index.NDIndex
Updates the NDIndex by appending indices as specified values on the NDArray.
addLoss(Loss) - Method in class ai.djl.training.loss.SimpleCompositeLoss
Adds a Loss that applies to all labels and predictions to this composite loss.
addLoss(Loss, int) - Method in class ai.djl.training.loss.SimpleCompositeLoss
Adds a Loss that applies to a single index of the label and predictions to this composite loss.
addMetric(Metric) - Method in class ai.djl.metric.Metrics
Adds a Metric to the collection.
addMetric(String, Number) - Method in class ai.djl.metric.Metrics
Adds a Metric given the metric's name and value.
addMetric(String, Number, String) - Method in class ai.djl.metric.Metrics
Adds a Metric given the metric's name, value, and unit.
addSliceDim(long, long) - Method in class ai.djl.ndarray.index.NDIndex
Appends a new index to slice the dimension and returns a range of values.
addSliceDim(long, long, long) - Method in class ai.djl.ndarray.index.NDIndex
Appends a new index to slice the dimension and returns a range of values.
addTrainingListeners(TrainingListener...) - Method in class ai.djl.training.DefaultTrainingConfig
Adds TrainingListeners for training.
ai.djl - package ai.djl
Contains top level, framework-agnostic classes for both inference and training.
ai.djl.engine - package ai.djl.engine
Contains classes responsible for loading a deep learning framework.
ai.djl.inference - package ai.djl.inference
Contains classes to implement inference tasks.
ai.djl.metric - package ai.djl.metric
Contains classes to collect metrics information.
ai.djl.modality - package ai.djl.modality
Contains utility classes for each of the predefined modalities.
ai.djl.modality.cv - package ai.djl.modality.cv
Contains utility classes for computer vision tasks and image processing.
ai.djl.modality.cv.output - package ai.djl.modality.cv.output
Contains output types used in various computer vision applications.
ai.djl.modality.cv.transform - package ai.djl.modality.cv.transform
Contains Transforms for working with Images.
ai.djl.modality.cv.translator - package ai.djl.modality.cv.translator
Contains translators used for computer vision applications.
ai.djl.modality.cv.util - package ai.djl.modality.cv.util
Contains utility classes for image manipulation.
ai.djl.modality.nlp - package ai.djl.modality.nlp
Contains utility classes for natural language processing tasks.
ai.djl.modality.nlp.preprocess - package ai.djl.modality.nlp.preprocess
Contains utility classes for natural language pre-processing tasks.
ai.djl.modality.nlp.qa - package ai.djl.modality.nlp.qa
Contains utility classes for question and answer processing.
ai.djl.ndarray - package ai.djl.ndarray
Contains classes and interfaces that define an n-dimensional array.
ai.djl.ndarray.index - package ai.djl.ndarray.index
Contains classes that help access NDArray's indices.
ai.djl.ndarray.types - package ai.djl.ndarray.types
Contains classes that define n-dimensional array data types.
ai.djl.nn - package ai.djl.nn
Contains classes to construct neural networks.
ai.djl.nn.convolutional - package ai.djl.nn.convolutional
Contains classes that define convolutional operations extending Convolution.
ai.djl.nn.core - package ai.djl.nn.core
Contains classes that define simple neural network operations.
ai.djl.nn.norm - package ai.djl.nn.norm
Contains classes that define normalizing neural network operations.
ai.djl.nn.pooling - package ai.djl.nn.pooling
Contains pooling neural network operations in Pool and helpers for it.
ai.djl.nn.recurrent - package ai.djl.nn.recurrent
Contains classes for recurrent neural network operations.
ai.djl.training - package ai.djl.training
Contains classes and implementations to train a neural network.
ai.djl.training.dataset - package ai.djl.training.dataset
Contains classes to download and prepare training and testing data.
ai.djl.training.evaluator - package ai.djl.training.evaluator
Contains classes for evaluating the effectiveness of models.
ai.djl.training.hyperparameter.optimizer - package ai.djl.training.hyperparameter.optimizer
Contains classes to optimize Hyperparameters.
ai.djl.training.hyperparameter.param - package ai.djl.training.hyperparameter.param
Contains different types of Hyperparameters.
ai.djl.training.initializer - package ai.djl.training.initializer
Contains classes for initializing a Block's Parameters.
ai.djl.training.listener - package ai.djl.training.listener
Contains classes providing functionality during training through TrainingListener.
ai.djl.training.loss - package ai.djl.training.loss
Contains classes for measuring the Loss of a model.
ai.djl.training.optimizer - package ai.djl.training.optimizer
Contains classes for optimizing a neural network Block.
ai.djl.training.optimizer.learningrate - package ai.djl.training.optimizer.learningrate
Contains classes for having a gradually changing learning rate.
ai.djl.training.util - package ai.djl.training.util
Contains utilities to use during training.
ai.djl.translate - package ai.djl.translate
Contains classes and interfaces that translate between java objects and NDArrays.
all() - Method in interface ai.djl.ndarray.NDArray
Returns true if all elements within this NDArray are non-zero or true.
allClose(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns true if two NDArrays are element-wise equal within a tolerance.
allClose(NDArray, double, double, boolean) - Method in interface ai.djl.ndarray.NDArray
Returns true if two NDArray are element-wise equal within a tolerance.
allClose(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns true if two NDArray are element-wise equal within a tolerance.
allClose(NDArray, NDArray, double, double, boolean) - Static method in class ai.djl.ndarray.NDArrays
Returns true if two NDArray are element-wise equal within a tolerance.
allocateDirect(int) - Method in interface ai.djl.ndarray.NDManager
Allocates a new engine specific direct byte buffer.
any() - Method in interface ai.djl.ndarray.NDArray
Returns true if any of the elements within this NDArray are non-zero or true.
Application - Class in ai.djl
A class contains common deep learning applications.
Application.CV - Interface in ai.djl
The common set of applications for computer vision.
Application.NLP - Interface in ai.djl
The common set of applications for natural language processing.
arange(int) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values starting from 0.
arange(float) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values starting from 0.
arange(int, int) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval with step 1.
arange(float, float) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval with step 1.
arange(int, int, int) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval.
arange(float, float, float) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval.
arange(int, int, int, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval.
arange(float, float, float, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced values within a given interval.
argMax() - Method in interface ai.djl.ndarray.NDArray
Returns the indices of the maximum values into the flattened NDArray.
argMax(int) - Method in interface ai.djl.ndarray.NDArray
Returns the indices of the maximum values along given axis.
argMin() - Method in interface ai.djl.ndarray.NDArray
Returns the indices of the minimum values into the flattened NDArray.
argMin(int) - Method in interface ai.djl.ndarray.NDArray
Returns the indices of the minimum values along given axis.
argSort() - Method in interface ai.djl.ndarray.NDArray
Returns the indices that would sort this NDArray.
argSort(int) - Method in interface ai.djl.ndarray.NDArray
Returns the indices that would sort this NDArray given the axis.
argSort(int, boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the indices that would sort this NDArray given the axis.
ArrayDataset - Class in ai.djl.training.dataset
ArrayDataset is an implementation of RandomAccessDataset that consist entirely of large NDArrays.
ArrayDataset(RandomAccessDataset.BaseBuilder<?>) - Constructor for class ai.djl.training.dataset.ArrayDataset
Creates a new instance of ArrayDataset with the arguments in ArrayDataset.Builder.
ArrayDataset.Builder - Class in ai.djl.training.dataset
The Builder to construct an ArrayDataset.
artifacts - Variable in class ai.djl.BaseModel
 
asDataType(ByteBuffer) - Method in enum ai.djl.ndarray.types.DataType
Converts a ByteBuffer to a buffer for this data type.
asin() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse trigonometric sine of this NDArray element-wise.
asInDevice(Device, boolean) - Method in class ai.djl.ndarray.NDList
Converts all the NDArray in NDList to a different Device.
asinh() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse hyperbolic sine of this NDArray element-wise.
atan() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse trigonometric tangent of this NDArray element-wise.
atanh() - Method in interface ai.djl.ndarray.NDArray
Returns the inverse hyperbolic tangent of this NDArray element-wise.
attach(String, AutoCloseable) - Method in class ai.djl.ndarray.BaseNDManager
Attaches a NDArray or NDManager to this NDManager.
attach(NDManager) - Method in interface ai.djl.ndarray.NDArray
Attaches this NDArray to the specified NDManager.
attach(NDManager) - Method in class ai.djl.ndarray.NDList
Attaches each ndarray in this list to the specified manager.
attach(String, AutoCloseable) - Method in interface ai.djl.ndarray.NDManager
Attaches a NDArray or NDManager to this NDManager.
attachGradient() - Method in interface ai.djl.ndarray.NDArray
Attaches a gradient NDArray to this NDArray and marks it so GradientCollector.backward(NDArray) can compute the gradient with respect to it.
availableSize() - Method in class ai.djl.training.dataset.ArrayDataset
Returns the number of records available to be read in this Dataset.
availableSize() - Method in class ai.djl.training.dataset.RandomAccessDataset
Returns the number of records available to be read in this Dataset.
avgPool(NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Performs Avg Pooling on the input.
avgPool1DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.
avgPool1DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.
avgPool1DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.
avgPool1DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.
avgPool1DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.
avgPool2DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.
avgPool2DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.
avgPool2DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.
avgPool2DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.
avgPool2DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.
avgPool3DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.
avgPool3DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.
avgPool3DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.
avgPool3DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.
avgPool3DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.
axis - Variable in class ai.djl.training.evaluator.AbstractAccuracy
 

B

backward(NDArray) - Method in interface ai.djl.training.GradientCollector
Calculate the gradient w.r.t previously marked variable (head).
BaseBuilder() - Constructor for class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
 
BaseBuilder() - Constructor for class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
BaseBuilder() - Constructor for class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
BaseHpOptimizer - Class in ai.djl.training.hyperparameter.optimizer
A base containing shared implementations for HpOptimizers.
BaseHpOptimizer(HpSet) - Constructor for class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
Constructs a BaseHpOptimizer.
BaseModel - Class in ai.djl
BaseModel is the basic implementation of Model.
BaseModel() - Constructor for class ai.djl.BaseModel
 
BaseNDManager - Class in ai.djl.ndarray
BaseNDManager is the default implementation of NDManager.
BaseNDManager(NDManager, Device) - Constructor for class ai.djl.ndarray.BaseNDManager
 
BasePredictor<I,O> - Class in ai.djl.inference
BasePredictor contains common code for implementations of the Predictor interface.
BasePredictor(Model, Translator<I, O>, boolean) - Constructor for class ai.djl.inference.BasePredictor
Creates a new instance of BasePredictor with the given Model and Translator.
basic() - Static method in interface ai.djl.training.listener.TrainingListener.Defaults
A basic TrainingListener set with minimal recommended functionality.
Batch - Class in ai.djl.training.dataset
A Batch is used to hold multiple items (data and label pairs) from a Dataset.
Batch(NDManager, NDList, NDList, int) - Constructor for class ai.djl.training.dataset.Batch
Creates a new instance of Batch with the given manager, data and labels.
Batch(NDManager, NDList, NDList, int, Batchifier) - Constructor for class ai.djl.training.dataset.Batch
Creates a new instance of Batch with the given manager, data and labels.
Batch(NDManager, NDList, NDList, int, Batchifier, long, long) - Constructor for class ai.djl.training.dataset.Batch
Creates a new instance of Batch with the given manager, data and labels.
BatchData(Batch, Map<Device, NDList>, Map<Device, NDList>) - Constructor for class ai.djl.training.listener.TrainingListener.BatchData
Constructs a new TrainingListener.BatchData.
batchFlattenBlock() - Static method in class ai.djl.nn.Blocks
Creates a Block whose forward function applies the batchFlatten method.
batchFlattenBlock(long) - Static method in class ai.djl.nn.Blocks
Creates a Block whose forward function applies the batchFlatten method.
batchFlattenBlock(long, long) - Static method in class ai.djl.nn.Blocks
Creates a Block whose forward function applies the batchFlatten method.
batchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
batchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Batchifier - Interface in ai.djl.translate
An interface that provides methods to convert an un-batched NDList into a batched NDList and vice versa.
batchify(NDList[]) - Method in interface ai.djl.translate.Batchifier
Converts an array of NDList into an NDList.
batchify(NDList[]) - Method in class ai.djl.translate.StackBatchifier
Converts an array of NDList into an NDList.
BatchNorm - Class in ai.djl.nn.norm
Normalizes a data batch by mean and variance, and applies a scale gamma as well as offset beta.
BatchNorm.Builder - Class in ai.djl.nn.norm
The Builder to construct a BatchNorm.
batchPredict(List<I>) - Method in class ai.djl.inference.BasePredictor
Predicts a batch for inference.
batchPredict(List<I>) - Method in interface ai.djl.inference.Predictor
Predicts a batch for inference.
BatchSampler - Class in ai.djl.training.dataset
BatchSampler is a Sampler that returns a single epoch over the data.
BatchSampler(Sampler.SubSampler, int) - Constructor for class ai.djl.training.dataset.BatchSampler
Creates a new instance of BatchSampler that samples from the given Sampler.SubSampler, and yields a mini-batch of indices.
BatchSampler(Sampler.SubSampler, int, boolean) - Constructor for class ai.djl.training.dataset.BatchSampler
Creates a new instance of BatchSampler that samples from the given Sampler.SubSampler, and yields a mini-batch of indices.
beforeInitialize(Shape[]) - Method in class ai.djl.nn.AbstractBlock
Performs any action necessary before initialization.
beforeInitialize(Shape[]) - Method in class ai.djl.nn.convolutional.Convolution
Performs any action necessary before initialization.
beforeInitialize(Shape[]) - Method in class ai.djl.nn.core.Linear
Performs any action necessary before initialization.
beforeInitialize(Shape[]) - Method in class ai.djl.nn.norm.BatchNorm
Performs any action necessary before initialization.
beforeInitialize(Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell
Performs any action necessary before initialization.
best() - Method in class ai.djl.modality.Classifications
Returns the most likely class for the classification.
bias - Variable in class ai.djl.nn.convolutional.Convolution
 
BinaryAccuracy - Class in ai.djl.training.evaluator
BinaryAccuracy is the AbstractAccuracy with two classes.
BinaryAccuracy(String, float, int, int) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy
Creates a binary (two class) accuracy evaluator.
BinaryAccuracy(String, float, int) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy
Creates a binary (two class) accuracy evaluator that computes accuracy across axis 1 along given index.
BinaryAccuracy(float) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy
Creates a binary (two class) accuracy evaluator that computes accuracy across axis 1 along the 0th index.
BinaryAccuracy() - Constructor for class ai.djl.training.evaluator.BinaryAccuracy
Creates a binary (two class) accuracy evaluator with 0 threshold.
block - Variable in class ai.djl.BaseModel
 
Block - Interface in ai.djl.nn
A Block is a composable function that forms a neural network.
BlockList - Class in ai.djl.nn
Represents a set of names and Blocks.
BlockList() - Constructor for class ai.djl.nn.BlockList
Creates an empty BlockList.
BlockList(int) - Constructor for class ai.djl.nn.BlockList
Constructs an empty BlockList with the specified initial capacity.
BlockList(List<String>, List<Block>) - Constructor for class ai.djl.nn.BlockList
Constructs a BlockList containing the elements of the specified keys and values.
BlockList(List<Pair<String, Block>>) - Constructor for class ai.djl.nn.BlockList
Constructs a BlockList containing the elements of the specified list of Pairs.
BlockList(Map<String, Block>) - Constructor for class ai.djl.nn.BlockList
Constructs a BlockList containing the elements of the specified map.
Blocks - Class in ai.djl.nn
Utility class that provides some useful blocks.
booleanMask(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns portion of this NDArray given the index boolean NDArray along first axis.
booleanMask(NDArray, int) - Method in interface ai.djl.ndarray.NDArray
Returns portion of this NDArray given the index boolean NDArray along given axis.
booleanMask(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns portion of the NDArray given the index boolean NDArray along first axis.
booleanMask(NDArray, NDArray, int) - Static method in class ai.djl.ndarray.NDArrays
Returns portion of the NDArray given the index boolean NDArray along given axis.
BoundingBox - Interface in ai.djl.modality.cv.output
An interface representing a bounding box for the detected object.
BoundingBoxError - Class in ai.djl.training.evaluator
BoundingBoxError is an Evaluator that computes the error in the prediction of bounding boxes in SingleShotDetection model.
BoundingBoxError(String) - Constructor for class ai.djl.training.evaluator.BoundingBoxError
Creates an BoundingBoxError evaluator.
broadcast(Shape) - Method in interface ai.djl.ndarray.NDArray
Broadcasts this NDArray to be the given shape.
broadcast(long...) - Method in interface ai.djl.ndarray.NDArray
Broadcasts this NDArray to be the given shape.
BufferedImageUtils - Class in ai.djl.modality.cv.util
BufferedImageUtils is an image processing utility to load, reshape and convert images using BufferedImage.
build() - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Builds a MultiBoxDetection block.
build() - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Builds a MultiBoxPrior block.
build() - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Builds a MultiBoxTarget block.
build() - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder
Builds the ImageClassificationTranslator with the provided data.
build() - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
Builds the translator.
build() - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder
Builds the translator.
build() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Builds the translator.
build() - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Builds the Vocabulary object with the set arguments.
build() - Method in class ai.djl.nn.convolutional.Conv1D.Builder
Builds a Conv1D block.
build() - Method in class ai.djl.nn.convolutional.Conv2D.Builder
Builds a Conv2D block.
build() - Method in class ai.djl.nn.convolutional.Conv3D.Builder
Builds a Conv3D block.
build() - Method in class ai.djl.nn.core.Embedding.Builder
Builds the Embedding.
build() - Method in class ai.djl.nn.core.Linear.Builder
Returns the constructed Linear.
build() - Method in class ai.djl.nn.norm.BatchNorm.Builder
Builds a BatchNorm block.
build() - Method in class ai.djl.nn.norm.Dropout.Builder
Builds a Dropout block.
build() - Method in class ai.djl.nn.recurrent.GRU.Builder
Builds a GRU block.
build() - Method in class ai.djl.nn.recurrent.LSTM.Builder
Builds a LSTM block.
build() - Method in class ai.djl.nn.recurrent.RNN.Builder
Builds a RNN block.
build() - Method in class ai.djl.training.dataset.ArrayDataset.Builder
Builds a new instance of ArrayDataset with the specified data and labels.
build() - Method in class ai.djl.training.optimizer.Adam.Builder
Builds a Adam block.
build() - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder
Builds a FactorTracker block.
build() - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder
Builds a MultiFactorTracker block.
build() - Method in class ai.djl.training.optimizer.Nag.Builder
Builds a Nag block.
build() - Method in class ai.djl.training.optimizer.Sgd.Builder
Builds a Sgd block.
builder() - Static method in class ai.djl.modality.cv.MultiBoxDetection
Creates a builder to build a MultiBoxDetection.
builder() - Static method in class ai.djl.modality.cv.MultiBoxPrior
Creates a builder to build a MultiBoxPrior.
builder() - Static method in class ai.djl.modality.cv.MultiBoxTarget
Creates a builder to build a MultiBoxTarget.
builder() - Static method in class ai.djl.modality.cv.translator.ImageClassificationTranslator
Creates a builder to build a ImageClassificationTranslator.
builder() - Static method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Creates a builder to build a InstanceSegmentationTranslator.
builder() - Static method in class ai.djl.modality.cv.translator.SimplePoseTranslator
Creates a builder to build a SimplePoseTranslator.
builder() - Static method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
Creates a builder to build a SingleShotDetectionTranslator.
Builder() - Constructor for class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
 
builder() - Static method in class ai.djl.nn.convolutional.Conv1D
Creates a builder to build a Conv1D.
builder() - Static method in class ai.djl.nn.convolutional.Conv2D
Creates a builder to build a Conv2D.
builder() - Static method in class ai.djl.nn.convolutional.Conv3D
Creates a builder to build a Conv3D.
builder() - Static method in class ai.djl.nn.core.Embedding
Creates a builder to build an Embedding.
builder() - Static method in class ai.djl.nn.core.Linear
Creates a builder to build a Linear.
builder() - Static method in class ai.djl.nn.norm.BatchNorm
Creates a builder to build a BatchNorm.
builder() - Static method in class ai.djl.nn.norm.Dropout
Creates a builder to build a Dropout.
builder() - Static method in class ai.djl.nn.recurrent.GRU
Creates a builder to build a GRU.
Builder() - Constructor for class ai.djl.nn.recurrent.GRU.Builder
 
builder() - Static method in class ai.djl.nn.recurrent.LSTM
Creates a builder to build a LSTM.
Builder() - Constructor for class ai.djl.nn.recurrent.LSTM.Builder
 
builder() - Static method in class ai.djl.nn.recurrent.RNN
Creates a builder to build a RNN.
Builder() - Constructor for class ai.djl.nn.recurrent.RNN.Builder
 
Builder() - Constructor for class ai.djl.training.dataset.ArrayDataset.Builder
 
builder() - Static method in class ai.djl.training.optimizer.Adam
Creates a builder to build a Adam.
buildSentence(List<String>) - Method in class ai.djl.modality.nlp.preprocess.SimpleTokenizer
 
buildSentence(List<String>) - Method in interface ai.djl.modality.nlp.preprocess.Tokenizer
Combines a list of tokens to form a sentence.

C

cast(DataType) - Method in interface ai.djl.Model
Casts the model to support a different precision level.
cast(DataType) - Method in class ai.djl.nn.AbstractBlock
Guaranteed to throw an exception.
cast(DataType) - Method in interface ai.djl.nn.Block
Guaranteed to throw an exception.
cbrt() - Method in interface ai.djl.ndarray.NDArray
Returns the cube-root of this NDArray element-wise.
ceil() - Method in interface ai.djl.ndarray.NDArray
Returns the ceiling of this NDArray element-wise.
CenterCrop - Class in ai.djl.modality.cv.transform
A Transform that crops the center of an image.
CenterCrop() - Constructor for class ai.djl.modality.cv.transform.CenterCrop
Creates a CenterCrop Transform that crops to size min(width, height).
CenterCrop(int, int) - Constructor for class ai.djl.modality.cv.transform.CenterCrop
Creates a CenterCrop Transform that crops the given width and height.
centerCrop(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Crops an image to a square of size min(width, height).
centerCrop(NDArray, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Crops an image to a given width and height from the center of the image.
checkConcatInput(NDList) - Static method in class ai.djl.ndarray.NDUtils
Check two criteria of concat input: 1.
checkLabelShapes(NDArray, NDArray, boolean) - Method in class ai.djl.training.evaluator.Evaluator
Checks if the two input NDArray have the same length or shape.
checkLabelShapes(NDArray, NDArray) - Method in class ai.djl.training.evaluator.Evaluator
Checks the length of NDArrays.
classes - Variable in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
 
Classification(String, double) - Constructor for class ai.djl.modality.Classifications.Classification
Constructs a single class result for a classification.
Classifications - Class in ai.djl.modality
Classifications is the container that stores the classification results for classification on a single input.
Classifications(List<String>, List<Double>) - Constructor for class ai.djl.modality.Classifications
Constructs a Classifications using a parallel list of classNames and probabilities.
Classifications(List<String>, NDArray) - Constructor for class ai.djl.modality.Classifications
Constructs a Classifications using list of classNames parallel to an NDArray of probabilities.
Classifications.Classification - Class in ai.djl.modality
A Classification stores the classification result for a single class on a single input.
classNames - Variable in class ai.djl.modality.Classifications
 
clear() - Method in class ai.djl.nn.AbstractBlock
Closes all the parameters of the block.
clear() - Method in interface ai.djl.nn.Block
Closes all the parameters of the block.
clip(Number, Number) - Method in interface ai.djl.ndarray.NDArray
Clips (limit) the values in this NDArray.
clipGrad - Variable in class ai.djl.training.optimizer.Optimizer
 
clipLstmState - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
close() - Method in class ai.djl.inference.BasePredictor
close() - Method in interface ai.djl.inference.Predictor
close() - Method in interface ai.djl.Model
close() - Method in class ai.djl.ndarray.BaseNDManager
close() - Method in interface ai.djl.ndarray.NDArray
close() - Method in class ai.djl.ndarray.NDList
close() - Method in interface ai.djl.ndarray.NDManager
close() - Method in class ai.djl.nn.Parameter
close() - Method in class ai.djl.training.dataset.Batch
close() - Method in interface ai.djl.training.GradientCollector
close() - Method in class ai.djl.training.LocalParameterServer
close() - Method in interface ai.djl.training.ParameterServer
close() - Method in interface ai.djl.training.Trainer
close() - Method in interface ai.djl.translate.TranslatorContext
closed - Variable in class ai.djl.ndarray.BaseNDManager
 
collectMemoryInfo(Metrics) - Static method in class ai.djl.training.listener.MemoryTrainingListener
Collect memory information.
components - Variable in class ai.djl.training.loss.AbstractCompositeLoss
 
concat(NDArray) - Method in interface ai.djl.ndarray.NDArray
Joins a NDArray along the first axis.
concat(NDArray, int) - Method in interface ai.djl.ndarray.NDArray
Joins a NDArray along an existing axis.
concat(NDList) - Static method in class ai.djl.ndarray.NDArrays
Joins a NDList along the first axis.
concat(NDList, int) - Static method in class ai.djl.ndarray.NDArrays
Joins a NDList along an existing axis.
contains(String) - Method in class ai.djl.ndarray.NDList
Returns true if this NDList contains an NDArray with the specified name.
contentEquals(Number) - Method in interface ai.djl.ndarray.NDArray
Returns true if all elements in this NDArray are equal to the Number.
contentEquals(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns true if all elements in this NDArray are equal to the other NDArray.
contentEquals(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns true if all elements in NDArray a are equal to NDArray b.
contentEquals(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns true if all elements in NDArray a are equal to NDArray b.
Conv1D - Class in ai.djl.nn.convolutional
Computes 1-D convolution on 3-D input.
Conv1D.Builder - Class in ai.djl.nn.convolutional
The Builder to construct a Conv1D type of Block.
Conv2D - Class in ai.djl.nn.convolutional
Computes 2-D convolution on 4-D input.
Conv2D.Builder - Class in ai.djl.nn.convolutional
The Builder to construct a Conv2D type of Block.
Conv3D - Class in ai.djl.nn.convolutional
Computes 3-D convolution on 5-D input.
Conv3D.Builder - Class in ai.djl.nn.convolutional
The Builder to construct a Conv3D type of Block.
Convolution - Class in ai.djl.nn.convolutional
Computes N-D convolution on (N+2)-D input.
Convolution(Convolution.ConvolutionBuilder<?>) - Constructor for class ai.djl.nn.convolutional.Convolution
Creates a Convolution object.
Convolution.ConvolutionBuilder<T extends Convolution.ConvolutionBuilder> - Class in ai.djl.nn.convolutional
A builder that can build any Convolution block.
ConvolutionBuilder() - Constructor for class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
copyTo(NDArray) - Method in interface ai.djl.ndarray.NDArray
Deep-copies the current NDArray to the one passed in.
correctInstances - Variable in class ai.djl.training.evaluator.AbstractAccuracy
 
cos() - Method in interface ai.djl.ndarray.NDArray
Returns the trigonometric cosine of this NDArray element-wise.
cosh() - Method in interface ai.djl.ndarray.NDArray
Returns the hyperbolic cosine of this NDArray element-wise.
countNonzero() - Method in interface ai.djl.ndarray.NDArray
Counts the number of non-zero values in this NDArray.
countNonzero(int) - Method in interface ai.djl.ndarray.NDArray
Counts the number of non-zero values in this NDArray along a given axis.
cpu() - Static method in class ai.djl.Device
Returns the default CPU Device.
CPU - Static variable in interface ai.djl.Device.Type
 
create(Shape) - Method in interface ai.djl.ndarray.NDManager
Creates an uninitialized instance of DataType.FLOAT32 NDArray with specified Shape.
create(Number) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(float) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(int) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(double) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(long) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(byte) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(boolean) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a scalar NDArray.
create(float[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(int[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(double[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(long[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(byte[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(boolean[]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 1D NDArray.
create(float[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2D NDArray.
create(int[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2D NDArray.
create(double[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2D NDArray.
create(long[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2-D NDArray.
create(byte[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2-D NDArray.
create(boolean[][]) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a 2-D NDArray.
create(Buffer, Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes a NDArray with specified Shape.
create(Shape, DataType) - Method in interface ai.djl.ndarray.NDManager
Creates an uninitialized instance of NDArray with specified Shape, and DataType.
create(Buffer, Shape, DataType) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and DataType.
create(float[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and float array.
create(int[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and int array.
create(double[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and double array.
create(long[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and long array.
create(byte[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and byte array.
create(boolean[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates and initializes an instance of NDArray with specified Shape and boolean array.
create(Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Creates an uninitialized instance of NDArray with specified Shape, DataType and Device.
createCSR(Buffer, long[], long[], Shape, Device) - Method in interface ai.djl.ndarray.NDManager
Creates a Compressed Sparse Row Storage (CSR) Format Matrix.
createCSR(Buffer, long[], long[], Shape) - Method in interface ai.djl.ndarray.NDManager
Creates a Compressed Sparse Row Storage (CSR) Format Matrix.
createMask(NDIndex) - Method in interface ai.djl.ndarray.NDArray
Returns a mask on whether each element matches the given index.
createMask(Predicate<Number>) - Method in interface ai.djl.ndarray.NDArray
Returns a mask on whether each element matches the given condition.
createRowSparse(Buffer, Shape, long[], Shape, Device) - Method in interface ai.djl.ndarray.NDManager
Stores the matrix in row sparse format.
createRowSparse(Buffer, Shape, long[], Shape) - Method in interface ai.djl.ndarray.NDManager
Stores the matrix in row sparse format.
Crop - Class in ai.djl.modality.cv.transform
A Transform that crops the image to a given location and size.
Crop(int, int, int, int) - Constructor for class ai.djl.modality.cv.transform.Crop
Creates a CenterCrop Transform.
crop(NDArray, int, int, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Crops an image with a given location and size.
CUDA - Static variable in class ai.djl.engine.StandardCapabilities
 
CUDNN - Static variable in class ai.djl.engine.StandardCapabilities
 
cumSum() - Method in interface ai.djl.ndarray.NDArray
Returns the cumulative sum of the elements in the flattened NDArray.
cumSum(int) - Method in interface ai.djl.ndarray.NDArray
Return the cumulative sum of the elements along a given axis.
currentPoint() - Method in interface ai.djl.modality.cv.output.PathIterator
Returns the coordinates and type of the current path segment in the iteration.

D

data - Variable in class ai.djl.training.dataset.ArrayDataset
 
DataDesc - Class in ai.djl.ndarray.types
A data descriptor class that encapsulates information of a NDArray.
DataDesc(Shape) - Constructor for class ai.djl.ndarray.types.DataDesc
Constructs and initializes a DataDesc with specified Shape.
DataDesc(Shape, String) - Constructor for class ai.djl.ndarray.types.DataDesc
Constructs and initializes a DataDesc with specified Shape and name.
DataDesc(Shape, DataType) - Constructor for class ai.djl.ndarray.types.DataDesc
Constructs and initializes a DataDesc with specified Shape and DataType.
DataDesc(Shape, DataType, String) - Constructor for class ai.djl.ndarray.types.DataDesc
Constructs and initializes a DataDesc with specified Shape, DataType, name, Device and SparseFormat.
DataIterable - Class in ai.djl.training.dataset
DataIterable is a data loader that combines Dataset, Batchifier, Pipeline, and Sampler to provide an iterable over the given RandomAccessDataset.
DataIterable(RandomAccessDataset, NDManager, Sampler, Batchifier, Pipeline, Pipeline, ExecutorService, int, Device) - Constructor for class ai.djl.training.dataset.DataIterable
Creates a new instance of DataIterable with the given parameters.
Dataset - Interface in ai.djl.training.dataset
An interface to represent a set of sample data/label pairs to train a model.
Dataset.Usage - Enum in ai.djl.training.dataset
An enum that indicates the mode - training, test or validation.
dataType - Variable in class ai.djl.BaseModel
 
DataType - Enum in ai.djl.ndarray.types
An enum representing the underlying NDArray's data type.
DataType.Format - Enum in ai.djl.ndarray.types
The general data type format categories.
debugDump(int) - Method in class ai.djl.ndarray.BaseNDManager
Prints information about this NDManager and all sub-managers to the console.
debugEnvironment() - Method in class ai.djl.engine.Engine
Logs debug information about the environment for use when debugging environment issues.
decode(DataInputStream) - Static method in class ai.djl.ndarray.types.Shape
Decodes the data in the given DataInputStream and converts it into the corresponding Shape object.
defaultDevice() - Static method in class ai.djl.Device
Returns the default context used in Engine.
defaultIfNull(Device) - Static method in class ai.djl.Device
Returns the given device or the default if it is null.
defaultIfNull(Device, Device) - Static method in class ai.djl.Device
Returns the given device or the passed in default if it is null.
DefaultTrainingConfig - Class in ai.djl.training
DefaultTrainingConfig is an implementation of the TrainingConfig interface.
DefaultTrainingConfig(Loss) - Constructor for class ai.djl.training.DefaultTrainingConfig
Creates an instance of DefaultTrainingConfig with the given Loss.
describeInput() - Method in class ai.djl.BaseModel
Returns the input descriptor of the model.
describeInput() - Method in interface ai.djl.Model
Returns the input descriptor of the model.
describeInput() - Method in class ai.djl.nn.AbstractBlock
Returns a PairList of input names, and shapes.
describeInput() - Method in interface ai.djl.nn.Block
Returns a PairList of input names, and shapes.
describeInput() - Method in class ai.djl.nn.core.Linear
Returns a PairList of input names, and shapes.
describeOutput() - Method in class ai.djl.BaseModel
Returns the output descriptor of the model.
describeOutput() - Method in interface ai.djl.Model
Returns the output descriptor of the model.
detach(String) - Method in class ai.djl.ndarray.BaseNDManager
Detaches a NDArray from this NDManager's lifecycle.
detach() - Method in interface ai.djl.ndarray.NDArray
Detaches the NDArray from current NDManager's lifecycle.
detach() - Method in class ai.djl.ndarray.NDList
Detaches each ndarray in this list from their current managers.
detach(String) - Method in interface ai.djl.ndarray.NDManager
Detaches a NDArray from this NDManager's lifecycle.
DetectedObject(String, double, BoundingBox) - Constructor for class ai.djl.modality.cv.output.DetectedObjects.DetectedObject
Constructs a bounding box with the given data.
DetectedObjects - Class in ai.djl.modality.cv.output
A class representing the detected objects results for a single image in an object detection case.
DetectedObjects(List<String>, List<Double>, List<BoundingBox>) - Constructor for class ai.djl.modality.cv.output.DetectedObjects
Constructs a DetectedObjects, usually during post-processing.
DetectedObjects.DetectedObject - Class in ai.djl.modality.cv.output
A DetectedObject represents a single potential detected Object for an image.
detection(NDList) - Method in class ai.djl.modality.cv.MultiBoxDetection
Converts multi-box detection predictions.
Device - Class in ai.djl
The Device class provides the specified assignment for CPU/GPU processing on the NDArray.
Device(String) - Constructor for class ai.djl.Device
Creates a Device with basic information.
Device(String, int) - Constructor for class ai.djl.Device
Creates a Device with basic information.
device - Variable in class ai.djl.ndarray.BaseNDManager
 
device - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
device - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Device.Type - Interface in ai.djl
Contains device type string constants.
DeviceType - Interface in ai.djl
DeviceType is a class used to map the Device name to their corresponding type number.
dilate - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
dilate - Variable in class ai.djl.nn.convolutional.Convolution
 
dimension() - Method in class ai.djl.ndarray.types.Shape
Returns the number of dimensions of this Shape.
div(Number) - Method in interface ai.djl.ndarray.NDArray
Divides this NDArray by a number element-wise.
div(NDArray) - Method in interface ai.djl.ndarray.NDArray
Divides this NDArray by the other NDArray element-wise.
div(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Divides the NDArray by a number element-wise.
div(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Divides a number by a NDArray element-wise.
div(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Divides a NDArray by a NDArray element-wise.
DivergenceCheckTrainingListener - Class in ai.djl.training.listener
TrainingListener that gives early warning if your training has failed by divergence.
DivergenceCheckTrainingListener() - Constructor for class ai.djl.training.listener.DivergenceCheckTrainingListener
 
divi(Number) - Method in interface ai.djl.ndarray.NDArray
Divides this NDArray by a number element-wise in place.
divi(NDArray) - Method in interface ai.djl.ndarray.NDArray
Divides this NDArray by the other NDArray element-wise in place.
divi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Divides a number by a NDArray element-wise in place.
divi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Divides a number by a NDArray element-wise.
divi(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Divides a NDArray by a NDArray element-wise.
dot(NDArray) - Method in interface ai.djl.ndarray.NDArray
Dot product of this NDArray and the other NDArray.
dot(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Dot product of NDArray a and NDArray b.
download(String, String) - Static method in class ai.djl.training.util.DownloadUtils
Downloads a file from specified url.
download(String, String, Progress) - Static method in class ai.djl.training.util.DownloadUtils
Downloads a file from specified url.
download(URL, Path, Progress) - Static method in class ai.djl.training.util.DownloadUtils
Downloads a file from specified url.
DownloadUtils - Class in ai.djl.training.util
A utility class downloads the file from specified url.
draw(Graphics2D, int, int) - Method in interface ai.djl.modality.cv.output.BoundingBox
Draws the bounding box using the Graphics2D.
draw(Graphics2D, int, int) - Method in class ai.djl.modality.cv.output.Mask
Draws the bounding box using the Graphics2D.
draw(Graphics2D, int, int) - Method in class ai.djl.modality.cv.output.Rectangle
Draws the bounding box using the Graphics2D.
drawBoundingBoxes(BufferedImage, DetectedObjects) - Static method in class ai.djl.modality.cv.ImageVisualization
Draws the bounding boxes on an image.
drawJoints(BufferedImage, Joints) - Static method in class ai.djl.modality.cv.ImageVisualization
Draws all joints of a body on an image.
Dropout - Class in ai.djl.nn.norm
Applies dropout operation to an input array.
Dropout.Builder - Class in ai.djl.nn.norm
The Builder to construct a Dropout type of Block.
dropRate - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
dropRate - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
dumpMemoryInfo(Metrics, String) - Static method in class ai.djl.training.listener.MemoryTrainingListener
Dump memory metrics into log directory.
duplicate() - Method in interface ai.djl.ndarray.NDArray
Returns a copy of this NDArray.

E

elu(NDArray, float) - Static method in class ai.djl.nn.Activation
Applies ELU activation on the input NDArray.
elu(NDList, float) - Static method in class ai.djl.nn.Activation
Applies ELU(Exponential Linear Unit) activation on the input singleton NDList.
eluBlock(float) - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the ELU activation function in its forward function.
embed(NDManager, T[]) - Method in class ai.djl.nn.core.Embedding
Embeds an array of items.
embed(NDManager, T) - Method in class ai.djl.nn.core.Embedding
Embeds an item.
Embedding<T> - Class in ai.djl.nn.core
An Embedding block map a collection of items to 1-Dimensional representative NDArrays.
Embedding(NDArray, List<T>) - Constructor for class ai.djl.nn.core.Embedding
Constructs a pretrained embedding.
Embedding.Builder<T> - Class in ai.djl.nn.core
The Builder to construct a Embedding type of Block.
EmbeddingException - Exception in ai.djl.modality.nlp
Thrown to indicate that there was some error while loading embeddings.
EmbeddingException(String) - Constructor for exception ai.djl.modality.nlp.EmbeddingException
Constructs a new exception with the specified detail message.
EmbeddingException(String, Throwable) - Constructor for exception ai.djl.modality.nlp.EmbeddingException
Constructs a new exception with the specified detail message and cause.
EmbeddingException(Throwable) - Constructor for exception ai.djl.modality.nlp.EmbeddingException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.
embedWord(NDManager, String) - Method in interface ai.djl.modality.nlp.WordEmbedding
Embeds a word.
embedWord(NDArray) - Method in interface ai.djl.modality.nlp.WordEmbedding
Embeds the word after preprocessed using WordEmbedding.preprocessWordToEmbed(NDManager, String).
end() - Method in class ai.djl.training.util.ProgressBar
endEpoch() - Method in interface ai.djl.training.Trainer
Runs the end epoch actions.
Engine - Class in ai.djl.engine
The Engine interface is the base of the provided implementation for DJL.
Engine() - Constructor for class ai.djl.engine.Engine
 
EngineException - Exception in ai.djl.engine
Thrown to indicate that a native error is raised from the underlying Engine.
EngineException(String) - Constructor for exception ai.djl.engine.EngineException
Constructs a new exception with the specified detail message.
EngineException(String, Throwable) - Constructor for exception ai.djl.engine.EngineException
Constructs a new exception with the specified detail message and cause.
EngineException(Throwable) - Constructor for exception ai.djl.engine.EngineException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) (which typically contains the class and detail message of cause).
EngineProvider - Interface in ai.djl.engine
The EngineProvider instance manufactures an Engine instance, which is available in the system.
EpochTrainingListener - Class in ai.djl.training.listener
EpochTrainingListener that tracks epochs.
EpochTrainingListener() - Constructor for class ai.djl.training.listener.EpochTrainingListener
 
eq(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Equals" comparison.
eq(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Equals" comparison.
eq(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Equals" comparison.
eq(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Equals" comparison.
eq(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Equals" comparison.
equals(Object) - Method in class ai.djl.Device
equals(Object) - Method in class ai.djl.ndarray.types.Shape
evaluate(NDList, NDList) - Method in class ai.djl.training.evaluator.AbstractAccuracy
 
evaluate(NDList, NDList) - Method in class ai.djl.training.evaluator.BoundingBoxError
 
evaluate(NDList, NDList) - Method in class ai.djl.training.evaluator.Evaluator
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.HingeLoss
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.L1Loss
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.L2Loss
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss
Calculates the evaluation between the labels and the predictions.
evaluate(NDList, NDList) - Method in class ai.djl.training.loss.SoftmaxCrossEntropyLoss
Calculates the evaluation between the labels and the predictions.
Evaluator - Class in ai.djl.training.evaluator
Base class for all Evaluators that can be used to evaluate the performance of a model.
Evaluator(String) - Constructor for class ai.djl.training.evaluator.Evaluator
Creates an evaluator with abstract update methods.
EvaluatorTrainingListener - Class in ai.djl.training.listener
TrainingListener that records evaluator results.
EvaluatorTrainingListener() - Constructor for class ai.djl.training.listener.EvaluatorTrainingListener
Constructs an EvaluatorTrainingListener that updates the training progress the default frequency.
EvaluatorTrainingListener(int) - Constructor for class ai.djl.training.listener.EvaluatorTrainingListener
Constructs an EvaluatorTrainingListener that updates the training progress the given frequency.
executor - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
executor - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
exp() - Method in interface ai.djl.ndarray.NDArray
Returns the exponential value of this NDArray element-wise.
expandDims(int) - Method in interface ai.djl.ndarray.NDArray
Expands the Shape of a NDArray.
eye(int) - Method in interface ai.djl.ndarray.NDManager
Returns a 2-D array with ones on the diagonal and zeros elsewhere.
eye(int, int) - Method in interface ai.djl.ndarray.NDManager
Returns a 2-D array with ones on the diagonal and zeros elsewhere.
eye(int, int, int) - Method in interface ai.djl.ndarray.NDManager
Returns a 2-D array with ones on the diagonal and zeros elsewhere.
eye(int, int, int, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Returns a 2-D array with ones on the diagonal and zeros elsewhere.

F

FactorTracker - Class in ai.djl.training.optimizer.learningrate
FactorTracker is an implementation of LearningRateTracker which is updated by a multiplicative factor, at a constant interval of steps, until it reaches a specified stop value.
FactorTracker(FactorTracker.Builder) - Constructor for class ai.djl.training.optimizer.learningrate.FactorTracker
Creates a new instance of FactorTracker.
factorTracker() - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker
Returns a new instance of FactorTracker.Builder that can build an FactorTracker.
FactorTracker.Builder - Class in ai.djl.training.optimizer.learningrate
The Builder to construct an FactorTracker object.
FileTranslator<T> - Class in ai.djl.modality.cv.translator
Built-in Translator that provides image pre-processing from file path.
FileTranslator(Translator<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.FileTranslator
Creates a FileTranslator instance.
FileTranslatorFactory<T> - Class in ai.djl.modality.cv.translator
A factory class creates Translator that can process image from a file path.
FileTranslatorFactory(TranslatorFactory<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.FileTranslatorFactory
Creates a FileTranslatorFactory instance.
filterByLayoutType(Predicate<LayoutType>) - Method in class ai.djl.ndarray.types.Shape
Returns only the axes of the Shape whose layout types match the predicate.
finalize() - Method in class ai.djl.BaseModel
fixedLearningRate(float) - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker
Returns a new instance of MultiFactorTracker.
flag - Variable in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
 
flatten() - Method in interface ai.djl.ndarray.NDArray
Flattens this NDArray into a 1-D NDArray in row-major order.
floor() - Method in interface ai.djl.ndarray.NDArray
Returns the floor of this NDArray element-wise.
forward(TranslatorContext, NDList) - Method in class ai.djl.inference.BasePredictor
 
forward(ParameterStore, NDList) - Method in interface ai.djl.nn.Block
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in interface ai.djl.nn.Block
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.convolutional.Convolution
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.core.Embedding
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.core.Linear
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.core.Prelu
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.LambdaBlock
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.norm.BatchNorm
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.norm.Dropout
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.ParallelBlock
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.recurrent.LSTM
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.recurrent.RecurrentCell
Applies the operating function of the block once.
forward(ParameterStore, NDList, PairList<String, Object>) - Method in class ai.djl.nn.SequentialBlock
Applies the operating function of the block once.
forward(NDList) - Method in interface ai.djl.training.Trainer
Applies the forward function of the model once on the given input NDList.
fromBuffer(Buffer) - Static method in enum ai.djl.ndarray.types.DataType
Returns the data type to use for a data buffer.
fromDeviceType(int) - Static method in interface ai.djl.DeviceType
Map device to its type number.
fromFile(Path) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Loads the image from the specified file.
fromUrl(URL) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Loads the image from the specified URL.
fromUrl(String) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Loads the image from the specified URL.
fromValue(char) - Static method in enum ai.djl.ndarray.types.LayoutType
Converts the character to the matching layout type.
fromValue(String) - Static method in enum ai.djl.ndarray.types.LayoutType
Converts each character to the matching layout type.
fromValue(int) - Static method in enum ai.djl.ndarray.types.SparseFormat
Gets the SparseFormat from it's integer value.

G

gates - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
gelu(NDArray) - Static method in class ai.djl.nn.Activation
Applies GELU(Gaussian Error Linear Unit) activation on the input NDArray.
gelu(NDList) - Static method in class ai.djl.nn.Activation
Applies GELU(Gaussian Error Linear Unit) activation on the input singleton NDList.
geluBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the GELU activation function in its forward function.
generateAnchorBoxes(NDArray) - Method in class ai.djl.modality.cv.MultiBoxPrior
Generates the anchorBoxes array in the input's manager and device.
get(String) - Method in class ai.djl.modality.Classifications
Returns the result for a particular class name.
get(int) - Method in class ai.djl.ndarray.index.NDIndex
Returns the index affecting the given dimension.
get(NDIndex) - Method in interface ai.djl.ndarray.NDArray
Returns a partial NDArray.
get(String) - Method in interface ai.djl.ndarray.NDArray
Returns a partial NDArray.
get(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a partial NDArray.
get(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns a partial NDArray.
get(int) - Method in class ai.djl.ndarray.types.Shape
Returns the shape in the given dimension.
get(NDManager, long) - Method in class ai.djl.training.dataset.ArrayDataset
Gets the Record for the given index from the dataset.
get(NDManager, long) - Method in class ai.djl.training.dataset.RandomAccessDataset
Gets the Record for the given index from the dataset.
getAccumulator(String) - Method in class ai.djl.training.evaluator.AbstractAccuracy
 
getAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError
 
getAccumulator(String) - Method in class ai.djl.training.evaluator.Evaluator
Returns the accumulated evaluator value.
getAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss
Returns the accumulated evaluator value.
getAccumulator(String) - Method in class ai.djl.training.loss.Loss
Returns the accumulated evaluator value.
getArray() - Method in class ai.djl.nn.Parameter
Gets the values of this Parameter as an NDArray.
getArtifact(String, Function<InputStream, T>) - Method in class ai.djl.BaseModel
Attempts to load the artifact using the given function and cache it if the specified artifact is not already cached.
getArtifact(String) - Method in class ai.djl.BaseModel
Finds an artifact resource with a given name in the model.
getArtifact(String, Function<InputStream, T>) - Method in interface ai.djl.Model
Attempts to load the artifact using the given function and cache it if the specified artifact is not already cached.
getArtifact(String) - Method in interface ai.djl.Model
Finds an artifact resource with a given name in the model.
getArtifactAsStream(String) - Method in class ai.djl.BaseModel
Finds an artifact resource with a given name in the model.
getArtifactAsStream(String) - Method in interface ai.djl.Model
Finds an artifact resource with a given name in the model.
getArtifactNames() - Method in interface ai.djl.Model
Returns the artifact names associated with the model.
getAsFullSlice(Shape) - Method in class ai.djl.ndarray.index.NDIndex
Returns this index as a full slice if it can be represented as one.
getAttachment(String) - Method in interface ai.djl.translate.TranslatorContext
Returns value of attached key-value pair to context.
getBatch() - Method in class ai.djl.training.listener.TrainingListener.BatchData
Returns the original batch.
getBatchifier() - Method in class ai.djl.translate.NoopTranslator
Gets the Batchifier.
getBatchifier() - Method in interface ai.djl.translate.Translator
Gets the Batchifier.
getBatchSize() - Method in class ai.djl.training.dataset.BatchSampler
Returns the batch size of the Sampler.
getBatchSize() - Method in interface ai.djl.training.dataset.Sampler
Returns the batch size of the Sampler.
getBest() - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
Returns the best hyperparameters and loss.
getBest() - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer
Returns the best hyperparameters and loss.
getBlock() - Method in class ai.djl.BaseModel
Gets the block from the Model.
getBlock() - Method in interface ai.djl.Model
Gets the block from the Model.
getBoolean(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a boolean element from this NDArray.
getBoundingBox() - Method in class ai.djl.modality.cv.output.DetectedObjects.DetectedObject
Returns the BoundingBox of the detected object.
getBounds() - Method in interface ai.djl.modality.cv.output.BoundingBox
Returns the bounding Rectangle of this BoundingBox.
getBounds() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the bounding Rectangle of this BoundingBox.
getByte(long...) - Method in interface ai.djl.ndarray.NDArray
Returns an byte element from this NDArray.
getChildren() - Method in interface ai.djl.nn.Block
Returns a list of all the children of the block.
getChildren() - Method in class ai.djl.nn.ParallelBlock
Returns a list of all the children of the block.
getChildren() - Method in class ai.djl.nn.ParameterBlock
Returns a list of all the children of the block.
getChildren() - Method in class ai.djl.nn.SequentialBlock
Returns a list of all the children of the block.
getClasses() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Get classes.
getClassName() - Method in class ai.djl.modality.Classifications.Classification
Returns the class name.
getComponents() - Method in class ai.djl.training.loss.AbstractCompositeLoss
Returns the component losses that make up the composite loss.
getConfidence() - Method in class ai.djl.modality.cv.output.Joints.Joint
Returns the confidence probability for the joint.
getData() - Method in class ai.djl.training.dataset.Batch
Gets the data of this Batch.
getData(NDManager) - Method in interface ai.djl.training.dataset.Dataset
Fetches an iterator that can iterate through the Dataset.
getData(NDManager) - Method in class ai.djl.training.dataset.RandomAccessDataset
Fetches an iterator that can iterate through the Dataset.
getData(NDManager, Sampler) - Method in class ai.djl.training.dataset.RandomAccessDataset
Fetches an iterator that can iterate through the Dataset with a custom sampler.
getData() - Method in class ai.djl.training.dataset.Record
Gets the data of this Record.
getDataType() - Method in class ai.djl.BaseModel
Returns the standard data type used within the model.
getDataType() - Method in interface ai.djl.Model
Returns the standard data type used within the model.
getDataType() - Method in interface ai.djl.ndarray.NDArray
Returns the DataType of this NDArray.
getDataType() - Method in class ai.djl.ndarray.types.DataDesc
Returns the DataType of the NDArray.
getDevice() - Method in class ai.djl.ndarray.BaseNDManager
Returns the default Device of this NDManager.
getDevice() - Method in interface ai.djl.ndarray.NDArray
Returns the Device of this NDArray.
getDevice() - Method in interface ai.djl.ndarray.NDManager
Returns the default Device of this NDManager.
getDeviceId() - Method in class ai.djl.Device
Returns the deviceId of the Device.
getDevices(int) - Static method in class ai.djl.Device
Returns an array of devices given the maximum number of GPUs to use.
getDevices() - Method in class ai.djl.training.DefaultTrainingConfig
Gets the Device that are available for computation.
getDevices() - Method in interface ai.djl.training.Trainer
Returns the devices used for training.
getDevices() - Method in interface ai.djl.training.TrainingConfig
Gets the Device that are available for computation.
getDeviceType() - Method in class ai.djl.Device
Returns the device type of the Device.
getDirectParameters() - Method in interface ai.djl.nn.Block
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.convolutional.Convolution
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.core.Embedding
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.core.Linear
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.core.Prelu
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.LambdaBlock
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.norm.BatchNorm
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.norm.Dropout
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.ParallelBlock
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.recurrent.RecurrentCell
Returns a list of all the direct parameters of the block.
getDirectParameters() - Method in class ai.djl.nn.SequentialBlock
Returns a list of all the direct parameters of the block.
getDouble(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a double element from this NDArray.
getEmbeddingType() - Method in class ai.djl.nn.core.Embedding.Builder
Returns the embedded type.
getEncoded() - Method in class ai.djl.ndarray.types.Shape
Gets the byte array representation of this Shape for serialization.
getEngine(String) - Static method in class ai.djl.engine.Engine
Returns the Engine with the given name.
getEngine() - Method in interface ai.djl.engine.EngineProvider
Returns the instance of the Engine class EngineProvider should bind to.
getEngine() - Method in interface ai.djl.ndarray.NDManager
Returns the Engine associated with this manager.
getEngineName() - Method in class ai.djl.engine.Engine
Returns the name of the Engine.
getEvaluator(Class<T>) - Method in interface ai.djl.training.Trainer
Gets the Evaluator that is an instance of the given Class.
getEvaluators() - Method in class ai.djl.training.DefaultTrainingConfig
Returns the list of Evaluators that should be computed during training.
getEvaluators() - Method in interface ai.djl.training.Trainer
Gets all Evaluators.
getEvaluators() - Method in interface ai.djl.training.TrainingConfig
Returns the list of Evaluators that should be computed during training.
getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv1D
Returns the expected layout of the input.
getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv2D
Returns the expected layout of the input.
getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv3D
Returns the expected layout of the input.
getExpectedLayout() - Method in class ai.djl.nn.convolutional.Convolution
Returns the expected layout of the input.
getFloat(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a float element from this NDArray.
getFormat() - Method in enum ai.djl.ndarray.types.DataType
Returns the format of the data type.
getGpuCount() - Static method in class ai.djl.Device
Returns the number of GPUs available in the system.
getGpuCount(String) - Static method in class ai.djl.Device
Returns the number of GPUs available with specified framework.
getGradient() - Method in interface ai.djl.ndarray.NDArray
Returns the gradient NDArray attached to this NDArray.
getHeight() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the height of the Rectangle.
getHParam(String) - Method in class ai.djl.training.hyperparameter.param.HpSet
Returns the hyperparameter in the set with the given name.
getId() - Method in class ai.djl.nn.Parameter
Gets the ID of this Parameter.
getImageHeight() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Get resized image height.
getImageWidth() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Get resized image width.
getIndex(String) - Method in class ai.djl.modality.nlp.Vocabulary
Returns the index of the given token.
getIndex() - Method in class ai.djl.ndarray.index.NDIndexBooleans
Returns the mask binary NDArray.
getIndex() - Method in class ai.djl.ndarray.index.NDIndexFixed
Returns the dimension of the index.
getIndices() - Method in class ai.djl.ndarray.index.NDIndex
Returns the indices.
getInitializer() - Method in enum ai.djl.nn.ParameterType
Gets the Initializer of this ParameterType.
getInitializer() - Method in class ai.djl.training.DefaultTrainingConfig
Gets the Initializer to initialize the parameters of the model.
getInitializer() - Method in interface ai.djl.training.TrainingConfig
Gets the Initializer to initialize the parameters of the model.
getInstance() - Static method in class ai.djl.engine.Engine
Returns the default Engine.
getInt(long...) - Method in interface ai.djl.ndarray.NDArray
Returns an int element from this NDArray.
getIoU(BoundingBox) - Method in interface ai.djl.modality.cv.output.BoundingBox
Gets the Intersection over Union (IoU) value between bounding boxes.
getIoU(BoundingBox) - Method in class ai.djl.modality.cv.output.Rectangle
Gets the Intersection over Union (IoU) value between bounding boxes.
getJoints() - Method in class ai.djl.modality.cv.output.Joints
Gets the joints for the image.
getLabels() - Method in class ai.djl.training.dataset.Batch
Gets the labels corresponding to the data of this Batch.
getLabels() - Method in class ai.djl.training.dataset.Record
Gets the labels that correspond to the data of this Record.
getLabels() - Method in class ai.djl.training.listener.TrainingListener.BatchData
Returns the labels for each device.
getLastValidLength() - Method in class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer
Returns the valid length of the sentence that was last served as input to SentenceLengthNormalizer.preprocess(List).
getLayout() - Method in class ai.djl.ndarray.types.Shape
Returns the layout type for each axis in this shape.
getLayoutType(int) - Method in class ai.djl.ndarray.types.Shape
Returns the layout type in the given dimension.
getLeadingOnes() - Method in class ai.djl.ndarray.types.Shape
Returns the number of leading ones in the array shape.
getLong(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a long element from this NDArray.
getLoss(HpSet) - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
Returns the recorded loss.
getLoss(HpSet) - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer
Returns the recorded loss.
getLoss() - Method in interface ai.djl.training.Trainer
Gets the training Loss function of the trainer.
getLossFunction() - Method in class ai.djl.training.DefaultTrainingConfig
Gets the Loss function to compute the loss against.
getLossFunction() - Method in interface ai.djl.training.TrainingConfig
Gets the Loss function to compute the loss against.
getManager() - Method in interface ai.djl.ndarray.NDArray
Returns the NDManager used to create this NDArray.
getManager() - Method in class ai.djl.training.dataset.Batch
Gets the NDManager that is attached to this Batch.
getManager() - Method in interface ai.djl.training.Trainer
Gets the NDManager from the model.
getMax() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the slice max for each axis.
getMax() - Method in class ai.djl.ndarray.index.NDIndexSlice
Returns the end of the range.
getMetric(String) - Method in class ai.djl.metric.Metrics
Returns all Metrics with the specified metric name.
getMetricName() - Method in class ai.djl.metric.Metric
Returns the name of the Metric.
getMetrics() - Method in interface ai.djl.training.Trainer
Returns the Metrics param used for benchmarking.
getMetrics() - Method in interface ai.djl.translate.TranslatorContext
Returns the Metric tool to do benchmark.
getMin() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the slice min for each axis.
getMin() - Method in class ai.djl.ndarray.index.NDIndexSlice
Returns the start of the range.
getModel() - Method in interface ai.djl.training.Trainer
Returns the model used to create this trainer.
getModel() - Method in interface ai.djl.translate.TranslatorContext
Returns the Model object to understand the input/output.
getName() - Method in class ai.djl.BaseModel
Gets the model name.
getName() - Method in interface ai.djl.Model
Gets the model name.
getName() - Method in interface ai.djl.ndarray.NDArray
Returns the name of this NDArray.
getName() - Method in class ai.djl.ndarray.types.DataDesc
Returns the name of the NDArray.
getName() - Method in class ai.djl.nn.Parameter
Gets the name of this Parameter.
getName() - Method in class ai.djl.training.evaluator.Evaluator
Returns the name of this Evaluator.
getName() - Method in class ai.djl.training.hyperparameter.param.Hyperparameter
Returns the name of the hyperparameter.
getNDArrayInternal() - Method in interface ai.djl.ndarray.NDArray
Returns an internal representative of Native NDArray.
getNDManager() - Method in class ai.djl.BaseModel
Gets the NDManager from the model.
getNDManager() - Method in interface ai.djl.Model
Gets the NDManager from the model.
getNDManager() - Method in interface ai.djl.translate.TranslatorContext
Returns the NDManager to create NDArray.
getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker
Fetches the value of the learning rate after the given number of steps/updates.
getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker
Fetches the value of the learning rate after the given number of steps/updates.
getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker
Fetches the value of the learning rate after the given number of steps/updates.
getNumberOfObjects() - Method in class ai.djl.modality.cv.output.DetectedObjects
Returns the number of objects found in an image.
getNumOfBytes() - Method in enum ai.djl.ndarray.types.DataType
Returns the number of bytes for each element.
getOptimizer() - Method in class ai.djl.training.DefaultTrainingConfig
Gets the Optimizer to use during training.
getOptimizer() - Method in interface ai.djl.training.TrainingConfig
Gets the Optimizer to use during training.
getOutput(long, long, long, long) - Method in enum ai.djl.nn.pooling.PoolingConvention
Computes the size of the output in the given input dimension for this PoolingConvention.
getOutputShapes(NDManager, Shape[]) - Method in interface ai.djl.nn.Block
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.convolutional.Convolution
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.Embedding
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.Linear
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.Prelu
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.LambdaBlock
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.norm.BatchNorm
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.norm.Dropout
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.ParallelBlock
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell
Returns the expected output shapes of the block for the specified input shapes.
getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.SequentialBlock
Returns the expected output shapes of the block for the specified input shapes.
getParagraph() - Method in class ai.djl.modality.nlp.qa.QAInput
Gets the resource document that contains the answer.
getParameters() - Method in class ai.djl.nn.AbstractBlock
Returns a list of all the parameters of the block, including the parameters of its children fetched recursively.
getParameters() - Method in interface ai.djl.nn.Block
Returns a list of all the parameters of the block, including the parameters of its children fetched recursively.
getParameterShape(String, Shape[]) - Method in interface ai.djl.nn.Block
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.convolutional.Convolution
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Embedding
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Linear
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Prelu
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.LambdaBlock
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.norm.BatchNorm
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.norm.Dropout
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.ParallelBlock
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParameterShape(String, Shape[]) - Method in class ai.djl.nn.SequentialBlock
Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.
getParentManager() - Method in class ai.djl.ndarray.BaseNDManager
Returns the parent NDManager.
getParentManager() - Method in interface ai.djl.ndarray.NDManager
Returns the parent NDManager.
getPath() - Method in class ai.djl.Application
Returns the repository path of the application.
getPath() - Method in interface ai.djl.modality.cv.output.BoundingBox
Returns an iterator object that iterates along the BoundingBox boundary and provides access to the geometry of the BoundingBox outline.
getPath() - Method in class ai.djl.modality.cv.output.Rectangle
Returns an iterator object that iterates along the BoundingBox boundary and provides access to the geometry of the BoundingBox outline.
getPipeline() - Method in class ai.djl.modality.cv.translator.ImageTranslator
Gets the Pipeline applied to the input.
getPipeline() - Method in interface ai.djl.translate.PreProcessor
Gets the Pipeline applied to the input.
getPoint() - Method in interface ai.djl.modality.cv.output.BoundingBox
Returns the top left point of the bounding box.
getPoint() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the top left point of the bounding box.
getPredictions() - Method in class ai.djl.training.listener.TrainingListener.BatchData
Returns the predictions for each device.
getProbability() - Method in class ai.djl.modality.Classifications.Classification
Returns the probability.
getProbDist() - Method in class ai.djl.modality.cv.output.Mask
Returns the probability for each pixel.
getProgress() - Method in class ai.djl.training.dataset.Batch
Returns the progress of the batch if it is part of some kind of iteration like a dataset iteration.
getProgressTotal() - Method in class ai.djl.training.dataset.Batch
Returns the total or end value for the progress of the batch if it is part of some kind of iteration like a dataset iteration.
getProperty(String) - Method in class ai.djl.BaseModel
Gets the property of the model based on property name.
getProperty(String) - Method in interface ai.djl.Model
Gets the property of the model based on property name.
getQuestion() - Method in class ai.djl.modality.nlp.qa.QAInput
Gets the question for the model.
getRank() - Method in class ai.djl.ndarray.index.NDIndex
Returns the number of dimensions specified in the Index.
getRank() - Method in class ai.djl.ndarray.index.NDIndexAll
Returns the number of dimensions occupied by this index element.
getRank() - Method in class ai.djl.ndarray.index.NDIndexBooleans
Returns the number of dimensions occupied by this index element.
getRank() - Method in interface ai.djl.ndarray.index.NDIndexElement
Returns the number of dimensions occupied by this index element.
getRank() - Method in class ai.djl.ndarray.index.NDIndexFixed
Returns the number of dimensions occupied by this index element.
getRank() - Method in class ai.djl.ndarray.index.NDIndexSlice
Returns the number of dimensions occupied by this index element.
getSampler() - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Gets the Sampler for the dataset.
getScalar(long...) - Method in interface ai.djl.ndarray.NDArray
Returns a scalar NDArray corresponding to a single element.
getSeqLength() - Method in class ai.djl.modality.nlp.qa.QAInput
Gets the sequence length.
getShape() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the slice shape without squeezing.
getShape() - Method in interface ai.djl.ndarray.NDArray
Returns the Shape of this NDArray.
getShape() - Method in class ai.djl.ndarray.types.DataDesc
Returns the Shape of the NDArray.
getShape() - Method in class ai.djl.ndarray.types.Shape
Returns the dimensions of the Shape.
getShapeFromEmptyNDArrayForReductionOp(Shape, int) - Static method in class ai.djl.ndarray.NDUtils
Get Shape of the empty NDArray after applying reduction operations.
getSize() - Method in class ai.djl.training.dataset.Batch
Returns the batchSize.
getSparseFormat() - Method in interface ai.djl.ndarray.NDArray
Returns the SparseFormat of this NDArray.
getSqueezedShape() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the slice shape with squeezing.
getStep() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the slice step for each axis.
getStep() - Method in class ai.djl.ndarray.index.NDIndexSlice
Returns the step between each slice.
getStringLayout() - Method in class ai.djl.nn.convolutional.Conv1D
Returns the string representing the layout of the input.
getStringLayout() - Method in class ai.djl.nn.convolutional.Conv2D
Returns the string representing the layout of the input.
getStringLayout() - Method in class ai.djl.nn.convolutional.Conv3D
Returns the string representing the layout of the input.
getStringLayout() - Method in class ai.djl.nn.convolutional.Convolution
Returns the string representing the layout of the input.
getSynsetArtifactName() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Get symset artifact name.
getThreshold() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Get threshold.
getTimestamp() - Method in class ai.djl.metric.Metric
Returns the timestamp of the Metric.
getToken(int) - Method in class ai.djl.modality.nlp.Vocabulary
Returns the token corresponding to the given index.
getToSqueeze() - Method in class ai.djl.ndarray.index.NDIndexFullSlice
Returns the list of axis to squeeze.
getTrailingOnes() - Method in class ai.djl.ndarray.types.Shape
Returns the number of trailing ones in the array shape.
getTrainingListeners() - Method in class ai.djl.training.DefaultTrainingConfig
Returns the list of TrainingListeners that should be used during training.
getTrainingListeners() - Method in interface ai.djl.training.TrainingConfig
Returns the list of TrainingListeners that should be used during training.
getType() - Method in enum ai.djl.ndarray.types.SparseFormat
Returns the SparseFormat name.
getType() - Method in class ai.djl.nn.Parameter
Gets the type of this Parameter.
getUid() - Method in interface ai.djl.ndarray.NDArray
Returns unique identifier of this NDArray.
getUint8(long...) - Method in interface ai.djl.ndarray.NDArray
Returns an integer element from this NDArray that represent unsigned integer with 8 bits.
getUnit() - Method in class ai.djl.metric.Metric
Returns the unit of the Metric.
getUnknownToken() - Method in class ai.djl.modality.nlp.Vocabulary
Return a String used for unseen or rarely-seen tokens.
getUnknownValueCount() - Method in class ai.djl.ndarray.types.Shape
Return the count of unknown value in this Shape.
getValue() - Method in class ai.djl.metric.Metric
Returns the value of the Metric.
getValue() - Method in enum ai.djl.ndarray.types.LayoutType
Returns the character representation of the layout type.
getValue() - Method in enum ai.djl.ndarray.types.SparseFormat
Returns the integer value of this SparseFormat.
getValue(Parameter, Device) - Method in class ai.djl.training.ParameterStore
Returns the value of a mirrored parameter on a device.
getVersion() - Method in class ai.djl.engine.Engine
Returns the version of the deep learning framework.
getWeightDecay() - Method in class ai.djl.training.optimizer.Optimizer
Gets the value of weight decay.
getWidth() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the width of the Rectangle.
getX() - Method in class ai.djl.modality.cv.output.Point
Returns the X coordinate of this Point in double precision.
getX() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the left x-coordinate of the Rectangle.
getY() - Method in class ai.djl.modality.cv.output.Point
Returns the Y coordinate of this Point in double precision.
getY() - Method in class ai.djl.modality.cv.output.Rectangle
Returns the top y-coordinate of the Rectangle.
globalAvgPool1DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalAvgPool1D pooling function.
globalAvgPool2DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalAvgPool2D pooling function.
globalAvgPool3DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalAvgPool3D pooling function.
globalLpPool1DBlock(int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalLpPool1D pooling function.
globalLpPool2DBlock(int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalLpPool2D pooling function.
globalLpPool3DBlock(int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalLpPool3D pooling function.
globalMaxPool1DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalmaxPool1DBlock pooling function.
globalMaxPool2DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalmaxPool2DBlock pooling function.
globalMaxPool3DBlock() - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the globalmaxPool3DBlock pooling function.
gpu() - Static method in class ai.djl.Device
Returns the default GPU Device.
gpu(int) - Static method in class ai.djl.Device
Returns a new instance of GPU Device with the specified deviceId.
GPU - Static variable in interface ai.djl.Device.Type
 
GradientCollector - Interface in ai.djl.training
An interface that provides a mechanism to collect gradients during training.
GRU - Class in ai.djl.nn.recurrent
Applies GRU (Gated Recurrent Unit) recurrent layer to input.
GRU.Builder - Class in ai.djl.nn.recurrent
The Builder to construct a GRU type of Block.
gt(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Greater" comparison.
gt(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Greater Than" comparison.
gt(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater Than" comparison.
gt(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater Than" comparison.
gt(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater Than" comparison.
gte(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Greater or equals" comparison.
gte(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Greater or equals" comparison.
gte(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater or equals" comparison.
gte(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater or equals" comparison.
gte(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Greater or equals" comparison.

H

hasCapability(String) - Method in class ai.djl.engine.Engine
Returns whether the engine has the specified capability.
hashCode() - Method in class ai.djl.Device
hashCode() - Method in class ai.djl.ndarray.types.Shape
hasItem(T) - Method in class ai.djl.nn.core.Embedding
Returns whether an item is in the embedding.
hasMetric(String) - Method in class ai.djl.metric.Metrics
Returns true if the metrics object has a metric with the given name.
hasNext() - Method in interface ai.djl.modality.cv.output.PathIterator
Tests if the iteration is complete.
hasNext() - Method in class ai.djl.training.dataset.DataIterable
head() - Method in class ai.djl.ndarray.NDList
Returns the head index of the NDList.
head() - Method in class ai.djl.ndarray.types.Shape
Returns the head index of the shape.
HingeLoss - Class in ai.djl.training.loss
HingeLoss is a type of Loss.
HingeLoss() - Constructor for class ai.djl.training.loss.HingeLoss
Calculates Hinge loss.
HingeLoss(String) - Constructor for class ai.djl.training.loss.HingeLoss
Calculates Hinge loss.
HingeLoss(String, int, float) - Constructor for class ai.djl.training.loss.HingeLoss
Calculates Hinge loss.
hingeLoss() - Static method in class ai.djl.training.loss.Loss
Returns a new instance of HingeLoss with default arguments.
hingeLoss(String) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of HingeLoss with default arguments.
hingeLoss(String, int, float) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of HingeLoss with the given arguments.
HpBool - Class in ai.djl.training.hyperparameter.param
A Hyperparameter for a boolean option.
HpBool(String) - Constructor for class ai.djl.training.hyperparameter.param.HpBool
Constructs a HpBool.
HpCategorical<T> - Class in ai.djl.training.hyperparameter.param
A Hyperparameter which is one of a fixed number of options (similar to an enum).
HpCategorical(String, List<T>) - Constructor for class ai.djl.training.hyperparameter.param.HpCategorical
Constructs a HpCategorical.
HpFloat - Class in ai.djl.training.hyperparameter.param
A Hyperparameter for a float.
HpFloat(String, float, float, boolean) - Constructor for class ai.djl.training.hyperparameter.param.HpFloat
Constructs a HpFloat.
HpInt - Class in ai.djl.training.hyperparameter.param
A Hyperparameter for an integer.
HpInt(String, int, int) - Constructor for class ai.djl.training.hyperparameter.param.HpInt
Constructs a HpInt.
HpOptimizer - Interface in ai.djl.training.hyperparameter.optimizer
An optimizer for Hyperparameters.
HpORandom - Class in ai.djl.training.hyperparameter.optimizer
A simple HpOptimizer that tries random hyperparameter choices within the range.
HpORandom(HpSet) - Constructor for class ai.djl.training.hyperparameter.optimizer.HpORandom
Constructs a new HpORandom.
HpSet - Class in ai.djl.training.hyperparameter.param
A nestable set of Hyperparameters.
HpSet(String, List<Hyperparameter<?>>) - Constructor for class ai.djl.training.hyperparameter.param.HpSet
Cosntructs a new HpSet.
HpSet(String) - Constructor for class ai.djl.training.hyperparameter.param.HpSet
Cosntructs a new empty HpSet.
HpVal<T> - Class in ai.djl.training.hyperparameter.param
A Hyperparameter with a known value instead of a range of possible values.
HpVal(String, T) - Constructor for class ai.djl.training.hyperparameter.param.HpVal
Cosntructs a new HpVal.
Hyperparameter<T> - Class in ai.djl.training.hyperparameter.param
A class representing an input to the network that can't be differentiated.
Hyperparameter(String) - Constructor for class ai.djl.training.hyperparameter.param.Hyperparameter
Constructs a hyperparameter with the given name.
hyperParams - Variable in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
 

I

identityBlock() - Static method in class ai.djl.nn.Blocks
Creates a LambdaBlock that performs the identity function.
IMAGE_CLASSIFICATION - Static variable in interface ai.djl.Application.CV
 
ImageClassificationTranslator - Class in ai.djl.modality.cv.translator
A generic Translator for Image Classification tasks.
ImageClassificationTranslator(ImageClassificationTranslator.Builder) - Constructor for class ai.djl.modality.cv.translator.ImageClassificationTranslator
Constructs an Image Classification using ImageClassificationTranslator.Builder.
ImageClassificationTranslator.Builder - Class in ai.djl.modality.cv.translator
A Builder to construct a ImageClassificationTranslator.
ImageTranslator<T> - Class in ai.djl.modality.cv.translator
Built-in Translator that provides default image pre-processing.
ImageTranslator(ImageTranslator.BaseBuilder<?>) - Constructor for class ai.djl.modality.cv.translator.ImageTranslator
Constructs an ImageTranslator with the provided builder.
ImageTranslator.BaseBuilder<T extends ImageTranslator.BaseBuilder> - Class in ai.djl.modality.cv.translator
A builder to extend for all classes extending the ImageTranslator.
ImageVisualization - Class in ai.djl.modality.cv
A Collection of utilities for visualizing the results of Computer Vision tasks.
includeBias - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
includeBias - Variable in class ai.djl.nn.convolutional.Convolution
 
increment(long) - Method in class ai.djl.training.util.ProgressBar
index - Variable in class ai.djl.training.evaluator.AbstractAccuracy
 
init(String, NDArray[]) - Method in class ai.djl.training.LocalParameterServer
Initializes the ParameterStore for the given parameter.
init(String, NDArray[]) - Method in interface ai.djl.training.ParameterServer
Initializes the ParameterStore for the given parameter.
initialize(NDManager, DataType, Shape...) - Method in interface ai.djl.nn.Block
Initializes the parameters of the block.
initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.ParallelBlock
Initializes the parameters of the block.
initialize(NDManager, DataType, Shape[]) - Method in class ai.djl.nn.Parameter
Initializes the parameter with the given NDManager, with given DataType for the given expected input shapes.
initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.ParameterBlock
Initializes the parameters of the block.
initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.SequentialBlock
Initializes the parameters of the block.
initialize(NDManager, Shape, DataType) - Method in interface ai.djl.training.initializer.Initializer
Initializes a single NDArray.
initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.NormalInitializer
Initializes a single NDArray.
initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.UniformInitializer
Initializes a single NDArray.
initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.XavierInitializer
Initializes a single NDArray.
initialize(Shape...) - Method in interface ai.djl.training.Trainer
Initializes the Model that the Trainer is going to train.
Initializer - Interface in ai.djl.training.initializer
An interface representing an initialization method.
inputData - Variable in class ai.djl.BaseModel
 
inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss
Returns the inputs to computing the loss for a component loss.
inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.SimpleCompositeLoss
 
inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.SingleShotDetectionLoss
Calculate loss between label and prediction.
inputNames - Variable in class ai.djl.nn.AbstractBlock
 
inputShapes - Variable in class ai.djl.nn.AbstractBlock
 
InputStreamTranslator<T> - Class in ai.djl.modality.cv.translator
Built-in Translator that provides image pre-processing from InputStream.
InputStreamTranslator(Translator<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.InputStreamTranslator
Creates a UrlTranslator instance.
InputStreamTranslatorFactory<T> - Class in ai.djl.modality.cv.translator
A factory class creates Translator that can process image from an InputStream.
InputStreamTranslatorFactory(TranslatorFactory<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.InputStreamTranslatorFactory
Creates a FileTranslatorFactory instance.
insert(int, Transform) - Method in class ai.djl.translate.Pipeline
Inserts the given Transform to the list of transforms at the given position.
insert(int, int, Transform) - Method in class ai.djl.translate.Pipeline
Inserts the given Transform to the list of transforms at the given position to be applied on the NDArray at the given index in the input NDList.
insert(int, String, Transform) - Method in class ai.djl.translate.Pipeline
Inserts the given Transform to the list of transforms at the given position to be applied on the NDArray with the given name in the input NDList.
INSTANCE_SEGMENTATION - Static variable in interface ai.djl.Application.CV
 
InstanceSegmentationTranslator - Class in ai.djl.modality.cv.translator
A ImageTranslator that post-process the NDArray into DetectedObjects with boundaries at the detailed pixel level.
InstanceSegmentationTranslator(InstanceSegmentationTranslator.Builder) - Constructor for class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Creates the Instance Segmentation translator from the given builder.
InstanceSegmentationTranslator.Builder - Class in ai.djl.modality.cv.translator
The builder for Instance Segmentation translator.
invoke(String, NDArray[], NDArray[], PairList<String, ?>) - Method in interface ai.djl.ndarray.NDManager
An engine specific generic invocation to native operation.
invoke(String, NDList, PairList<String, ?>) - Method in interface ai.djl.ndarray.NDManager
An engine specific generic invocation to native operation.
isBidirectional() - Method in class ai.djl.nn.recurrent.RecurrentCell
 
isEmpty() - Method in interface ai.djl.ndarray.NDArray
Returns true if this NDArray is special case: no-value NDArray.
isFloating() - Method in enum ai.djl.ndarray.types.DataType
Checks whether it is a floating data type.
isInfinite() - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray with value true where this NDArray's entries are infinite, or false where they are not infinite.
isInitialized() - Method in class ai.djl.nn.AbstractBlock
Returns a boolean whether the block is initialized.
isInitialized() - Method in interface ai.djl.nn.Block
Returns a boolean whether the block is initialized.
isInitialized() - Method in class ai.djl.nn.Parameter
Checks if this Parameter is initialized.
isInteger() - Method in enum ai.djl.ndarray.types.DataType
Checks whether it is an integer data type.
isKnownToken(String) - Method in class ai.djl.modality.nlp.Vocabulary
Returns whether the given token is a known word.
isLayoutKnown() - Method in class ai.djl.ndarray.types.Shape
Returns true if a layout is set.
isNaN() - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray with value true where this NDArray's entries are NaN, or false where they are not NaN.
isOpen() - Method in class ai.djl.ndarray.BaseNDManager
Check if the manager is still valid.
isOpen() - Method in interface ai.djl.ndarray.NDManager
Check if the manager is still valid.
isScalar() - Method in interface ai.djl.ndarray.NDArray
Returns true if this NDArray is a scalar NDArray with empty Shape.
isScalar() - Method in class ai.djl.ndarray.types.Shape
Returns true if the NDArray is a scalar.
isSparse() - Method in interface ai.djl.ndarray.NDArray
Returns true if this NDArray is a SparseNDArray.
item(int) - Method in class ai.djl.modality.Classifications
Returns the item at a given index based on the order used to construct the Classifications.
item(int) - Method in class ai.djl.modality.cv.output.DetectedObjects
Returns the item at a given index based on the order used to construct the Classifications.
items() - Method in class ai.djl.modality.Classifications
Returns a classification item for each potential class for the input.
iterateDataset(Dataset) - Method in interface ai.djl.training.Trainer
Fetches an iterator that can iterate through the given Dataset.
iterator() - Method in class ai.djl.training.dataset.DataIterable

J

Joint(double, double, double) - Constructor for class ai.djl.modality.cv.output.Joints.Joint
Constructs a Joint with given data.
Joints - Class in ai.djl.modality.cv.output
A result of all joints found during Human Pose Estimation on a single image.
Joints(List<Joints.Joint>) - Constructor for class ai.djl.modality.cv.output.Joints
Constructs the Joints with the provided joints.
Joints.Joint - Class in ai.djl.modality.cv.output
A joint that was detected using Human Pose Estimation on an image.

K

kernel - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
kernel - Variable in class ai.djl.nn.convolutional.Convolution
 

L

L1Loss - Class in ai.djl.training.loss
L1Loss calculates L1 loss between label and prediction.
L1Loss() - Constructor for class ai.djl.training.loss.L1Loss
Calculates L1 Loss between the label and prediction, a.k.a.
L1Loss(String) - Constructor for class ai.djl.training.loss.L1Loss
Calculates L1 Loss between the label and prediction, a.k.a.
L1Loss(String, float) - Constructor for class ai.djl.training.loss.L1Loss
Calculates L1 Loss between the label and prediction, a.k.a.
l1Loss() - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L1Loss with default weight and batch axis.
l1Loss(String) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L1Loss with default weight and batch axis.
l1Loss(String, float) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L1Loss with given weight and batch axis.
L2Loss - Class in ai.djl.training.loss
Calculates L2Loss between label and prediction, a.k.a.
L2Loss() - Constructor for class ai.djl.training.loss.L2Loss
Calculate L2Loss between the label and prediction, a.k.a.
L2Loss(String) - Constructor for class ai.djl.training.loss.L2Loss
Calculate L2Loss between the label and prediction, a.k.a.
L2Loss(String, float) - Constructor for class ai.djl.training.loss.L2Loss
Calculates L2Loss between the label and prediction, a.k.a.
l2Loss() - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L2Loss with default weight and batch axis.
l2Loss(String) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L2Loss with default weight and batch axis.
l2Loss(String, float) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of L2Loss with given weight and batch axis.
labels - Variable in class ai.djl.training.dataset.ArrayDataset
 
LambdaBlock - Class in ai.djl.nn
LambdaBlock is a Block with no parameters or children.
LambdaBlock(Function<NDList, NDList>) - Constructor for class ai.djl.nn.LambdaBlock
Creates a LambdaBlock that can apply the specified function.
latestMetric(String) - Method in class ai.djl.metric.Metrics
Returns the latest Metric with the specified metric name.
LayoutType - Enum in ai.djl.ndarray.types
An enum to represent the meaning of a particular axis in an NDArray.
leakyRelu(NDArray, float) - Static method in class ai.djl.nn.Activation
Applies Leaky ReLU activation on the input NDArray.
leakyRelu(NDList, float) - Static method in class ai.djl.nn.Activation
Applies Leaky ReLU activation on the input singleton NDList.
leakyReluBlock(float) - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the LeakyReLU activation function in its forward function.
LearningRateTracker - Class in ai.djl.training.optimizer.learningrate
A LearningRateTracker tracks the evolution of the learning rate through the training process.
LearningRateTracker.LrBaseBuilder<T extends LearningRateTracker.LrBaseBuilder> - Class in ai.djl.training.optimizer.learningrate
The Builder to construct a LearningRateTracker.
like() - Method in interface ai.djl.ndarray.NDArray
Returns an uninitialized NDArray with the same Shape, DataType and SparseFormat as the input NDArray.
limit - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
limit - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Linear - Class in ai.djl.nn.core
A Linear block applies a linear transformation \(Y = XW^T + b\).
Linear.Builder - Class in ai.djl.nn.core
The Builder to construct a Linear type of Block.
linspace(int, int, int) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced numbers over a specified interval.
linspace(float, float, int) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced numbers over a specified interval.
linspace(int, int, int, boolean, Device) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced numbers over a specified interval.
linspace(float, float, int, boolean, Device) - Method in interface ai.djl.ndarray.NDManager
Returns evenly spaced numbers over a specified interval.
load(Path) - Method in interface ai.djl.Model
Loads the model from the modelPath.
load(Path, String) - Method in interface ai.djl.Model
Loads the model from the modelPath and the given name.
load(Path, String, Map<String, String>) - Method in interface ai.djl.Model
Loads the model from the modelPath with the name and options provided.
load(Path) - Method in interface ai.djl.ndarray.NDManager
Loads the NDArrays saved to a file.
load(Path, Device) - Method in interface ai.djl.ndarray.NDManager
Loads the NDArrays saved to a file.
load(NDManager, DataInputStream) - Method in class ai.djl.nn.Parameter
Loads parameter NDArrays from InputStream.
loadParameters(NDManager, DataInputStream) - Method in interface ai.djl.nn.Block
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.convolutional.Convolution
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Embedding
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Linear
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Prelu
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.LambdaBlock
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.norm.BatchNorm
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.norm.Dropout
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.ParallelBlock
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.recurrent.LSTM
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.recurrent.RecurrentCell
Loads the parameters from the given input stream.
loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.SequentialBlock
Loads the parameters from the given input stream.
LocalParameterServer - Class in ai.djl.training
LocalParameterServer is an implementation of the ParameterServer interface.
LocalParameterServer(Optimizer) - Constructor for class ai.djl.training.LocalParameterServer
Create a new instance of LocalParameterServer for the given optimizer.
log() - Method in interface ai.djl.ndarray.NDArray
Returns the natural logarithmic value of this NDArray element-wise.
log10() - Method in interface ai.djl.ndarray.NDArray
Returns the base 10 logarithm of this NDArray element-wise.
log2() - Method in interface ai.djl.ndarray.NDArray
Returns the base 2 logarithm of this NDArray element-wise.
logging() - Static method in interface ai.djl.training.listener.TrainingListener.Defaults
A default TrainingListener set including batch output logging.
logging(String) - Static method in interface ai.djl.training.listener.TrainingListener.Defaults
A default TrainingListener set including batch output logging and output directory.
LoggingTrainingListener - Class in ai.djl.training.listener
TrainingListener that outputs the progress of training each batch and epoch into logs.
LoggingTrainingListener() - Constructor for class ai.djl.training.listener.LoggingTrainingListener
 
logicalAnd(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the truth value of this NDArray AND the other NDArray element-wise.
logicalAnd(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the truth value of NDArray a AND NDArray b element-wise.
logicalNot() - Method in interface ai.djl.ndarray.NDArray
Computes the truth value of NOT this NDArray element-wise.
logicalOr(NDArray) - Method in interface ai.djl.ndarray.NDArray
Computes the truth value of this NDArray OR the other NDArray element-wise.
logicalOr(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Computes the truth value of NDArray a AND NDArray b element-wise.
logicalXor(NDArray) - Method in interface ai.djl.ndarray.NDArray
Computes the truth value of this NDArray XOR the other NDArray element-wise.
logicalXor(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Computes the truth value of NDArray a AND NDArray b element-wise.
logSoftmax(int) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function followed by a logarithm.
logSoftmax(int, float) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function followed by a logarithm.
logSoftmax(int[]) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function followed by a logarithm.
logSoftmax(int[], float) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function followed by a logarithm.
Loss - Class in ai.djl.training.loss
Loss functions (or Cost functions) are used to evaluate the model predictions against true labels for optimization.
Loss(String) - Constructor for class ai.djl.training.loss.Loss
Base class for metric with abstract update methods.
LowerCaseConvertor - Class in ai.djl.modality.nlp.preprocess
LowerCaseConvertor converts every character of the input tokens to it's respective lower case character.
LowerCaseConvertor(Locale) - Constructor for class ai.djl.modality.nlp.preprocess.LowerCaseConvertor
Creates a TextProcessor that converts input text into lower case character given the Locale.
LowerCaseConvertor() - Constructor for class ai.djl.modality.nlp.preprocess.LowerCaseConvertor
Creates a TextProcessor that converts input text into lower case character with the default english Locale.
lpPool1DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.
lpPool1DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.
lpPool1DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.
lpPool1DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.
lpPool2DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.
lpPool2DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.
lpPool2DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.
lpPool2DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.
lpPool3DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.
lpPool3DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.
lpPool3DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the LpPoo3D pooling function in its forward function.
lpPool3DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.
LrBaseBuilder() - Constructor for class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
 
LSTM - Class in ai.djl.nn.recurrent
Applies Long Short-Term Memory recurrent layer to input.
LSTM.Builder - Class in ai.djl.nn.recurrent
The Builder to construct a LSTM type of Block.
lstmStateClipMax - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
lstmStateClipMin - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
lt(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Less" comparison.
lt(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Less" comparison.
lt(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less" comparison.
lt(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less" comparison.
lt(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less" comparison.
lte(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Less or equals" comparison.
lte(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Less or equals" comparison.
lte(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less or equals" comparison.
lte(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less or equals" comparison.
lte(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Less or equals" comparison.

M

MACHINE_TRANSLATION - Static variable in interface ai.djl.Application.NLP
 
MalformedModelException - Exception in ai.djl
Thrown to indicate Model parameters are not in expected format or are malformed.
MalformedModelException(String) - Constructor for exception ai.djl.MalformedModelException
Constructs a new exception with the specified detail message.
MalformedModelException(String, Throwable) - Constructor for exception ai.djl.MalformedModelException
Constructs a new exception with the specified detail message and cause.
MalformedModelException(Throwable) - Constructor for exception ai.djl.MalformedModelException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.
manager - Variable in class ai.djl.BaseModel
 
manager - Variable in class ai.djl.inference.BasePredictor
 
map(Function<Pair<Long, LayoutType>, Pair<Long, LayoutType>>) - Method in class ai.djl.ndarray.types.Shape
Returns a mapped shape.
Mask - Class in ai.djl.modality.cv.output
A mask with a probability for each pixel within a bounding rectangle.
Mask(double, double, double, double, float[][]) - Constructor for class ai.djl.modality.cv.output.Mask
Constructs a Mask with the given data.
max() - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray.
max(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray along given axes.
max(int[], boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray along given axes.
maximum(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray and a number element-wise.
maximum(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray and the other NDArray element-wise.
maximum(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the maximum of a NDArray and a number element-wise.
maximum(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the maximum of a number and a NDArray element-wise.
maximum(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the maximum of NDArray a and NDArray b element-wise.
maxPool1DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool1D pooling function in its forward function.
maxPool1DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.
maxPool1DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.
maxPool1DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.
maxPool2DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.
maxPool2DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.
maxPool2DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.
maxPool2DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.
maxPool3DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.
maxPool3DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.
maxPool3DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.
maxPool3DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool
Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.
mean(String) - Method in class ai.djl.metric.Metrics
Returns the average value of the specified metric.
mean() - Method in interface ai.djl.ndarray.NDArray
Returns the average of this NDArray.
mean(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns the average of this NDArray along given axes.
mean(int[], boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the average of this NDArray along given axes.
median() - Method in interface ai.djl.ndarray.NDArray
Returns median value for this NDArray.
median(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns median value along given axes.
MemoryTrainingListener - Class in ai.djl.training.listener
TrainingListener that collects the memory usage information.
MemoryTrainingListener() - Constructor for class ai.djl.training.listener.MemoryTrainingListener
Constructs a MemoryTrainingListener that does not output data to a file.
MemoryTrainingListener(String) - Constructor for class ai.djl.training.listener.MemoryTrainingListener
Constructs a MemoryTrainingListener that outputs data in the given directory.
Metric - Class in ai.djl.metric
A class representing a single recorded Metric value.
Metric(String, Number) - Constructor for class ai.djl.metric.Metric
Constructs a Metric instance with the specified metricName and value.
Metric(String, Number, String) - Constructor for class ai.djl.metric.Metric
Constructs a Metric instance with the specified metricName, value , and unit.
metricName(Evaluator, String) - Static method in class ai.djl.training.listener.EvaluatorTrainingListener
Returns the metric created with the evaluator for the given stage.
Metrics - Class in ai.djl.metric
A collection of Metric objects organized by metric name.
Metrics() - Constructor for class ai.djl.metric.Metrics
Constructs an empty Metrics instance.
min() - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray.
min(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray along given axes.
min(int[], boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray along given axes.
minFrequency - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
 
minimum(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray and a number element-wise.
minimum(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the maximum of this NDArray and the other NDArray element-wise.
minimum(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the minimum of a NDArray and a number element-wise.
minimum(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the minimum of a number and a NDArray element-wise.
minimum(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the minimum of NDArray a and NDArray b element-wise.
mish(NDArray) - Static method in class ai.djl.nn.Activation
Applies Mish activation on the input NDArray.
mish(NDList) - Static method in class ai.djl.nn.Activation
Applies Mish activation on the input singleton NDList.
mishBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the Mish activation function in its forward function.
MKL - Static variable in class ai.djl.engine.StandardCapabilities
 
MKLDNN - Static variable in class ai.djl.engine.StandardCapabilities
 
mod(Number) - Method in interface ai.djl.ndarray.NDArray
Returns element-wise remainder of division.
mod(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns element-wise remainder of division.
mod(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division.
mod(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division.
mod(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division.
mode - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
model - Variable in class ai.djl.inference.BasePredictor
 
Model - Interface in ai.djl
A model is a collection of artifacts that is created by the training process.
modelDir - Variable in class ai.djl.BaseModel
 
ModelException - Exception in ai.djl
Thrown to indicate Model parameter or load exceptions parent to Model Exceptions.
ModelException(String) - Constructor for exception ai.djl.ModelException
Constructs a new exception with the specified detail message.
ModelException(String, Throwable) - Constructor for exception ai.djl.ModelException
Constructs a new exception with the specified detail message and cause.
ModelException(Throwable) - Constructor for exception ai.djl.ModelException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.
modelName - Variable in class ai.djl.BaseModel
 
modi(Number) - Method in interface ai.djl.ndarray.NDArray
Returns element-wise remainder of division in place.
modi(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns in place element-wise remainder of division in place.
modi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division in place.
modi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division in place.
modi(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns element-wise remainder of division.
mul(Number) - Method in interface ai.djl.ndarray.NDArray
Multiplies this NDArray by a number element-wise.
mul(NDArray) - Method in interface ai.djl.ndarray.NDArray
Multiplies this NDArray by other NDArrays element-wise.
mul(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Multiplies the NDArray by a number element-wise.
mul(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Multiplies a number by a NDArray element-wise.
mul(NDArray...) - Static method in class ai.djl.ndarray.NDArrays
Multiplies all of the NDArrays together element-wise.
muli(Number) - Method in interface ai.djl.ndarray.NDArray
Multiplies this NDArray by a number element-wise in place.
muli(NDArray) - Method in interface ai.djl.ndarray.NDArray
Multiplies this NDArray by other NDArray element-wise in place.
muli(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Multiplies the NDArray by a number element-wise in place.
muli(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Multiplies a number by a NDArray element-wise.
muli(NDArray...) - Static method in class ai.djl.ndarray.NDArrays
Multiplies all of the NDArrays together element-wise in place.
MultiBoxDetection - Class in ai.djl.modality.cv
MultiBoxDetection is the class that takes the output of a multi-box detection model, and converts it into an NDList that contains the object detections.
MultiBoxDetection(MultiBoxDetection.Builder) - Constructor for class ai.djl.modality.cv.MultiBoxDetection
Creates a new instance of MultiBoxDetection with the arguments from the given MultiBoxDetection.Builder.
MultiBoxDetection.Builder - Class in ai.djl.modality.cv
The Builder to construct a MultiBoxDetection object.
MultiBoxPrior - Class in ai.djl.modality.cv
MultiBoxPrior is the class that generates anchor boxes that act as priors for object detection.
MultiBoxPrior(MultiBoxPrior.Builder) - Constructor for class ai.djl.modality.cv.MultiBoxPrior
Creates a new instance of MultiBoxPrior with the arguments from the given MultiBoxPrior.Builder.
MultiBoxPrior.Builder - Class in ai.djl.modality.cv
The Builder to construct a MultiBoxPrior object.
MultiBoxTarget - Class in ai.djl.modality.cv
MultiBoxTarget is the class that computes the training targets for training a Single Shot Detection (SSD) models.
MultiBoxTarget(MultiBoxTarget.Builder) - Constructor for class ai.djl.modality.cv.MultiBoxTarget
Creates a new instance of MultiBoxTarget with the arguments from the given MultiBoxTarget.Builder.
MultiBoxTarget.Builder - Class in ai.djl.modality.cv
The Builder to construct a MultiBoxTarget object.
multiFactorTracker() - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker
Returns a new instance of MultiFactorTracker.Builder that can build an MultiFactorTracker.
MultiFactorTracker - Class in ai.djl.training.optimizer.learningrate
MultiFactorTracker is an implementation of LearningRateTracker which is updated by a multiplicative factor, at an uneven interval of steps, until it reaches a specified stop value.
MultiFactorTracker(MultiFactorTracker.Builder) - Constructor for class ai.djl.training.optimizer.learningrate.MultiFactorTracker
Creates a new instance of MultiFactorTracker.
MultiFactorTracker.Builder - Class in ai.djl.training.optimizer.learningrate
The Builder to construct an MultiFactorTracker object.

N

Nag - Class in ai.djl.training.optimizer
Nag is a Nesterov accelerated gradient optimizer.
Nag(Nag.Builder) - Constructor for class ai.djl.training.optimizer.Nag
Creates a new instance of Nag optimizer.
nag() - Static method in class ai.djl.training.optimizer.Optimizer
Returns a new instance of Nag.Builder that can build an Nag optimizer.
Nag.Builder - Class in ai.djl.training.optimizer
The Builder to construct an Nag object.
name - Variable in class ai.djl.training.hyperparameter.param.Hyperparameter
 
NDArray - Interface in ai.djl.ndarray
An interface representing an n-dimensional array.
NDArrays - Class in ai.djl.ndarray
This class contains various methods for manipulating NDArrays.
NDImageUtils - Class in ai.djl.modality.cv.util
NDImageUtils is an image processing utility to load, reshape, and convert images using NDArray images.
NDImageUtils.Flag - Enum in ai.djl.modality.cv.util
Flag indicates the color channel options for images.
NDIndex - Class in ai.djl.ndarray.index
The NDIndex allows you to specify a subset of an NDArray that can be used for fetching or updating.
NDIndex() - Constructor for class ai.djl.ndarray.index.NDIndex
Creates an empty NDIndex to append values to.
NDIndex(String) - Constructor for class ai.djl.ndarray.index.NDIndex
Creates a NDIndex given the index values.
NDIndex(long...) - Constructor for class ai.djl.ndarray.index.NDIndex
Creates an NDIndex with the given indices as specified values on the NDArray.
NDIndexAll - Class in ai.djl.ndarray.index
An NDIndexElement to return all values in a particular dimension.
NDIndexAll() - Constructor for class ai.djl.ndarray.index.NDIndexAll
 
NDIndexBooleans - Class in ai.djl.ndarray.index
An NDIndexElement to return values based on a mask binary NDArray.
NDIndexBooleans(NDArray) - Constructor for class ai.djl.ndarray.index.NDIndexBooleans
Constructs a NDIndexBooleans instance with specified mask binary NDArray.
NDIndexElement - Interface in ai.djl.ndarray.index
An index for particular dimensions created by NDIndex.
NDIndexFixed - Class in ai.djl.ndarray.index
An NDIndexElement that returns only a specific value in the corresponding dimension.
NDIndexFixed(long) - Constructor for class ai.djl.ndarray.index.NDIndexFixed
Constructs a NDIndexFixed instance with specified dimension.
NDIndexFullSlice - Class in ai.djl.ndarray.index
An index as a slice on all dimensions where some dimensions can be squeezed.
NDIndexSlice - Class in ai.djl.ndarray.index
An NDIndexElement that returns a range of values in the specified dimension.
NDIndexSlice(Long, Long, Long) - Constructor for class ai.djl.ndarray.index.NDIndexSlice
Constructs a NDIndexSlice instance with specified range and step.
NDList - Class in ai.djl.ndarray
An NDList represents a sequence of NDArrays with names.
NDList() - Constructor for class ai.djl.ndarray.NDList
Constructs an empty NDList.
NDList(int) - Constructor for class ai.djl.ndarray.NDList
Constructs an empty NDList with the specified initial capacity.
NDList(NDArray...) - Constructor for class ai.djl.ndarray.NDList
Constructs and initiates an NDList with the specified NDArrays.
NDList(Collection<NDArray>) - Constructor for class ai.djl.ndarray.NDList
Constructs and initiates an NDList with the specified NDArrays.
NDManager - Interface in ai.djl.ndarray
NDArray managers are used to create NDArrays (n-dimensional array on native engine).
NDUtils - Class in ai.djl.ndarray
A class containing utility methods for NDArray operations.
neg() - Method in interface ai.djl.ndarray.NDArray
Returns the numerical negative NDArray element-wise.
negi() - Method in interface ai.djl.ndarray.NDArray
Returns the numerical negative NDArray element-wise in place.
neq(Number) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Not equals" comparison.
neq(NDArray) - Method in interface ai.djl.ndarray.NDArray
Returns the boolean NDArray for element-wise "Not equals" comparison.
neq(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Not equals" comparison.
neq(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Not equals" comparison.
neq(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns the boolean NDArray for element-wise "Not equals" comparison.
newBaseManager() - Method in class ai.djl.engine.Engine
Creates a new top-level NDManager.
newBaseManager(Device) - Method in class ai.djl.engine.Engine
Creates a new top-level NDManager with specified Device.
newBaseManager() - Static method in interface ai.djl.ndarray.NDManager
Creates a new top-level NDManager.
newBaseManager(Device) - Static method in interface ai.djl.ndarray.NDManager
Creates a new top-level NDManager with specified Device.
newBaseManager(Device, String) - Static method in interface ai.djl.ndarray.NDManager
Creates a new top-level NDManager with specified Device and engine.
newGradientCollector() - Method in interface ai.djl.training.Trainer
Returns a new instance of GradientCollector.
newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.FileTranslatorFactory
Returns a new instance of the Translator class.
newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.InputStreamTranslatorFactory
Returns a new instance of the Translator class.
newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.UrlTranslatorFactory
Returns a new instance of the Translator class.
newInstance() - Static method in interface ai.djl.Model
Creates an empty model instance.
newInstance(Device) - Static method in interface ai.djl.Model
Creates an empty model instance on the specified Device.
newInstance(Device, String) - Static method in interface ai.djl.Model
Creates an empty model instance on the specified Device and engine.
newInstance(Map<String, Object>) - Method in interface ai.djl.translate.TranslatorFactory
Returns a new instance of the Translator class.
newModel(Device) - Method in class ai.djl.engine.Engine
Constructs a new model.
newPredictor(Translator<I, O>) - Method in interface ai.djl.Model
Creates a new Predictor based on the model.
newSubManager() - Method in interface ai.djl.ndarray.NDManager
Creates a child NDManager.
newSubManager(Device) - Method in interface ai.djl.ndarray.NDManager
Creates a child NDManager with specified default Device.
newTrainer(TrainingConfig) - Method in interface ai.djl.Model
Creates a new Trainer instance for a Model.
next() - Method in interface ai.djl.modality.cv.output.PathIterator
Moves the iterator to the next segment of the path forwards along the primary direction of traversal as long as there are more points in that direction.
next() - Method in class ai.djl.training.dataset.DataIterable
nextConfig() - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer
Returns the next hyperparameters to test.
nextConfig() - Method in class ai.djl.training.hyperparameter.optimizer.HpORandom
Returns the next hyperparameters to test.
none() - Method in interface ai.djl.ndarray.NDArray
Returns true if none of the elements within this NDArray are non-zero or true.
nonzero() - Method in interface ai.djl.ndarray.NDArray
Returns the indices of elements that are non-zero.
NoopTranslator - Class in ai.djl.translate
A no operational Translator implementation.
NoopTranslator() - Constructor for class ai.djl.translate.NoopTranslator
 
NormalInitializer - Class in ai.djl.training.initializer
NormalInitializer initializes weights with random values sampled from a normal distribution with a mean of zero and standard deviation of sigma.
NormalInitializer() - Constructor for class ai.djl.training.initializer.NormalInitializer
Creates an instance of NormalInitializer with a default sigma of 0.01.
NormalInitializer(float) - Constructor for class ai.djl.training.initializer.NormalInitializer
Creates a Normal initializer.
Normalize - Class in ai.djl.modality.cv.transform
A Transform that normalizes an image NDArray of shape CHW or NCHW.
Normalize(float[], float[]) - Constructor for class ai.djl.modality.cv.transform.Normalize
Creates a Normalize Transform that normalizes.
normalize(NDArray, float, float) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Normalizes an image NDArray of shape CHW or NCHW with a single mean and standard deviation to apply to all channels.
normalize(NDArray, float[], float[]) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Normalizes an image NDArray of shape CHW or NCHW with mean and standard deviation.
numDimensions() - Method in class ai.djl.nn.convolutional.Conv1D
Returns the number of dimensions of the input.
numDimensions() - Method in class ai.djl.nn.convolutional.Conv2D
Returns the number of dimensions of the input.
numDimensions() - Method in class ai.djl.nn.convolutional.Conv3D
Returns the number of dimensions of the input.
numDimensions() - Method in class ai.djl.nn.convolutional.Convolution
Returns the number of dimensions of the input.
numDirections - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
numFilters - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
numFilters - Variable in class ai.djl.nn.convolutional.Convolution
 
numGroups - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
numGroups - Variable in class ai.djl.nn.convolutional.Convolution
 
numStackedLayers - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
numStackedLayers - Variable in class ai.djl.nn.recurrent.RecurrentCell
 

O

OBJECT_DETECTION - Static variable in interface ai.djl.Application.CV
 
onEpoch(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener
Listens to the end of an epoch during training.
onEpoch(Trainer) - Method in interface ai.djl.training.listener.TrainingListener
Listens to the end of an epoch during training.
ones(Shape, DataType) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Shape filled with ones.
ones(Shape) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Shape filled with ones.
ones(Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Device, Shape, and DataType filled with ones.
ONES - Static variable in interface ai.djl.training.initializer.Initializer
 
onesLike() - Method in interface ai.djl.ndarray.NDArray
Returns an NDArray of ones with the same Shape, DataType and SparseFormat as the input NDArray.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EpochTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EvaluatorTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.LoggingTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.MemoryTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener
Listens to the end of training one batch of data during training.
onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in interface ai.djl.training.listener.TrainingListener
Listens to the end of training one batch of data during training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener
Listens to the beginning of training.
onTrainingBegin(Trainer) - Method in interface ai.djl.training.listener.TrainingListener
Listens to the beginning of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener
Listens to the end of training.
onTrainingEnd(Trainer) - Method in interface ai.djl.training.listener.TrainingListener
Listens to the end of training.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EpochTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EvaluatorTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.LoggingTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.MemoryTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener
Listens to the end of validating one batch of data during validation.
onValidationBatch(Trainer, TrainingListener.BatchData) - Method in interface ai.djl.training.listener.TrainingListener
Listens to the end of validating one batch of data during validation.
OPENMP - Static variable in class ai.djl.engine.StandardCapabilities
 
opInputs(ParameterStore, NDList) - Method in class ai.djl.nn.recurrent.LSTM
 
opInputs(ParameterStore, NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell
 
optApplySoftmax(boolean) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder
Sets whether to apply softmax when processing output.
optAxis(int) - Method in class ai.djl.nn.norm.BatchNorm.Builder
Set the axis in which channel is specified.
optBackgroundId(int) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the class ID for the background.
optBaseLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
Sets the base learning rate.
optBatchier(Batchifier) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Batchifier for the dataset.
optBeginNumUpdate(int) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
Sets the initial value of the number of updates.
optBeta1(float) - Method in class ai.djl.training.optimizer.Adam.Builder
Sets the decay rate for the first moment estimates.
optBeta2(float) - Method in class ai.djl.training.optimizer.Adam.Builder
Sets the decay rate for the second moment estimates.
optBias(boolean) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the optional parameter of whether to include a bias vector.
optBias(boolean) - Method in class ai.djl.nn.core.Linear.Builder
Sets the optional parameter that indicates whether to include a bias vector with default value of true.
optBidrectional(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the optional parameter that indicates whether to use bidirectional recurrent layers.
optCenter(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder
If True, add offset of `beta` to normalized tensor.
optClip(boolean) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.
optClip(boolean) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.
optClipGrad(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
Sets the value of the \(clipGrad\).
optDataType(DataType) - Method in class ai.djl.nn.core.Embedding.Builder
Sets the data type of the embedding arrays (default is Float32).
optDevice(Device) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Device.
optDevices(Device[]) - Method in class ai.djl.training.DefaultTrainingConfig
Sets the array of Device available for training.
optDilate(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the dilation along each dimension.
optDropRate(float) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the drop rate of the dropout on the outputs of each RNN layer, except the last layer.
optEpsilon(float) - Method in class ai.djl.nn.norm.BatchNorm.Builder
Sets the epsilon value to prevent division by 0.
optEpsilon(float) - Method in class ai.djl.training.optimizer.Adam.Builder
Sets \(epsilon\) - a small quantity for numerical stability.
optExecutor(ExecutorService, int) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the ExecutorService to spawn threads to fetch data.
optFactor(float) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder
Sets the value of the multiplicative factor.
optFactor(float) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder
Set the value of the multiplicative factor.
optFlag(NDImageUtils.Flag) - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
Sets the optional NDImageUtils.Flag (default is NDImageUtils.Flag#COLOR).
optForceSuppress(boolean) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the boolean parameter that indicates whether to suppress all detections regardless of class_id.
optIgnoreLabel(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Sets the label for ignored anchors.
Optimizer - Class in ai.djl.training.optimizer
An Optimizer updates the weight parameters to minimize the loss function.
Optimizer(Optimizer.OptimizerBuilder<?>) - Constructor for class ai.djl.training.optimizer.Optimizer
Creates a new instance of Optimizer.
Optimizer.OptimizerBuilder<T extends Optimizer.OptimizerBuilder> - Class in ai.djl.training.optimizer
The Builder to construct an Optimizer.
OptimizerBuilder() - Constructor for class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
 
optInitializer(Initializer) - Method in class ai.djl.training.DefaultTrainingConfig
Sets the Initializer to use for the parameters (default from paper).
optIouThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Sets the anchor-GroundTruth overlap threshold to be regarded as a positive match.
optLabels(NDArray...) - Method in class ai.djl.training.dataset.ArrayDataset.Builder
Sets the labels for the data in the ArrayDataset.
optLearningRateTracker(LearningRateTracker) - Method in class ai.djl.training.optimizer.Adam.Builder
Sets the LearningRateTracker for this optimizer.
optLimit(long) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets this dataset's limit.
optLstmStateClipMin(float, float) - Method in class ai.djl.nn.recurrent.LSTM.Builder
Sets the minimum and maximum clip value of LSTM states.
optMaxEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
Sets the maximum edge length of the rescaled image.
optMinFrequency(int) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Sets the optional parameter that specifies the minimum frequency to consider a token to be part of the Vocabulary.
optMinNegativeSamples(int) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Sets the minimum number of negative samples.
optMomentum(float) - Method in class ai.djl.nn.norm.BatchNorm.Builder
Set the momentum for moving average.
optMomentum(float) - Method in class ai.djl.training.optimizer.Sgd.Builder
Sets the momentum for Sgd.
optNegativeMinigRatio(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Sets the max negative to positive samples ratio.
optNegativeMiningThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder
Sets the threshold used for negative mining.
optNmsThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the non-maximum suppression(NMS) threshold.
optNmsTopK(int) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.
optNumGroups(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the number of group partitions.
optOffsets(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Sets the value of the center-box offsets across \(x\) and \(y\) dimensions.
optOptimizer(Optimizer) - Method in class ai.djl.training.DefaultTrainingConfig
Sets the Optimizer used during training (default Adam).
optPad(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the padding along each dimension.
optPipeline(Pipeline) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Pipeline of Transform to be applied on the data.
optProbability(float) - Method in class ai.djl.nn.norm.Dropout.Builder
Sets the probability or the fraction of the input that gets dropped out during training time.
optRescaleSize(double, double) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Sets the optional rescale size.
optReservedTokens(Collection<String>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Sets the optional parameter that sets the list of reserved tokens.
optScale(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder
If True, multiply result by `gamma`.
optSharedAxes(int[]) - Method in class ai.djl.nn.norm.Dropout.Builder
Sets the axes for variational dropout kernel.
optShortEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
Sets the shorter edge length of the rescaled image.
optStateOutput(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the optional parameter that indicates whether to have the states as symbol outputs.
optSteps(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Sets the step across \(x\) and \(y\) dimensions.
optStopFactorLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder
Sets the stop value after which the learning rate should remain constant.
optStride(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the stride of the convolution.
optTargetPipeline(Pipeline) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Pipeline of Transform to be applied on the labels.
optThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder
Sets the threshold score for a detection to be a positive prediction.
optThreshold(float) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
Sets the threshold for prediction accuracy.
optThreshold(float) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder
Sets the threshold for prediction accuracy.
optThreshold(float) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Sets the threshold for prediction accuracy.
optUnknownToken(String) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
Sets the optional parameter that specifies the unknown token's string value.
optUseDefault(boolean) - Method in class ai.djl.nn.core.Embedding.Builder
Sets whether to use a default embedding for undefined items (default true).
optWarmUpBeginLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
Sets the value of the learning rate at the beginning of warm-up mode.
optWarmUpMode(WarmUpMode) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
optWarmUpSteps(int) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
Sets the number of steps until the point the learning rate is updated in warm-up mode.
optWeightDecays(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
Sets the value of weight decay.

P

pad - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
pad - Variable in class ai.djl.nn.convolutional.Convolution
 
ParallelBlock - Class in ai.djl.nn
ParallelBlock is a Block whose children form a parallel branch in the network and are combined to produce a single output.
ParallelBlock(Function<List<NDList>, NDList>) - Constructor for class ai.djl.nn.ParallelBlock
Creates a parallel block whose branches are combined to form a single output by the given function.
ParallelBlock(Function<List<NDList>, NDList>, List<Block>) - Constructor for class ai.djl.nn.ParallelBlock
Creates a parallel block whose branches are formed by each block in the list of blocks, and are combined to form a single output by the given function.
Parameter - Class in ai.djl.nn
Parameter is a container class that holds a learnable parameter of a model.
Parameter(String, Block, ParameterType) - Constructor for class ai.djl.nn.Parameter
Creates a Parameter with the given name, and parameter type, and associated with the given Block.
Parameter(String, Block, ParameterType, boolean) - Constructor for class ai.djl.nn.Parameter
Creates a Parameter with the given name, and parameter type, and associated with the given Block.
ParameterBlock - Class in ai.djl.nn
ParameterBlock is an abstract implementation of Block.
ParameterBlock() - Constructor for class ai.djl.nn.ParameterBlock
 
ParameterList - Class in ai.djl.nn
Represents a set of names and Parameters.
ParameterList() - Constructor for class ai.djl.nn.ParameterList
Create an empty ParameterList.
ParameterList(int) - Constructor for class ai.djl.nn.ParameterList
Constructs an empty ParameterList with the specified initial capacity.
ParameterList(List<String>, List<Parameter>) - Constructor for class ai.djl.nn.ParameterList
Constructs a ParameterList containing the elements of the specified keys and values.
ParameterList(List<Pair<String, Parameter>>) - Constructor for class ai.djl.nn.ParameterList
Constructs a ParameterList containing the elements of the specified list of Pairs.
ParameterList(Map<String, Parameter>) - Constructor for class ai.djl.nn.ParameterList
Constructs a ParameterList containing the elements of the specified map.
parameters - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
ParameterServer - Interface in ai.djl.training
An interface for a key-value store to store parameters, and their corresponding gradients.
ParameterStore - Class in ai.djl.training
The ParameterStore contains a map from a parameter to the mirrors of it on other devices.
ParameterStore(NDManager, boolean) - Constructor for class ai.djl.training.ParameterStore
Constructs an empty ParameterStore.
ParameterType - Enum in ai.djl.nn
Enumerates the types of Parameter.
paramPathResolver(Map<String, String>) - Method in class ai.djl.BaseModel
 
parent - Variable in class ai.djl.ndarray.BaseNDManager
 
PathIterator - Interface in ai.djl.modality.cv.output
A sequence of points used to outline an object in an image.
percentile(String, int) - Method in class ai.djl.metric.Metrics
Returns a percentile Metric object for the specified metric name.
percentile(Number) - Method in interface ai.djl.ndarray.NDArray
Returns percentile for this NDArray.
percentile(Number, int[]) - Method in interface ai.djl.ndarray.NDArray
Returns median along given dimension(s).
pipeline - Variable in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
 
pipeline - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
pipeline - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Pipeline - Class in ai.djl.translate
Pipeline allows applying multiple transforms on an input NDList.
Pipeline() - Constructor for class ai.djl.translate.Pipeline
Creates a new instance of Pipeline that has no Transform defined yet.
Pipeline(Transform...) - Constructor for class ai.djl.translate.Pipeline
Creates a new instance of Pipeline that can apply the given transforms on its input.
Point - Class in ai.djl.modality.cv.output
A point representing a location in (x,y) coordinate space, specified in double precision.
Point(double, double) - Constructor for class ai.djl.modality.cv.output.Point
Constructs and initializes a point at the specified (x,y) location in the coordinate space.
Pool - Class in ai.djl.nn.pooling
Utility class that provides Block and methods for different pooling functions.
PoolingConvention - Enum in ai.djl.nn.pooling
An enum that enumerates the different pooling conventions.
POSE_ESTIMATION - Static variable in interface ai.djl.Application.CV
 
PostProcessor<O> - Interface in ai.djl.translate
An interface that provides post-processing functionality.
pow(Number) - Method in interface ai.djl.ndarray.NDArray
Takes the power of this NDArray with a number element-wise.
pow(NDArray) - Method in interface ai.djl.ndarray.NDArray
Takes the power of this NDArray with the other NDArray element-wise.
pow(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of the NDArray with a number element-wise.
pow(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of a number with a NDArray element-wise.
pow(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of a NDArray with a NDArray element-wise.
powi(Number) - Method in interface ai.djl.ndarray.NDArray
Takes the power of this NDArray with a number element-wise in place.
powi(NDArray) - Method in interface ai.djl.ndarray.NDArray
Takes the power of this NDArray with the other NDArray element-wise in place.
powi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of the NDArray with a number element-wise in place.
powi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of a number with a NDArray element-wise in place.
powi(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Takes the power of a NDArray with a NDArray element-wise.
predict(I) - Method in class ai.djl.inference.BasePredictor
Predicts an item for inference.
predict(I) - Method in interface ai.djl.inference.Predictor
Predicts an item for inference.
Predictor<I,O> - Interface in ai.djl.inference
The Predictor interface provides a session for model inference.
prefetchNumber - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
prefetchNumber - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Prelu - Class in ai.djl.nn.core
Applies Leaky Parametric ReLU activation element-wise to the input.
Prelu() - Constructor for class ai.djl.nn.core.Prelu
Creates a Parametric ReLU Block.
preluBlock() - Static method in class ai.djl.nn.Activation
Returns a Prelu block.
prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator
Prepares the translator with the manager and model to use.
prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Prepares the translator with the manager and model to use.
prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
 
prepare(NDManager, Model) - Method in interface ai.djl.translate.Translator
Prepares the translator with the manager and model to use.
prepared - Variable in class ai.djl.inference.BasePredictor
 
preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.LowerCaseConvertor
Applies the preprocessing defined to the given input tokens.
preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.PunctuationSeparator
Applies the preprocessing defined to the given input tokens.
preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer
Applies the preprocessing defined to the given input tokens.
preprocess(List<String>) - Method in interface ai.djl.modality.nlp.preprocess.TextProcessor
Applies the preprocessing defined to the given input tokens.
PreProcessor<I> - Interface in ai.djl.translate
An interface that provides pre-processing functionality.
preprocessWordToEmbed(NDManager, String) - Method in interface ai.djl.modality.nlp.WordEmbedding
Preprocesses the word to embed into an NDArray to pass into the model.
probabilities - Variable in class ai.djl.modality.Classifications
 
processInput(TranslatorContext, Path) - Method in class ai.djl.modality.cv.translator.FileTranslator
Processes the input and converts it to NDList.
processInput(TranslatorContext, BufferedImage) - Method in class ai.djl.modality.cv.translator.ImageTranslator
Processes the BufferedImage input and converts it to NDList.
processInput(TranslatorContext, InputStream) - Method in class ai.djl.modality.cv.translator.InputStreamTranslator
Processes the input and converts it to NDList.
processInput(TranslatorContext, BufferedImage) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Processes the BufferedImage input and converts it to NDList.
processInput(TranslatorContext, URL) - Method in class ai.djl.modality.cv.translator.UrlTranslator
Processes the input and converts it to NDList.
processInput(TranslatorContext, NDList) - Method in class ai.djl.translate.NoopTranslator
Processes the input and converts it to NDList.
processInput(TranslatorContext, I) - Method in interface ai.djl.translate.PreProcessor
Processes the input and converts it to NDList.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.FileTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.InputStreamTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.UrlTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in class ai.djl.translate.NoopTranslator
Processes the output NDList to the corresponding output object.
processOutput(TranslatorContext, NDList) - Method in interface ai.djl.translate.PostProcessor
Processes the output NDList to the corresponding output object.
prod() - Method in interface ai.djl.ndarray.NDArray
Returns the product of this NDArray.
prod(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns the product of this NDArray elements over the given axes.
prod(int[], boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the product of this NDArray elements over the given axes.
ProgressBar - Class in ai.djl.training.util
ProgressBar is an implementation of Progress.
ProgressBar() - Constructor for class ai.djl.training.util.ProgressBar
Creates an instance of ProgressBar with a maximum value of 1.
ProgressBar(String, long) - Constructor for class ai.djl.training.util.ProgressBar
Creates an instance of ProgressBar with the given maximum value, and displays the given message.
ProgressBar(String, long, String) - Constructor for class ai.djl.training.util.ProgressBar
Creates an instance of ProgressBar with the given maximum value, and displays the given message.
properties - Variable in class ai.djl.BaseModel
 
pull(String, NDArray[], int) - Method in class ai.djl.training.LocalParameterServer
Pulls the value of a key from Parameter Server to NDArrays.
pull(String, NDArray[], int) - Method in interface ai.djl.training.ParameterServer
Pulls the value of a key from Parameter Server to NDArrays.
PunctuationSeparator - Class in ai.djl.modality.nlp.preprocess
PunctuationSeparator converts every character of the input tokens to it's respective lower case character.
PunctuationSeparator(String) - Constructor for class ai.djl.modality.nlp.preprocess.PunctuationSeparator
Creates a TextProcessor that separates the given punctuations into distinct tokens in the text.
PunctuationSeparator() - Constructor for class ai.djl.modality.nlp.preprocess.PunctuationSeparator
Creates a TextProcessor that separates the given punctuations into distinct tokens in the text.
push(String, NDArray[], int) - Method in class ai.djl.training.LocalParameterServer
Updates values of a key in Parameter Server.
push(String, NDArray[], int) - Method in interface ai.djl.training.ParameterServer
Updates values of a key in Parameter Server.

Q

QAInput - Class in ai.djl.modality.nlp.qa
The input container for BERT QA model.
QAInput(String, String, int) - Constructor for class ai.djl.modality.nlp.qa.QAInput
Creates the BERT QA model.
quantize() - Method in interface ai.djl.Model
Converts the model to use a lower precision quantized network.
QUESTION_ANSWER - Static variable in interface ai.djl.Application.NLP
 

R

random() - Method in class ai.djl.training.hyperparameter.param.HpCategorical
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
random() - Method in class ai.djl.training.hyperparameter.param.HpFloat
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
random() - Method in class ai.djl.training.hyperparameter.param.HpInt
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
random() - Method in class ai.djl.training.hyperparameter.param.HpSet
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
random() - Method in class ai.djl.training.hyperparameter.param.HpVal
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
random() - Method in class ai.djl.training.hyperparameter.param.Hyperparameter
Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.
RandomAccessDataset - Class in ai.djl.training.dataset
RandomAccessDataset represent the dataset that support random access reads.
RandomAccessDataset(RandomAccessDataset.BaseBuilder<?>) - Constructor for class ai.djl.training.dataset.RandomAccessDataset
Creates a new instance of RandomAccessDataset with the given necessary configurations.
RandomAccessDataset.BaseBuilder<T extends RandomAccessDataset.BaseBuilder> - Class in ai.djl.training.dataset
The Builder to construct a RandomAccessDataset.
randomColor() - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Returns a random color.
randomMultinomial(int, NDArray) - Method in interface ai.djl.ndarray.NDManager
Draw samples from a multinomial distribution.
randomMultinomial(int, NDArray, Shape) - Method in interface ai.djl.ndarray.NDManager
Draw samples from a multinomial distribution.
randomNormal(Shape) - Method in interface ai.djl.ndarray.NDManager
Draws random samples from a normal (Gaussian) distribution with mean 0 and standard deviation 1.
randomNormal(Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Draws random samples from a normal (Gaussian) distribution with mean 0 and standard deviation 1.
randomNormal(float, float, Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Draws random samples from a normal (Gaussian) distribution.
RandomSampler - Class in ai.djl.training.dataset
RandomSampler is an implementation of the Sampler.SubSampler interface.
RandomSampler() - Constructor for class ai.djl.training.dataset.RandomSampler
Creates a new instance of RandomSampler.
RandomSampler(int) - Constructor for class ai.djl.training.dataset.RandomSampler
Creates a new instance of RandomSampler with the given seed.
randomSplit(int...) - Method in class ai.djl.training.dataset.RandomAccessDataset
Splits the dataset set into multiple portions.
randomUniform(float, float, Shape) - Method in interface ai.djl.ndarray.NDManager
Draws samples from a uniform distribution.
randomUniform(float, float, Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Draws samples from a uniform distribution.
readFileToArray(NDManager, Path) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Reads an image file and converts to NDArray.
readFileToArray(NDManager, Path, NDImageUtils.Flag) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Reads an image file and converts to NDArray.
readInputShapes(DataInputStream) - Method in class ai.djl.nn.AbstractBlock
 
readParameters(Map<String, String>) - Method in class ai.djl.BaseModel
 
Record - Class in ai.djl.training.dataset
Record represents a single element of data and labels from Dataset.
Record(NDList, NDList) - Constructor for class ai.djl.training.dataset.Record
Creates a new instance of Record with a single element of data and its corresponding labels.
Rectangle - Class in ai.djl.modality.cv.output
A Rectangle specifies an area in a coordinate space that is enclosed by the Rectangle object's upper-left point Point in the coordinate space, its width, and its height.
Rectangle(double, double, double, double) - Constructor for class ai.djl.modality.cv.output.Rectangle
Constructs a new Rectangle whose upper-left corner is specified as (x,y) and whose width and height are specified by the arguments of the same name.
Rectangle(Point, double, double) - Constructor for class ai.djl.modality.cv.output.Rectangle
Constructs a new Rectangle whose upper-left corner is specified as coordinate point and whose width and height are specified by the arguments of the same name.
RecurrentCell - Class in ai.djl.nn.recurrent
Applies recurrent layers to input data.
RecurrentCell(RecurrentCell.BaseBuilder<?>) - Constructor for class ai.djl.nn.recurrent.RecurrentCell
Creates a RecurrentCell object.
RecurrentCell.BaseBuilder<T extends RecurrentCell.BaseBuilder> - Class in ai.djl.nn.recurrent
The Builder to construct a RecurrentCell type of Block.
relu(NDArray) - Static method in class ai.djl.nn.Activation
Applies ReLU activation on the input NDArray.
relu(NDList) - Static method in class ai.djl.nn.Activation
Applies ReLU activation on the input singleton NDList.
reluBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the ReLU activation function in its forward function.
remove(String) - Method in class ai.djl.ndarray.NDList
Removes the first occurrence of the specified element from this NDList if it is present.
removeLastBlock() - Method in class ai.djl.nn.SequentialBlock
Removes the Block added last from the sequence of blocks.
removeLastBlock() - Method in interface ai.djl.nn.SymbolBlock
Removes the last block in the symbolic graph.
repeat(long) - Method in interface ai.djl.ndarray.NDArray
Repeats element of this NDArray the number of times given repeats.
repeat(int, long) - Method in interface ai.djl.ndarray.NDArray
Repeats element of this NDArray the number of times given repeats along given axis.
repeat(long[]) - Method in interface ai.djl.ndarray.NDArray
Repeats element of this NDArray the number of times given repeats along each axis.
repeat(Shape) - Method in interface ai.djl.ndarray.NDArray
Repeats element of this NDArray to match the desired shape.
replaceLastBlock(Block) - Method in class ai.djl.nn.SequentialBlock
Replaces the Block last added from the sequence of blocks, and adds the given block.
requireGradient() - Method in class ai.djl.nn.Parameter
Returns whether this parameter needs gradients to be computed.
rescaleGrad - Variable in class ai.djl.training.optimizer.Optimizer
 
reservedTokens - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
 
reset(String, long, String) - Method in class ai.djl.training.util.ProgressBar
resetAccumulator(String) - Method in class ai.djl.training.evaluator.AbstractAccuracy
 
resetAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError
 
resetAccumulator(String) - Method in class ai.djl.training.evaluator.Evaluator
Resets the evaluator value with the given key.
resetAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss
Resets the evaluator value with the given key.
resetAccumulator(String) - Method in class ai.djl.training.loss.Loss
Resets the evaluator value with the given key.
reshape(long...) - Method in interface ai.djl.ndarray.NDArray
Reshapes this NDArray to the given Shape.
reshape(Shape) - Method in interface ai.djl.ndarray.NDArray
Reshapes this NDArray to the given Shape.
reshapeLike(NDArray) - Method in interface ai.djl.ndarray.NDArray
Reshapes this NDArray to the match the Shape of the given NDArray.
Resize - Class in ai.djl.modality.cv.transform
A Transform that resizes the image.
Resize(int) - Constructor for class ai.djl.modality.cv.transform.Resize
Creates a Resize Transform that resizes to the given size.
Resize(int, int) - Constructor for class ai.djl.modality.cv.transform.Resize
Creates a Resize Transform that resizes to the given width and height.
resize(NDArray, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Resizes an image to the given size.
resize(NDArray, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Resizes an image to the given width and height.
resources - Variable in class ai.djl.ndarray.BaseNDManager
 
results - Variable in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
 
RNN - Class in ai.djl.nn.recurrent
Applies a single-gate recurrent layer to input.
RNN.Activation - Enum in ai.djl.nn.recurrent
An enum that enumerates the type of activation.
RNN.Builder - Class in ai.djl.nn.recurrent
The Builder to construct a RNN type of Block.
round() - Method in interface ai.djl.ndarray.NDArray
Returns the round of this NDArray element-wise.

S

sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.BatchSampler
Fetches an iterator that iterates through the given RandomAccessDataset in mini-batches of indices.
sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.RandomSampler
Fetches an iterator that iterates through the indices of the given RandomAccessDataset.
sample(RandomAccessDataset) - Method in interface ai.djl.training.dataset.Sampler
Fetches an iterator that iterates through the given RandomAccessDataset in mini-batches of indices.
sample(RandomAccessDataset) - Method in interface ai.djl.training.dataset.Sampler.SubSampler
Fetches an iterator that iterates through the indices of the given RandomAccessDataset.
sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.SequenceSampler
Fetches an iterator that iterates through the indices of the given RandomAccessDataset.
sampler - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
sampler - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
Sampler - Interface in ai.djl.training.dataset
An interface for sampling data items from a RandomAccessDataset.
Sampler.SubSampler - Interface in ai.djl.training.dataset
An interface that samples a single data item at a time.
save(Path, String) - Method in class ai.djl.BaseModel
Saves the model to the specified modelPath with the name provided.
save(Path, String) - Method in interface ai.djl.Model
Saves the model to the specified modelPath with the name provided.
save(DataOutputStream) - Method in class ai.djl.nn.Parameter
Writes the parameter NDArrays to the given output stream.
saveInputShapes(DataOutputStream) - Method in class ai.djl.nn.AbstractBlock
 
saveParameters(DataOutputStream) - Method in interface ai.djl.nn.Block
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.convolutional.Convolution
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Embedding
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Linear
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Prelu
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.LambdaBlock
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.norm.BatchNorm
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.norm.Dropout
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.ParallelBlock
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.recurrent.LSTM
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.recurrent.RecurrentCell
Writes the parameters of the block to the given outputStream.
saveParameters(DataOutputStream) - Method in class ai.djl.nn.SequentialBlock
Writes the parameters of the block to the given outputStream.
self() - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder
self() - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
 
self() - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
self() - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder
self() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
self() - Method in class ai.djl.nn.convolutional.Conv1D.Builder
self() - Method in class ai.djl.nn.convolutional.Conv2D.Builder
self() - Method in class ai.djl.nn.convolutional.Conv3D.Builder
self() - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
self() - Method in class ai.djl.nn.recurrent.GRU.Builder
self() - Method in class ai.djl.nn.recurrent.LSTM.Builder
self() - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
self() - Method in class ai.djl.nn.recurrent.RNN.Builder
self() - Method in class ai.djl.training.dataset.ArrayDataset.Builder
Returns this {code Builder} object.
self() - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Returns this {code Builder} object.
self() - Method in class ai.djl.training.optimizer.Adam.Builder
self() - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder
self() - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder
 
self() - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder
self() - Method in class ai.djl.training.optimizer.Nag.Builder
self() - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
 
self() - Method in class ai.djl.training.optimizer.Sgd.Builder
selu(NDArray) - Static method in class ai.djl.nn.Activation
Applies Scaled ELU activation on the input NDArray.
selu(NDList) - Static method in class ai.djl.nn.Activation
Applies Scaled ELU activation on the input singleton NDList.
seluBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the SELU activation function in its forward function.
SEMANTIC_SEGMENTATION - Static variable in interface ai.djl.Application.CV
 
SentenceLengthNormalizer - Class in ai.djl.modality.nlp.preprocess
SentenceLengthNormalizer normalizes the length of all the input sentences to the specified number of tokens.
SentenceLengthNormalizer() - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer
Creates a TextProcessor that normalizes the length of the input.
SentenceLengthNormalizer(int, boolean) - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer
Creates a TextProcessor that normalizes the length of the input to the given sentence length.
SentenceLengthNormalizer(int, boolean, String, String, String) - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer
Creates a TextProcessor that normalizes the length of the input to the given sentence length.
sentences - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
 
SequenceSampler - Class in ai.djl.training.dataset
SequenceSampler is an implementation of the Sampler.SubSampler interface.
SequenceSampler() - Constructor for class ai.djl.training.dataset.SequenceSampler
 
SequentialBlock - Class in ai.djl.nn
SequentialBlock is a Block whose children form a chain of blocks with each child block feeding its output to the next.
SequentialBlock() - Constructor for class ai.djl.nn.SequentialBlock
 
set(Buffer) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from Buffer.
set(float[]) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from an array of floats.
set(int[]) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from an array of ints.
set(double[]) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from an array of doubles.
set(long[]) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from an array of longs.
set(byte[]) - Method in interface ai.djl.ndarray.NDArray
Sets this NDArray value from an array of bytes.
set(NDIndex, NDArray) - Method in interface ai.djl.ndarray.NDArray
Sets the specified index in this NDArray with the given values.
set(NDIndex, Number) - Method in interface ai.djl.ndarray.NDArray
Sets the specified index in this NDArray with the given value.
setActivation(RNN.Activation) - Method in class ai.djl.nn.recurrent.RNN.Builder
Sets the activation for the RNN - ReLu or Tanh.
setArray(NDArray) - Method in class ai.djl.nn.Parameter
Sets the values of this Parameter.
setAttachment(String, Object) - Method in interface ai.djl.translate.TranslatorContext
Set a key-value pair of attachments.
setBlock(Block) - Method in class ai.djl.BaseModel
Sets the block for the Model for training and inference.
setBlock(Block) - Method in interface ai.djl.Model
Sets the block for the Model for training and inference.
setClasses(List<String>) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Sets the class list.
setData(NDArray...) - Method in class ai.djl.training.dataset.ArrayDataset.Builder
Sets the data as an NDArray for the ArrayDataset.
setDataType(DataType) - Method in class ai.djl.BaseModel
Sets the standard data type used within the model.
setDataType(DataType) - Method in interface ai.djl.Model
Sets the standard data type used within the model.
setDataType(DataType) - Method in class ai.djl.ndarray.types.DataDesc
Sets the DataType of the NDArray.
setEmbeddingSize(int) - Method in class ai.djl.nn.core.Embedding.Builder
Sets the size of the embeddings.
setInitializer(Initializer) - Method in class ai.djl.nn.AbstractBlock
Sets an Initializer to the block.
setInitializer(Initializer, String) - Method in class ai.djl.nn.AbstractBlock
Sets an Initializer to the specified direct parameter of the block, overriding the initializer of the parameter, if already set.
setInitializer(Initializer) - Method in interface ai.djl.nn.Block
Sets an Initializer to the block.
setInitializer(Initializer, String) - Method in interface ai.djl.nn.Block
Sets an Initializer to the specified direct parameter of the block, overriding the initializer of the parameter, if already set.
setInitializer(Initializer, boolean) - Method in class ai.djl.nn.Parameter
Sets the Initializer for this Parameter, if not already set.
setItems(Collection<T>) - Method in class ai.djl.nn.core.Embedding.Builder
Sets the collection of items that should feature embeddings.
setKernel(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the shape of the kernel.
setLearningRateTracker(LearningRateTracker) - Method in class ai.djl.training.optimizer.Nag.Builder
Sets the LearningRateTracker for this optimizer.
setLearningRateTracker(LearningRateTracker) - Method in class ai.djl.training.optimizer.Sgd.Builder
Sets the LearningRateTracker for this optimizer.
setMandatoryDataType(DataType) - Method in class ai.djl.nn.Parameter
Sets the mandatory data type for this Parameter.
setMetrics(Metrics) - Method in class ai.djl.inference.BasePredictor
Attaches a Metrics param to use for benchmark.
setMetrics(Metrics) - Method in interface ai.djl.inference.Predictor
Attaches a Metrics param to use for benchmark.
setMetrics(Metrics) - Method in interface ai.djl.training.Trainer
Attaches a Metrics param to use for benchmarking.
setModelDir(Path) - Method in class ai.djl.BaseModel
 
setModelName(String) - Method in class ai.djl.BaseModel
 
setMomentum(float) - Method in class ai.djl.training.optimizer.Nag.Builder
Sets the momentum for Nag.
setName(String) - Method in interface ai.djl.ndarray.NDArray
Sets name of this NDArray.
setName(String) - Method in class ai.djl.ndarray.types.DataDesc
Sets the name of the NDArray.
setNumFilters(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Sets the Required number of filters.
setNumStackedLayers(int) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the Required number of stacked layers.
setOutChannels(long) - Method in class ai.djl.nn.core.Linear.Builder
Sets the number of output channels.
setParameterServer(ParameterServer, Device[]) - Method in class ai.djl.training.ParameterStore
Sets the parameterServer used to apply updates to the parameters.
setPipeline(Pipeline) - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder
Sets the Pipeline to use for pre-processing the image.
setProperty(String, String) - Method in class ai.djl.BaseModel
Sets a property to the model.
setProperty(String, String) - Method in interface ai.djl.Model
Sets a property to the model.
setRandomSeed(int) - Method in class ai.djl.engine.Engine
Seeds the random number generator in DJL Engine.
setRatios(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Sets the aspect ratios of the anchor boxes to be generated around each pixel.
setRescaleGrad(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder
Sets the value used to rescale the gradient.
setSampling(int, boolean) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Sampler with the given batch size.
setSampling(int, boolean, boolean) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Sampler with the given batch size.
setSampling(Sampler) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
Sets the Sampler for the dataset.
setScalar(NDIndex, Number) - Method in interface ai.djl.ndarray.NDArray
Sets the specified index in this NDArray with the given value.
setSequenceLength(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the optional parameter that indicates whether to include an extra input parameter sequence_length to specify variable length sequence.
setShape(Shape) - Method in class ai.djl.ndarray.types.DataDesc
Sets the Shape of the NDArray.
setSizes(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder
Sets the sizes of the anchor boxes to be generated around each pixel.
setStateSize(int) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
Sets the Required size of the state for each layer.
setStep(int) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder
Sets the number of steps after which the multiplicative factor must be applied once.
setSteps(int[]) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder
Sets an array of integers indicating when the learning rate should be changed, usually in an uneven interval of steps.
setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder
Sets the name of the synset file listing the potential classes for an image.
setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder
Sets the name for the synset.
setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder
Sets the name for the synset.
setType(Class<T>) - Method in class ai.djl.nn.core.Embedding.Builder
Creates a new Conv2D.Builder with the specified embedding type.
sgd() - Static method in class ai.djl.training.optimizer.Optimizer
Returns a new instance of Sgd.Builder that can build an Sgd optimizer.
Sgd - Class in ai.djl.training.optimizer
Sgd is a Stochastic Gradient Descent (SDG) optimizer.
Sgd(Sgd.Builder) - Constructor for class ai.djl.training.optimizer.Sgd
Creates a new instance of Sgd.
Sgd.Builder - Class in ai.djl.training.optimizer
The Builder to construct an Sgd object.
Shape - Class in ai.djl.ndarray.types
A class that presents the NDArray's shape information.
Shape(long...) - Constructor for class ai.djl.ndarray.types.Shape
Constructs and initializes a Shape with specified dimension as {@code (long...
Shape(List<Long>) - Constructor for class ai.djl.ndarray.types.Shape
Constructs and initializes a Shape with specified dimension.
Shape(PairList<Long, LayoutType>) - Constructor for class ai.djl.ndarray.types.Shape
Constructs and initializes a Shape with specified shape and layout pairList.
Shape(long[], String) - Constructor for class ai.djl.ndarray.types.Shape
Constructs and initializes a Shape with specified dimension and layout.
Shape(long[], LayoutType[]) - Constructor for class ai.djl.ndarray.types.Shape
Constructs and initializes a Shape with specified dimension and layout.
shapeEquals(NDArray) - Method in interface ai.djl.ndarray.NDArray
Checks 2 NDArrays for equal shapes.
shapeEquals(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Checks 2 NDArrays for equal shapes.
sigmoid(NDArray) - Static method in class ai.djl.nn.Activation
Applies Sigmoid activation on the input NDArray.
sigmoid(NDList) - Static method in class ai.djl.nn.Activation
Applies Sigmoid activation on the input singleton NDList.
sigmoidBinaryCrossEntropyLoss() - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SigmoidBinaryCrossEntropyLoss with default arguments.
sigmoidBinaryCrossEntropyLoss(String) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SigmoidBinaryCrossEntropyLoss with default arguments.
sigmoidBinaryCrossEntropyLoss(String, float, boolean) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SigmoidBinaryCrossEntropyLoss with the given arguments.
SigmoidBinaryCrossEntropyLoss - Class in ai.djl.training.loss
SigmoidBinaryCrossEntropyLoss is a type of Loss.
SigmoidBinaryCrossEntropyLoss() - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss
Performs Sigmoid cross-entropy loss for binary classification.
SigmoidBinaryCrossEntropyLoss(String) - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss
Performs Sigmoid cross-entropy loss for binary classification.
SigmoidBinaryCrossEntropyLoss(String, float, boolean) - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss
Performs Sigmoid cross-entropy loss for binary classification.
sigmoidBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the Sigmoid activation function in its forward function.
SimpleCompositeLoss - Class in ai.djl.training.loss
SimpleCompositeLoss is an implementation of the Loss abstract class that can combine different Loss functions by adding the individual losses together.
SimpleCompositeLoss() - Constructor for class ai.djl.training.loss.SimpleCompositeLoss
Creates a new empty instance of CompositeLoss that can combine the given Loss components.
SimpleCompositeLoss(String) - Constructor for class ai.djl.training.loss.SimpleCompositeLoss
Creates a new empty instance of CompositeLoss that can combine the given Loss components.
SimplePoseTranslator - Class in ai.djl.modality.cv.translator
A ImageTranslator that post-process the NDArray into human Joints.
SimplePoseTranslator(SimplePoseTranslator.Builder) - Constructor for class ai.djl.modality.cv.translator.SimplePoseTranslator
Creates the Pose Estimation translator from the given builder.
SimplePoseTranslator.Builder - Class in ai.djl.modality.cv.translator
The builder for Pose Estimation translator.
SimpleTokenizer - Class in ai.djl.modality.nlp.preprocess
SimpleTokenizer is an implementation of the Tokenizer interface that converts sentences into token by splitting them by a given delimiter.
SimpleTokenizer(String) - Constructor for class ai.djl.modality.nlp.preprocess.SimpleTokenizer
Creates an instance of SimpleTokenizer with the given delimiter.
SimpleTokenizer() - Constructor for class ai.djl.modality.nlp.preprocess.SimpleTokenizer
Creates an instance of SimpleTokenizer with the default delimiter.
sin() - Method in interface ai.djl.ndarray.NDArray
Returns the trigonometric sine of this NDArray element-wise.
SingleShotDetectionAccuracy - Class in ai.djl.training.evaluator
SingleShotDetectionAccuracy is an implementation of AbstractAccuracy.
SingleShotDetectionAccuracy(String) - Constructor for class ai.djl.training.evaluator.SingleShotDetectionAccuracy
Creates a new instance of SingleShotDetectionAccuracy with the given name.
SingleShotDetectionLoss - Class in ai.djl.training.loss
SingleShotDetectionLoss is an implementation of Loss.
SingleShotDetectionLoss() - Constructor for class ai.djl.training.loss.SingleShotDetectionLoss
Base class for metric with abstract update methods.
SingleShotDetectionTranslator - Class in ai.djl.modality.cv.translator
A ImageTranslator that post-process the NDArray into DetectedObjects with boundaries.
SingleShotDetectionTranslator(SingleShotDetectionTranslator.Builder) - Constructor for class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
Creates the SSD translator from the given builder.
SingleShotDetectionTranslator.Builder - Class in ai.djl.modality.cv.translator
The builder for SSD translator.
singletonOrThrow() - Method in class ai.djl.ndarray.NDList
Returns the only element if this is a singleton NDList or throws an exception if multiple elements.
sinh() - Method in interface ai.djl.ndarray.NDArray
Returns the hyperbolic sine of this NDArray element-wise.
size() - Method in class ai.djl.modality.nlp.Vocabulary
Returns the size of the Vocabulary.
size(int) - Method in interface ai.djl.ndarray.NDArray
Returns the size of this NDArray along a given axis.
size() - Method in interface ai.djl.ndarray.NDArray
Returns the total number of elements in this NDArray.
size(int...) - Method in class ai.djl.ndarray.types.Shape
Returns the size of a specific dimension or several specific dimensions.
size() - Method in class ai.djl.ndarray.types.Shape
Returns the total size.
size() - Method in class ai.djl.training.dataset.RandomAccessDataset
Returns the size of this Dataset.
slice(int) - Method in class ai.djl.ndarray.types.Shape
Creates a new Shape whose content is a slice of this shape.
slice(int, int) - Method in class ai.djl.ndarray.types.Shape
Creates a new Shape whose content is a slice of this shape.
softmax(int) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function along the given axis.
softmax(int, float) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function along the given axis.
softmax(int[]) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function along the given axes.
softmax(int[], float) - Method in interface ai.djl.ndarray.NDArray
Applies the softmax function along the given axes.
softmaxCrossEntropyLoss() - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SoftmaxCrossEntropyLoss with default arguments.
softmaxCrossEntropyLoss(String) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SoftmaxCrossEntropyLoss with default arguments.
softmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Static method in class ai.djl.training.loss.Loss
Returns a new instance of SoftmaxCrossEntropyLoss with the given arguments.
SoftmaxCrossEntropyLoss - Class in ai.djl.training.loss
SoftmaxCrossEntropyLoss is a type of Loss that calculates the softmax cross entropy loss.
SoftmaxCrossEntropyLoss() - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss
Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.
SoftmaxCrossEntropyLoss(String) - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss
Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.
SoftmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss
Creates a new instance of SoftmaxCrossEntropyLoss with the given parameters.
softrelu(NDArray) - Static method in class ai.djl.nn.Activation
Applies soft ReLU activation on the input NDArray.
softrelu(NDList) - Static method in class ai.djl.nn.Activation
Applies soft ReLU activation on the input singleton NDList.
softreluBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the SoftReLU activation function in its forward function.
sort() - Method in interface ai.djl.ndarray.NDArray
Sorts the flattened NDArray.
sort(int) - Method in interface ai.djl.ndarray.NDArray
Sorts the flattened NDArray.
SparseFormat - Enum in ai.djl.ndarray.types
An enum representing Sparse matrix storage formats.
SparseNDArray - Interface in ai.djl.ndarray
An interface representing a Sparse NDArray.
split(long) - Method in interface ai.djl.ndarray.NDArray
Splits this NDArray into multiple subNDArrays given sections along first axis.
split(long[]) - Method in interface ai.djl.ndarray.NDArray
Splits this NDArray into multiple sub-NDArrays given indices along first axis.
split(long, int) - Method in interface ai.djl.ndarray.NDArray
Splits this NDArray into multiple subNDArrays given sections along the given axis.
split(long[], int) - Method in interface ai.djl.ndarray.NDArray
Splits this NDArray into multiple sub-NDArrays given indices along given axis.
split(Device[], boolean) - Method in class ai.djl.training.dataset.Batch
Splits the data and labels in the Batch across the given devices.
split(NDList, int, boolean) - Method in interface ai.djl.translate.Batchifier
Splits the given NDList into the given number of slices.
split(NDList, int, boolean) - Method in class ai.djl.translate.StackBatchifier
Splits the given NDList into the given number of slices.
sqrt() - Method in interface ai.djl.ndarray.NDArray
Returns the square root of this NDArray element-wise.
square() - Method in interface ai.djl.ndarray.NDArray
Returns the square of this NDArray element-wise.
squeeze() - Method in interface ai.djl.ndarray.NDArray
Removes all singleton dimensions from this NDArray Shape.
squeeze(int) - Method in interface ai.djl.ndarray.NDArray
Removes a singleton dimension at the given axis.
squeeze(int[]) - Method in interface ai.djl.ndarray.NDArray
Removes singleton dimensions at the given axes.
stack(NDArray) - Method in interface ai.djl.ndarray.NDArray
Joins a NDArray along the first axis.
stack(NDArray, int) - Method in interface ai.djl.ndarray.NDArray
Joins a NDArray along a new axis.
stack(NDList) - Static method in class ai.djl.ndarray.NDArrays
Joins a sequence of NDArrays in NDList along the first axis.
stack(NDList, int) - Static method in class ai.djl.ndarray.NDArrays
Joins a sequence of NDArrays in NDList along a new axis.
STACK - Static variable in interface ai.djl.translate.Batchifier
 
StackBatchifier - Class in ai.djl.translate
StackBatchifier is used to merge a list of samples to form a mini-batch of NDArray(s).
StackBatchifier() - Constructor for class ai.djl.translate.StackBatchifier
 
StandardCapabilities - Class in ai.djl.engine
Constant definitions for the standard capability.
start(long) - Method in class ai.djl.training.util.ProgressBar
stateOutputs - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
stateOutputs - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
stateShape - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
stateSize - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
stateSize - Variable in class ai.djl.nn.recurrent.RecurrentCell
 
step() - Method in interface ai.djl.training.Trainer
Updates all of the parameters of the model once.
stream() - Method in class ai.djl.ndarray.index.NDIndex
Returns a stream of the NDIndexElements.
stream() - Method in class ai.djl.ndarray.types.Shape
Returns a stream of the Shape.
stride - Variable in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
 
stride - Variable in class ai.djl.nn.convolutional.Convolution
 
sub(Number) - Method in interface ai.djl.ndarray.NDArray
Subtracts a number from this NDArray element-wise.
sub(NDArray) - Method in interface ai.djl.ndarray.NDArray
Subtracts the other NDArray from this NDArray element-wise.
sub(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a number from the NDArray element-wise.
sub(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a NDArray from a number element-wise.
sub(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a NDArray from a NDArray element-wise.
subi(Number) - Method in interface ai.djl.ndarray.NDArray
Subtracts a number from this NDArray element-wise in place.
subi(NDArray) - Method in interface ai.djl.ndarray.NDArray
Subtracts the other NDArray from this NDArray element-wise in place.
subi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a number from the NDArray element-wise in place.
subi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a NDArray from a number element-wise in place.
subi(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Subtracts a NDArray from a NDArray element-wise in place.
subNDList(int) - Method in class ai.djl.ndarray.NDList
Returns a view of the portion of this NDList between the specified fromIndex, inclusive, and to the end.
sum() - Method in interface ai.djl.ndarray.NDArray
Returns the sum of this NDArray.
sum(int[]) - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray along given axes.
sum(int[], boolean) - Method in interface ai.djl.ndarray.NDArray
Returns the minimum of this NDArray along given axes.
swapAxes(int, int) - Method in interface ai.djl.ndarray.NDArray
Interchanges two axes of this NDArray.
swish(NDArray, float) - Static method in class ai.djl.nn.Activation
Applies Swish activation on the input NDArray.
swish(NDList, float) - Static method in class ai.djl.nn.Activation
Applies SWish activation on the input singleton NDList.
swishBlock(float) - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the Swish activation function in its forward function.
SymbolBlock - Interface in ai.djl.nn
SymbolBlock is a Block is used to load models that were exported directly from the engine in its native format.
sync() - Method in class ai.djl.training.ParameterStore
Synchronizes the values on all mirrors with the main parameter.
synsetArtifactName - Variable in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
 

T

tan() - Method in interface ai.djl.ndarray.NDArray
Returns the trigonometric tangent of this NDArray element-wise.
tanh() - Method in interface ai.djl.ndarray.NDArray
Returns the hyperbolic tangent of this NDArray element-wise.
tanh(NDArray) - Static method in class ai.djl.nn.Activation
Applies Tanh activation on the input NDArray.
tanh(NDList) - Static method in class ai.djl.nn.Activation
Applies Tanh activation on the input singleton NDList.
tanhBlock() - Static method in class ai.djl.nn.Activation
Creates a LambdaBlock that applies the Tanh activation function in its forward function.
target(NDList) - Method in class ai.djl.modality.cv.MultiBoxTarget
Computes multi-box training targets.
targetPipeline - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder
 
targetPipeline - Variable in class ai.djl.training.dataset.RandomAccessDataset
 
TEXT_CLASSIFICATION - Static variable in interface ai.djl.Application.NLP
 
TextProcessor - Interface in ai.djl.modality.nlp.preprocess
TextProcessor allows applying pre-processing to input tokens for natural language applications.
threshold - Variable in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator
 
tile(long) - Method in interface ai.djl.ndarray.NDArray
Constructs a NDArray by repeating this NDArray the number of times given repeats.
tile(int, long) - Method in interface ai.djl.ndarray.NDArray
Constructs a NDArray by repeating this NDArray the number of times given by repeats along given axis.
tile(long[]) - Method in interface ai.djl.ndarray.NDArray
Constructs a NDArray by repeating this NDArray the number of times given by repeats.
tile(Shape) - Method in interface ai.djl.ndarray.NDArray
Constructs a NDArray by repeating this NDArray the number of times to match the desired shape.
TimeMeasureTrainingListener - Class in ai.djl.training.listener
TrainingListener that outputs the training time metrics after training is done.
TimeMeasureTrainingListener(String) - Constructor for class ai.djl.training.listener.TimeMeasureTrainingListener
toArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a Number array based on its DataType.
toBooleanArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a boolean array.
toByteArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a byte array.
toByteBuffer() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a ByteBuffer.
toDebugString(int, int, int, int) - Method in interface ai.djl.ndarray.NDArray
Runs the debug string representation of this NDArray.
toDegrees() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray from radians to degrees element-wise.
toDense() - Method in interface ai.djl.ndarray.NDArray
Returns a dense representation of the sparse NDArray.
toDevice(Device, boolean) - Method in interface ai.djl.ndarray.NDArray
Moves this NDArray to a different Device.
toDeviceType(Device) - Static method in interface ai.djl.DeviceType
Map device to its type number.
toDoubleArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a double array.
toFloatArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a float array.
toIntArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to an int array.
tokenize(String) - Method in class ai.djl.modality.nlp.preprocess.SimpleTokenizer
 
tokenize(String) - Method in interface ai.djl.modality.nlp.preprocess.Tokenizer
Breaks down the given sentence into a list of tokens that can be represented by embeddings.
Tokenizer - Interface in ai.djl.modality.nlp.preprocess
Tokenizer interface provides the ability to break-down sentences into embeddable tokens.
toLayoutString() - Method in class ai.djl.ndarray.types.Shape
Returns the string layout type for each axis in this shape.
toLongArray() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a long array.
toNDArray(NDManager, BufferedImage) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Converts a BufferedImage to an RGB NDArray.
toNDArray(NDManager, BufferedImage, NDImageUtils.Flag) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils
Converts a BufferedImage to an NDArray with designated color mode.
topK(int) - Method in class ai.djl.modality.Classifications
Returns a list of the top k best classes.
TopKAccuracy - Class in ai.djl.training.evaluator
TopKAccuracy is an Evaluator that computes the accuracy of the top k predictions.
TopKAccuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy
Creates a TopKAccuracy instance.
TopKAccuracy(int, int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy
Creates an instance of TopKAccuracy evaluator that computes topK accuracy across axis 1 along the given index.
TopKAccuracy(int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy
Creates an instance of TopKAccuracy evaluator that computes topK accuracy across axis 1 along the 0th index.
toRadians() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray from degrees to radians element-wise.
toSparse(SparseFormat) - Method in interface ai.djl.ndarray.NDArray
Returns a sparse representation of NDArray.
toString() - Method in class ai.djl.Device
toString() - Method in class ai.djl.metric.Metric
toString() - Method in class ai.djl.modality.Classifications.Classification
toString() - Method in class ai.djl.modality.Classifications
toString() - Method in class ai.djl.modality.cv.output.DetectedObjects.DetectedObject
toString() - Method in class ai.djl.modality.cv.output.Joints.Joint
toString() - Method in class ai.djl.modality.cv.output.Joints
toString() - Method in class ai.djl.modality.cv.output.Rectangle
toString() - Method in class ai.djl.ndarray.BaseNDManager
toString() - Method in class ai.djl.ndarray.NDList
toString() - Method in class ai.djl.ndarray.types.DataDesc
toString() - Method in enum ai.djl.ndarray.types.DataType
toString(LayoutType[]) - Static method in enum ai.djl.ndarray.types.LayoutType
Converts a layout type array to a string of the character representations.
toString() - Method in class ai.djl.ndarray.types.Shape
toString() - Method in class ai.djl.nn.LambdaBlock
toString() - Method in class ai.djl.nn.ParallelBlock
toString() - Method in class ai.djl.nn.ParameterBlock
toString() - Method in class ai.djl.nn.SequentialBlock
toString() - Method in class ai.djl.training.hyperparameter.param.HpCategorical
toString() - Method in class ai.djl.training.hyperparameter.param.HpFloat
toString() - Method in class ai.djl.training.hyperparameter.param.HpInt
toString() - Method in class ai.djl.training.hyperparameter.param.HpSet
toString() - Method in class ai.djl.training.hyperparameter.param.HpVal
totalInstances - Variable in class ai.djl.training.evaluator.Evaluator
 
ToTensor - Class in ai.djl.modality.cv.transform
A Transform that converts an image NDArray from preprocessing format to Neural Network format.
ToTensor() - Constructor for class ai.djl.modality.cv.transform.ToTensor
 
toTensor(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils
Converts an image NDArray from preprocessing format to Neural Network format.
toType(DataType, boolean) - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a different DataType.
toUint8Array() - Method in interface ai.djl.ndarray.NDArray
Converts this NDArray to a uint8 array.
trace() - Method in interface ai.djl.ndarray.NDArray
Returns the sum along diagonals of this NDArray.
trace(int) - Method in interface ai.djl.ndarray.NDArray
Returns the sum along diagonals of this NDArray.
trace(int, int, int) - Method in interface ai.djl.ndarray.NDArray
Returns the sum along diagonals of this NDArray.
TRAIN_ALL - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener
 
TRAIN_EPOCH - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener
 
TRAIN_PROGRESS - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener
 
trainBatch(Batch) - Method in interface ai.djl.training.Trainer
Trains the model with one iteration of the given Batch of data.
Trainer - Interface in ai.djl.training
The Trainer interface provides a session for model training.
TrainingConfig - Interface in ai.djl.training
An interface that is responsible for holding the configuration required by Trainer.
TrainingDivergedException - Exception in ai.djl
Thrown to indicate when there is a divergence during Training.
TrainingDivergedException(String) - Constructor for exception ai.djl.TrainingDivergedException
Constructs a new exception with the specified detail message.
TrainingDivergedException(String, Throwable) - Constructor for exception ai.djl.TrainingDivergedException
Constructs a new exception with the specified detail message and cause.
TrainingDivergedException(Throwable) - Constructor for exception ai.djl.TrainingDivergedException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.
TrainingListener - Interface in ai.djl.training.listener
TrainingListener offers an interface that allows performing some actions when certain events have occurred in the Trainer.
TrainingListener.BatchData - Class in ai.djl.training.listener
A class to pass data from the batch into the training listeners.
TrainingListener.Defaults - Interface in ai.djl.training.listener
Contains default TrainingListener sets.
transform(NDArray) - Method in class ai.djl.modality.cv.transform.CenterCrop
Applies the Transform to the given NDArray.
transform(NDArray) - Method in class ai.djl.modality.cv.transform.Crop
Applies the Transform to the given NDArray.
transform(NDArray) - Method in class ai.djl.modality.cv.transform.Normalize
Applies the Transform to the given NDArray.
transform(NDArray) - Method in class ai.djl.modality.cv.transform.Resize
Applies the Transform to the given NDArray.
transform(NDArray) - Method in class ai.djl.modality.cv.transform.ToTensor
Applies the Transform to the given NDArray.
transform(NDArray) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator
Applies the Transform to the given NDArray.
transform(NDList) - Method in class ai.djl.translate.Pipeline
Applies the transforms configured in this object on the input NDList.
Transform - Interface in ai.djl.translate
An interface to apply various transforms to the input.
transform(NDArray) - Method in interface ai.djl.translate.Transform
Applies the Transform to the given NDArray.
TranslateException - Exception in ai.djl.translate
Thrown to indicate that an error is raised during processing of the input or output.
TranslateException(String) - Constructor for exception ai.djl.translate.TranslateException
Constructs a new exception with the specified detail message.
TranslateException(String, Throwable) - Constructor for exception ai.djl.translate.TranslateException
Constructs a new exception with the specified detail message and cause.
TranslateException(Throwable) - Constructor for exception ai.djl.translate.TranslateException
Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.
Translator<I,O> - Interface in ai.djl.translate
The Translator interface provides model pre-processing and postprocessing functionality.
TranslatorContext - Interface in ai.djl.translate
The TranslatorContext interface provides a toolkit for pre-processing and postprocessing functionality.
TranslatorFactory<I,O> - Interface in ai.djl.translate
A utility class creates Translator instances.
transpose() - Method in interface ai.djl.ndarray.NDArray
Returns this NDArray with axes transposed.
transpose(int...) - Method in interface ai.djl.ndarray.NDArray
Returns this NDArray with given axes transposed.
trunc() - Method in interface ai.djl.ndarray.NDArray
Returns the truncated value of this NDArray element-wise.

U

uid - Variable in class ai.djl.ndarray.BaseNDManager
 
unbatchify(NDList) - Method in interface ai.djl.translate.Batchifier
Reverses the batchify operation.
unbatchify(NDList) - Method in class ai.djl.translate.StackBatchifier
Reverses the batchify operation.
unembedWord(NDArray) - Method in interface ai.djl.modality.nlp.WordEmbedding
Returns the closest matching word for a given embedding.
UniformInitializer - Class in ai.djl.training.initializer
UniformInitializer initializes weights with random values uniformly sampled from a given range.
UniformInitializer() - Constructor for class ai.djl.training.initializer.UniformInitializer
Creates an instance of UniformInitializer with a default scale of 0.07.
UniformInitializer(float) - Constructor for class ai.djl.training.initializer.UniformInitializer
Initializes a uniform initializer.
unknownToken - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
 
update(HpSet, float) - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer
Updates the optimizer with the results of a hyperparameter test.
update(HpSet, float) - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer
Updates the optimizer with the results of a hyperparameter test.
update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Adam
Updates the parameters according to the gradients.
update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Nag
Updates the parameters according to the gradients.
update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Optimizer
Updates the parameters according to the gradients.
update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Sgd
Updates the parameters according to the gradients.
update(long, String) - Method in class ai.djl.training.util.ProgressBar
updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.evaluator.AbstractAccuracy
 
updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.evaluator.BoundingBoxError
 
updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.evaluator.Evaluator
Updates the evaluator with the given key based on a NDList of labels and predictions.
updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss
Updates the evaluator with the given key based on a NDList of labels and predictions.
updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.loss.Loss
Updates the evaluator with the given key based on a NDList of labels and predictions.
updateAllParameters() - Method in class ai.djl.training.ParameterStore
Updates all the mirrored parameters.
updateCount(String) - Method in class ai.djl.training.optimizer.Optimizer
 
updateInputLayoutToTNC(NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell
 
UrlTranslator<T> - Class in ai.djl.modality.cv.translator
Built-in Translator that provides image pre-processing from URL.
UrlTranslator(Translator<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.UrlTranslator
Creates a UrlTranslator instance.
UrlTranslatorFactory<T> - Class in ai.djl.modality.cv.translator
A factory class creates Translator that can process image from a URL.
UrlTranslatorFactory(TranslatorFactory<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.UrlTranslatorFactory
Creates a FileTranslatorFactory instance.
useBidirectional - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
useSequenceLength - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder
 
useSequenceLength - Variable in class ai.djl.nn.recurrent.RecurrentCell
 

V

validate() - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder
Validates that the required arguments are set.
VALIDATE_EPOCH - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener
 
validateBatch(Batch) - Method in interface ai.djl.training.Trainer
Validates the given batch of data.
validateInputSize(NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell
 
validateLayout(LayoutType[], LayoutType[]) - Static method in interface ai.djl.nn.Block
Validates that actual layout matches the expected layout.
valueOf(String) - Static method in enum ai.djl.modality.cv.util.NDImageUtils.Flag
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.ndarray.types.DataType.Format
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.ndarray.types.DataType
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.ndarray.types.LayoutType
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.ndarray.types.SparseFormat
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.nn.ParameterType
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.nn.pooling.PoolingConvention
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.nn.recurrent.RNN.Activation
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.training.dataset.Dataset.Usage
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.training.initializer.XavierInitializer.FactorType
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.training.initializer.XavierInitializer.RandomType
Returns the enum constant of this type with the specified name.
valueOf(String) - Static method in enum ai.djl.training.optimizer.learningrate.WarmUpMode
Returns the enum constant of this type with the specified name.
values() - Static method in enum ai.djl.modality.cv.util.NDImageUtils.Flag
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.ndarray.types.DataType.Format
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.ndarray.types.DataType
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.ndarray.types.LayoutType
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.ndarray.types.SparseFormat
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.nn.ParameterType
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.nn.pooling.PoolingConvention
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.nn.recurrent.RNN.Activation
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.training.dataset.Dataset.Usage
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.training.initializer.XavierInitializer.FactorType
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.training.initializer.XavierInitializer.RandomType
Returns an array containing the constants of this enum type, in the order they are declared.
values() - Static method in enum ai.djl.training.optimizer.learningrate.WarmUpMode
Returns an array containing the constants of this enum type, in the order they are declared.
Vocabulary - Class in ai.djl.modality.nlp
Vocabulary is a collection of tokens.
Vocabulary(Vocabulary.VocabularyBuilder) - Constructor for class ai.djl.modality.nlp.Vocabulary
Create a Vocabulary object with ta Vocabulary.VocabularyBuilder.
Vocabulary.VocabularyBuilder - Class in ai.djl.modality.nlp
Builder class that is used to build the Vocabulary.
VocabularyBuilder() - Constructor for class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder
 
vocabularyContains(String) - Method in interface ai.djl.modality.nlp.WordEmbedding
Returns whether an embedding exists for a word.

W

waitToRead(NDList) - Method in class ai.djl.inference.BasePredictor
 
WarmUpMode - Enum in ai.djl.training.optimizer.learningrate
An enum that enumerates the types of warm-up modes for a LearningRateTracker.
weight - Variable in class ai.djl.nn.convolutional.Convolution
 
where(NDArray, NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays
Returns elements chosen from the NDArray or the other NDArray depending on condition.
withDefaultState(Map<String, Map<Device, NDArray>>, String, Device, Function<String, NDArray>) - Method in class ai.djl.training.optimizer.Optimizer
 
WORD_EMBEDDING - Static variable in interface ai.djl.Application.NLP
 
WordEmbedding - Interface in ai.djl.modality.nlp
A class to manage 1-D NDArray representations of words.

X

XavierInitializer - Class in ai.djl.training.initializer
XavierInitializer is an Initializer that performs "Xavier" initialization for parameters.
XavierInitializer(XavierInitializer.RandomType, XavierInitializer.FactorType, float) - Constructor for class ai.djl.training.initializer.XavierInitializer
Initializes a Xavier initializer.
XavierInitializer() - Constructor for class ai.djl.training.initializer.XavierInitializer
Creates a new instance of XavierInitializer.
XavierInitializer.FactorType - Enum in ai.djl.training.initializer
Enum for different types of factor type.
XavierInitializer.RandomType - Enum in ai.djl.training.initializer
Enum for different types of random distributions.

Z

zeros(Shape) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Shape filled with zeros.
zeros(Shape, DataType) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Shape filled with zeros.
zeros(Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager
Creates an instance of NDArray with specified Device, Shape, and DataType filled with zeros.
ZEROS - Static variable in interface ai.djl.training.initializer.Initializer
 
zerosLike() - Method in interface ai.djl.ndarray.NDArray
Returns an NDArray of zeros with the same Shape, DataType and SparseFormat as the input NDArray.
A B C D E F G H I J K L M N O P Q R S T U V W X Z 
Skip navigation links