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

A

abs() - Method in interface ai.djl.ndarray.NDArray

Returns the absolute value of this NDArray element-wise.

abs() - Method in interface ai.djl.ndarray.NDArrayAdapter

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(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(byte) - Constructor for class ai.djl.nn.AbstractBlock

Builds an empty block with the given version for parameter serialization.

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.

AbstractEmbedding<T> - Interface in ai.djl.nn.core

An Embedding maps elements of type T to a 1-Dimensional representative NDArrays.

AbstractIndexedEmbedding<T> - Interface in ai.djl.nn.core

An AbstractEmbedding where each embedded item can be assigned an integer index.

AbstractRepository - Class in ai.djl.repository

The AbstractRepository is the shared base for implementers of the Repository interface.

AbstractRepository() - Constructor for class ai.djl.repository.AbstractRepository

AbstractSymbolBlock - Class in ai.djl.nn

AbstractSymbolBlock is an abstract implementation of SymbolBlock.

AbstractSymbolBlock(byte) - Constructor for class ai.djl.nn.AbstractSymbolBlock

Builds an empty block with the given version for parameter serialization.

accuracy(NDList, NDList) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelLoss

Calculates the percentage of correctly predicted masked tokens.

accuracy(NDList, NDList) - Method in class ai.djl.nn.transformer.BertNextSentenceLoss

Calculates the fraction of correct predictions.

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

A helper for classes extending AbstractAccuracy.

accuracyHelper(NDList, NDList) - Method in class ai.djl.training.evaluator.TopKAccuracy

A helper for classes extending AbstractAccuracy.

acos() - Method in interface ai.djl.ndarray.NDArray

Returns the inverse trigonometric cosine of this NDArray element-wise.

acos() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

acosh() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the inverse hyperbolic cosine of this NDArray element-wise.

ACTION_RECOGNITION - Static variable in interface ai.djl.Application.CV

An application that accepts an image or video and classifies the action being done in it.

ActionRecognitionModelLoader - Class in ai.djl.modality.cv.zoo

Model loader for Action Recognition models.

ActionRecognitionModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.ActionRecognitionModelLoader

Creates the Model loader from the given repository.

ActionSpace - Class in ai.djl.modality.rl

Contains the available actions that can be taken in an RlEnv.

ActionSpace() - Constructor for class ai.djl.modality.rl.ActionSpace

Activation - Class in ai.djl.nn

Utility class that provides activation functions and blocks.

activation - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Adadelta - Class in ai.djl.training.optimizer

Adadelta is an Adadelta Optimizer.

Adadelta(Adadelta.Builder) - Constructor for class ai.djl.training.optimizer.Adadelta

Creates a new instance of Adadelta.

adadelta() - Static method in class ai.djl.training.optimizer.Optimizer

Returns a new instance of Adadelta.Builder that can build an Adadelta optimizer.

Adadelta.Builder - Class in ai.djl.training.optimizer

The Builder to construct an Adadelta object.

Adagrad - Class in ai.djl.training.optimizer

Adagrad is an AdaGrad Optimizer.

Adagrad(Adagrad.Builder) - Constructor for class ai.djl.training.optimizer.Adagrad

Creates a new instance of Adam optimizer.

adagrad() - Static method in class ai.djl.training.optimizer.Optimizer

Returns a new instance of Adagrad.Builder that can build an Adagrad optimizer.

Adagrad.Builder - Class in ai.djl.training.optimizer

The Builder to construct an Adagrad object.

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.SimpleVocabulary.Builder

Adds the given sentence to the SimpleVocabulary.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Adds a number to this NDArray element-wise.

add(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(long...) - Method in class ai.djl.ndarray.types.Shape

Joins this shape with axes.

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(NDList...) - Method in class ai.djl.nn.transformer.MemoryScope

Adds all arrays in the given lists to this memory scope.

add(NDArray...) - Method in class ai.djl.nn.transformer.MemoryScope

Adds the given arrays to this scopes sub manager.

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

Adds an accumulator for the results of the evaluation with the given key.

addAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError

Adds an accumulator for the results of the evaluation with the given key.

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

Adds an accumulator for the results of the evaluation with the given key.

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.SimpleVocabulary.Builder

Adds the given list of sentences to the SimpleVocabulary.

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 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.

addAllDim() - Method in class ai.djl.ndarray.index.NDIndex

Appends a new index to get all values in the dimension.

addAllDim(int) - Method in class ai.djl.ndarray.index.NDIndex

Appends multiple new index to get all values in the dimension.

addArtifact(Artifact) - Method in class ai.djl.repository.Metadata

Adds one artifact for the metadata.

addBooleanIndex(NDArray) - Method in class ai.djl.ndarray.index.NDIndex

Updates the NDIndex by appending a boolean NDArray.

addChildBlock(String, B) - Method in class ai.djl.nn.AbstractBlock

Use this to add a child block to this block.

addData(byte[]) - Method in class ai.djl.modality.Input

Appends an item at the end of the input.

addData(String, byte[]) - Method in class ai.djl.modality.Input

Adds a key/value pair to the input content.

addData(int, byte[]) - Method in class ai.djl.modality.Input

Inserts the specified element at the specified position in the input.

addEvaluator(Evaluator) - Method in class ai.djl.training.DefaultTrainingConfig

Adds an Evaluator that needs to be computed during training.

addFromCustomizedFile(URL, Function<URL, List<String>>) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Adds a customized vocabulary to the SimpleVocabulary.

addFromTextFile(Path) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Adds a text vocabulary to the SimpleVocabulary.

addFromTextFile(URL) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Adds a text vocabulary to the SimpleVocabulary.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Adds a number to this NDArray element-wise in place.

addi(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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, Object...) - 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.

addLicense(License) - Method in class ai.djl.repository.Metadata

Adds one License.

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.

addMetric(String, long) - Method in class ai.djl.training.Trainer

Helper to add a metric for a time difference.

addPad(int, int, NDArraySupplier) - Method in class ai.djl.translate.PaddingStackBatchifier.Builder

Adds a new dimension to be padded in the input NDList.

addPad(int, int, NDArraySupplier, int) - Method in class ai.djl.translate.PaddingStackBatchifier.Builder

Adds a new dimension to be padded in the input NDList.

addParameter(P) - Method in class ai.djl.nn.AbstractBlock

Adds a parameter to this block.

addParameter(P, Shape) - Method in class ai.djl.nn.AbstractBlock

Adds a parameter to this block.

addParameter(P, Function<Shape[], Shape>) - Method in class ai.djl.nn.AbstractBlock

Adds a parameter to this block.

addPickDim(NDArray) - Method in class ai.djl.ndarray.index.NDIndex

Appends a picking index that gets values by index in the axis.

addProperty(String, String) - Method in class ai.djl.modality.Input

Adds a property to the input.

addProperty(String, String) - Method in class ai.djl.modality.Output

Adds a property to the output.

addSingleton(Function<NDArray, NDArray>) - Method in class ai.djl.nn.SequentialBlock

Adds a LambdaBlock.singleton(Function) that applies the given function to the sequence of blocks.

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.

addStep(RlEnv.Step) - Method in class ai.djl.modality.rl.LruReplayBuffer

Adds a new step to the buffer.

addStep(RlEnv.Step) - Method in interface ai.djl.modality.rl.ReplayBuffer

Adds a new step to the buffer.

addTargetTransform(Transform) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Adds the Transform to the target Pipeline to be applied on the labels.

addTrainingListeners(TrainingListener...) - Method in class ai.djl.training.DefaultTrainingConfig

Adds TrainingListeners for training.

addTransform(Transform) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

Adds the Transform to the Pipeline use for pre-processing the image.

addTransform(Transform) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Adds the Transform to the Pipeline to be applied on the data.

ai.djl - package ai.djl

Contains top level, engine-agnostic classes for both inference and training.

ai.djl.engine - package ai.djl.engine

Contains classes responsible for loading a deep learning engine.

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.translator.wrapper - package ai.djl.modality.cv.translator.wrapper

Contains wrappers to for multiple input formats to a BaseImageTranslator.

ai.djl.modality.cv.util - package ai.djl.modality.cv.util

Contains utility classes for image manipulation.

ai.djl.modality.cv.zoo - package ai.djl.modality.cv.zoo

Contains standard ModelLoaders for the Application.CV application.

ai.djl.modality.nlp - package ai.djl.modality.nlp

Contains utility classes for natural language processing tasks.

ai.djl.modality.nlp.bert - package ai.djl.modality.nlp.bert

Contains classes that deal with BERT for natural language pre-processing tasks.

ai.djl.modality.nlp.embedding - package ai.djl.modality.nlp.embedding

Contains classes that deal with word embeddings for natural language pre-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.modality.nlp.translator - package ai.djl.modality.nlp.translator

Contains utility classes for each of the predefined translator.

ai.djl.modality.rl - package ai.djl.modality.rl

Contains utility classes for reinforcement learning.

ai.djl.modality.rl.agent - package ai.djl.modality.rl.agent

Contains agents to learn using reinforcement learning.

ai.djl.modality.rl.env - package ai.djl.modality.rl.env

Contains environments to train reinforcement learning in.

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.index.dim - package ai.djl.ndarray.index.dim

Contains classes that represent an index element in a NDArray's indices.

ai.djl.ndarray.index.full - package ai.djl.ndarray.index.full

Contains classes that represent simplified representations of an 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 and Deconvolution.

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.nn.transformer - package ai.djl.nn.transformer

Contains blocks for transformer models.

ai.djl.repository - package ai.djl.repository

Contains a Maven-based Repository format for creating repositories of artifacts such as datasets and model zoos.

ai.djl.repository.zoo - package ai.djl.repository.zoo

Contains classes for working with model zoo repositories.

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 - package ai.djl.training.hyperparameter

Contains utilities to train, describe, and manipulate Hyperparameters.

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.tracker - package ai.djl.training.tracker

Contains classes for having a gradually changing hyper-parameter.

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 - Static variable in interface ai.djl.Application.CV

Any computer vision application, including those in Application.CV.

ANY - Static variable in interface ai.djl.Application.NLP

Any NLP application, including those in Application.NLP.

ANY - Static variable in interface ai.djl.Application.Tabular

Any tabular application, including those in Application.Tabular.

any() - Method in interface ai.djl.ndarray.NDArray

Returns true if any of the elements within this NDArray are non-zero or true.

apache() - Static method in class ai.djl.repository.License

The default Apache License.

Application - Class in ai.djl

A class contains common tasks that can be completed using deep learning.

Application.CV - Interface in ai.djl

The common set of applications for computer vision (image and video data).

Application.NLP - Interface in ai.djl

The common set of applications for natural language processing (text data).

Application.Tabular - Interface in ai.djl

The common set of applications for tabular data.

arange(float, float, float, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Returns evenly spaced values within a given interval.

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) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced values within a given interval.

arange(float, float, float, DataType) - 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.

argMax() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the indices of the maximum values into the flattened NDArray.

argMax(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

argMin() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the indices of the minimum values into the flattened NDArray.

argMin(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

argSort(int, boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

Artifact - Class in ai.djl.repository

An Artifact is a set of data files such as a model or dataset.

Artifact() - Constructor for class ai.djl.repository.Artifact

Artifact.Item - Class in ai.djl.repository

A file (possibly compressed) within an Artifact.

Artifact.VersionComparator - Class in ai.djl.repository

A Comparator to compare artifacts based on their version numbers.

artifactId - Variable in class ai.djl.repository.Metadata

artifacts - Variable in class ai.djl.BaseModel

artifacts - Variable in class ai.djl.repository.Metadata

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.

asin() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the inverse trigonometric sine of this NDArray element-wise.

asinh() - Method in interface ai.djl.ndarray.NDArray

Returns the inverse hyperbolic sine of this NDArray element-wise.

asinh() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

atan() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

atanh() - Method in interface ai.djl.ndarray.NDArrayAdapter

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 interface ai.djl.ndarray.NDArrayAdapter

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(List<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.

attachGradient(SparseFormat) - Method in interface ai.djl.ndarray.NDArray

Attaches a gradient NDArray with the given SparseFormat to this NDArray and marks it so GradientCollector.backward(NDArray) can compute the gradient with respect to it.

attachGradient() - Method in interface ai.djl.ndarray.NDArrayAdapter

Attaches a gradient NDArray to this NDArray and marks it so GradientCollector.backward(NDArray) can compute the gradient with respect to it.

attachGradient(SparseFormat) - Method in interface ai.djl.ndarray.NDArrayAdapter

Attaches a gradient NDArray with the given SparseFormat 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.

avgPool1d(NDArray, Shape, Shape, Shape, boolean, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D Avg Pooling on the input.

avgPool1dBlock(Shape, Shape, Shape, boolean, 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, 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) - 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.

avgPool2d(NDArray, Shape, Shape, Shape, boolean, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D Avg Pooling on the input.

avgPool2dBlock(Shape, Shape, Shape, boolean, 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, 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) - 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.

avgPool3d(NDArray, Shape, Shape, Shape, boolean, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D Avg Pooling on the input.

avgPool3dBlock(Shape, Shape, Shape, boolean, 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, 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) - 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).

base() - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets this builder's params to the BASE config of the original BERT paper.

BaseBuilder() - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

BaseBuilder() - Constructor for class ai.djl.modality.nlp.translator.QATranslator.BaseBuilder

BaseBuilder() - Constructor for class ai.djl.nn.core.Embedding.BaseBuilder

BaseBuilder() - Constructor for class ai.djl.nn.recurrent.RecurrentBlock.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.

BaseImageTranslator<T> - Class in ai.djl.modality.cv.translator

Built-in Translator that provides default image pre-processing.

BaseImageTranslator(BaseImageTranslator.BaseBuilder<?>) - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator

Constructs an ImageTranslator with the provided builder.

BaseImageTranslator.BaseBuilder<T extends BaseImageTranslator.BaseBuilder> - Class in ai.djl.modality.cv.translator

A builder to extend for all classes extending the BaseImageTranslator.

BaseImageTranslator.ClassificationBuilder<T extends BaseImageTranslator.BaseBuilder> - Class in ai.djl.modality.cv.translator

A Builder to construct a ImageClassificationTranslator.

BaseImageTranslator.SynsetLoader - Class in ai.djl.modality.cv.translator

BaseModel - Class in ai.djl

BaseModel is the basic implementation of Model.

BaseModel(String) - Constructor for class ai.djl.BaseModel

BaseModelLoader - Class in ai.djl.repository.zoo

Shared code for the ModelLoader implementations.

BaseModelLoader(Repository, MRL, String, ModelZoo) - Constructor for class ai.djl.repository.zoo.BaseModelLoader

Constructs a ModelLoader given the repository, mrl, and version.

BaseNDManager - Class in ai.djl.ndarray

BaseNDManager is the default implementation of NDManager.

BaseNDManager(NDManager, Device) - Constructor for class ai.djl.ndarray.BaseNDManager

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, Batchifier, 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.

batchFirst - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

batchFirst - Variable in class ai.djl.nn.recurrent.RecurrentBlock

batchFlatten(NDArray) - Static method in class ai.djl.nn.Blocks

Inflates the NDArray provided as input to a 2-D NDArray of shape (batch, size).

batchFlatten(NDArray, long) - Static method in class ai.djl.nn.Blocks

Inflates the NDArray provided as input to a 2-D NDArray of shape (batch, size).

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.

batchifier - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

batchifier - Variable in class ai.djl.modality.nlp.translator.QATranslator

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.PaddingStackBatchifier

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

In batch training (training with more than one samples per iteration), a batch normalization layer works by normalizing the values of input data to have mean of 0 and variance of 1.

batchNorm(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.norm.BatchNorm

Applies Batch Normalization for each channel across a batch of data.

batchNorm(NDArray, NDArray, NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.norm.BatchNorm

Applies Batch Normalization for each channel across a batch of data.

batchNorm(NDArray, NDArray, NDArray, NDArray, NDArray, int) - Static method in class ai.djl.nn.norm.BatchNorm

Applies Batch Normalization for each channel across a batch of data.

batchNorm(NDArray, NDArray, NDArray, NDArray, NDArray, int, float, float, boolean) - Static method in class ai.djl.nn.norm.BatchNorm

Applies Batch Normalization for each channel across a batch of data.

BatchNorm.Builder - Class in ai.djl.nn.norm

The Builder to construct a BatchNorm.

batchPredict(List<I>) - Method in class 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.convolutional.Deconvolution

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.RecurrentBlock

Performs any action necessary before initialization.

BertBlock - Class in ai.djl.nn.transformer

Implements the core bert model (without next sentence and masked language task) of bert.

BertBlock.Builder - Class in ai.djl.nn.transformer

The Builder to construct a BertBlock type of Block.

BertFullTokenizer - Class in ai.djl.modality.nlp.bert

BertFullTokenizer runs end to end tokenization of input text

BertFullTokenizer(SimpleVocabulary, boolean) - Constructor for class ai.djl.modality.nlp.bert.BertFullTokenizer

Creates an instance of BertFullTokenizer.

BertMaskedLanguageModelBlock - Class in ai.djl.nn.transformer

Block for the bert masked language task.

BertMaskedLanguageModelBlock(BertBlock, Function<NDArray, NDArray>) - Constructor for class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Creates a new block that applies the masked language task.

BertMaskedLanguageModelLoss - Class in ai.djl.nn.transformer

The loss for the bert masked language model task.

BertMaskedLanguageModelLoss(int, int, int) - Constructor for class ai.djl.nn.transformer.BertMaskedLanguageModelLoss

Creates an MLM loss.

BertNextSentenceBlock - Class in ai.djl.nn.transformer

Block to perform the Bert next-sentence-prediction task.

BertNextSentenceBlock() - Constructor for class ai.djl.nn.transformer.BertNextSentenceBlock

Creates a next sentence block.

BertNextSentenceLoss - Class in ai.djl.nn.transformer

Calculates the loss for the next sentence prediction task.

BertNextSentenceLoss(int, int) - Constructor for class ai.djl.nn.transformer.BertNextSentenceLoss

Creates a new bert next sentence loss.

BertPretrainingBlock - Class in ai.djl.nn.transformer

Creates a block that performs all bert pretraining tasks (next sentence and masked language).

BertPretrainingBlock(BertBlock.Builder) - Constructor for class ai.djl.nn.transformer.BertPretrainingBlock

Creates a new Bert pretraining block fitting the given bert configuration.

BertPretrainingLoss - Class in ai.djl.nn.transformer

Loss that combines the next sentence and masked language losses of bert pretraining.

BertPretrainingLoss() - Constructor for class ai.djl.nn.transformer.BertPretrainingLoss

Creates a loss combining the next sentence and masked language loss for bert pretraining.

BertToken - Class in ai.djl.modality.nlp.bert

BertToken contains all the information for Bert model after encoding question and paragraph.

BertToken(List<String>, List<Long>, List<Long>, int) - Constructor for class ai.djl.modality.nlp.bert.BertToken

Creates an instance of BertToken which includes information for Bert model.

BertTokenizer - Class in ai.djl.modality.nlp.bert

BertTokenizer is a class to help you encode question and paragraph sentence.

BertTokenizer() - Constructor for class ai.djl.modality.nlp.bert.BertTokenizer

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

bias - Variable in class ai.djl.nn.convolutional.Deconvolution

bidirectional - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

bidirectional - Variable in class ai.djl.nn.recurrent.RecurrentBlock

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 - Variable in class ai.djl.inference.Predictor

block - Variable in class ai.djl.modality.nlp.Decoder

block - Variable in class ai.djl.modality.nlp.Encoder

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, int) - Method in interface ai.djl.ndarray.NDArrayAdapter

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 around an object inside an image.

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.

boxIntersection(Rectangle, Rectangle) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

boxIou(Rectangle, Rectangle) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

boxUnion(Rectangle, Rectangle) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

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.

broadcast(Shape) - Method in interface ai.djl.ndarray.NDArrayAdapter

Broadcasts this NDArray to be the given shape.

BufferedImageFactory - Class in ai.djl.modality.cv

BufferedImageFactory is the default implementation of ImageFactory.

BufferedImageFactory() - Constructor for class ai.djl.modality.cv.BufferedImageFactory

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.cv.translator.YoloTranslator.Builder

Builds the translator.

build() - Method in class ai.djl.modality.cv.translator.YoloV5Translator.Builder

Builds the translator.

build() - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding.Builder

Builds a new instance of TrainableWordEmbedding based on the arguments in this builder.

build() - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Builds the SimpleVocabulary 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.Conv1dTranspose.Builder

Builds a Conv1dTranspose block.

build() - Method in class ai.djl.nn.convolutional.Conv2d.Builder

Builds a Conv2d block.

build() - Method in class ai.djl.nn.convolutional.Conv2dTranspose.Builder

Builds a Conv2dTranspose block.

build() - Method in class ai.djl.nn.convolutional.Conv3d.Builder

Builds a Conv3d block.

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.nn.transformer.BertBlock.Builder

Returns a new BertBlock with the parameters of this builder.

build() - Method in class ai.djl.nn.transformer.IdEmbedding.Builder

Builds the IdEmbedding.

build() - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock.Builder

Creates a new ScaledDotProductAttentionBlock with the current configuration.

build() - Method in class ai.djl.repository.zoo.Criteria.Builder

Builds a Criteria instance.

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.Adadelta.Builder

Builds a Adadelta block.

build() - Method in class ai.djl.training.optimizer.Adagrad.Builder

Builds a Adagrad block.

build() - Method in class ai.djl.training.optimizer.Adam.Builder

Builds a Adam block.

build() - Method in class ai.djl.training.optimizer.Nag.Builder

Builds a Nag block.

build() - Method in class ai.djl.training.optimizer.RmsProp.Builder

Builds a RmsProp block.

build() - Method in class ai.djl.training.optimizer.Sgd.Builder

Builds a Sgd block.

build() - Method in class ai.djl.training.tracker.CosineTracker.Builder

Builds a CosineTracker block.

build() - Method in class ai.djl.training.tracker.FactorTracker.Builder

Builds a FactorTracker block.

build() - Method in class ai.djl.training.tracker.LinearTracker.Builder

Builds a LinearTracker block.

build() - Method in class ai.djl.training.tracker.MultiFactorTracker.Builder

Builds a MultiFactorTracker block.

build() - Method in class ai.djl.training.tracker.PolynomialDecayTracker.Builder

Builds a PolynomialDecayTracker.

build() - Method in class ai.djl.training.tracker.WarmUpTracker.Builder

Builds a WarmUpTracker block.

build() - Method in class ai.djl.translate.PaddingStackBatchifier.Builder

Builds the PaddingStackBatchifier.

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(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.ImageClassificationTranslator

Creates a builder to build a ImageClassificationTranslator with specified arguments.

builder() - Static method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Creates a builder to build a InstanceSegmentationTranslator.

builder(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Creates a builder to build a InstanceSegmentationTranslator with specified arguments.

builder() - Static method in class ai.djl.modality.cv.translator.SimplePoseTranslator

Creates a builder to build a SimplePoseTranslator.

builder(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.SimplePoseTranslator

Creates a builder to build a SimplePoseTranslator with specified arguments.

builder() - Static method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

Creates a builder to build a SingleShotDetectionTranslator.

builder(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

Creates a builder to build a SingleShotDetectionTranslator with specified arguments.

Builder() - Constructor for class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

builder() - Static method in class ai.djl.modality.cv.translator.YoloTranslator

Creates a builder to build a YoloTranslator.

builder(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.YoloTranslator

Creates a builder to build a YoloTranslator with specified arguments.

Builder() - Constructor for class ai.djl.modality.cv.translator.YoloTranslator.Builder

builder() - Static method in class ai.djl.modality.cv.translator.YoloV5Translator

Creates a builder to build a YoloV5Translator.

builder(Map<String, ?>) - Static method in class ai.djl.modality.cv.translator.YoloV5Translator

Creates a builder to build a YoloV5Translator with specified arguments.

Builder() - Constructor for class ai.djl.modality.cv.translator.YoloV5Translator.Builder

builder() - Static method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Creates a builder to build an Embedding.

builder() - Static method in class ai.djl.modality.nlp.SimpleVocabulary

Creates a new builder to build a SimpleVocabulary.

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.Conv1dTranspose

Creates a builder to build a Conv1dTranspose.

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.Conv2dTranspose

Creates a builder to build a Conv2dTranspose.

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.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() - Static method in class ai.djl.nn.transformer.BertBlock

Returns a new BertBlock builder.

Builder() - Constructor for class ai.djl.nn.transformer.IdEmbedding.Builder

builder() - Static method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Creates a new Builder to build an Attention Block with.

builder() - Static method in class ai.djl.repository.zoo.Criteria

Creates a builder to build a Criteria.

Builder() - Constructor for class ai.djl.training.dataset.ArrayDataset.Builder

builder() - Static method in class ai.djl.training.optimizer.Adagrad

Creates a builder to build a Adam.

builder() - Static method in class ai.djl.training.optimizer.Adam

Creates a builder to build a Adam.

builder() - Static method in class ai.djl.training.optimizer.RmsProp

Creates a builder to build a RMSProp.

builder() - Static method in class ai.djl.training.tracker.CosineTracker

Creates a new builder.

builder() - Static method in class ai.djl.training.tracker.FactorTracker

Creates a new builder.

builder() - Static method in class ai.djl.training.tracker.LinearTracker

Creates a new builder.

builder() - Static method in class ai.djl.training.tracker.MultiFactorTracker

Creates a new builder.

builder() - Static method in class ai.djl.training.tracker.PolynomialDecayTracker

Creates a new builder.

builder() - Static method in class ai.djl.training.tracker.WarmUpTracker

Creates a new builder.

builder() - Static method in class ai.djl.translate.PaddingStackBatchifier

Returns a PaddingStackBatchifier.Builder.

buildModel(HpSet) - Method in class ai.djl.training.hyperparameter.EasyHpo

Builds the Model and Block to train.

buildSentence(List<String>) - Method in class ai.djl.modality.nlp.preprocess.SimpleTokenizer

Combines a list of tokens to form a sentence.

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.

cast(DataType) - Method in class ai.djl.repository.zoo.ZooModel

Casts the model to support a different precision level.

cbrt() - Method in interface ai.djl.ndarray.NDArray

Returns the cube-root of this NDArray element-wise.

cbrt() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

ceil() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

children - Variable in class ai.djl.nn.AbstractBlock

All direct children of this Block.

chooseAction(RlEnv, boolean) - Method in class ai.djl.modality.rl.agent.EpsilonGreedy

Chooses the next action to take within the RlEnv.

chooseAction(RlEnv, boolean) - Method in class ai.djl.modality.rl.agent.QAgent

Chooses the next action to take within the RlEnv.

chooseAction(RlEnv, boolean) - Method in interface ai.djl.modality.rl.agent.RlAgent

Chooses the next action to take within the RlEnv.

classes - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator

Classification(String, double) - Constructor for class ai.djl.modality.Classifications.Classification

Constructs a single class result for a classification.

ClassificationBuilder() - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

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.

Classifications.ClassificationsSerializer - Class in ai.djl.modality

A customized Gson serializer to serialize the Classifications object.

ClassificationsSerializer() - Constructor for class ai.djl.modality.Classifications.ClassificationsSerializer

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.

clip(Number, Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Clips (limit) the values in this NDArray.

clipGrad - Variable in class ai.djl.training.optimizer.Optimizer

close() - Method in class ai.djl.BaseModel

close() - Method in class ai.djl.inference.Predictor

close() - Method in class ai.djl.modality.nlp.embedding.ModelZooTextEmbedding

close() - Method in interface ai.djl.modality.rl.env.RlEnv

close() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

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.nn.transformer.MemoryScope

Closes this scope by closing the sub manager used to manage it.

close() - Method in class ai.djl.repository.zoo.ZooModel

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 class 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.

compare(Artifact, Artifact) - Method in class ai.djl.repository.Artifact.VersionComparator

compareTo(Version) - Method in class ai.djl.repository.Version

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.

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder

configPostProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.YoloV5Translator.Builder

configPreProcess(Map<String, ?>) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

ConstantEmbedding - Class in ai.djl.nn.core

An AbstractIndexedEmbedding that always returns a constant value.

ConstantEmbedding(NDArray) - Constructor for class ai.djl.nn.core.ConstantEmbedding

Constructs a constant embedding with the given constant.

ConstantInitializer - Class in ai.djl.training.initializer

Initializer that generates tensors with constant values.

ConstantInitializer(float) - Constructor for class ai.djl.training.initializer.ConstantInitializer

Creates a Constant Initializer.

contains(String) - Method in class ai.djl.modality.nlp.SimpleVocabulary

Check if the vocabulary contains a token.

contains(String) - Method in interface ai.djl.modality.nlp.Vocabulary

Check if the vocabulary contains a token.

contains(String) - Method in class ai.djl.ndarray.NDList

Returns true if this NDList contains an NDArray with the specified name.

contains(Version) - Method in class ai.djl.repository.VersionRange

Returns true if a version falls within this range.

contains(Artifact) - Method in class ai.djl.repository.VersionRange

Returns true if the artifact's version falls within this range.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns true if all elements in this NDArray are equal to the Number.

contentEquals(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

A Conv1d layer works similar to Convolution layer with the exception of the number of dimension it operates on being only one, which is LayoutType.WIDTH.

conv1d(NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

conv1d(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

conv1d(NDArray, NDArray, NDArray, Shape) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

conv1d(NDArray, NDArray, NDArray, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

conv1d(NDArray, NDArray, NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

conv1d(NDArray, NDArray, NDArray, Shape, Shape, Shape, int) - Static method in class ai.djl.nn.convolutional.Conv1d

Applies 1D convolution over an input signal composed of several input planes.

Conv1d.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv1d type of Block.

Conv1dTranspose - Class in ai.djl.nn.convolutional

A Conv1dTranspose layer works similar to Deconvolution layer with the exception of the number of dimension it operates on being only one, which is LayoutType.WIDTH.

conv1dTranspose(NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray, Shape) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D deconvolution over an input signal composed of several input planes.

conv1dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape, Shape, int) - Static method in class ai.djl.nn.convolutional.Conv1dTranspose

Applies 1D convolution over an input signal composed of several input planes.

Conv1dTranspose.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv1dTranspose type of Block.

Conv2d - Class in ai.djl.nn.convolutional

Being the pioneer of convolution layers, Conv2d layer works on two dimensions of input, LayoutType.WIDTH and LayoutType.HEIGHT as usually a Conv2d layer is used to process data with two spatial dimensions, namely image.

conv2d(NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

conv2d(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

conv2d(NDArray, NDArray, NDArray, Shape) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

conv2d(NDArray, NDArray, NDArray, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

conv2d(NDArray, NDArray, NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

conv2d(NDArray, NDArray, NDArray, Shape, Shape, Shape, int) - Static method in class ai.djl.nn.convolutional.Conv2d

Applies 2D convolution over an input signal composed of several input planes.

Conv2d.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv2d type of Block.

Conv2dTranspose - Class in ai.djl.nn.convolutional

The input to a Conv2dTranspose is an NDList with a single 4-D NDArray.

conv2dTranspose(NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray, Shape) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

conv2dTranspose(NDArray, NDArray, NDArray, Shape, Shape, Shape, Shape, int) - Static method in class ai.djl.nn.convolutional.Conv2dTranspose

Applies 2D deconvolution over an input signal composed of several input planes.

Conv2dTranspose.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv2dTranspose type of Block.

Conv3d - Class in ai.djl.nn.convolutional

Conv3d layer behaves just as Convolution does, with the distinction being it operates of 3-dimensional data such as medical images or video data.

conv3d(NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

conv3d(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

conv3d(NDArray, NDArray, NDArray, Shape) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

conv3d(NDArray, NDArray, NDArray, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

conv3d(NDArray, NDArray, NDArray, Shape, Shape, Shape) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

conv3d(NDArray, NDArray, NDArray, Shape, Shape, Shape, int) - Static method in class ai.djl.nn.convolutional.Conv3d

Applies 3D convolution over an input signal composed of several input planes.

Conv3d.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv3d type of Block.

Convolution - Class in ai.djl.nn.convolutional

A convolution layer does a dot product calculation on each channel of \(k\)-channel input data by specified number of filters, each containing \(k\) kernels for calculating each channel in the input data and then summed per filter, hence the number of filters denote the number of output channels of a convolution layer.

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.

copyTo(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

cos() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

cosh() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the hyperbolic cosine of this NDArray element-wise.

cosine() - Static method in interface ai.djl.training.tracker.Tracker

Returns a new instance of CosineTracker.Builder that can build an CosineTracker.

CosineTracker - Class in ai.djl.training.tracker

CosineTracker is an implementation of Tracker which is updated by taking sections of a cosine curve to smoothly reduce learning rate until a specified step and base learning rate.

CosineTracker(CosineTracker.Builder) - Constructor for class ai.djl.training.tracker.CosineTracker

Creates a new instance of CosineTracker.

CosineTracker.Builder - Class in ai.djl.training.tracker

The Builder to construct an CosineTracker object.

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(String) - Method in class ai.djl.ndarray.BaseNDManager

Creates and initializes a scalar NDArray.

create(Shape, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Creates an uninitialized instance of NDArray with specified Shape, and DataType.

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(String) - 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.

createAttentionMaskFromInputMask(NDArray, NDArray) - Static method in class ai.djl.nn.transformer.BertBlock

Creates a 3D attention mask from a 2D tensor mask.

createCoo(Buffer, long[][], Shape) - Method in class ai.djl.ndarray.BaseNDManager

Creates a Coordinate Format (COO) Matrix.

createCoo(Buffer, long[][], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates a Coordinate Format (COO) Matrix.

createCSR(Buffer, long[], long[], Shape) - Method in class ai.djl.ndarray.BaseNDManager

Creates a Compressed Sparse Row Storage (CSR) Format Matrix.

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.

createModel(String, Device, Artifact, Map<String, Object>, String) - Method in class ai.djl.repository.zoo.BaseModelLoader

createRowSparse(Buffer, Shape, long[], Shape) - Method in class ai.djl.ndarray.BaseNDManager

Stores the matrix in row sparse format.

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.

Criteria<I,O> - Class in ai.djl.repository.zoo

The Criteria class contains search criteria to look up a ZooModel.

Criteria.Builder<I,O> - Class in ai.djl.repository.zoo

A Builder to construct a Criteria.

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.

cumSum() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the cumulative sum of the elements in the flattened NDArray.

cumSum(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

dataBatchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

dataBatchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset

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, Batchifier, Pipeline, Pipeline, ExecutorService, int, Device) - Constructor for class ai.djl.training.dataset.DataIterable

Creates a new instance of DataIterable with the given parameters.

DataManager - Class in ai.djl.training

DataManager is an class that is used primarily in the Trainer to manipulate data from the dataset to be appropriate for the current model.

DataManager() - Constructor for class ai.djl.training.DataManager

dataset(Application, String, String) - Static method in class ai.djl.repository.MRL

Creates a dataset MRL with specified application.

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 - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

dataType - Variable in class ai.djl.nn.core.Embedding

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() - Static method in class ai.djl.engine.Engine

Prints debug information about the environment for debugging environment issues.

decode(byte[]) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Decodes the given byte array into an object of input parameter type.

decode(NDManager, byte[]) - Static method in interface ai.djl.ndarray.NDArray

Decodes NDArray from bytes.

decode(NDManager, byte[]) - Static method in class ai.djl.ndarray.NDList

Decodes NDList from byte array.

decode(byte[]) - Method in interface ai.djl.ndarray.NDManager

Decodes NDArray through byte array.

decode(InputStream) - Method in interface ai.djl.ndarray.NDManager

Decodes NDArray through DataInputStream.

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.

decode(byte[]) - Method in interface ai.djl.nn.core.AbstractIndexedEmbedding

Decodes the given byte array into an object of input parameter type.

decode(byte[]) - Method in class ai.djl.nn.core.ConstantEmbedding

Decodes the given byte array into an object of input parameter type.

decode(byte[]) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Decodes the given byte array into an object of input parameter type.

decode(byte[]) - Method in class ai.djl.nn.core.Embedding.DefaultItem

Decodes the given byte array into an object of input parameter type.

Decoder - Class in ai.djl.modality.nlp

Decoder is an abstract block that be can used as decoder in encoder-decoder architecture.

Decoder(byte, Block) - Constructor for class ai.djl.modality.nlp.Decoder

Constructs a new instance of Decoder with the given block.

decoder - Variable in class ai.djl.modality.nlp.EncoderDecoder

Deconvolution - Class in ai.djl.nn.convolutional

Transposed convolution, also named fractionally-strided convolution Dumoulin & Visin or deconvolution Long et al., 2015, serves this purpose.

Deconvolution(Deconvolution.DeconvolutionBuilder<?>) - Constructor for class ai.djl.nn.convolutional.Deconvolution

Creates a Deconvolution object.

Deconvolution.DeconvolutionBuilder<T extends Deconvolution.DeconvolutionBuilder> - Class in ai.djl.nn.convolutional

A builder that can build any Deconvolution block.

DeconvolutionBuilder() - Constructor for class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

DEFAULT_DATA_MANAGER - Static variable in class ai.djl.training.DataManager

DEFAULT_FORM - Static variable in class ai.djl.modality.nlp.preprocess.UnicodeNormalizer

defaultDevice() - Static method in class ai.djl.Device

Returns the default context used in Engine.

defaultDevice(Engine) - Static method in class ai.djl.Device

Returns the default context used in Engine.

defaultDevice() - Method in class ai.djl.engine.Engine

Returns the engine's default Device.

DefaultEmbedding() - Constructor for class ai.djl.nn.core.Embedding.DefaultEmbedding

defaultIfNull(Device, Engine) - Static method in class ai.djl.Device

Returns the given device or the default if it is null.

defaultItem - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

DefaultItem(T) - Constructor for class ai.djl.nn.core.Embedding.DefaultItem

DefaultModelZoo - Class in ai.djl.repository.zoo

A ModelZoo that contains models in specified locations.

DefaultModelZoo(String) - Constructor for class ai.djl.repository.zoo.DefaultModelZoo

Creates the LocalModelZoo instance from the given search locations.

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.

DefaultZooProvider - Class in ai.djl.repository.zoo

An ZooProvider implementation can load models from specified locations.

DefaultZooProvider() - Constructor for class ai.djl.repository.zoo.DefaultZooProvider

describeInput() - Method in class ai.djl.BaseModel

Returns the input descriptor of the model.

describeInput() - Method in class ai.djl.modality.nlp.EncoderDecoder

Returns a PairList of input names, and shapes.

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.

describeInput() - Method in class ai.djl.repository.zoo.ZooModel

Returns the input descriptor of the model.

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.

describeOutput() - Method in interface ai.djl.nn.SymbolBlock

Returns a PairList of output names and shapes stored in model file.

describeOutput() - Method in class ai.djl.repository.zoo.ZooModel

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 interface ai.djl.ndarray.NDArrayAdapter

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 Application.CV.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 - 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.

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

dilation - Variable in class ai.djl.nn.convolutional.Convolution

dilation - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

dilation - Variable in class ai.djl.nn.convolutional.Deconvolution

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Divides this NDArray by a number element-wise.

div(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Divides this NDArray by a number element-wise in place.

divi(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Path, URI, Artifact.Item, Progress) - Method in class ai.djl.repository.AbstractRepository

download(Path, URI, Artifact.Item, Progress) - Method in class ai.djl.repository.JarRepository

download(Path, URI, Artifact.Item, Progress) - Method in class ai.djl.repository.SimpleRepository

download(Path, URI, Artifact.Item, Progress) - Method in class ai.djl.repository.SimpleUrlRepository

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.

drawBoundingBoxes(DetectedObjects) - Method in interface ai.djl.modality.cv.Image

Draws the bounding boxes on the image.

drawJoints(Joints) - Method in interface ai.djl.modality.cv.Image

Draws all joints of a body on an image.

Dropout - Class in ai.djl.nn.norm

A dropout layer benefits a network by allowing some units (neurons), and hence their respective connections, of a network to be randomly and temporarily removed by setting its value to 0 only during training by specified probability \(p\), usually set to 0.5.

dropout(NDArray) - Static method in class ai.djl.nn.norm.Dropout

Applies Dropout to the input.

dropout(NDArray, float) - Static method in class ai.djl.nn.norm.Dropout

Applies Dropout to the input.

dropout(NDArray, float, boolean) - Static method in class ai.djl.nn.norm.Dropout

Applies Dropout to the input.

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.RecurrentBlock.BaseBuilder

dropRate - Variable in class ai.djl.nn.recurrent.RecurrentBlock

dumpMemoryInfo(Metrics, String) - Static method in class ai.djl.training.listener.MemoryTrainingListener

Dump memory metrics into log directory.

duplicate(Image.Type) - Method in interface ai.djl.modality.cv.Image

Gets a deep copy of the original image given the type.

duplicate() - Method in interface ai.djl.ndarray.NDArray

Returns a copy of this NDArray.

E

EasyHpo - Class in ai.djl.training.hyperparameter

Helper for easy training with hyperparameters.

EasyHpo() - Constructor for class ai.djl.training.hyperparameter.EasyHpo

EasyTrain - Class in ai.djl.training

Helper for easy training of a whole model, a trainining batch, or a validation batch.

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(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Embeds an item.

embed(NDManager, T[]) - Method in interface ai.djl.nn.core.AbstractEmbedding

Embeds an array of items.

embed(T) - Method in interface ai.djl.nn.core.AbstractIndexedEmbedding

Embeds an item.

embed(Object) - Method in class ai.djl.nn.core.ConstantEmbedding

Embeds an item.

embed(NDManager, Object[]) - Method in class ai.djl.nn.core.ConstantEmbedding

Embeds an array of items.

embed(NDManager, T[]) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Embeds an array of items.

embed(T) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Embeds an item.

embed(NDManager, T[]) - Method in class ai.djl.nn.core.Embedding.DefaultItem

Embeds an array of items.

embed(T) - Method in class ai.djl.nn.core.Embedding.DefaultItem

Embeds an item.

embed(NDManager, T[]) - Method in class ai.djl.nn.core.Embedding

Embeds an array of items.

embedding - Variable in class ai.djl.nn.core.ConstantEmbedding

Embedding<T> - Class in ai.djl.nn.core

An Embedding block map a collection of items to 1-Dimensional representative NDArrays.

Embedding(Embedding.BaseBuilder<T, ?>) - Constructor for class ai.djl.nn.core.Embedding

Embedding(NDArray) - Constructor for class ai.djl.nn.core.Embedding

Constructs a pretrained embedding.

Embedding(NDArray, boolean) - Constructor for class ai.djl.nn.core.Embedding

Constructs a pretrained embedding.

embedding - Variable in class ai.djl.nn.core.Embedding

Embedding.BaseBuilder<T,B extends Embedding.BaseBuilder<T,B>> - Class in ai.djl.nn.core

The Builder to construct a Embedding type of Block.

Embedding.DefaultEmbedding - Class in ai.djl.nn.core

Embedding.DefaultItem - Class in ai.djl.nn.core

EmbeddingException - Exception in ai.djl.modality.nlp.embedding

Thrown to indicate that there was some error while loading embeddings.

EmbeddingException(String) - Constructor for exception ai.djl.modality.nlp.embedding.EmbeddingException

Constructs a new exception with the specified detail message.

EmbeddingException(String, Throwable) - Constructor for exception ai.djl.modality.nlp.embedding.EmbeddingException

Constructs a new exception with the specified detail message and cause.

EmbeddingException(Throwable) - Constructor for exception ai.djl.modality.nlp.embedding.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.

embeddingSize - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

embeddingSize - Variable in class ai.djl.nn.core.Embedding

embeddingType - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

embedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.ModelZooTextEmbedding

Embeds the text after preprocessed using TextEmbedding.preprocessTextToEmbed(List).

embedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.SimpleTextEmbedding

Embeds the text after preprocessed using TextEmbedding.preprocessTextToEmbed(List).

embedText(NDManager, List<String>) - Method in interface ai.djl.modality.nlp.embedding.TextEmbedding

Embeds a text.

embedText(NDManager, long[]) - Method in interface ai.djl.modality.nlp.embedding.TextEmbedding

Embeds the text after preprocessed using TextEmbedding.preprocessTextToEmbed(List).

embedText(NDArray) - Method in interface ai.djl.modality.nlp.embedding.TextEmbedding

Embeds the text after preprocessed using TextEmbedding.preprocessTextToEmbed(List).

embedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Embeds the text after preprocessed using TextEmbedding.preprocessTextToEmbed(List).

embedWord(NDArray) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Embeds the word after preprocessed using WordEmbedding.preprocessWordToEmbed(String).

embedWord(NDManager, String) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Embeds a word.

embedWord(NDManager, long) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Embeds the word after preprocessed using WordEmbedding.preprocessWordToEmbed(String).

embedWord(NDArray) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Embeds the word after preprocessed using WordEmbedding.preprocessWordToEmbed(String).

encode(String, String) - Method in class ai.djl.modality.nlp.bert.BertTokenizer

Encodes questions and paragraph sentences.

encode(String, String, int) - Method in class ai.djl.modality.nlp.bert.BertTokenizer

Encodes questions and paragraph sentences with max length.

encode(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Encodes an object of input type into a byte array.

encode() - Method in interface ai.djl.ndarray.NDArray

Encodes NDArray to byte array.

encode() - Method in class ai.djl.ndarray.NDList

Encodes the NDList to byte array.

encode(T) - Method in interface ai.djl.nn.core.AbstractIndexedEmbedding

Encodes an object of input type into a byte array.

encode(Object) - Method in class ai.djl.nn.core.ConstantEmbedding

Encodes an object of input type into a byte array.

encode(T) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Encodes an object of input type into a byte array.

encode(T) - Method in class ai.djl.nn.core.Embedding.DefaultItem

Encodes an object of input type into a byte array.

Encoder - Class in ai.djl.modality.nlp

Encoder is an abstract block that be can used as encoder in encoder-decoder architecture.

Encoder(byte, Block) - Constructor for class ai.djl.modality.nlp.Encoder

Constructs a new instance of Encoder with the given block.

encoder - Variable in class ai.djl.modality.nlp.EncoderDecoder

EncoderDecoder - Class in ai.djl.modality.nlp

EncoderDecoder is a general implementation of the very popular encoder-decoder architecture.

EncoderDecoder(Encoder, Decoder) - Constructor for class ai.djl.modality.nlp.EncoderDecoder

Constructs a new instance of EncoderDecoder class with the given Encoder and Decoder.

end() - Method in class ai.djl.training.util.ProgressBar

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

EpsilonGreedy - Class in ai.djl.modality.rl.agent

The EpsilonGreedy is a simple exploration/excitation agent.

EpsilonGreedy(RlAgent, Tracker) - Constructor for class ai.djl.modality.rl.agent.EpsilonGreedy

Constructs an EpsilonGreedy.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Equals" comparison.

eq(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.Application

equals(Object) - Method in class ai.djl.Device

equals(Object) - Method in class ai.djl.ndarray.types.Shape

equals(Object) - Method in class ai.djl.repository.Version

erfinv() - Method in interface ai.djl.ndarray.NDArray

Returns element-wise inverse gauss error function of the NDArray.

erfinv() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns element-wise inverse gauss error function of the NDArray.

erfinv(NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns element-wise inverse gauss error function of the input NDArray.

evaluate(NDList, NDList) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelLoss

evaluate(NDList, NDList) - Method in class ai.djl.nn.transformer.BertNextSentenceLoss

evaluate(NDList, NDList) - Method in class ai.djl.training.evaluator.AbstractAccuracy

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.evaluator.BoundingBoxError

Calculates the evaluation between the labels and the predictions.

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.MaskedSoftmaxCrossEntropyLoss

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.

evaluate(NDList) - Method in class ai.djl.training.Trainer

Evaluates function of the model once on the given input NDList.

evaluateDataset(Trainer, Dataset) - Static method in class ai.djl.training.EasyTrain

Evaluates the test dataset.

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.

exp() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the exponential value of this NDArray element-wise.

expandDims(int) - Method in interface ai.djl.ndarray.NDArray

Expands the Shape of a NDArray.

expandDims(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

Expands the Shape of a NDArray.

eye(int, int, int, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Returns a 2-D array with ones on the diagonal and zeros elsewhere.

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) - 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

factor() - Static method in interface ai.djl.training.tracker.Tracker

Returns a new instance of FactorTracker.Builder that can build an FactorTracker.

factories - Variable in class ai.djl.repository.zoo.BaseModelLoader

FactorTracker - Class in ai.djl.training.tracker

FactorTracker is an implementation of Tracker which is updated by a multiplicative factor.

FactorTracker(FactorTracker.Builder) - Constructor for class ai.djl.training.tracker.FactorTracker

Creates a new instance of FactorTracker.

FactorTracker.Builder - Class in ai.djl.training.tracker

The Builder to construct an FactorTracker object.

fallthrough - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

fallthroughEmbedding - Variable in class ai.djl.nn.core.Embedding

FilenameUtils - Class in ai.djl.repository

A class containing utility methods.

FileTranslator<T> - Class in ai.djl.modality.cv.translator.wrapper

Built-in Translator that provides image pre-processing from file path.

FileTranslator(Translator<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.FileTranslator

Creates a FileTranslator instance.

FileTranslatorFactory<T> - Class in ai.djl.modality.cv.translator.wrapper

A factory class creates Translator that can process image from a file path.

FileTranslatorFactory(TranslatorFactory<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.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.

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

filters - Variable in class ai.djl.nn.convolutional.Convolution

filters - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

filters - Variable in class ai.djl.nn.convolutional.Deconvolution

finalize() - Method in class ai.djl.BaseModel

finalize() - Method in class ai.djl.inference.Predictor

finalize() - Method in class ai.djl.training.Trainer

fit(Trainer, int, Dataset, Dataset) - Static method in class ai.djl.training.EasyTrain

Runs a basic epoch training experience with a given trainer.

fit() - Method in class ai.djl.training.hyperparameter.EasyHpo

Fits the model given the implemented abstract methods.

fixed(float) - Static method in interface ai.djl.training.tracker.Tracker

Returns a new instance of Tracker with a fixed value.

flag - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

flatten() - Method in interface ai.djl.ndarray.NDArray

Flattens this NDArray into a 1-D NDArray in row-major order.

flatten() - Method in interface ai.djl.ndarray.NDArrayAdapter

Flattens this NDArray into a 1-D NDArray in row-major order.

flip(int...) - Method in interface ai.djl.ndarray.NDArray

Returns the reverse order of elements in an array along the given axis.

flip(int...) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the reverse order of elements in an array along the given axis.

floor() - Method in interface ai.djl.ndarray.NDArray

Returns the floor of this NDArray element-wise.

floor() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the floor of this NDArray element-wise.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.modality.nlp.EncoderDecoder

Applies the operating function of the block once.

forward(ParameterStore, NDList, NDList, PairList<String, Object>) - Method in class ai.djl.modality.nlp.EncoderDecoder

A forward call using both training data and labels.

forward(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.AbstractBlock

Applies the operating function of the block once.

forward(ParameterStore, NDList, boolean) - Method in interface ai.djl.nn.Block

Applies the operating function of the block once.

forward(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in interface ai.djl.nn.Block

Applies the operating function of the block once.

forward(ParameterStore, NDList, NDList, PairList<String, Object>) - Method in interface ai.djl.nn.Block

A forward call using both training data and labels.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.BertBlock

Applies the operating function of the block once.

forward(ParameterStore, NDArray, NDArray, NDArray, boolean) - Method in class ai.djl.nn.transformer.BertBlock

Embeds the input, runs it through a transformer encoder stack and outputs the resulting embeddings and the pooled classifier.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Applies the operating function of the block once.

forward(ParameterStore, NDArray, NDArray, NDArray, boolean) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Calculates the result of the masked language task.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.BertNextSentenceBlock

Applies the operating function of the block once.

forward(ParameterStore, NDArray, boolean) - Method in class ai.djl.nn.transformer.BertNextSentenceBlock

Applies the internal binary classifier.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.BertPretrainingBlock

Applies the operating function of the block once.

forward(ParameterStore, NDArray, NDArray, NDArray, NDArray, boolean) - Method in class ai.djl.nn.transformer.BertPretrainingBlock

Applies one bert pretraining step.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.IdEmbedding

Applies the operating function of the block once.

forward(ParameterStore, NDArray, boolean) - Method in class ai.djl.nn.transformer.IdEmbedding

Performs a lookup into the embedding using the given input ids.

forward(ParameterStore, NDList, boolean) - Method in class ai.djl.nn.transformer.TransformerEncoderBlock

Applies the operating function of the block once.

forward(NDList) - Method in class ai.djl.training.Trainer

Applies the forward function of the model once on the given input NDList.

forward(NDList, NDList) - Method in class ai.djl.training.Trainer

Applies the forward function of the model once with both data and labels.

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.modality.nlp.Decoder

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.modality.nlp.Encoder

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.modality.nlp.EncoderDecoder

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.AbstractBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.convolutional.Convolution

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.convolutional.Deconvolution

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.core.ConstantEmbedding

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.core.Embedding

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.core.Linear

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.core.Prelu

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.LambdaBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.norm.BatchNorm

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.norm.Dropout

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.ParallelBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.recurrent.GRU

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.recurrent.LSTM

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.recurrent.RNN

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.SequentialBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.BertBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.BertNextSentenceBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.BertPretrainingBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.IdEmbedding

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.PointwiseFeedForwardBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

forwardInternal(ParameterStore, NDList, boolean, PairList<String, Object>) - Method in class ai.djl.nn.transformer.TransformerEncoderBlock

from(NDArray) - Static method in class ai.djl.nn.transformer.MemoryScope

Creates a new memory scope for the device of the given array and adds the array.

from(NDList) - Static method in class ai.djl.nn.transformer.MemoryScope

Creates a new memory scope that fits the device of the first array in the given list, adds all arrays in the given list.

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) - Method in class ai.djl.modality.cv.BufferedImageFactory

Gets Image from file.

fromFile(Path) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from file.

fromImage(Object) - Method in class ai.djl.modality.cv.BufferedImageFactory

Gets Image from varies Java image types.

fromImage(Object) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from varies Java image types.

fromIndex(NDIndex, Shape) - Static method in class ai.djl.ndarray.index.full.NDIndexFullPick

Returns (if possible) the NDIndexFullPick representation of an NDIndex.

fromIndex(NDIndex, Shape) - Static method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns (if possible) the NDIndexFullSlice representation of an NDIndex.

fromInputStream(InputStream) - Method in class ai.djl.modality.cv.BufferedImageFactory

Gets Image from InputStream.

fromInputStream(InputStream) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from InputStream.

fromNDArray(NDArray) - Method in class ai.djl.modality.cv.BufferedImageFactory

Gets Image from NDArray.

fromNDArray(NDArray) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from NDArray.

fromString(String) - Static method in interface ai.djl.translate.Batchifier

Returns a batchifier from a batchifier name.

fromUrl(URL) - Method in class ai.djl.modality.cv.BufferedImageFactory

Gets Image from URL.

fromUrl(URL) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from URL.

fromUrl(String) - Method in class ai.djl.modality.cv.ImageFactory

Gets Image from 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.

full(Shape, float, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Return a new NDArray of given shape, filled with value.

full(Shape, int) - Method in interface ai.djl.ndarray.NDManager

Return a new NDArray of given shape, filled with value.

full(Shape, float) - Method in interface ai.djl.ndarray.NDManager

Return a new NDArray of given shape, filled with value.

full(Shape, float, DataType) - Method in interface ai.djl.ndarray.NDManager

Return a new NDArray of given shape, filled with value.

full(Shape, float, DataType, Device) - Method in interface ai.djl.ndarray.NDManager

Return a new NDArray of given shape, device, filled with value.

G

gates - Variable in class ai.djl.nn.recurrent.RecurrentBlock

gatherFromIndices(NDArray, NDArray) - Static method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Given a 3D array of shape (B, S, E) and a 2D array of shape (B, I) returns the flattened lookup result of shape (B * I * E).

gatherNd(NDArray, NDArray) - Static method in class ai.djl.nn.transformer.MissingOps

Applies the mxnet gather_nd operator.

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(NDArray, NDIndexFullPick) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Returns a subarray by picking the elements.

get(NDArray, NDIndexFullSlice) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Returns a subarray at the slice.

get(NDArray, NDIndex) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Returns a subarray at the given index.

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, Object...) - 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.

get(NDManager) - Method in interface ai.djl.translate.NDArraySupplier

Gets an NDArray from the given NDManager.

getAccumulator(String) - Method in class ai.djl.training.evaluator.AbstractAccuracy

Returns the accumulated evaluator value.

getAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError

Returns the accumulated evaluator value.

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.

getAction() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns the action taken.

getActionSpace() - Method in interface ai.djl.modality.rl.env.RlEnv

Returns the current actions that can be taken in the environment.

getAllEngines() - Static method in class ai.djl.engine.Engine

Returns a set of engine names that are loaded.

getApplication() - Method in class ai.djl.repository.Metadata

Returns the Application.

getApplication() - Method in class ai.djl.repository.MRL

Returns the resource application.

getApplication() - Method in class ai.djl.repository.zoo.BaseModelLoader

Returns the application of the ModelLoader.

getApplication() - Method in class ai.djl.repository.zoo.Criteria

Returns the application of the model.

getApplication() - Method in interface ai.djl.repository.zoo.ModelLoader

Returns the application of the ModelLoader.

getApplicationName() - Method in class ai.djl.repository.Metadata

Returns the Application name.

getArguments(Map<String, Object>) - Method in class ai.djl.repository.Artifact

Returns the artifact arguments.

getArguments() - Method in class ai.djl.repository.zoo.Criteria

Returns the override configurations of the model loading arguments.

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.

getArtifact() - Method in class ai.djl.repository.Artifact.Item

Returns the artifact associated with this item.

getArtifact(String, Function<InputStream, T>) - Method in class ai.djl.repository.zoo.ZooModel

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.repository.zoo.ZooModel

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.

getArtifactAsStream(String) - Method in class ai.djl.repository.zoo.ZooModel

Finds an artifact resource with a given name in the model.

getArtifactId() - Method in class ai.djl.repository.Metadata

Returns the artifactId.

getArtifactId() - Method in class ai.djl.repository.MRL

Returns the artifactId.

getArtifactId() - Method in class ai.djl.repository.zoo.BaseModelLoader

Returns the artifact ID of the ModelLoader.

getArtifactId() - Method in class ai.djl.repository.zoo.Criteria

Returns the artifactId of the ModelLoader to be searched.

getArtifactId() - Method in interface ai.djl.repository.zoo.ModelLoader

Returns the artifact ID of the ModelLoader.

getArtifactNames() - Method in class ai.djl.BaseModel

Returns the artifact names associated with the model.

getArtifactNames() - Method in interface ai.djl.Model

Returns the artifact names associated with the model.

getArtifactNames() - Method in class ai.djl.repository.zoo.ZooModel

Returns the artifact names associated with the model.

getArtifacts() - Method in class ai.djl.repository.Metadata

Returns all the artifacts in the metadata.

getAttachment(String) - Method in interface ai.djl.translate.TranslatorContext

Returns value of attached key-value pair to context.

getAttentionMask() - Method in class ai.djl.modality.nlp.bert.BertToken

Gets the mask to avoid performing attention on padding token indices.

getAxis() - Method in class ai.djl.ndarray.index.full.NDIndexFullPick

Returns the axis to pick.

getBaseUri() - Method in class ai.djl.repository.JarRepository

Returns the URI to the base of the repository.

getBaseUri() - Method in class ai.djl.repository.LocalRepository

Returns the URI to the base of the repository.

getBaseUri() - Method in class ai.djl.repository.RemoteRepository

Returns the URI to the base of the repository.

getBaseUri() - Method in interface ai.djl.repository.Repository

Returns the URI to the base of the repository.

getBaseUri() - Method in class ai.djl.repository.SimpleRepository

Returns the URI to the base of the repository.

getBaseUri() - Method in class ai.djl.repository.SimpleUrlRepository

Returns the URI to the base of the repository.

getBatch() - Method in interface ai.djl.modality.rl.env.RlEnv

Returns a batch of steps from the environment ReplayBuffer.

getBatch() - Method in class ai.djl.modality.rl.LruReplayBuffer

Returns a batch of steps from this buffer.

getBatch() - Method in interface ai.djl.modality.rl.ReplayBuffer

Returns a batch of steps from this buffer.

getBatch() - Method in class ai.djl.training.listener.TrainingListener.BatchData

Returns the original batch.

getBatchifier() - Method in class ai.djl.modality.cv.translator.BaseImageTranslator

Gets the Batchifier.

getBatchifier() - Method in class ai.djl.modality.cv.translator.wrapper.FileTranslator

Gets the Batchifier.

getBatchifier() - Method in class ai.djl.modality.cv.translator.wrapper.InputStreamTranslator

Gets the Batchifier.

getBatchifier() - Method in class ai.djl.modality.cv.translator.wrapper.UrlTranslator

Gets the Batchifier.

getBatchifier() - Method in class ai.djl.modality.nlp.translator.QATranslator

Gets the Batchifier.

getBatchifier() - Method in class ai.djl.modality.nlp.translator.SimpleText2TextTranslator

Gets the Batchifier.

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.

getBertMaskedLanguageModelLoss() - Method in class ai.djl.nn.transformer.BertPretrainingLoss

gets BertMaskedLanguageModelLoss.

getBertNextSentenceLoss() - Method in class ai.djl.nn.transformer.BertPretrainingLoss

gets BertNextSentenceLoss.

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.

getBlock() - Method in class ai.djl.repository.zoo.Criteria

Returns the optional Block to be used for ZooModel.

getBlock() - Method in class ai.djl.repository.zoo.ZooModel

Gets the block from the Model.

getBoolean(long...) - Method in interface ai.djl.ndarray.NDArray

Returns a boolean element from this NDArray.

getBooleanValue(Map<String, ?>, String, boolean) - Static method in class ai.djl.modality.cv.translator.BaseImageTranslator

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.

getCacheDirectory() - Method in class ai.djl.repository.AbstractRepository

Returns the cache directory for the repository.

getCacheDirectory() - Method in interface ai.djl.repository.Repository

Returns the cache directory for the repository.

getCacheDirectory() - Method in class ai.djl.repository.SimpleRepository

Returns the cache directory for the repository.

getCheckpoint() - Method in class ai.djl.training.listener.SaveModelTrainingListener

Returns the checkpoint frequency (or -1 for no checkpointing).

getChildren() - Method in class ai.djl.nn.AbstractBlock

Returns a list of all the children of the block.

getChildren() - Method in interface ai.djl.nn.Block

Returns a list of all the children of the block.

getClassName() - Method in class ai.djl.modality.Classifications.Classification

Returns the class name.

getCode() - Method in class ai.djl.modality.Output

Returns the status code of the output.

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.

getContent() - Method in class ai.djl.modality.Input

Returns the content of the input.

getContent() - Method in class ai.djl.modality.Output

Returns the content of the input.

getData(Batch) - Method in class ai.djl.training.DataManager

Fetches data from the given Batch in required form.

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.

getDataManager() - Method in class ai.djl.training.DefaultTrainingConfig

Gets the DataManager that computes data and labels from the output of dataset.

getDataManager() - Method in class ai.djl.training.Trainer

Returns the DataManager.

getDataManager() - Method in interface ai.djl.training.TrainingConfig

Gets the DataManager that computes data and labels from the output of dataset.

getDataset(Dataset.Usage) - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the dataset to train with.

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 interface ai.djl.ndarray.NDArrayAdapter

Returns the DataType of this NDArray.

getDataType() - Method in class ai.djl.ndarray.types.DataDesc

Returns the DataType of the NDArray.

getDataType() - Method in class ai.djl.repository.zoo.ZooModel

Returns the standard data type used within the model.

getDefaultArtifact() - Method in class ai.djl.repository.Resource

Returns the default artifact.

getDescription() - Method in class ai.djl.repository.Metadata

Returns the description.

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.NDArrayAdapter

Returns the Device of this NDArray.

getDevice() - Method in interface ai.djl.ndarray.NDManager

Returns the default Device of this NDManager.

getDevice() - Method in class ai.djl.repository.zoo.Criteria

Returns the Device of the model to be loaded on.

getDeviceId() - Method in class ai.djl.Device

Returns the deviceId of the Device.

getDevices() - Static method in class ai.djl.Device

Returns an array of devices.

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 class 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 class ai.djl.nn.AbstractBlock

Returns a list of all the direct parameters of the block.

getDirectParameters() - Method in interface ai.djl.nn.Block

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.

getEllipsisIndex() - Method in class ai.djl.ndarray.index.NDIndex

Returns the index of the ellipsis.

getEmbeddingSize() - Method in class ai.djl.nn.transformer.BertBlock

Returns the embedding size used for tokens.

getEmbeddingType() - Method in class ai.djl.nn.core.Embedding.BaseBuilder

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.

getEngine() - Method in class ai.djl.repository.zoo.Criteria

Returns the engine name.

getEngineName() - Method in class ai.djl.engine.Engine

Returns the name of the Engine.

getEpoch() - Method in class ai.djl.training.TrainingResult

Returns the actual number of epoch.

getEvaluations() - Method in class ai.djl.training.TrainingResult

Returns the raw evaluation metrics.

getEvaluators() - Method in class ai.djl.training.DefaultTrainingConfig

Returns the list of Evaluators that should be computed during training.

getEvaluators() - Method in class 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.Conv1dTranspose

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.Conv2dTranspose

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.

getExpectedLayout() - Method in class ai.djl.nn.convolutional.Deconvolution

Returns the expected layout of the input.

getExtension() - Method in class ai.djl.repository.Artifact.Item

Returns the type of file extension.

getFile(Artifact.Item, String) - Method in class ai.djl.repository.AbstractRepository

Returns the path to a file for the item.

getFile(Artifact.Item, String) - Method in interface ai.djl.repository.Repository

Returns the path to a file for the item.

getFileExtension(String) - Static method in class ai.djl.repository.FilenameUtils

Returns the file name extension of the file.

getFiles() - Method in class ai.djl.repository.Artifact

Returns all the file items in the artifact.

getFileType(String) - Static method in class ai.djl.repository.FilenameUtils

Returns the type of the file.

getFilters() - Method in class ai.djl.repository.zoo.Criteria

Returns the search filters that must match the properties of the model.

getFloat(long...) - Method in interface ai.djl.ndarray.NDArray

Returns a float element from this NDArray.

getFloatValue(Map<String, ?>, String, float) - Static method in class ai.djl.modality.cv.translator.BaseImageTranslator

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.

getGradient() - Method in interface ai.djl.ndarray.NDArray

Returns the gradient NDArray attached to this NDArray.

getGradient() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the gradient NDArray attached to this NDArray.

getGroupId() - Method in class ai.djl.repository.Metadata

Returns the groupId.

getGroupId() - Method in class ai.djl.repository.MRL

Returns the groupId.

getGroupId() - Method in class ai.djl.repository.zoo.Criteria

Returns the groupId of the ModelZoo to be searched.

getGroupId() - Method in class ai.djl.repository.zoo.DefaultModelZoo

Returns the global unique identifier of the ModelZoo.

getGroupId() - Method in interface ai.djl.repository.zoo.ModelZoo

Returns the global unique identifier of the ModelZoo.

getHeight() - Method in interface ai.djl.modality.cv.Image

Gets the height of the image.

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.

getId() - Method in class ai.djl.repository.License

Returns the identifier of the license.

getImageHeight() - Method in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

Get resized image height.

getImageWidth() - Method in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

Get resized image width.

getIncrementalVersion() - Method in class ai.djl.repository.Version

Returns the incremental version (assuming major.minor.incremental...) of the version.

getIndex(String) - Method in class ai.djl.modality.nlp.SimpleVocabulary

Returns the index of the given token.

getIndex(String) - Method in interface ai.djl.modality.nlp.Vocabulary

Returns the index of the given token.

getIndex() - Method in class ai.djl.ndarray.index.dim.NDIndexBooleans

Returns the mask binary NDArray.

getIndex() - Method in class ai.djl.ndarray.index.dim.NDIndexFixed

Returns the dimension of the index.

getIndices() - Method in class ai.djl.ndarray.index.dim.NDIndexPick

Returns the indices to pick.

getIndices() - Method in class ai.djl.ndarray.index.full.NDIndexFullPick

Returns the indices to pick.

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.

getInputClass() - Method in class ai.djl.repository.zoo.Criteria

Returns the input data type.

getInstance() - Static method in class ai.djl.engine.Engine

Returns the default Engine.

getInstance() - Static method in class ai.djl.modality.cv.ImageFactory

Gets new instance of Image factory from the provided factory implementation.

getInt(long...) - Method in interface ai.djl.ndarray.NDArray

Returns an int element from this NDArray.

getIntValue(Map<String, ?>, String, int) - Static method in class ai.djl.modality.cv.translator.BaseImageTranslator

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.

getKeyProjection() - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Pointwise Linear projection of the keys.

getLabels(Batch) - Method in class ai.djl.training.DataManager

Fetches labels from the given Batch in required form.

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.

getLastUpdated() - Method in class ai.djl.repository.Metadata

Returns the last update date for the metadata.

getLatestEvaluations() - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Returns the latest evaluations.

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.

getLicenses() - Method in class ai.djl.repository.Metadata

Returns the License.

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 class 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.

getMajorVersion() - Method in class ai.djl.repository.Version

Returns the major version (assuming major.minor.incremental...) of the version.

getManager() - Method in interface ai.djl.ndarray.NDArray

Returns the NDManager used to create this NDArray.

getManager() - Method in interface ai.djl.ndarray.NDArrayAdapter

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 class ai.djl.training.ParameterStore

Get the NDManager associated with ParameterStore.

getManager() - Method in class ai.djl.training.Trainer

Gets the NDManager from the model.

getMax() - Method in class ai.djl.ndarray.index.dim.NDIndexSlice

Returns the end of the range.

getMax() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the slice max for each axis.

getMessage() - Method in class ai.djl.modality.Output

Returns the status code of the output.

getMetadata() - Method in class ai.djl.repository.Artifact

Returns the metadata containing this artifact.

getMetadataVersion() - Method in class ai.djl.repository.Artifact

Returns the metadata format version.

getMetadataVersion() - Method in class ai.djl.repository.Metadata

Returns the metadata format version.

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.

getMetricNames() - Method in class ai.djl.metric.Metrics

Returns a set of String metric names.

getMetrics() - Method in class 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.dim.NDIndexSlice

Returns the start of the range.

getMin() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the slice min for each axis.

getMinorVersion() - Method in class ai.djl.repository.Version

Returns the minor version (assuming major.minor.incremental...) of the version.

getModel() - Method in class 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.

getModelLoader(String) - Method in interface ai.djl.repository.zoo.ModelZoo

Returns the ModelLoader based on the model name.

getModelLoaders() - Method in class ai.djl.repository.zoo.DefaultModelZoo

Lists the available model families in the ModelZoo.

getModelLoaders() - Method in interface ai.djl.repository.zoo.ModelZoo

Lists the available model families in the ModelZoo.

getModelName() - Method in class ai.djl.repository.zoo.Criteria

Returns the optional model name to be used for ZooModel.

getModelPath() - Method in class ai.djl.BaseModel

Returns the directory from where the model is loaded.

getModelPath() - Method in interface ai.djl.Model

Returns the directory from where the model is loaded.

getModelPath() - Method in class ai.djl.repository.zoo.ZooModel

Returns the directory from where the model is loaded.

getModelZoo() - Method in class ai.djl.repository.zoo.Criteria

Returns the ModelZoo to be searched.

getModelZoo() - Method in class ai.djl.repository.zoo.DefaultZooProvider

Returns the instance of the ModelZoo.

getModelZoo() - Method in interface ai.djl.repository.zoo.ZooProvider

Returns the instance of the ModelZoo.

getMrl() - Method in class ai.djl.repository.Resource

Returns the MRL of the resource.

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 class ai.djl.ndarray.BaseNDManager

Gets the name of the NDManager.

getName() - Method in interface ai.djl.ndarray.NDArray

Returns the name of this NDArray.

getName() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the name of this NDArray.

getName() - Method in interface ai.djl.ndarray.NDManager

Gets the name of the NDManager.

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.repository.Artifact

Returns the artifact name.

getName() - Method in class ai.djl.repository.Artifact.Item

Returns the item name.

getName() - Method in class ai.djl.repository.JarRepository

Returns the repository name.

getName() - Method in class ai.djl.repository.License

Returns the name of the license.

getName() - Method in class ai.djl.repository.LocalRepository

Returns the repository name.

getName() - Method in class ai.djl.repository.Metadata

Returns the metadata-level name.

getName() - Method in class ai.djl.repository.RemoteRepository

Returns the repository name.

getName() - Method in interface ai.djl.repository.Repository

Returns the repository name.

getName() - Method in class ai.djl.repository.SimpleRepository

Returns the repository name.

getName() - Method in class ai.djl.repository.SimpleUrlRepository

Returns the repository name.

getName() - Method in class ai.djl.repository.zoo.ZooModel

Gets the model name.

getName() - Method in interface ai.djl.repository.zoo.ZooProvider

Gets the name of the ModelZoo.

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.

getNamePart(String) - Static method in class ai.djl.repository.FilenameUtils

Returns the name of the file without file extension.

getNDArrayInternal() - Method in interface ai.djl.ndarray.NDArray

Returns an internal representative of Native NDArray.

getNDArrayInternal() - Method in interface ai.djl.ndarray.NDArrayAdapter

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 class ai.djl.repository.zoo.ZooModel

Gets the NDManager from the model.

getNDManager() - Method in interface ai.djl.translate.TranslatorContext

Returns the NDManager to create NDArray.

getNewValue(int) - Method in class ai.djl.training.tracker.CosineTracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in class ai.djl.training.tracker.FactorTracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in class ai.djl.training.tracker.LinearTracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in class ai.djl.training.tracker.MultiFactorTracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in class ai.djl.training.tracker.PolynomialDecayTracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in interface ai.djl.training.tracker.Tracker

Fetches the value after the given number of steps/updates.

getNewValue(int) - Method in class ai.djl.training.tracker.WarmUpTracker

Fetches the value 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.

getNumDirections() - Method in class ai.djl.nn.recurrent.RecurrentBlock

getNumEpochs() - Method in class ai.djl.training.listener.EpochTrainingListener

Returns the number of epochs.

getNumOfBytes() - Method in enum ai.djl.ndarray.types.DataType

Returns the number of bytes for each element.

getObservation() - Method in interface ai.djl.modality.rl.env.RlEnv

Returns the observation detailing the current state of the environment.

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.

getOptions(Map<String, String>) - Method in class ai.djl.repository.Artifact

Returns the artifact options.

getOptions() - Method in class ai.djl.repository.zoo.Criteria

Returns the model loading options.

getOutputClass() - Method in class ai.djl.repository.zoo.Criteria

Returns the output data type.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.modality.nlp.Decoder

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.modality.nlp.Encoder

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.modality.nlp.EncoderDecoder

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.AbstractSymbolBlock

Returns the expected output shapes of the block for the specified input shapes.

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.convolutional.Deconvolution

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.ConstantEmbedding

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.RecurrentBlock

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.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.BertBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.BertNextSentenceBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.BertPretrainingBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.IdEmbedding

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.PointwiseFeedForwardBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.transformer.TransformerEncoderBlock

Returns the expected output shapes of the block for the specified input shapes.

getOverrideModelName() - Method in class ai.djl.training.listener.SaveModelTrainingListener

Returns the override model name to save checkpoints with.

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 class ai.djl.nn.AbstractBlock

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 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.recurrent.RecurrentBlock

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.

getParsedVersion() - Method in class ai.djl.repository.Artifact

Returns the artifact version as a Version.

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.BaseImageTranslator

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.

getPostActionSpace() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns the available actions after the step.

getPostObservation() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns the observation detailing the state after the action.

getPredictions() - Method in class ai.djl.training.listener.TrainingListener.BatchData

Returns the predictions for each device.

getPreObservation() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns the observation detailing the state before the action.

getPreprocessors(boolean) - Static method in class ai.djl.modality.nlp.bert.BertFullTokenizer

Get a list of TextProcessors to process input text for Bert models.

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.repository.zoo.Criteria

Returns the optional Progress for the model loading.

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.

getProperties() - Method in class ai.djl.modality.Input

Returns the properties of the input.

getProperties() - Method in class ai.djl.modality.Output

Returns the properties of the output.

getProperties() - Method in class ai.djl.repository.Artifact

Returns the artifact properties.

getProperty(String) - Method in class ai.djl.BaseModel

Gets the property of the model based on property name.

getProperty(String, String) - Method in class ai.djl.modality.Input

Returns the value to which the specified key is mapped.

getProperty(String) - Method in interface ai.djl.Model

Gets the property of the model based on property name.

getProperty(String) - Method in class ai.djl.repository.zoo.ZooModel

Gets the property of the model based on property name.

getQueryProjection() - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Pointwise Linear projection of the queries.

getQuestion() - Method in class ai.djl.modality.nlp.qa.QAInput

Gets the question for the model.

getRank() - Method in class ai.djl.engine.Engine

Return the rank of the Engine.

getRank() - Method in class ai.djl.ndarray.index.dim.NDIndexAll

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.dim.NDIndexBooleans

Returns the number of dimensions occupied by this index element.

getRank() - Method in interface ai.djl.ndarray.index.dim.NDIndexElement

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.dim.NDIndexFixed

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.dim.NDIndexPick

getRank() - Method in class ai.djl.ndarray.index.dim.NDIndexSlice

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.NDIndex

Returns the number of dimensions specified in the Index.

getRecommendedVersion() - Method in class ai.djl.repository.VersionRange

Returns the recommended version in the range.

getRepository() - Method in class ai.djl.repository.Resource

Returns the Repository of the resource.

getRepositoryUri() - Method in class ai.djl.repository.Metadata

Returns the URI to the repository storing the metadata.

getRequestId() - Method in class ai.djl.modality.Input

Returns the requestId of the input.

getRequestId() - Method in class ai.djl.modality.Output

Returns the requestId of the output.

getResourceDirectory(Artifact) - Method in interface ai.djl.repository.Repository

Returns the resource directory for the an artifact.

getResourceDirectory(Artifact) - Method in class ai.djl.repository.SimpleRepository

Returns the resource directory for the an artifact.

getResources() - Method in class ai.djl.repository.AbstractRepository

Returns a list of MRLs in the repository.

getResources() - Method in class ai.djl.repository.JarRepository

Returns a list of MRLs in the repository.

getResources() - Method in class ai.djl.repository.LocalRepository

Returns a list of MRLs in the repository.

getResources() - Method in interface ai.djl.repository.Repository

Returns a list of MRLs in the repository.

getResources() - Method in class ai.djl.repository.SimpleRepository

Returns a list of MRLs in the repository.

getResources() - Method in class ai.djl.repository.SimpleUrlRepository

Returns a list of MRLs in the repository.

getResourceType() - Method in class ai.djl.repository.Metadata

Returns the resource type.

getResourceUri() - Method in class ai.djl.repository.Artifact

Returns the location of the resource directory.

getRestrictions() - Method in class ai.djl.repository.VersionRange

Returns the restrictions that compose the range.

getResultProjection() - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Pointwise Linear projection of the results.

getReward() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns the reward given for the action.

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.

getScopeManager() - Method in class ai.djl.nn.transformer.MemoryScope

Returns the NDManager used to manage this scopes resources.

getSha1Hash() - Method in class ai.djl.repository.Artifact.Item

Returns the hash of the item.

getShape() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the slice shape without squeezing.

getShape() - Method in interface ai.djl.ndarray.NDArray

Returns the Shape of this NDArray.

getShape() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

getShapes() - Method in class ai.djl.ndarray.NDList

Gets all of shapes in the NDList.

getSize() - Method in class ai.djl.repository.Artifact.Item

Returns the file size.

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.

getSparseFormat() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the SparseFormat of this NDArray.

getSqueezedShape() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the slice shape with squeezing.

getStates(NDList) - Method in class ai.djl.modality.nlp.Encoder

Gets the state of the encoder from the given encoder output.

getStep() - Method in class ai.djl.ndarray.index.dim.NDIndexSlice

Returns the step between each slice.

getStep() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the slice step for each axis.

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.Conv1dTranspose

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.Conv2dTranspose

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.

getStringLayout() - Method in class ai.djl.nn.convolutional.Deconvolution

Returns the string representing the layout of the input.

getStringValue(Map<String, ?>, String, String) - Static method in class ai.djl.modality.cv.translator.BaseImageTranslator

getSubimage(int, int, int, int) - Method in interface ai.djl.modality.cv.Image

Gets the subimage defined by a specified rectangular region.

getSupportedEngines() - Method in class ai.djl.repository.zoo.DefaultModelZoo

Returns all supported engine names.

getSupportedEngines() - Method in interface ai.djl.repository.zoo.ModelZoo

Returns all supported engine names.

getSupportedScheme() - Method in interface ai.djl.repository.RepositoryFactory

Returns a set of URI scheme that the RepositoryFactory supports.

getTimestamp() - Method in class ai.djl.metric.Metric

Returns the timestamp of the Metric.

getToken(long) - Method in class ai.djl.modality.nlp.SimpleVocabulary

Returns the token corresponding to the given index.

getToken(long) - Method in interface ai.djl.modality.nlp.Vocabulary

Returns the token corresponding to the given index.

getTokenDictionarySize() - Method in class ai.djl.nn.transformer.BertBlock

Returns the size of the token dictionary.

getTokenEmbedding() - Method in class ai.djl.nn.transformer.BertBlock

Returns the token embedding used by this Bert model.

getTokens() - Method in class ai.djl.modality.nlp.bert.BertToken

Gets the indices of input sequence tokens in the vocabulary.

getTokenTypes() - Method in class ai.djl.modality.nlp.bert.BertToken

Gets segment token indices to indicate first and second portions of the inputs.

getToSqueeze() - Method in class ai.djl.ndarray.index.full.NDIndexFullSlice

Returns the squeeze array of axis.

getTrailingOnes() - Method in class ai.djl.ndarray.types.Shape

Returns the number of trailing ones in the array shape.

getTrainEvaluation(String) - Method in class ai.djl.training.TrainingResult

Returns the evaluation to which the specified key is mapped.

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.

getTrainingResult() - Method in class ai.djl.training.Trainer

Returns the TrainingResult.

getTrainLoss() - Method in class ai.djl.training.TrainingResult

Returns the train loss.

getTranslator() - Method in class ai.djl.repository.zoo.ZooModel

Returns the default translator.

getTranslatorFactory() - Method in class ai.djl.repository.zoo.Criteria

Returns the optional TranslatorFactory to be used for ZooModel.

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.

getType() - Method in class ai.djl.repository.Artifact.Item

Sets the type of the item.

getTypeDictionarySize() - Method in class ai.djl.nn.transformer.BertBlock

Returns the size of the type dictionary.

getUid() - Method in interface ai.djl.ndarray.NDArray

Returns unique identifier of this NDArray.

getUid() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

getUnknownValueCount() - Method in class ai.djl.ndarray.types.Shape

Return the count of unknown value in this Shape.

getUri() - Method in class ai.djl.repository.Artifact.Item

Returns the URI of the item.

getUrl() - Method in class ai.djl.repository.License

Returns the url of the license.

getValidateEvaluation(String) - Method in class ai.djl.training.TrainingResult

Returns the evaluation to which the specified key is mapped.

getValidateLoss() - Method in class ai.djl.training.TrainingResult

Returns the validate loss.

getValidLength() - Method in class ai.djl.modality.nlp.bert.BertToken

Gets the length of the original sentence which has question and paragraph.

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(ParameterStore, Device, boolean) - Method in class ai.djl.nn.transformer.IdEmbedding

Quick hack for bert model to acces embedding table, replace by a proper function to calculate raw logits from embeddings.

getValue(Parameter, Device, boolean) - Method in class ai.djl.training.ParameterStore

Returns the value of a mirrored parameter on a device.

getValueProjection() - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Pointwise Linear projection of the values.

getVersion() - Method in class ai.djl.engine.Engine

Returns the version of the deep learning engine.

getVersion() - Method in class ai.djl.repository.Artifact

Returns the artifact version.

getVersion() - Method in class ai.djl.repository.Resource

Returns the version of the resource.

getVocabulary() - Method in class ai.djl.modality.nlp.bert.BertFullTokenizer

Returns the SimpleVocabulary used for tokenization.

getWebsite() - Method in class ai.djl.repository.Metadata

Returns the website.

getWeightDecay() - Method in class ai.djl.training.optimizer.Optimizer

Gets the value of weight decay.

getWidth() - Method in interface ai.djl.modality.cv.Image

Gets the width of the image.

getWidth() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the width of the Rectangle.

getWrappedImage() - Method in interface ai.djl.modality.cv.Image

Gets the wrapped image.

getWrappedModel() - Method in class ai.djl.repository.zoo.ZooModel

Returns the wrapped model.

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.

globalAvgPool1d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D Global Avg Pooling on the input.

globalAvgPool1dBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool1d pooling function.

globalAvgPool2d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D Global Avg Pooling on the input.

globalAvgPool2dBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool2d pooling function.

globalAvgPool3d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D Global Avg Pooling on the input.

globalAvgPool3dBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool3d pooling function.

globalLpPool1d(NDArray, float) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D Global LP Pooling on the input.

globalLpPool1dBlock(float) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool1d pooling function.

globalLpPool2d(NDArray, float) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D Global LP Pooling on the input.

globalLpPool2dBlock(float) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool2d pooling function.

globalLpPool3d(NDArray, float) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D Global LP Pooling on the input.

globalLpPool3dBlock(float) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool3d pooling function.

globalMaxPool1d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D Global Max Pooling on the input.

globalMaxPool1dBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalmaxPool1dBlock pooling function.

globalMaxPool2d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D Global Max Pooling on the input.

globalMaxPool2dBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalmaxPool2dBlock pooling function.

globalMaxPool3d(NDArray) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D Global Max Pooling on the input.

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.

GROUP_ID - Static variable in class ai.djl.repository.zoo.DefaultModelZoo

groupId - Variable in class ai.djl.repository.Metadata

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

groups - Variable in class ai.djl.nn.convolutional.Convolution

groups - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

groups - Variable in class ai.djl.nn.convolutional.Deconvolution

GRU - Class in ai.djl.nn.recurrent

GRU is an abstract implementation of recurrent neural networks which 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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Greater" comparison.

gt(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

gte(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

hasBiases - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

hasBiases - Variable in class ai.djl.nn.recurrent.RecurrentBlock

hasCapability(String) - Method in class ai.djl.engine.Engine

Returns whether the engine has the specified capability.

hasEngine(String) - Static method in class ai.djl.engine.Engine

Returns if the specified engine is available.

hasGradient() - Method in interface ai.djl.ndarray.NDArray

Returns true if the gradient calculation is required for this NDArray.

hasGradient() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns true if the gradient calculation is required for this NDArray.

hashCode() - Method in class ai.djl.Application

hashCode() - Method in class ai.djl.Device

hashCode() - Method in class ai.djl.ndarray.types.Shape

hashCode() - Method in class ai.djl.repository.Version

hasItem(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Returns whether an item is in the embedding.

hasItem(T) - Method in interface ai.djl.nn.core.AbstractEmbedding

Returns whether an item is in the embedding.

hasItem(Object) - Method in class ai.djl.nn.core.ConstantEmbedding

Returns whether an item is in the embedding.

hasItem(T) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Returns whether an item is in the embedding.

hasItem(T) - Method in class ai.djl.nn.core.Embedding.DefaultItem

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.

hasModelZoo(String) - Static method in interface ai.djl.repository.zoo.ModelZoo

Returns whether a model zoo with the group id is available.

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

hasProperties(Map<String, String>) - Method in class ai.djl.repository.Artifact

Returns true if every filter matches the corresponding property.

hasZeroDimension() - Method in class ai.djl.ndarray.types.Shape

Returns true if the NDArray contains zero dimensions.

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.

height - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

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

HyphenNormalizer - Class in ai.djl.modality.nlp.preprocess

Unicode normalization does not take care of "exotic" hyphens that we normally do not want in NLP input.

HyphenNormalizer() - Constructor for class ai.djl.modality.nlp.preprocess.HyphenNormalizer

I

IdEmbedding - Class in ai.djl.nn.transformer

An Embedding from integer ids to float vectors.

IdEmbedding.Builder - Class in ai.djl.nn.transformer

The Builder to construct an IdEmbedding type of Block.

identityBlock() - Static method in class ai.djl.nn.Blocks

Creates a LambdaBlock that performs the identity function.

Image - Interface in ai.djl.modality.cv

Image is a container of an image in DJL.

Image.Flag - Enum in ai.djl.modality.cv

Flag indicates the color channel options for images.

Image.Interpolation - Enum in ai.djl.modality.cv

Interpolation indicates the Interpolation options for resizinig an image.

Image.Type - Enum in ai.djl.modality.cv

Type indicates the type options for images.

IMAGE_CLASSIFICATION - Static variable in interface ai.djl.Application.CV

An application where images are assigned a single class name.

ImageClassificationModelLoader - Class in ai.djl.modality.cv.zoo

Model loader for Image Classification models.

ImageClassificationModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.ImageClassificationModelLoader

Creates the Model loader from the given repository.

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.

ImageFactory - Class in ai.djl.modality.cv

ImageFactory contains image creation mechanism on top of different platforms like PC and Android.

ImageFactory() - Constructor for class ai.djl.modality.cv.ImageFactory

imageHeight - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator

imageHeight - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

imageWidth - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator

imageWidth - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

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

includeBias - Variable in class ai.djl.nn.convolutional.Convolution

includeBias - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

includeBias - Variable in class ai.djl.nn.convolutional.Deconvolution

increment(long) - Method in class ai.djl.training.util.ProgressBar

index - Variable in class ai.djl.training.evaluator.AbstractAccuracy

init() - Method in class ai.djl.repository.Metadata

Restores artifacts state.

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 class ai.djl.modality.nlp.EncoderDecoder

Initializes the parameters of the block.

initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.AbstractBlock

Initializes the parameters of the block.

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.Parameter

Initializes the parameter with the given NDManager, with given DataType for the given expected input shapes.

initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.ConstantInitializer

Initializes a single NDArray.

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.TruncatedNormalInitializer

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 class ai.djl.training.Trainer

Initializes the Model that the Trainer is going to train.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.modality.nlp.Decoder

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.modality.nlp.Encoder

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.AbstractBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.ParallelBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.SequentialBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.BertBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.BertMaskedLanguageModelBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.BertNextSentenceBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.BertPretrainingBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.IdEmbedding

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.PointwiseFeedForwardBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock

Initializes the Child blocks of this block.

initializeChildBlocks(NDManager, DataType, Shape...) - Method in class ai.djl.nn.transformer.TransformerEncoderBlock

Initializes the Child blocks of this block.

Initializer - Interface in ai.djl.training.initializer

An interface representing an initialization method.

Input - Class in ai.djl.modality

A class stores the generic input data for inference.

Input(String) - Constructor for class ai.djl.modality.Input

Constructs a Input with specified requestId.

inputData - Variable in class ai.djl.BaseModel

inputForComponent(int, NDList, NDList) - Method in class ai.djl.nn.transformer.BertPretrainingLoss

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

Returns the inputs to computing the loss for a component loss.

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

List of names for the input, named inputs should be manually set in sub class.

inputShape(HpSet) - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the input shape for the model.

inputShapes - Variable in class ai.djl.nn.AbstractBlock

The shape of the input for this block, set by the initialization process.

InputStreamTranslator<T> - Class in ai.djl.modality.cv.translator.wrapper

Built-in Translator that provides image pre-processing from InputStream.

InputStreamTranslator(Translator<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.InputStreamTranslator

Creates a UrlTranslator instance.

InputStreamTranslatorFactory<T> - Class in ai.djl.modality.cv.translator.wrapper

A factory class creates Translator that can process image from an InputStream.

InputStreamTranslatorFactory(TranslatorFactory<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.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

An application that finds zero or more objects in an image, the object class (see image classification), and their location as a pixel map.

InstanceSegmentationModelLoader - Class in ai.djl.modality.cv.zoo

Model loader for Instance Segmentation models.

InstanceSegmentationModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.InstanceSegmentationModelLoader

Creates the Model loader from the given repository.

InstanceSegmentationTranslator - Class in ai.djl.modality.cv.translator

A BaseImageTranslator 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 class ai.djl.ndarray.BaseNDManager

An engine specific generic invocation to native operation.

invoke(String, NDList, PairList<String, ?>) - Method in class ai.djl.ndarray.BaseNDManager

An engine specific generic invocation to native operation.

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.

isArchiveFile(String) - Static method in class ai.djl.repository.FilenameUtils

Returns if the the file is an archive file.

isCHW(Shape) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Check if the shape of the image follows CHW/NCHW.

isControl(char) - Static method in class ai.djl.modality.nlp.NlpUtils

Check whether a character is is considered as a control character.

isDone() - Method in interface ai.djl.modality.rl.env.RlEnv.Step

Returns whether the environment is finished or can accept further actions.

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.

isHyphenLike(Integer) - Static method in class ai.djl.modality.nlp.preprocess.HyphenNormalizer

Returns whether the given code point is a hyphen-like codepoint.

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.

isInfinite() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

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.

isNaN() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

isPunctuation(char) - Static method in class ai.djl.modality.nlp.NlpUtils

Check whether a character is considered as a punctuation.

isRemote() - Method in class ai.djl.repository.JarRepository

Returns whether the repository is remote repository.

isRemote() - Method in class ai.djl.repository.LocalRepository

Returns whether the repository is remote repository.

isRemote() - Method in class ai.djl.repository.RemoteRepository

Returns whether the repository is remote repository.

isRemote() - Method in interface ai.djl.repository.Repository

Returns whether the repository is remote repository.

isRemote() - Method in class ai.djl.repository.SimpleRepository

Returns whether the repository is remote repository.

isRemote() - Method in class ai.djl.repository.SimpleUrlRepository

Returns whether the repository is remote repository.

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.

isSnapshot() - Method in class ai.djl.repository.Artifact

Returns true if the artifact is a snapshot.

isSnapshot() - Method in class ai.djl.repository.Version

Returns true if this is a snapshot version.

isSparse() - Method in interface ai.djl.ndarray.NDArray

Returns true if this NDArray is a SparseNDArray.

isWhiteSpace(char) - Static method in class ai.djl.modality.nlp.NlpUtils

Check whether a character is is considered as a whitespace.

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.

Item() - Constructor for class ai.djl.repository.Artifact.Item

items() - Method in class ai.djl.modality.Classifications

Returns a classification item for each potential class for the input.

iterateDataset(Dataset) - Method in class ai.djl.training.Trainer

Fetches an iterator that can iterate through the given Dataset.

iterator() - Method in class ai.djl.training.dataset.DataIterable

J

JarRepository - Class in ai.djl.repository

A JarRepository is a Repository contains an archive file from classpath.

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

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

kernelShape - Variable in class ai.djl.nn.convolutional.Convolution

kernelShape - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

kernelShape - Variable in class ai.djl.nn.convolutional.Deconvolution

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.

labelBatchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

labelBatchifier - Variable in class ai.djl.training.dataset.RandomAccessDataset

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.

LambdaProcessor - Class in ai.djl.modality.nlp.preprocess

TextProcessor will apply user defined lambda function on input tokens.

LambdaProcessor(Function<String, String>) - Constructor for class ai.djl.modality.nlp.preprocess.LambdaProcessor

Creates a LambdaProcessor and specify the function to apply.

large() - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets this builder's params to the LARGE config of the original BERT paper.

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.

LazyNDArray - Interface in ai.djl.ndarray

An NDArray that waits to compute values until they are needed.

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.

License - Class in ai.djl.repository

A License is a container to save the license information.

License() - Constructor for class ai.djl.repository.License

licenses - Variable in class ai.djl.repository.Metadata

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(NDArray, NDArray) - Static method in class ai.djl.nn.core.Linear

Applies a linear transformation to the incoming data.

linear(NDArray, NDArray, NDArray) - Static method in class ai.djl.nn.core.Linear

Applies a linear transformation to the incoming data.

Linear.Builder - Class in ai.djl.nn.core

The Builder to construct a Linear type of Block.

LINEAR_REGRESSION - Static variable in interface ai.djl.Application.Tabular

An application that takes a feature vector (table row) and predicts a numerical feature based on it.

LinearTracker - Class in ai.djl.training.tracker

FactorTracker is an implementation of Tracker which is updated by a constant factor.

LinearTracker(LinearTracker.Builder) - Constructor for class ai.djl.training.tracker.LinearTracker

Creates a new instance of FactorTracker.

LinearTracker.Builder - Class in ai.djl.training.tracker

The Builder to construct an LinearTracker object.

linspace(float, float, int, boolean) - Method in class ai.djl.ndarray.BaseNDManager

Returns evenly spaced numbers over a specified interval.

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) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced numbers over a specified interval.

linspace(float, float, int, boolean) - 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.

listArtifacts() - Method in class ai.djl.repository.Resource

Returns a list of artifacts in this resource.

listDirectory(Artifact.Item, String) - Method in class ai.djl.repository.AbstractRepository

Returns the list of files directly within a specified directory in a zipped directory item.

listDirectory(Artifact.Item, String) - Method in interface ai.djl.repository.Repository

Returns the list of files directly within a specified directory in a zipped directory item.

listModels() - Method in class ai.djl.repository.zoo.BaseModelLoader

Returns a list of the available artifacts that can be loaded.

listModels() - Method in interface ai.djl.repository.zoo.ModelLoader

Returns a list of the available artifacts that can be loaded.

listModels() - Static method in interface ai.djl.repository.zoo.ModelZoo

Returns the available Application and their model artifact metadata.

listModels(Criteria<?, ?>) - Static method in interface ai.djl.repository.zoo.ModelZoo

Returns the available Application and their model artifact metadata.

load(Model) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.SynsetLoader

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, ?>) - Method in interface ai.djl.Model

Loads the model from the modelPath with the name and options provided.

load(Path) - Method in class ai.djl.ndarray.BaseNDManager

Loads the NDArrays saved to a file.

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.

load(Path, String, Map<String, ?>) - Method in class ai.djl.repository.zoo.ZooModel

Loads the model from the modelPath with the name and options provided.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.AbstractBlock

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.convolutional.Convolution

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.convolutional.Deconvolution

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.core.Linear

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.core.Prelu

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.norm.BatchNorm

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.norm.Dropout

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.ParallelBlock

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.recurrent.RecurrentBlock

Overwrite this to load additional metadata with the parameter values.

loadMetadata(byte, DataInputStream) - Method in class ai.djl.nn.SequentialBlock

Overwrite this to load additional metadata with the parameter values.

loadModel(Map<String, String>, Device, Progress) - Method in class ai.djl.modality.cv.zoo.ActionRecognitionModelLoader

Loads the model with the given search filters.

loadModel() - Method in class ai.djl.modality.cv.zoo.ImageClassificationModelLoader

Loads the model.

loadModel(Progress) - Method in class ai.djl.modality.cv.zoo.ImageClassificationModelLoader

Loads the model.

loadModel(Map<String, String>, Device, Progress) - Method in class ai.djl.modality.cv.zoo.ImageClassificationModelLoader

Loads the model with the given search filters.

loadModel() - Method in class ai.djl.modality.cv.zoo.InstanceSegmentationModelLoader

Loads the model.

loadModel(Progress) - Method in class ai.djl.modality.cv.zoo.InstanceSegmentationModelLoader

Loads the model.

loadModel(Map<String, String>, Device, Progress) - Method in class ai.djl.modality.cv.zoo.InstanceSegmentationModelLoader

Loads the model with the given search filters.

loadModel() - Method in class ai.djl.modality.cv.zoo.ObjectDetectionModelLoader

Loads the model.

loadModel(Progress) - Method in class ai.djl.modality.cv.zoo.ObjectDetectionModelLoader

Loads the model.

loadModel(Map<String, String>, Device, Progress) - Method in class ai.djl.modality.cv.zoo.ObjectDetectionModelLoader

Loads the model with the given search filters.

loadModel() - Method in class ai.djl.modality.cv.zoo.SimplePoseModelLoader

Loads the model.

loadModel(Progress) - Method in class ai.djl.modality.cv.zoo.SimplePoseModelLoader

Loads the model.

loadModel(Map<String, String>, Device, Progress) - Method in class ai.djl.modality.cv.zoo.SimplePoseModelLoader

Loads the model with the given search filters.

loadModel(Criteria<I, O>) - Method in class ai.djl.repository.zoo.BaseModelLoader

Loads the model with the given criteria.

loadModel(Criteria<I, O>) - Method in interface ai.djl.repository.zoo.ModelLoader

Loads the model with the given criteria.

loadModel(Criteria<I, O>) - Static method in interface ai.djl.repository.zoo.ModelZoo

Gets the ModelLoader based on the model name.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.modality.nlp.Decoder

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.modality.nlp.Encoder

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.modality.nlp.EncoderDecoder

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.AbstractBlock

Loads the parameters from the given input stream.

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.core.ConstantEmbedding

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.LambdaBlock

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.

LocalRepository - Class in ai.djl.repository

A LocalRepository is a Repository located in a filesystem directory.

LocalRepository(String, Path) - Constructor for class ai.djl.repository.LocalRepository

(Internal) Constructs a LocalRepository from the path with inferred name.

locate(MRL) - Method in class ai.djl.repository.JarRepository

Returns the metadata at a mrl.

locate(MRL) - Method in class ai.djl.repository.LocalRepository

Returns the metadata at a mrl.

locate(MRL) - Method in class ai.djl.repository.RemoteRepository

Returns the metadata at a mrl.

locate(MRL) - Method in interface ai.djl.repository.Repository

Returns the metadata at a mrl.

locate(MRL) - Method in class ai.djl.repository.SimpleRepository

Returns the metadata at a mrl.

locate(MRL) - Method in class ai.djl.repository.SimpleUrlRepository

Returns the metadata at a mrl.

log() - Method in interface ai.djl.ndarray.NDArray

Returns the natural logarithmic value of this NDArray element-wise.

log() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

log10() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

log2() - Method in interface ai.djl.ndarray.NDArrayAdapter

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(int) - 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

Constructs a LoggingTrainingListener instance.

LoggingTrainingListener(int) - Constructor for class ai.djl.training.listener.LoggingTrainingListener

Constructs a LoggingTrainingListener instance with specified steps.

logicalAnd(NDArray) - Method in interface ai.djl.ndarray.NDArray

Returns the truth value of this NDArray AND the other NDArray element-wise.

logicalAnd(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

logicalNot() - Method in interface ai.djl.ndarray.NDArrayAdapter

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) - Method in interface ai.djl.ndarray.NDArrayAdapter

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) - Method in interface ai.djl.ndarray.NDArrayAdapter

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) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

lpPool1d(NDArray, float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D LP Pooling on the input.

lpPool1dBlock(float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1dBlock pooling function in its forward function.

lpPool1dBlock(float, Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1dBlock pooling function in its forward function.

lpPool1dBlock(float, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1dBlock pooling function in its forward function.

lpPool2d(NDArray, float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D LP Pooling on the input.

lpPool2dBlock(float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2dBlock pooling function in its forward function.

lpPool2dBlock(float, Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2dBlock pooling function in its forward function.

lpPool2dBlock(float, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2dBlock pooling function in its forward function.

lpPool2dBlock(float, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2dBlock pooling function in its forward function.

lpPool3d(NDArray, float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D LP Pooling on the input.

lpPool3dBlock(float, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3dBlock pooling function in its forward function.

lpPool3dBlock(float, Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3dBlock pooling function in its forward function.

lpPool3dBlock(float, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the LpPoo3D pooling function in its forward function.

lpPool3dBlock(float, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3dBlock pooling function in its forward function.

LruReplayBuffer - Class in ai.djl.modality.rl

A simple ReplayBuffer that randomly selects across the whole buffer, but always removes the oldest items in the buffer once it is full.

LruReplayBuffer(int, int) - Constructor for class ai.djl.modality.rl.LruReplayBuffer

Constructs a LruReplayBuffer.

LSTM - Class in ai.djl.nn.recurrent

LSTM is an implementation of recurrent neural networks which 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.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Less" comparison.

lt(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Less or equals" comparison.

lte(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

An application that translates text from one language to another.

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

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.

maskedSoftmaxCrossEntropyLoss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of MaskedSoftmaxCrossEntropyLoss with default arguments.

maskedSoftmaxCrossEntropyLoss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of MaskedSoftmaxCrossEntropyLoss with default arguments.

maskedSoftmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of MaskedSoftmaxCrossEntropyLoss with the given arguments.

MaskedSoftmaxCrossEntropyLoss - Class in ai.djl.training.loss

MaskedSoftmaxCrossEntropyLoss is an implementation of Loss that only considers a specific number of values for the loss computations, and masks the rest according to the given sequence.

MaskedSoftmaxCrossEntropyLoss() - Constructor for class ai.djl.training.loss.MaskedSoftmaxCrossEntropyLoss

Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.

MaskedSoftmaxCrossEntropyLoss(String) - Constructor for class ai.djl.training.loss.MaskedSoftmaxCrossEntropyLoss

Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.

MaskedSoftmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Constructor for class ai.djl.training.loss.MaskedSoftmaxCrossEntropyLoss

Creates a new instance of MaskedSoftmaxCrossEntropyLoss with the given parameters.

match(Map<String, String>) - Method in class ai.djl.repository.Resource

Returns the first artifact that matches a given criteria.

MatchAllMetadata() - Constructor for class ai.djl.repository.Metadata.MatchAllMetadata

Creates a MatchAllMetadata instance.

matches(Application) - Method in class ai.djl.Application

Returns whether this application matches the test application set.

matches(List<Artifact>) - Method in class ai.djl.repository.VersionRange

Filters the provided artifacts to those that match the version range.

matMul(NDArray) - Method in interface ai.djl.ndarray.NDArray

Product matrix of this NDArray and the other NDArray.

matMul(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

Product matrix of this NDArray and the other NDArray.

matMul(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Product matrix of this NDArray and the other NDArray.

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.

max() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the maximum of this NDArray.

max(int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the maximum of this NDArray and a number element-wise.

maximum(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

maxPool1d(NDArray, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 1-D Max Pooling on the input.

maxPool1dBlock(Shape, Shape, Shape, boolean) - 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.

maxPool2d(NDArray, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 2-D Max Pooling on the input.

maxPool2dBlock(Shape, Shape, Shape, boolean) - 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.

maxPool3d(NDArray, Shape, Shape, Shape, boolean) - Static method in class ai.djl.nn.pooling.Pool

Performs 3-D Max Pooling on the input.

maxPool3dBlock(Shape, Shape, Shape, boolean) - 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.

md5hash(String) - Static method in class ai.djl.repository.AbstractRepository

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.

mean() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the average of this NDArray.

mean(int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

median() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns median value for this NDArray.

median(int[]) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns median value along given axes.

MemoryScope - Class in ai.djl.nn.transformer

Helper class for more complicated memory management scenarios.

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.

Metadata - Class in ai.djl.repository

A Metadata is a collection of Artifacts with unified metadata (including MRL) that are stored in the same "metadata.json" file.

Metadata() - Constructor for class ai.djl.repository.Metadata

Metadata.MatchAllMetadata - Class in ai.djl.repository

A Metadata class that matches all any search criteria.

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.

micro() - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets this builder's params to a minimal configuration that nevertheless performs quite well.

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.

min() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the minimum of this NDArray.

min(int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the minimum of this NDArray along given axes.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the minimum of this NDArray and a number element-wise.

minimum(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

MissingOps - Class in ai.djl.nn.transformer

Operators missing from NDArray that are necessary to implement Bert pretraining.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns element-wise remainder of division.

mod(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

Model - Interface in ai.djl

A model is a collection of artifacts that is created by the training process.

model(Application, String, String) - Static method in class ai.djl.repository.MRL

Creates a model MRL with specified application.

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.

ModelLoader - Interface in ai.djl.repository.zoo

A ModelLoader loads a particular ZooModel from a Repository for a model zoo.

modelName - Variable in class ai.djl.BaseModel

ModelNotFoundException - Exception in ai.djl.repository.zoo

Thrown when an application tries to load a model from repository search path.

ModelNotFoundException(String) - Constructor for exception ai.djl.repository.zoo.ModelNotFoundException

Constructs a new exception with the specified detail message.

ModelNotFoundException(String, Throwable) - Constructor for exception ai.djl.repository.zoo.ModelNotFoundException

Constructs a new exception with the specified detail message and cause.

ModelNotFoundException(Throwable) - Constructor for exception ai.djl.repository.zoo.ModelNotFoundException

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).

modelZoo - Variable in class ai.djl.repository.zoo.BaseModelLoader

ModelZoo - Interface in ai.djl.repository.zoo

An interface represents a collection of models.

ModelZooTextEmbedding - Class in ai.djl.modality.nlp.embedding

A WordEmbedding using a ZooModel.

ModelZooTextEmbedding(Model) - Constructor for class ai.djl.modality.nlp.embedding.ModelZooTextEmbedding

Constructs a ModelZooTextEmbedding.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns element-wise remainder of division in place.

modi(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

MRL - Class in ai.djl.repository

The MRL (Machine learning Resource Locator) is a pointer to a Metadata "resource" on a machine learning Repository.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Multiplies this NDArray by a number element-wise.

mul(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Multiplies this NDArray by a number element-wise in place.

muli(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

multiFactor() - Static method in interface ai.djl.training.tracker.Tracker

Returns a new instance of MultiFactorTracker.Builder that can build an MultiFactorTracker.

MultiFactorTracker - Class in ai.djl.training.tracker

MultiFactorTracker is an implementation of Tracker which returns piecewise constant values for fixed numbers of steps.

MultiFactorTracker(MultiFactorTracker.Builder) - Constructor for class ai.djl.training.tracker.MultiFactorTracker

Creates a new instance of MultiFactorTracker.

MultiFactorTracker.Builder - Class in ai.djl.training.tracker

The Builder to construct an MultiFactorTracker object.

MULTIPLE_CHOICE - Static variable in interface ai.djl.Application.NLP

An application to represent a multiple choice question.

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.ndarray.BaseNDManager

name - Variable in class ai.djl.training.hyperparameter.param.Hyperparameter

nano() - Method in class ai.djl.nn.transformer.BertBlock.Builder

Tiny config for testing on laptops.

NDArray - Interface in ai.djl.ndarray

An interface representing an n-dimensional array.

NDArrayAdapter - Interface in ai.djl.ndarray

A base implementation of the NDArray that does nothing.

NDArrayIndexer - Class in ai.djl.ndarray.index

A helper class for NDArray implementations for operations with an NDIndex.

NDArrayIndexer() - Constructor for class ai.djl.ndarray.index.NDArrayIndexer

NDArrays - Class in ai.djl.ndarray

This class contains various methods for manipulating NDArrays.

NDArraySupplier - Interface in ai.djl.translate

Represents a supplier of NDArray.

NDImageUtils - Class in ai.djl.modality.cv.util

NDImageUtils is an image processing utility to load, reshape, and convert images using NDArray 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, Object...) - 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.dim

An NDIndexElement to return all values in a particular dimension.

NDIndexAll() - Constructor for class ai.djl.ndarray.index.dim.NDIndexAll

NDIndexBooleans - Class in ai.djl.ndarray.index.dim

An NDIndexElement to return values based on a mask binary NDArray.

NDIndexBooleans(NDArray) - Constructor for class ai.djl.ndarray.index.dim.NDIndexBooleans

Constructs a NDIndexBooleans instance with specified mask binary NDArray.

NDIndexElement - Interface in ai.djl.ndarray.index.dim

An index for particular dimensions created by NDIndex.

NDIndexFixed - Class in ai.djl.ndarray.index.dim

An NDIndexElement that returns only a specific value in the corresponding dimension.

NDIndexFixed(long) - Constructor for class ai.djl.ndarray.index.dim.NDIndexFixed

Constructs a NDIndexFixed instance with specified dimension.

NDIndexFullPick - Class in ai.djl.ndarray.index.full

A simplified representation of a pick-based NDIndex.

NDIndexFullSlice - Class in ai.djl.ndarray.index.full

An index as a slice on all dimensions where some dimensions can be squeezed.

NDIndexPick - Class in ai.djl.ndarray.index.dim

An NDIndexElement that gets elements by index in the specified axis.

NDIndexPick(NDArray) - Constructor for class ai.djl.ndarray.index.dim.NDIndexPick

Constructs a pick.

NDIndexSlice - Class in ai.djl.ndarray.index.dim

An NDIndexElement that returns a range of values in the specified dimension.

NDIndexSlice(Long, Long, Long) - Constructor for class ai.djl.ndarray.index.dim.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.

neg() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the numerical negative NDArray element-wise.

negi() - Method in interface ai.djl.ndarray.NDArray

Returns the numerical negative NDArray element-wise in place.

negi() - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the boolean NDArray for element-wise "Not equals" comparison.

neq(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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 class ai.djl.engine.Engine

Returns a new instance of GradientCollector.

newGradientCollector() - Method in class ai.djl.training.Trainer

Returns a new instance of GradientCollector.

newInstance(Model, Map<String, ?>) - Method in class ai.djl.modality.cv.translator.wrapper.FileTranslatorFactory

Returns a new instance of the Translator class.

newInstance(Model, Map<String, ?>) - Method in class ai.djl.modality.cv.translator.wrapper.InputStreamTranslatorFactory

Returns a new instance of the Translator class.

newInstance(Model, Map<String, ?>) - Method in class ai.djl.modality.cv.translator.wrapper.UrlTranslatorFactory

Returns a new instance of the Translator class.

newInstance(String) - Static method in interface ai.djl.Model

Creates an empty model instance.

newInstance(String, Device) - Static method in interface ai.djl.Model

Creates an empty model instance on the specified Device.

newInstance(String, String) - Static method in interface ai.djl.Model

Creates an empty model instance on the specified Device and engine.

newInstance(String, Device, String) - Static method in interface ai.djl.Model

Creates an empty model instance on the specified Device and engine.

newInstance(String, Path) - Static method in interface ai.djl.repository.Repository

Creates a new instance of a repository with a name and url.

newInstance(String, String) - Static method in interface ai.djl.repository.Repository

Creates a new instance of a repository with a name and url.

newInstance(String, String) - Method in interface ai.djl.repository.RepositoryFactory

Creates a new instance of a repository with a name and url.

newInstance(Model, Map<String, ?>) - Method in class ai.djl.translate.ServingTranslatorFactory

Returns a new instance of the Translator class.

newInstance(Model, Map<String, ?>) - Method in interface ai.djl.translate.TranslatorFactory

Returns a new instance of the Translator class.

newModel(String, Device) - Method in class ai.djl.engine.Engine

Constructs a new model.

newParameterServer(Optimizer) - Method in class ai.djl.engine.Engine

Returns a new instance of ParameterServer.

newPredictor(Translator<I, O>) - Method in class ai.djl.BaseModel

Creates a new Predictor based on the model.

newPredictor(Translator<I, O>) - Method in interface ai.djl.Model

Creates a new Predictor based on the model.

newPredictor() - Method in class ai.djl.repository.zoo.ZooModel

Creates a new Predictor based on the model with the default translator.

newPredictor(Translator<P, Q>) - Method in class ai.djl.repository.zoo.ZooModel

Creates a new Predictor based on the model.

newSubManager() - Method in class ai.djl.ndarray.BaseNDManager

Creates a child NDManager.

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 class ai.djl.BaseModel

Creates a new Trainer instance for a Model.

newTrainer(TrainingConfig) - Method in interface ai.djl.Model

Creates a new Trainer instance for a Model.

newTrainer(TrainingConfig) - Method in class ai.djl.repository.zoo.ZooModel

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.

NlpUtils - Class in ai.djl.modality.nlp

Utility functions for processing String and Characters in NLP problems.

nms(List<YoloV5Translator.IntermediateResult>) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

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.

nonzero() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the indices of elements that are non-zero.

NoopTranslator - Class in ai.djl.translate

A no operational Translator implementation.

NoopTranslator(Batchifier) - Constructor for class ai.djl.translate.NoopTranslator

Constructs a NoopTranslator with the given Batchifier.

NoopTranslator() - Constructor for class ai.djl.translate.NoopTranslator

Constructs a NoopTranslator.

norm() - Method in interface ai.djl.ndarray.NDArray

Returns the norm of this NDArray.

norm(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the norm of this NDArray.

norm(boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the norm of this NDArray.

norm(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the norm of this NDArray.

norm(int, int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the norm of this NDArray.

norm(boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the norm of this NDArray.

norm(int, int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the norm of this NDArray.

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.

normalizeDefault(String) - Static method in class ai.djl.modality.nlp.preprocess.UnicodeNormalizer

Normalizes a String using a sensible default normal form.

normalizeHyphens(String) - Static method in class ai.djl.modality.nlp.preprocess.HyphenNormalizer

Replaces hyphen like codepoints by ASCII "-", removes soft hyphens.

notifyListeners(Consumer<TrainingListener>) - Method in class ai.djl.training.Trainer

Executes a method on each of the TrainingListeners.

numChannels() - Method in enum ai.djl.modality.cv.Image.Flag

Returns the number of channels for this flag.

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.Conv1dTranspose

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.Conv2dTranspose

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.

numDimensions() - Method in class ai.djl.nn.convolutional.Deconvolution

Returns the number of dimensions of the input.

numEpochs(HpSet) - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the number of epochs to train for the current hyperparameter set.

numHyperParameterTests() - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the number of hyperparameter sets to train with.

numItems - Variable in class ai.djl.nn.core.Embedding

numLayers - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

numLayers - Variable in class ai.djl.nn.recurrent.RecurrentBlock

O

OBJECT_DETECTION - Static variable in interface ai.djl.Application.CV

An application that finds zero or more objects in an image, the object class (see image classification), and their locations as a BoundingBox.

ObjectDetectionBuilder() - Constructor for class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

ObjectDetectionModelLoader - Class in ai.djl.modality.cv.zoo

A ModelLoader for Single Shot Detection (SSD) models.

ObjectDetectionModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.ObjectDetectionModelLoader

Creates the Model loader from the given repository.

ObjectDetectionTranslator - Class in ai.djl.modality.cv.translator

A BaseImageTranslator that post-process the NDArray into DetectedObjects with boundaries.

ObjectDetectionTranslator(ObjectDetectionTranslator.ObjectDetectionBuilder<?>) - Constructor for class ai.djl.modality.cv.translator.ObjectDetectionTranslator

Creates the ObjectDetectionTranslator from the given builder.

ObjectDetectionTranslator.ObjectDetectionBuilder<T extends ObjectDetectionTranslator.ObjectDetectionBuilder> - Class in ai.djl.modality.cv.translator

The base builder for the object detection translator.

of(String) - Static method in class ai.djl.Application

Converts a path string to a Application.

of(String, int) - Static method in class ai.djl.Device

Returns a Device with device type and device id.

oneHot(int, NDArray) - Static method in class ai.djl.nn.transformer.MissingOps

Creates a one-hot-encoding from the given data.

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.SaveModelTrainingListener

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.

onEpoch(Trainer) - Method in class ai.djl.training.listener.TrainingListenerAdapter

Listens to the end of an epoch during training.

ones(Shape, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Creates an instance of NDArray with specified Shape filled with ones.

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.

onesLike() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.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.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TrainingListenerAdapter

Listens to the end of training one batch of data during 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 interface ai.djl.training.listener.TrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.TrainingListenerAdapter

Listens to the beginning 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.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.SaveModelTrainingListener

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.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.TrainingListenerAdapter

Listens to the end of training.

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.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TrainingListenerAdapter

Listens to the end of validating one batch of data during validation.

OPENMP - Static variable in class ai.djl.engine.StandardCapabilities

openStream(Artifact.Item, String) - Method in class ai.djl.repository.AbstractRepository

Returns an InputStream for an item in a repository.

openStream(Artifact.Item, String) - Method in interface ai.djl.repository.Repository

Returns an InputStream for an item in a repository.

optApplication(Application) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the model application for this criteria.

optApplySoftmax(boolean) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

Sets whether to apply softmax when processing output.

optArgument(String, Object) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the optional model loading argument for this criteria.

optArguments(Map<String, Object>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets an extra model loading argument for this criteria.

optArtifactId(String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets optional artifactId of the ModelLoader for this criteria.

optAttentionHeadCount(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the number of attention heads to use in each transformer block.

optAttentionProbsDropoutProb(float) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock.Builder

Sets the probability of applying dropout to the attention probability distribution.

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.

optBatchFirst(boolean) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the optional batchFirst flag that indicates whether the input is batch major or not.

optBatchifier(Batchifier) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

Sets the Batchifier for the Translator.

optBatchifier(Batchifier) - Method in class ai.djl.modality.nlp.translator.QATranslator.BaseBuilder

Sets the Batchifier for the Translator.

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.convolutional.Deconvolution.DeconvolutionBuilder

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.

optBidirectional(boolean) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the optional parameter that indicates whether to use bidirectional recurrent layers.

optBlock(Block) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets an optional model Block for this criteria.

optCenter(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder

If True, add offset of `beta` to normalized tensor.

optCentered(boolean) - Method in class ai.djl.training.optimizer.RmsProp.Builder

Sets which version of RMSProp to use.

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 \(clipGrad\).

optDataBatchifier(Batchifier) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Batchifier for the data.

optDataManager(DataManager) - Method in class ai.djl.training.DefaultTrainingConfig

Sets the DataManager to be used during training.

optDataType(DataType) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets the data type of the embedding arrays (default is Float32).

optDefaultItem(T) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets whether to use a default item's embedding for undefined items.

optDevice(Device) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the Device for this criteria.

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.

optDilation(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the dilation along each dimension.

optDilation(Shape) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the dilation along each dimension.

optDropRate(float) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the drop rate of the dropout on the outputs of each RNN layer, except the last layer.

optEmbeddingSize(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the embedding size to use for input tokens.

optEngine(String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the engine name for this criteria.

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.Adadelta.Builder

Sets \(epsilon\) - a small quantity for numerical stability.

optEpsilon(float) - Method in class ai.djl.training.optimizer.Adagrad.Builder

Sets \(epsilon\) - a small quantity for numerical stability.

optEpsilon(float) - Method in class ai.djl.training.optimizer.Adam.Builder

Sets \(epsilon\) - a small quantity for numerical stability.

optEpsilon(float) - Method in class ai.djl.training.optimizer.RmsProp.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.tracker.MultiFactorTracker.Builder

Set the value of the multiplicative factor.

optFallthrough(AbstractIndexedEmbedding<T>) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets a custom handler for items not found in the embedding.

optFilter(String, String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets an extra search filter for this criteria.

optFilters(Map<String, String>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the extra search filters for this criteria.

optFinalValue(float) - Method in class ai.djl.training.tracker.CosineTracker.Builder

Sets the final value that the learning rate will remain constant as after the specified max number of updates.

optFlag(Image.Flag) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

Sets the optional Image.Flag (default is Image.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.

optGroupId(String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets optional groupId of the ModelZoo for this criteria.

optGroups(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the number of group partitions.

optGroups(int) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the number of group partitions.

optHasBiases(boolean) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the optional biases flag that indicates whether to use biases or not.

optHiddenDropoutProbability(float) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the dropout probabilty in the hidden fully connected networks.

optHiddenSize(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the size of the hidden layers in the fully connected networks used.

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

optIncludeValidLengths(boolean) - Method in class ai.djl.translate.PaddingStackBatchifier.Builder

Sets whether to include the valid lengths (length of non-padded data) for each array.

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.

optLabelBatchifier(Batchifier) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Batchifier for the labels.

optLabels(NDArray...) - Method in class ai.djl.training.dataset.ArrayDataset.Builder

Sets the labels for the data in the ArrayDataset.

optLearningRateTracker(Tracker) - Method in class ai.djl.training.optimizer.Adagrad.Builder

Sets the Tracker for this optimizer.

optLearningRateTracker(Tracker) - Method in class ai.djl.training.optimizer.Adam.Builder

Sets the Tracker for this optimizer.

optLearningRateTracker(Tracker) - Method in class ai.djl.training.optimizer.RmsProp.Builder

Sets the Tracker for this optimizer.

optLimit(long) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets this dataset's limit.

optMaxEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the maximum edge length of the rescaled image.

optMaxSequenceLength(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the maximum sequence length this model can process.

optMaxUpdates(int) - Method in class ai.djl.training.tracker.FactorTracker.Builder

Sets the maximum number of updates after which the value should remain constant.

optMaxUpdates(int) - Method in class ai.djl.training.tracker.LinearTracker.Builder

Sets the maximum number of updates after which the value should remain constant.

optMaxValue(float) - Method in class ai.djl.training.tracker.LinearTracker.Builder

Sets the maximum value for a positive slope.

optMinFrequency(int) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Sets the optional parameter that specifies the minimum frequency to consider a token to be part of the SimpleVocabulary.

optMinNegativeSamples(int) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the minimum number of negative samples.

optMinValue(float) - Method in class ai.djl.training.tracker.FactorTracker.Builder

Sets the minimum value.

optMinValue(float) - Method in class ai.djl.training.tracker.LinearTracker.Builder

Sets the minimum value for a negative slope.

optModelName(String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets an optional model name for this criteria.

optModelPath(Path) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the optional model path of the ModelLoader for this criteria.

optModelUrls(String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets optional model urls of the ModelLoader for this criteria.

optModelZoo(ModelZoo) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets optional ModelZoo of the ModelLoader for this criteria.

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.RmsProp.Builder

Sets the momentum factor.

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.

optNmsThreshold(float) - Method in class ai.djl.modality.cv.translator.YoloV5Translator.Builder

Sets the 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.

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).

optOption(String, String) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the optional model loading option for this criteria.

optOptions(Map<String, String>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the model loading options for this criteria.

optOutPadding(Shape) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the out_padding along each dimension.

optOutputType(YoloV5Translator.YoloOutputType) - Method in class ai.djl.modality.cv.translator.YoloV5Translator.Builder

Sets the YoloOutputType.

optPadding(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the padding along each dimension.

optPadding(Shape) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

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.

optPower(float) - Method in class ai.djl.training.tracker.PolynomialDecayTracker.Builder

Sets the power of the polynomial to decay by.

optProgress(Progress) - Method in class ai.djl.repository.zoo.Criteria.Builder

Set the optional Progress.

optRate(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.ObjectDetectionTranslator.ObjectDetectionBuilder

Sets the optional rescale size.

optReservedTokens(Collection<String>) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Sets the optional parameter that sets the list of reserved tokens.

optReturnState(boolean) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the optional flag that indicates whether to return state or not.

optRho(float) - Method in class ai.djl.training.optimizer.Adadelta.Builder

Sets the rho for Adadelta.

optRho(float) - Method in class ai.djl.training.optimizer.RmsProp.Builder

Sets the decay factor for the moving average over the past squared gradient.

optScale(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder

If True, multiply result by `gamma`.

optShortEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the shorter edge length of the rescaled image.

optSlope(float) - Method in class ai.djl.training.tracker.LinearTracker.Builder

Sets the value of the linear slope.

optSparseGrad(boolean) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets the optional parameter whether to compute row sparse gradient in the backward calculation.

optSteps(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the step across \(x\) and \(y\) dimensions.

optStride(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the stride of the convolution.

optStride(Shape) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the stride of the deconvolution.

optSynset(List<String>) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

Sets the potential classes for an image.

optSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

Sets the name of the synset file listing the potential classes for an image.

optSynsetUrl(String) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

Sets the URL of the synset file.

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.ObjectDetectionTranslator.ObjectDetectionBuilder

Sets the threshold for prediction accuracy.

optThreshold(float) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder

Sets the threshold for prediction accuracy.

optTransformerBlockCount(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the number of transformer blocks to use.

optTranslator(Translator<I, O>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the optional Translator to override default Translator.

optTranslatorFactory(TranslatorFactory<I, O>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Sets the optional TranslatorFactory to override default Translator.

optTypeDictionarySize(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the number of possible token types.

optUnknownToken(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding.Builder

Sets the optional String value for the unknown token.

optUnknownToken(String) - Method in class ai.djl.modality.nlp.SimpleVocabulary.Builder

Sets the optional parameter that specifies the unknown token's string value.

optUseDefault(boolean) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets whether to use a default embedding for undefined items (default true).

optWarmUpBeginValue(float) - Method in class ai.djl.training.tracker.WarmUpTracker.Builder

Sets the value at the beginning of warm-up mode.

optWarmUpMode(WarmUpTracker.Mode) - Method in class ai.djl.training.tracker.WarmUpTracker.Builder

Sets the WarmUpTracker.Mode for the WarmUpTracker.

optWarmUpSteps(int) - Method in class ai.djl.training.tracker.WarmUpTracker.Builder

Sets the number of steps until the point the value is updated in warm-up mode.

optWeightDecays(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the value of weight decay.

outPadding - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

outPadding - Variable in class ai.djl.nn.convolutional.Deconvolution

Output - Class in ai.djl.modality

A class stores the generic inference results.

Output(String) - Constructor for class ai.djl.modality.Output

Constructs a Output with specified requestId.

Output(String, int, String) - Constructor for class ai.djl.modality.Output

Constructs a Output with specified requestId, code and message.

overlap(double, double, double, double) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

P

pad(List<E>, E, int) - Method in class ai.djl.modality.nlp.bert.BertTokenizer

Pads the tokens to the required length.

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

padding - Variable in class ai.djl.nn.convolutional.Convolution

padding - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

padding - Variable in class ai.djl.nn.convolutional.Deconvolution

PaddingStackBatchifier - Class in ai.djl.translate

The padding stack batchifier is a StackBatchifier that also pads elements to reach the same length.

PaddingStackBatchifier.Builder - Class in ai.djl.translate

Builder to build a PaddingStackBatchifier.

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.

ParallelTrain - Class in ai.djl.training

Helper for easy training of a whole model on multiple GPUs in parallel.

ParallelTrain(Device[]) - Constructor for class ai.djl.training.ParallelTrain

Build a ParallelTrain for the given devices.

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.

Parameter(String, Block, ParameterType, boolean, SparseFormat) - Constructor for class ai.djl.nn.Parameter

Creates a Parameter with the given name, and parameter type, and associated with the given Block.

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.AbstractBlock

All direct parameters of this Block.

ParameterServer - Interface in ai.djl.training

An interface for a key-value store to store parameters, and their corresponding gradients.

parameterShapeCallbacks - Variable in class ai.djl.nn.AbstractBlock

Callbacks to determine the shape of a parameter.

parameterStore - Variable in class ai.djl.inference.Predictor

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(String, Map<String, ?>) - Method in class ai.djl.BaseModel

parent - Variable in class ai.djl.ndarray.BaseNDManager

parse(String) - Static method in class ai.djl.repository.VersionRange

Creates a new version range from a string version range.

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).

percentile(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns percentile for this NDArray.

percentile(Number, int[]) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns median along given dimension(s).

pipeline - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.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.

PointwiseFeedForwardBlock - Class in ai.djl.nn.transformer

Fully connected Feed-Forward network, only applied to the last dimension of the input.

PointwiseFeedForwardBlock(List<Integer>, int, Function<NDList, NDList>) - Constructor for class ai.djl.nn.transformer.PointwiseFeedForwardBlock

Creates a pointwise feed-forward block.

PolynomialDecayTracker - Class in ai.djl.training.tracker

Polynomial decay Tracker.

PolynomialDecayTracker(PolynomialDecayTracker.Builder) - Constructor for class ai.djl.training.tracker.PolynomialDecayTracker

Builds a PolynomialDecayTracker.

PolynomialDecayTracker.Builder - Class in ai.djl.training.tracker

Builder for PolynomialDecayTracker.

Pool - Class in ai.djl.nn.pooling

Utility class that provides Block and methods for different pooling functions.

POSE_ESTIMATION - Static variable in interface ai.djl.Application.CV

An application that accepts an image of a single person and returns the Joints locations of the person.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Takes the power of this NDArray with a number element-wise.

pow(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Takes the power of this NDArray with a number element-wise in place.

powi(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.Predictor

Predicts an item for inference.

Predictor<I,O> - Class in ai.djl.inference

The Predictor interface provides a session for model inference.

Predictor(Model, Translator<I, O>, boolean) - Constructor for class ai.djl.inference.Predictor

Creates a new instance of BasePredictor with the given Model and Translator.

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.

prelu(NDArray, NDArray) - Static method in class ai.djl.nn.core.Prelu

Applies a Prelu activation on the input NDArray.

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.ObjectDetectionTranslator

Prepares the translator with the manager and model to use.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.wrapper.FileTranslator

Prepares the translator with the manager and model to use.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.wrapper.InputStreamTranslator

Prepares the translator with the manager and model to use.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.wrapper.UrlTranslator

Prepares the translator with the manager and model to use.

prepare(Artifact, Progress) - Method in class ai.djl.repository.AbstractRepository

Prepares the artifact for use with progress tracking.

prepare(Artifact) - Method in interface ai.djl.repository.Repository

Prepares the artifact for use.

prepare(Artifact, Progress) - Method in interface ai.djl.repository.Repository

Prepares the artifact for use with progress tracking.

prepare(Artifact) - Method in class ai.djl.repository.Resource

Prepares the artifact for use.

prepare(Artifact, Progress) - Method in class ai.djl.repository.Resource

Prepares the artifact for use with progress tracking.

prepare(Artifact, Progress) - Method in class ai.djl.repository.SimpleRepository

Prepares the artifact for use with progress tracking.

prepare(Progress) - Method in class ai.djl.training.dataset.ArrayDataset

Prepares the dataset for use with tracked progress.

prepare() - Method in interface ai.djl.training.dataset.Dataset

Prepares the dataset for use.

prepare(Progress) - Method in interface ai.djl.training.dataset.Dataset

Prepares the dataset for use with tracked progress.

prepare(NDManager, Model) - Method in interface ai.djl.translate.Translator

Prepares the translator with the manager and model to use.

preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.HyphenNormalizer

Applies the preprocessing defined to the given input tokens.

preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.LambdaProcessor

Applies the preprocessing defined to the given input tokens.

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.TextCleaner

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.

preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.TextTerminator

Applies the preprocessing defined to the given input tokens.

preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.TextTruncator

Applies the preprocessing defined to the given input tokens.

preprocess(List<String>) - Method in interface ai.djl.modality.nlp.preprocess.Tokenizer

Applies the preprocessing defined to the given input tokens.

preprocess(List<String>) - Method in class ai.djl.modality.nlp.preprocess.UnicodeNormalizer

Applies the preprocessing defined to the given input tokens.

PreProcessor<I> - Interface in ai.djl.translate

An interface that provides pre-processing functionality.

preprocessTextToEmbed(List<String>) - Method in class ai.djl.modality.nlp.embedding.ModelZooTextEmbedding

Preprocesses the text to embed into an array to pass into the model.

preprocessTextToEmbed(List<String>) - Method in class ai.djl.modality.nlp.embedding.SimpleTextEmbedding

Preprocesses the text to embed into an array to pass into the model.

preprocessTextToEmbed(List<String>) - Method in interface ai.djl.modality.nlp.embedding.TextEmbedding

Preprocesses the text to embed into an array to pass into the model.

preprocessTextToEmbed(List<String>) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Preprocesses the text to embed into an array to pass into the model.

preprocessWordToEmbed(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Pre-processes the word to embed into an array to pass into the model.

preprocessWordToEmbed(String) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Pre-processes the word to embed into an array to pass into the model.

probabilities - Variable in class ai.djl.modality.Classifications

probabilities(ParameterStore, NDArray, boolean) - Method in class ai.djl.nn.transformer.IdEmbedding

Turns an array of embeddings of shape (d0 ...

processInput(TranslatorContext, Image) - Method in class ai.djl.modality.cv.translator.BaseImageTranslator

Processes the Image input and converts it to NDList.

processInput(TranslatorContext, Image) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Processes the Image input and converts it to NDList.

processInput(TranslatorContext, Path) - Method in class ai.djl.modality.cv.translator.wrapper.FileTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, InputStream) - Method in class ai.djl.modality.cv.translator.wrapper.InputStreamTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, URL) - Method in class ai.djl.modality.cv.translator.wrapper.UrlTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, String) - Method in class ai.djl.modality.nlp.translator.SimpleText2TextTranslator

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.ImageClassificationTranslator

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.wrapper.FileTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.wrapper.InputStreamTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.wrapper.UrlTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.YoloTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.YoloV5Translator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.nlp.translator.SimpleText2TextTranslator

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.

prod() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the product of this NDArray.

prod(int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

PunctuationSeparator - Class in ai.djl.modality.nlp.preprocess

PunctuationSeparator separates punctuation into a separate token.

PunctuationSeparator() - Constructor for class ai.djl.modality.nlp.preprocess.PunctuationSeparator

Q

QAgent - Class in ai.djl.modality.rl.agent

An RlAgent that implements Q or Deep-Q Learning.

QAgent(Trainer, float) - Constructor for class ai.djl.modality.rl.agent.QAgent

Constructs a QAgent.

QAgent(Trainer, float, Batchifier) - Constructor for class ai.djl.modality.rl.agent.QAgent

Constructs a QAgent with a custom Batchifier.

QAInput - Class in ai.djl.modality.nlp.qa

The input container for a Application.NLP.QUESTION_ANSWER model.

QAInput(String, String) - Constructor for class ai.djl.modality.nlp.qa.QAInput

Creates the BERT QA model.

QATranslator - Class in ai.djl.modality.nlp.translator

An abstract class to define the question answering translator.

QATranslator(QATranslator.BaseBuilder<?>) - Constructor for class ai.djl.modality.nlp.translator.QATranslator

QATranslator.BaseBuilder<T extends QATranslator.BaseBuilder> - Class in ai.djl.modality.nlp.translator

The builder for question answering translator.

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

An application that a reference document and a question about the document and returns text answering the question.

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.

RANDOM_FOREST - Static variable in interface ai.djl.Application.Tabular

This is erroneous because random forest is a technique (not deep learning), not an application.

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<T>> - Class in ai.djl.training.dataset

The Builder to construct a RandomAccessDataset.

randomAction() - Method in class ai.djl.modality.rl.ActionSpace

Returns a random action.

RandomBrightness - Class in ai.djl.modality.cv.transform

A Transform that randomly jitters image brightness with a factor chosen from [max(0, 1 - brightness), 1 + brightness].

RandomBrightness(float) - Constructor for class ai.djl.modality.cv.transform.RandomBrightness

Creates a RandomBrightness Transform.

randomBrightness(NDArray, float) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Randomly jitters image brightness with a factor chosen from [max(0, 1 - brightness), 1 + brightness].

RandomColorJitter - Class in ai.djl.modality.cv.transform

A Transform that randomly jitters the brightness, contrast, saturation, and hue of an image.

RandomColorJitter(float, float, float, float) - Constructor for class ai.djl.modality.cv.transform.RandomColorJitter

Creates a RandomColorJitter Transform.

randomColorJitter(NDArray, float, float, float, float) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Randomly jitters the brightness, contrast, saturation, and hue of an image.

RandomFlipLeftRight - Class in ai.djl.modality.cv.transform

A Transform that randomly flip the input image left to right with a probability of 0.5.

RandomFlipLeftRight() - Constructor for class ai.djl.modality.cv.transform.RandomFlipLeftRight

randomFlipLeftRight(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Randomly flip the input image left to right with a probability of 0.5.

RandomFlipTopBottom - Class in ai.djl.modality.cv.transform

A Transform that randomly flip the input image top to bottom with a probability of 0.5.

RandomFlipTopBottom() - Constructor for class ai.djl.modality.cv.transform.RandomFlipTopBottom

randomFlipTopBottom(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Randomly flip the input image top to bottom with a probability of 0.5.

RandomHue - Class in ai.djl.modality.cv.transform

A Transform that randomly jitters image hue with a factor chosen from [max(0, 1 - hue), 1 + hue].

RandomHue(float) - Constructor for class ai.djl.modality.cv.transform.RandomHue

Creates a RandomHue Transform.

randomHue(NDArray, float) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Randomly jitters image hue with a factor chosen from [max(0, 1 - hue), 1 + hue].

randomInteger(long, long, Shape, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Returns random integer values from low (inclusive) to high (exclusive).

randomInteger(long, long, Shape, DataType) - Method in interface ai.djl.ndarray.NDManager

Returns random integer values from low (inclusive) to high (exclusive).

randomMultinomial(int, NDArray) - Method in class ai.djl.ndarray.BaseNDManager

Draw samples from a multinomial distribution.

randomMultinomial(int, NDArray, Shape) - Method in class ai.djl.ndarray.BaseNDManager

Draw samples from a multinomial distribution.

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(float, float, Shape, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Draws random samples from a normal (Gaussian) 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) - 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) - Method in interface ai.djl.ndarray.NDManager

Draws random samples from a normal (Gaussian) distribution.

randomNormal(float, float, Shape, DataType, Device) - Method in interface ai.djl.ndarray.NDManager

Draws random samples from a normal (Gaussian) distribution.

RandomResizedCrop - Class in ai.djl.modality.cv.transform

A Transform that crop the input image with random scale and aspect ratio.

RandomResizedCrop(int, int, double, double, double, double) - Constructor for class ai.djl.modality.cv.transform.RandomResizedCrop

Creates a RandomResizedCrop Transform.

randomResizedCrop(NDArray, int, int, double, double, double, double) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Crop the input image with random scale and aspect ratio.

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, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Draws samples from a uniform distribution.

randomUniform(float, float, Shape) - Method in interface ai.djl.ndarray.NDManager

Draws samples from a uniform distribution.

randomUniform(float, float, Shape, DataType) - 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.

readInputShapes(DataInputStream) - Method in class ai.djl.nn.AbstractBlock

readParameters(Path, Map<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.

RecurrentBlock - Class in ai.djl.nn.recurrent

RecurrentBlock is an abstract implementation of recurrent neural networks.

RecurrentBlock(RecurrentBlock.BaseBuilder<?>) - Constructor for class ai.djl.nn.recurrent.RecurrentBlock

Creates a RecurrentBlock object.

RecurrentBlock.BaseBuilder<T extends RecurrentBlock.BaseBuilder> - Class in ai.djl.nn.recurrent

The Builder to construct a RecurrentBlock type of Block.

registerRepositoryFactory(RepositoryFactory) - Static method in interface ai.djl.repository.Repository

Registers a RepositoryFactory to handle the specified url scheme.

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.

RemoteRepository - Class in ai.djl.repository

A RemoteRepository is a Repository located on a remote web server.

RemoteRepository(String, URI) - Constructor for class ai.djl.repository.RemoteRepository

(Internal) Constructs a remote repository.

remove(String) - Method in class ai.djl.ndarray.NDList

Removes the first occurrence of the specified element from this NDList if it is present.

remove(NDList...) - Method in class ai.djl.nn.transformer.MemoryScope

Remove the given arrays from this scope and attach them back to this scopes parent NDManager.

remove(NDArray...) - Method in class ai.djl.nn.transformer.MemoryScope

Remove the given arrays from this scope and attach them back to this scopes parent NDManager.

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.

repeat(long) - Method in interface ai.djl.ndarray.NDArrayAdapter

Repeats element of this NDArray the number of times given repeats.

repeat(int, long) - Method in interface ai.djl.ndarray.NDArrayAdapter

Repeats element of this NDArray the number of times given repeats along given axis.

repeat(long[]) - Method in interface ai.djl.ndarray.NDArrayAdapter

Repeats element of this NDArray the number of times given repeats along each axis.

repeat(Shape) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

ReplayBuffer - Interface in ai.djl.modality.rl

Records RlEnv.Steps so that they can be trained on.

Repository - Interface in ai.djl.repository

Repository is a format for storing data Artifacts for various uses including deep learning models and datasets.

RepositoryFactory - Interface in ai.djl.repository

A interface responsible to create Repository instances.

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

reset() - Method in interface ai.djl.modality.rl.env.RlEnv

Resets the environment to it's default state.

reset(String, long, String) - Method in class ai.djl.training.util.ProgressBar

resetAccumulator(String) - Method in class ai.djl.training.evaluator.AbstractAccuracy

Resets the evaluator value with the given key.

resetAccumulator(String) - Method in class ai.djl.training.evaluator.BoundingBoxError

Resets the evaluator value with the given key.

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.

reshape(Shape) - Method in interface ai.djl.ndarray.NDArrayAdapter

Reshapes this NDArray to the given Shape.

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(int, int, Image.Interpolation) - Constructor for class ai.djl.modality.cv.transform.Resize

Creates a Resize Transform that resizes to the given width and height with given interpolation.

resize(NDArray, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Resizes an image to the given width and height.

resize(NDArray, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Resizes an image to the given width and height.

resize(NDArray, int, int, Image.Interpolation) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Resizes an image to the given width and height with given interpolation.

resolve(MRL, String, Map<String, String>) - Method in class ai.djl.repository.JarRepository

Returns the artifact matching a mrl, version, and property filter.

resolve(MRL, String, Map<String, String>) - Method in class ai.djl.repository.LocalRepository

Returns the artifact matching a mrl, version, and property filter.

resolve(MRL, String, Map<String, String>) - Method in class ai.djl.repository.RemoteRepository

Returns the artifact matching a mrl, version, and property filter.

resolve(MRL, String, Map<String, String>) - Method in interface ai.djl.repository.Repository

Returns the artifact matching a mrl, version, and property filter.

resolve(MRL, String, Map<String, String>) - Method in class ai.djl.repository.SimpleRepository

Returns the artifact matching a mrl, version, and property filter.

resolve(MRL, String, Map<String, String>) - Method in class ai.djl.repository.SimpleUrlRepository

Returns the artifact matching a mrl, version, and property filter.

resolvePath(Artifact.Item, String) - Method in class ai.djl.repository.AbstractRepository

resolvePath(Artifact.Item, String) - Method in class ai.djl.repository.SimpleRepository

Resource - Class in ai.djl.repository

A class represents a resource in a Repository.

Resource(Repository, MRL, String) - Constructor for class ai.djl.repository.Resource

Constructs a Resource instance.

resource - Variable in class ai.djl.repository.zoo.BaseModelLoader

resources - Variable in class ai.djl.ndarray.BaseNDManager

results - Variable in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

returnState - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

returnState - Variable in class ai.djl.nn.recurrent.RecurrentBlock

RlAgent - Interface in ai.djl.modality.rl.agent

An RlAgent is the model or technique to decide the actions to take in an RlEnv.

RlEnv - Interface in ai.djl.modality.rl.env

An environment to use for reinforcement learning.

RlEnv.Step - Interface in ai.djl.modality.rl.env

A record of taking a step in the environment.

rmsprop() - Static method in class ai.djl.training.optimizer.Optimizer

Returns a new instance of RmsProp.Builder that can build an RmsProp optimizer.

RmsProp - Class in ai.djl.training.optimizer

The RMSProp Optimizer.

RmsProp(RmsProp.Builder) - Constructor for class ai.djl.training.optimizer.RmsProp

Creates a new instance of RMSProp optimizer.

RmsProp.Builder - Class in ai.djl.training.optimizer

The Builder to construct an RmsProp object.

RNN - Class in ai.djl.nn.recurrent

RNN is an implementation of recurrent neural networks which 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.

rotate90(NDArray, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Rotate an image NDArray counter-clockwise 90 degree.

rotate90(int, int[]) - Method in interface ai.djl.ndarray.NDArray

Rotates an array by 90 degrees in the plane specified by axes.

rotate90(int, int[]) - Method in interface ai.djl.ndarray.NDArrayAdapter

Rotates an array by 90 degrees in the plane specified by axes.

round() - Method in interface ai.djl.ndarray.NDArray

Returns the round of this NDArray element-wise.

round() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the round of this NDArray element-wise.

runEnvironment(RlAgent, boolean) - Method in interface ai.djl.modality.rl.env.RlEnv

Runs the environment from reset until done.

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(BufferedImage, OutputStream, String) - Method in class ai.djl.modality.cv.BufferedImageFactory

save(OutputStream, String) - Method in interface ai.djl.modality.cv.Image

Save the image to file.

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.

save(InputStream, Path, URI, Artifact.Item, Progress) - Method in class ai.djl.repository.AbstractRepository

save(Path, String) - Method in class ai.djl.repository.zoo.ZooModel

Saves the model to the specified modelPath with the name provided.

saveInputShapes(DataOutputStream) - Method in class ai.djl.nn.AbstractBlock

saveMetadata(DataOutputStream) - Method in class ai.djl.nn.AbstractBlock

Override this method to save additional data apart from parameter values.

saveMetadata(DataOutputStream) - Method in class ai.djl.nn.core.Linear

Override this method to save additional data apart from parameter values.

saveMetadata(DataOutputStream) - Method in class ai.djl.nn.norm.BatchNorm

Override this method to save additional data apart from parameter values.

saveModel(Model, TrainingResult) - Method in class ai.djl.training.hyperparameter.EasyHpo

Saves the best hyperparameter set.

saveModel(Trainer) - Method in class ai.djl.training.listener.SaveModelTrainingListener

SaveModelTrainingListener - Class in ai.djl.training.listener

A TrainingListener that saves a model and can save checkpoints.

SaveModelTrainingListener(String) - Constructor for class ai.djl.training.listener.SaveModelTrainingListener

Constructs a SaveModelTrainingListener using the model's name.

SaveModelTrainingListener(String, String) - Constructor for class ai.djl.training.listener.SaveModelTrainingListener

Constructs a SaveModelTrainingListener.

SaveModelTrainingListener(String, String, int) - Constructor for class ai.djl.training.listener.SaveModelTrainingListener

Constructs a SaveModelTrainingListener.

saveParameters(DataOutputStream) - Method in class ai.djl.modality.nlp.Decoder

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.modality.nlp.Encoder

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.modality.nlp.EncoderDecoder

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.AbstractBlock

Writes the parameters of the block to the given outputStream.

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.core.ConstantEmbedding

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.

ScaledDotProductAttentionBlock - Class in ai.djl.nn.transformer

A Block implementing scaled product attention according to Vaswani et.

ScaledDotProductAttentionBlock.Builder - Class in ai.djl.nn.transformer

A builder for ScaledDotProductAttentionBlocks.

scaleGradient(double) - Method in interface ai.djl.ndarray.NDArray

Returns an NDArray equal to this that magnifies the gradient propagated to this by a constant.

search(VersionRange, Map<String, String>) - Method in class ai.djl.repository.Metadata.MatchAllMetadata

Returns the artifacts matching the version and property requirements.

search(VersionRange, Map<String, String>) - Method in class ai.djl.repository.Metadata

Returns the artifacts matching the version and property requirements.

self() - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

self() - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

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.modality.cv.translator.YoloTranslator.Builder

self() - Method in class ai.djl.modality.cv.translator.YoloV5Translator.Builder

self() - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding.Builder

Returns this {code Builder} object.

self() - Method in class ai.djl.modality.nlp.translator.QATranslator.BaseBuilder

self() - Method in class ai.djl.nn.convolutional.Conv1d.Builder

self() - Method in class ai.djl.nn.convolutional.Conv1dTranspose.Builder

self() - Method in class ai.djl.nn.convolutional.Conv2d.Builder

self() - Method in class ai.djl.nn.convolutional.Conv2dTranspose.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.convolutional.Deconvolution.DeconvolutionBuilder

self() - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Returns this {code Builder} object.

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.RecurrentBlock.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.Adadelta.Builder

self() - Method in class ai.djl.training.optimizer.Adagrad.Builder

self() - Method in class ai.djl.training.optimizer.Adam.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.RmsProp.Builder

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

An application that classifies each pixel in an image into a category.

SENTIMENT_ANALYSIS - Static variable in interface ai.djl.Application.NLP

An application that classifies text into positive or negative, an specific case of Application.NLP.TEXT_CLASSIFICATION.

sequenceMask(NDArray, float) - Method in interface ai.djl.ndarray.NDArray

Sets all elements outside the sequence to a constant value.

sequenceMask(NDArray) - Method in interface ai.djl.ndarray.NDArray

Sets all elements outside the sequence to 0.

sequenceMask(NDArray, float) - Method in interface ai.djl.ndarray.NDArrayAdapter

Sets all elements outside the sequence to a constant value.

sequenceMask(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

Sets all elements outside the sequence to 0.

sequenceMask(NDArray, NDArray, float) - Static method in class ai.djl.ndarray.NDArrays

Sets all elements of the given NDArray outside the sequence NDArray to a constant value.

sequenceMask(NDArray, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Sets all elements of the given NDArray outside the sequence NDArray to 0.

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

Creates an empty sequential block.

serialize(Classifications, Type, JsonSerializationContext) - Method in class ai.djl.modality.Classifications.ClassificationsSerializer

ServingTranslator - Interface in ai.djl.translate

A Translator that can handle generic Input and Output.

ServingTranslatorFactory - Class in ai.djl.translate

A TranslatorFactory that creates an generic Translator.

ServingTranslatorFactory() - Constructor for class ai.djl.translate.ServingTranslatorFactory

set(NDArray, NDIndexFullSlice, NDArray) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets the values of the array at the fullSlice with an array.

set(NDArray, NDIndexBooleans, NDArray) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets the values of the array at the boolean locations with an array.

set(NDArray, NDIndex, NDArray) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets the values of the array at the index locations with an array.

set(NDArray, NDIndexFullSlice, Number) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets the values of the array at the fullSlice with a number.

set(NDArray, NDIndex, Number) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets the values of the array at the index locations with a number.

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.

set(NDIndex, Function<NDArray, NDArray>) - Method in interface ai.djl.ndarray.NDArray

Sets the specific index by a function.

set(Buffer) - Method in interface ai.djl.ndarray.NDArrayAdapter

Sets this NDArray value from Buffer.

setActivation(RNN.Activation) - Method in class ai.djl.nn.recurrent.RNN.Builder

Sets the activation for the RNN - ReLu or Tanh.

setApplication(Application) - Method in class ai.djl.repository.Metadata

Sets the Application.

setArguments(Map<String, Object>) - Method in class ai.djl.repository.Artifact

Sets the artifact arguments.

setArguments(Map<String, Object>) - Method in interface ai.djl.translate.ServingTranslator

Sets the configurations for the Translator instance.

setArray(NDArray) - Method in class ai.djl.nn.Parameter

Sets the values of this Parameter.

setArtifact(Artifact) - Method in class ai.djl.repository.Artifact.Item

Sets the artifact associated with this item.

setArtifactId(String) - Method in class ai.djl.repository.Metadata

Sets the artifactId.

setArtifacts(List<Artifact>) - Method in class ai.djl.repository.Metadata

Sets the artifacts for the metadata.

setAttachment(String, Object) - Method in interface ai.djl.translate.TranslatorContext

Set a key-value pair of attachments.

setBaseValue(float) - Method in class ai.djl.training.tracker.CosineTracker.Builder

Sets the initial value after no steps.

setBaseValue(float) - Method in class ai.djl.training.tracker.FactorTracker.Builder

Sets the initial value after no steps.

setBaseValue(float) - Method in class ai.djl.training.tracker.LinearTracker.Builder

Sets the initial value after no steps.

setBaseValue(float) - Method in class ai.djl.training.tracker.MultiFactorTracker.Builder

Sets the initial value after no steps.

setBaseValue(float) - Method in class ai.djl.training.tracker.PolynomialDecayTracker.Builder

Sets the initial value after no steps.

setBatchifier(Batchifier) - Method in class ai.djl.translate.NoopTranslator

Sets the Batchifier for the Translator.

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.

setBlock(Block) - Method in class ai.djl.repository.zoo.ZooModel

Sets the block for the Model for training and inference.

setCheckpoint(int) - Method in class ai.djl.training.listener.SaveModelTrainingListener

Sets the checkpoint frequency.

setCode(int) - Method in class ai.djl.modality.Output

Sets the status code of the output.

setContent(PairList<String, byte[]>) - Method in class ai.djl.modality.Input

Sets the content of the input.

setContent(byte[]) - Method in class ai.djl.modality.Output

Sets the content of the input.

setContent(String) - Method in class ai.djl.modality.Output

Sets the content of the input with string value.

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.

setDataType(DataType) - Method in class ai.djl.repository.zoo.ZooModel

Sets the standard data type used within the model.

setDecaySteps(int) - Method in class ai.djl.training.tracker.PolynomialDecayTracker.Builder

Sets the number of training steps to decay learning rate in.

setDescription(String) - Method in class ai.djl.repository.Metadata

Sets the description.

setDictionarySize(int) - Method in class ai.djl.nn.transformer.IdEmbedding.Builder

Sets the number of ids that should be embedded.

setEmbeddingSize(int) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Sets the size of the embeddings.

setEmbeddingSize(int) - Method in class ai.djl.nn.transformer.IdEmbedding.Builder

Sets the size of the embeddings.

setEmbeddingSize(int) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock.Builder

Sets the embedding Size to be used for the internal token representation.

setEndLearningRate(float) - Method in class ai.djl.training.tracker.PolynomialDecayTracker.Builder

Sets the learning rate at which to end rate decay.

setEpoch(int) - Method in class ai.djl.training.TrainingResult

Sets the actual number of epoch.

setEvaluations(Map<String, Float>) - Method in class ai.djl.training.TrainingResult

Sets the raw evaluation metrics.

setExtension(String) - Method in class ai.djl.repository.Artifact.Item

Sets the file extension.

setFactor(float) - Method in class ai.djl.training.tracker.FactorTracker.Builder

Sets the value of the multiplicative factor.

setFiles(Map<String, Artifact.Item>) - Method in class ai.djl.repository.Artifact

Sets the file items.

setFilters(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the Required number of filters.

setFilters(int) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the Required number of filters.

setGroupId(String) - Method in class ai.djl.repository.Metadata

Sets the groupId.

setHeadCount(int) - Method in class ai.djl.nn.transformer.ScaledDotProductAttentionBlock.Builder

Sets the number of attention Heads, must divide the embedding size without rest.

setId(String) - Method in class ai.djl.repository.License

Sets the identifier of the license.

setImageFactory(ImageFactory) - Static method in class ai.djl.modality.cv.ImageFactory

Sets a custom instance of ImageFactory.

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.

setKernelShape(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the shape of the kernel.

setKernelShape(Shape) - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Sets the shape of the kernel.

setLastUpdated(Date) - Method in class ai.djl.repository.Metadata

Sets the last update date for the metadata.

setLearningRateTracker(Tracker) - Method in class ai.djl.training.optimizer.Nag.Builder

Sets the Tracker for this optimizer.

setLearningRateTracker(Tracker) - Method in class ai.djl.training.optimizer.Sgd.Builder

Sets the Tracker for this optimizer.

setLicense(Map<String, License>) - Method in class ai.djl.repository.Metadata

Sets the License.

setMainTracker(Tracker) - Method in class ai.djl.training.tracker.WarmUpTracker.Builder

Sets the base value.

setMandatoryDataType(DataType) - Method in class ai.djl.nn.Parameter

Sets the mandatory data type for this Parameter.

setMaxUpdates(int) - Method in class ai.djl.training.tracker.CosineTracker.Builder

Sets the maximum number of updates after which the value should remain constant.

setMessage(String) - Method in class ai.djl.modality.Output

Sets the status message of the output.

setMetadata(Metadata) - Method in class ai.djl.repository.Artifact

Sets the associated metadata.

setMetadataVersion(String) - Method in class ai.djl.repository.Artifact

Sets the metadata format version.

setMetadataVersion(String) - Method in class ai.djl.repository.Metadata

Sets the metadata format version.

setMetrics(Metrics) - Method in class ai.djl.inference.Predictor

Attaches a Metrics param to use for benchmark.

setMetrics(Metrics) - Method in class ai.djl.training.Trainer

Attaches a Metrics param to use for benchmarking.

setModelDir(Path) - 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 class ai.djl.ndarray.BaseNDManager

Sets the name for the NDManager.

setName(String) - Method in interface ai.djl.ndarray.NDArray

Sets name of this NDArray.

setName(String) - Method in interface ai.djl.ndarray.NDArrayAdapter

Sets name of this NDArray.

setName(String) - Method in interface ai.djl.ndarray.NDManager

Sets the name for the NDManager.

setName(String) - Method in class ai.djl.ndarray.types.DataDesc

Sets the name of the NDArray.

setName(String) - Method in class ai.djl.repository.Artifact.Item

Sets the item name.

setName(String) - Method in class ai.djl.repository.Artifact

Sets the artifact name.

setName(String) - Method in class ai.djl.repository.License

Sets the name of the license.

setName(String) - Method in class ai.djl.repository.Metadata

Sets the metadata-level name.

setNumLayers(int) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the Required number of stacked layers.

setOptions(Map<String, String>) - Method in class ai.djl.repository.Artifact

Sets the artifact arguments.

setOverrideModelName(String) - Method in class ai.djl.training.listener.SaveModelTrainingListener

Sets the override model name to save checkpoints with.

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.BaseImageTranslator.BaseBuilder

Sets the Pipeline to use for pre-processing the image.

setProperties(Map<String, String>) - Method in class ai.djl.modality.Input

Sets the properties of the input.

setProperties(Map<String, String>) - Method in class ai.djl.modality.Output

Sets the properties of the output.

setProperties(Map<String, String>) - Method in class ai.djl.repository.Artifact

Sets the artifact properties.

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.

setProperty(String, String) - Method in class ai.djl.repository.zoo.ZooModel

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.

setRepositoryUri(URI) - Method in class ai.djl.repository.Metadata

Sets the repository URI.

setRescaleGrad(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the value used to rescale the gradient.

setResourceType(String) - Method in class ai.djl.repository.Metadata

Returns the resource type.

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.

setSaveModelCallback(Consumer<Trainer>) - Method in class ai.djl.training.listener.SaveModelTrainingListener

Sets the callback function on model saving.

setScalar(NDArray, NDIndex, Number) - Method in class ai.djl.ndarray.index.NDArrayIndexer

Sets a scalar value in the array at the indexed location.

setScalar(NDIndex, Number) - Method in interface ai.djl.ndarray.NDArray

Sets the specified scalar in this NDArray with the given value.

setSha1Hash(String) - Method in class ai.djl.repository.Artifact.Item

Sets the sha1hash of the item.

setShape(Shape) - Method in class ai.djl.ndarray.types.DataDesc

Sets the Shape of the NDArray.

setSize(long) - Method in class ai.djl.repository.Artifact.Item

Sets the file size.

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.

setSnapshot(boolean) - Method in class ai.djl.repository.Artifact

Sets if the artifact is a snapshot.

setStateSize(int) - Method in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

Sets the Required size of the state for each layer.

setSteps(int[]) - Method in class ai.djl.training.tracker.MultiFactorTracker.Builder

Sets an array of integers indicating when the value should be changed, usually in an uneven interval of steps.

setTokenDictionarySize(int) - Method in class ai.djl.nn.transformer.BertBlock.Builder

Sets the number of tokens in the dictionary.

setType(Class<String>) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding.Builder

Creates a new Embedding.BaseBuilder with the specified embedding type.

setType(Class<T>) - Method in class ai.djl.nn.core.Embedding.BaseBuilder

Creates a new Embedding.BaseBuilder with the specified embedding type.

setType(String) - Method in class ai.djl.repository.Artifact.Item

Sets the type of the item.

setTypes(Class<P>, Class<Q>) - Method in class ai.djl.repository.zoo.Criteria.Builder

Creates a new @{code Builder} class with the specified input and output data type.

setUnits(long) - Method in class ai.djl.nn.core.Linear.Builder

Sets the number of output channels.

setupHyperParams() - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the initial hyperparameters.

setupTrainingConfig(HpSet) - Method in class ai.djl.training.hyperparameter.EasyHpo

Returns the TrainingConfig to use to train each hyperparameter set.

setUri(String) - Method in class ai.djl.repository.Artifact.Item

Sets the URI of the item.

setUrl(String) - Method in class ai.djl.repository.License

Sets the url of the license.

setVersion(String) - Method in class ai.djl.repository.Artifact

Sets the artifact version.

setVocabulary(Vocabulary) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding.Builder

Sets the Vocabulary to be used.

setWebsite(String) - Method in class ai.djl.repository.Metadata

Sets the website.

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 (SGD) 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.

sign() - Method in interface ai.djl.ndarray.NDArray

Returns the element-wise sign.

sign() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the element-wise sign.

signi() - Method in interface ai.djl.ndarray.NDArray

Returns the element-wise sign in-place.

signi() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the element-wise sign in-place.

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.

SimplePoseModelLoader - Class in ai.djl.modality.cv.zoo

The translator for Simple Pose models.

SimplePoseModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.SimplePoseModelLoader

Creates the Model loader from the given repository.

SimplePoseTranslator - Class in ai.djl.modality.cv.translator

A BaseImageTranslator 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.

SimpleRepository - Class in ai.djl.repository

A SimpleRepository is a Repository containing only a single artifact without requiring a "metadata.json" file.

SimpleRepository(String, Path, String, String) - Constructor for class ai.djl.repository.SimpleRepository

(Internal) Constructs a SimpleRepository.

SimpleText2TextTranslator - Class in ai.djl.modality.nlp.translator

A Translator that performs pre-process and post-processing for a sequence-to-sequence text model.

SimpleText2TextTranslator() - Constructor for class ai.djl.modality.nlp.translator.SimpleText2TextTranslator

SimpleTextEmbedding - Class in ai.djl.modality.nlp.embedding

A TextEmbedding that applies a WordEmbedding to each word independently.

SimpleTextEmbedding(WordEmbedding) - Constructor for class ai.djl.modality.nlp.embedding.SimpleTextEmbedding

Constructs a SimpleTextEmbedding.

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.

SimpleUrlRepository - Class in ai.djl.repository

A SimpleUrlRepository is a Repository contains an archive file from a HTTP URL.

SimpleVocabulary - Class in ai.djl.modality.nlp

The simple implementation of Vocabulary.

SimpleVocabulary(SimpleVocabulary.Builder) - Constructor for class ai.djl.modality.nlp.SimpleVocabulary

Create a SimpleVocabulary object with a SimpleVocabulary.Builder.

SimpleVocabulary(List<String>) - Constructor for class ai.djl.modality.nlp.SimpleVocabulary

Create a SimpleVocabulary object with the given list of tokens.

SimpleVocabulary.Builder - Class in ai.djl.modality.nlp

Builder class that is used to build the SimpleVocabulary.

sin() - Method in interface ai.djl.ndarray.NDArray

Returns the trigonometric sine of this NDArray element-wise.

sin() - Method in interface ai.djl.ndarray.NDArrayAdapter

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 BaseImageTranslator 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.

singleton(Function<NDArray, NDArray>) - Static method in class ai.djl.nn.LambdaBlock

Creates a LambdaBlock for a singleton function.

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.

sinh() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the hyperbolic sine of this NDArray element-wise.

size() - Method in class ai.djl.modality.nlp.SimpleVocabulary

Returns the size of the Vocabulary.

size() - Method in interface 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.

sliceAxis(int, long, long) - Static method in class ai.djl.ndarray.index.NDIndex

Creates an NDIndex that just has one slice in the given axis.

softmax(int) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function along the given axis.

softmax(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

Applies the softmax function along the given axis.

SOFTMAX_REGRESSION - Static variable in interface ai.djl.Application.Tabular

An application that takes a feature vector (table row) and predicts a categorical feature based on it.

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.

softPlus(NDArray) - Static method in class ai.djl.nn.Activation

Applies softPlus activation on the input NDArray.

softPlus(NDList) - Static method in class ai.djl.nn.Activation

Applies softPlus activation on the input singleton NDList.

softPlusBlock() - 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.

sort() - Method in interface ai.djl.ndarray.NDArrayAdapter

Sorts the flattened NDArray.

sort(int) - Method in interface ai.djl.ndarray.NDArrayAdapter

Sorts the flattened NDArray.

SparseFormat - Enum in ai.djl.ndarray.types

An enum representing Sparse matrix storage formats.

sparseGrad - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

sparseGrad - Variable in class ai.djl.nn.core.Embedding

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(long[], int) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.PaddingStackBatchifier

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.

sqrt() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

square() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

squeeze(int[]) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

stateSize - Variable in class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

stateSize - Variable in class ai.djl.nn.recurrent.RecurrentBlock

step(NDList, boolean) - Method in interface ai.djl.modality.rl.env.RlEnv

Takes a step by performing an action in this environment.

step() - Method in class ai.djl.training.Trainer

Updates all of the parameters of the model once.

stopGradient() - Method in interface ai.djl.ndarray.NDArray

Returns an NDArray equal to this that stop gradient propagation through it.

stopGradient() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns an NDArray equal to this that stop gradient propagation through it.

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

stride - Variable in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

stride - Variable in class ai.djl.nn.convolutional.Deconvolution

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Subtracts a number from this NDArray element-wise.

sub(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

subDataset(int, int) - Method in class ai.djl.training.dataset.RandomAccessDataset

Returns a view of the portion of this data between the specified fromIndex, inclusive, and toIndex, exclusive.

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(Number) - Method in interface ai.djl.ndarray.NDArrayAdapter

Subtracts a number from this NDArray element-wise in place.

subi(NDArray) - Method in interface ai.djl.ndarray.NDArrayAdapter

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 sum of this NDArray along given axes.

sum(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the sum of this NDArray along given axes.

sum() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the sum of this NDArray.

sum(int[], boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the sum 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.

synsetLoader - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

SynsetLoader(List<String>) - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator.SynsetLoader

SynsetLoader(URL) - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator.SynsetLoader

SynsetLoader(String) - Constructor for class ai.djl.modality.cv.translator.BaseImageTranslator.SynsetLoader

T

tail() - Method in class ai.djl.ndarray.types.Shape

Returns the tail index of the shape.

tan() - Method in interface ai.djl.ndarray.NDArray

Returns the trigonometric tangent of this NDArray element-wise.

tan() - Method in interface ai.djl.ndarray.NDArrayAdapter

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() - Method in interface ai.djl.ndarray.NDArrayAdapter

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

An application that classifies text data.

TEXT_EMBEDDING - Static variable in interface ai.djl.Application.NLP

An application that takes text and returns a feature vector that represents the text.

TextCleaner - Class in ai.djl.modality.nlp.preprocess

Applies remove or replace of certain characters based on condition.

TextCleaner(Function<Character, Boolean>) - Constructor for class ai.djl.modality.nlp.preprocess.TextCleaner

Remove a character if it meets the condition supplied.

TextCleaner(Function<Character, Boolean>, char) - Constructor for class ai.djl.modality.nlp.preprocess.TextCleaner

Replace a character if it meets the condition supplied.

TextEmbedding - Interface in ai.djl.modality.nlp.embedding

A class to manage 1-D NDArray representations of multiple words.

TextProcessor - Interface in ai.djl.modality.nlp.preprocess

TextProcessor allows applying pre-processing to input tokens for natural language applications.

TextTerminator - Class in ai.djl.modality.nlp.preprocess

A TextProcessor that adds a beginning of string and end of string token.

TextTerminator() - Constructor for class ai.djl.modality.nlp.preprocess.TextTerminator

Constructs a default TextTerminator.

TextTerminator(boolean, boolean) - Constructor for class ai.djl.modality.nlp.preprocess.TextTerminator

Constructs a TextTerminator using the default tokens.

TextTerminator(boolean, boolean, String, String) - Constructor for class ai.djl.modality.nlp.preprocess.TextTerminator

Constructs a TextTerminator.

TextTruncator - Class in ai.djl.modality.nlp.preprocess

TextProcessor that truncates text to a maximum size.

TextTruncator(int) - Constructor for class ai.djl.modality.nlp.preprocess.TextTruncator

Constructs a TextTruncator.

threshold - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator.ObjectDetectionBuilder

threshold - Variable in class ai.djl.modality.cv.translator.ObjectDetectionTranslator

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.

tile(long) - Method in interface ai.djl.ndarray.NDArrayAdapter

Constructs a NDArray by repeating this NDArray the number of times given repeats.

tile(int, long) - Method in interface ai.djl.ndarray.NDArrayAdapter

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.NDArrayAdapter

Constructs a NDArray by repeating this NDArray the number of times given by repeats.

tile(Shape) - Method in interface ai.djl.ndarray.NDArrayAdapter

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

Constructs a TimeMeasureTrainingListener.

toArray() - Method in interface ai.djl.ndarray.NDArray

Converts this NDArray to a Number array based on its DataType.

toArray() - Method in class ai.djl.training.dataset.RandomAccessDataset

Returns the dataset contents as a Java array.

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.

toByteBuffer() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

toDegrees() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

toDense() - Method in interface ai.djl.ndarray.NDArrayAdapter

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.

toDevice(Device, boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Moves this NDArray to a different Device.

toDevice(Device, boolean) - Method in class ai.djl.ndarray.NDList

Converts all the NDArray in NDList 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.bert.BertFullTokenizer

Breaks down the given sentence into a list of tokens that can be represented by embeddings.

tokenize(String) - Method in class ai.djl.modality.nlp.bert.BertTokenizer

Breaks down the given sentence into a list of tokens that can be represented by embeddings.

tokenize(String) - Method in class ai.djl.modality.nlp.bert.WordpieceTokenizer

Breaks down the given sentence into a list of tokens that can be represented by embeddings.

tokenize(String) - Method in class ai.djl.modality.nlp.preprocess.SimpleTokenizer

Breaks down the given sentence into a list of tokens that can be represented by embeddings.

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) - Method in interface ai.djl.modality.cv.Image

Converts image to a RGB NDArray.

toNDArray(NDManager, Image.Flag) - Method in interface ai.djl.modality.cv.Image

Converts image to a NDArray.

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.

toRadians() - Method in interface ai.djl.ndarray.NDArrayAdapter

Converts this NDArray from degrees to radians element-wise.

toSparse(SparseFormat) - Method in interface ai.djl.ndarray.NDArray

Returns a sparse representation of NDArray.

toSparse(SparseFormat) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns a sparse representation of NDArray.

toString() - Method in class ai.djl.Application

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.AbstractBlock

toString() - Method in class ai.djl.nn.LambdaBlock

toString() - Method in class ai.djl.nn.norm.Dropout

toString() - Method in class ai.djl.nn.ParallelBlock

toString() - Method in class ai.djl.nn.SequentialBlock

toString() - Method in class ai.djl.repository.Artifact

toString() - Method in class ai.djl.repository.MRL

toString() - Method in class ai.djl.repository.Version

toString() - Method in class ai.djl.repository.VersionRange

toString() - Method in class ai.djl.repository.zoo.BaseModelLoader

toString() - Method in class ai.djl.repository.zoo.Criteria

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.

toType(DataType, boolean) - Method in interface ai.djl.ndarray.NDArrayAdapter

Converts this NDArray to a different DataType.

toUint8Array() - Method in interface ai.djl.ndarray.NDArray

Converts this NDArray to a uint8 array.

toURI() - Method in class ai.djl.repository.MRL

Returns the URI to the metadata location (used for Repository implementations).

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.

trace(int, int, int) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the sum along diagonals of this NDArray.

Tracker - Interface in ai.djl.training.tracker

A Tracker represents a hyper-parameter that changes gradually through the training process.

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

TrainableTextEmbedding - Class in ai.djl.modality.nlp.embedding

TrainableTextEmbedding is an implementation of TextEmbedding based on TrainableWordEmbedding block.

TrainableTextEmbedding(TrainableWordEmbedding) - Constructor for class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Constructs a TrainableTextEmbedding.

TrainableWordEmbedding - Class in ai.djl.modality.nlp.embedding

TrainableWordEmbedding is an implementation of WordEmbedding and Embedding based on a SimpleVocabulary.

TrainableWordEmbedding(TrainableWordEmbedding.Builder) - Constructor for class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Constructs a new instance of TrainableWordEmbedding from the TrainableWordEmbedding.Builder.

TrainableWordEmbedding(Vocabulary, int) - Constructor for class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Constructs a new instance of TrainableWordEmbedding from a SimpleVocabulary and a given embedding size.

TrainableWordEmbedding(NDArray, List<String>) - Constructor for class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Constructs a pretrained embedding.

TrainableWordEmbedding(NDArray, List<String>, boolean) - Constructor for class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Constructs a pretrained embedding.

TrainableWordEmbedding.Builder - Class in ai.djl.modality.nlp.embedding

A builder for a TrainableWordEmbedding.

trainBatch(RlEnv.Step[]) - Method in class ai.djl.modality.rl.agent.EpsilonGreedy

Trains this RlAgent on a batch of RlEnv.Steps.

trainBatch(RlEnv.Step[]) - Method in class ai.djl.modality.rl.agent.QAgent

Trains this RlAgent on a batch of RlEnv.Steps.

trainBatch(RlEnv.Step[]) - Method in interface ai.djl.modality.rl.agent.RlAgent

Trains this RlAgent on a batch of RlEnv.Steps.

trainBatch(Trainer, Batch) - Static method in class ai.djl.training.EasyTrain

Trains the model with one iteration of the given Batch of data.

trainBatch(Trainer, Batch) - Method in class ai.djl.training.ParallelTrain

Trains the model with one iteration of the given Batch of data.

Trainer - Class in ai.djl.training

The Trainer interface provides a session for model training.

Trainer(Model, TrainingConfig) - Constructor for class ai.djl.training.Trainer

Creates an instance of Trainer with the given Model and TrainingConfig.

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 performs 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.

TrainingListenerAdapter - Class in ai.djl.training.listener

Base implementation of the training listener that does nothing.

TrainingListenerAdapter() - Constructor for class ai.djl.training.listener.TrainingListenerAdapter

TrainingResult - Class in ai.djl.training

A class that is responsible for holding the training result produced by Trainer.

TrainingResult() - Constructor for class ai.djl.training.TrainingResult

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.RandomBrightness

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.RandomColorJitter

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.RandomFlipLeftRight

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.RandomFlipTopBottom

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.RandomHue

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.RandomResizedCrop

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.

TransformerEncoderBlock - Class in ai.djl.nn.transformer

Self-Attention based transformer encoder block.

TransformerEncoderBlock(int, int, int, float, Function<NDList, NDList>) - Constructor for class ai.djl.nn.transformer.TransformerEncoderBlock

Creates a transformer encoder block.

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.

transpose() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns this NDArray with axes transposed.

transpose(int...) - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns this NDArray with given axes transposed.

trunc() - Method in interface ai.djl.ndarray.NDArray

Returns the truncated value of this NDArray element-wise.

trunc() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns the truncated value of this NDArray element-wise.

TruncatedNormalInitializer - Class in ai.djl.training.initializer

Naive implementation of a truncated normal initializer.

TruncatedNormalInitializer() - Constructor for class ai.djl.training.initializer.TruncatedNormalInitializer

Creates an instance of TruncatedNormalInitializer with a default sigma of 0.01.

TruncatedNormalInitializer(float) - Constructor for class ai.djl.training.initializer.TruncatedNormalInitializer

Creates a TruncatedNormalInitializer initializer.

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.PaddingStackBatchifier

Reverses the batchify operation.

unbatchify(NDList) - Method in class ai.djl.translate.StackBatchifier

Reverses the batchify operation.

UNDEFINED - Static variable in class ai.djl.Application

undefined(String, String) - Static method in class ai.djl.repository.MRL

Creates a dataset MRL with specified application.

unembed(long) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Returns the item corresponding to the given index.

unembed(long) - Method in interface ai.djl.nn.core.AbstractIndexedEmbedding

Returns the item corresponding to the given index.

unembed(long) - Method in class ai.djl.nn.core.ConstantEmbedding

Returns the item corresponding to the given index.

unembed(long) - Method in class ai.djl.nn.core.Embedding.DefaultEmbedding

Returns the item corresponding to the given index.

unembed(long) - Method in class ai.djl.nn.core.Embedding.DefaultItem

Returns the item corresponding to the given index.

unembedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.ModelZooTextEmbedding

Returns the closest matching text for a given embedding.

unembedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.SimpleTextEmbedding

Returns the closest matching text for a given embedding.

unembedText(NDArray) - Method in interface ai.djl.modality.nlp.embedding.TextEmbedding

Returns the closest matching text for a given embedding.

unembedText(NDArray) - Method in class ai.djl.modality.nlp.embedding.TrainableTextEmbedding

Returns the closest matching text for a given embedding.

unembedWord(NDArray) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Returns the closest matching word for the given index.

unembedWord(NDArray) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Returns the closest matching word for the given index.

UnicodeNormalizer - Class in ai.djl.modality.nlp.preprocess

Applies unicode normalization to input strings.

UnicodeNormalizer(Normalizer.Form) - Constructor for class ai.djl.modality.nlp.preprocess.UnicodeNormalizer

Unicode normalizer with a configurable normal form.

UnicodeNormalizer() - Constructor for class ai.djl.modality.nlp.preprocess.UnicodeNormalizer

Default version of the Unicode Normalizer using NFKC normal form.

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.

UNSUPPORTED_MSG - Static variable in interface ai.djl.ndarray.NDArrayAdapter

update(Shape, int, long) - Static method in class ai.djl.ndarray.types.Shape

Returns a new shape altering the given dimension.

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.LocalParameterServer

Updates the parameter of a key from Parameter Server.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Adadelta

Updates the parameters according to the gradients.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Adagrad

Updates the parameters according to the gradients.

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.RmsProp

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(String, NDArray[]) - Method in interface ai.djl.training.ParameterServer

Updates the parameter of a key from Parameter Server.

update(String, NDArray[], NDArray[]) - Method in interface ai.djl.training.ParameterServer

Updates the parameter of a key from Parameter Server.

update(long, String) - Method in class ai.djl.training.util.ProgressBar

updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.evaluator.AbstractAccuracy

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.evaluator.BoundingBoxError

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.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

UrlTranslator<T> - Class in ai.djl.modality.cv.translator.wrapper

Built-in Translator that provides image pre-processing from URL.

UrlTranslator(Translator<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.UrlTranslator

Creates a UrlTranslator instance.

UrlTranslatorFactory<T> - Class in ai.djl.modality.cv.translator.wrapper

A factory class creates Translator that can process image from a URL.

UrlTranslatorFactory(TranslatorFactory<Image, T>) - Constructor for class ai.djl.modality.cv.translator.wrapper.UrlTranslatorFactory

Creates a FileTranslatorFactory instance.

useDefault - Variable in class ai.djl.nn.core.Embedding.BaseBuilder

V

validate() - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

validate() - Method in class ai.djl.modality.cv.translator.BaseImageTranslator.ClassificationBuilder

validate() - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Validates that the required arguments are set.

validate() - Method in class ai.djl.nn.convolutional.Deconvolution.DeconvolutionBuilder

Validates that the required arguments are set.

VALIDATE_EPOCH - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener

validateBatch(Trainer, Batch) - Static method in class ai.djl.training.EasyTrain

Validates the given batch of data.

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.Image.Flag

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.modality.cv.Image.Interpolation

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.modality.cv.Image.Type

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.modality.cv.translator.YoloV5Translator.YoloOutputType

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.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.tracker.WarmUpTracker.Mode

Returns the enum constant of this type with the specified name.

values() - Static method in enum ai.djl.modality.cv.Image.Flag

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.modality.cv.Image.Interpolation

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.modality.cv.Image.Type

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.modality.cv.translator.YoloV5Translator.YoloOutputType

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.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.tracker.WarmUpTracker.Mode

Returns an array containing the constants of this enum type, in the order they are declared.

version - Variable in class ai.djl.nn.AbstractBlock

The model version of this block, used for checking if parameters are still valid during parameter loading.

Version - Class in ai.djl.repository

A Version is used for the versioning of an Artifact.

Version(String) - Constructor for class ai.djl.repository.Version

Constructs a version with the version string.

VersionComparator() - Constructor for class ai.djl.repository.Artifact.VersionComparator

VersionRange - Class in ai.djl.repository

A VersionRange is a set of Restrictions that match some Versions.

Vocabulary - Interface in ai.djl.modality.nlp

Vocabulary is a collection of tokens.

vocabularyContains(String) - Method in class ai.djl.modality.nlp.embedding.TrainableWordEmbedding

Returns whether an embedding exists for a word.

vocabularyContains(String) - Method in interface ai.djl.modality.nlp.embedding.WordEmbedding

Returns whether an embedding exists for a word.

W

waitAll() - Method in interface ai.djl.ndarray.LazyNDArray

Runs all NDArrays and sleeps until their values are fully computed.

waitToRead() - Method in interface ai.djl.ndarray.LazyNDArray

Runs the current NDArray and sleeps until the value is ready to read.

waitToRead(NDArray...) - Method in class ai.djl.nn.transformer.MemoryScope

Waits for all given arrays to be ready to read, i.e.

waitToRead(NDList...) - Method in class ai.djl.nn.transformer.MemoryScope

Waits for all arrays in all given lists to be ready to be read, i.e.

waitToWrite() - Method in interface ai.djl.ndarray.LazyNDArray

Runs the current NDArray and sleeps until the value is ready to write.

warmUp() - Static method in interface ai.djl.training.tracker.Tracker

Returns a new instance of WarmUpTracker.Builder that can build an WarmUpTracker.

WarmUpTracker - Class in ai.djl.training.tracker

A WarmUpTracker applies a simple warm-up before executing a main Tracker.

WarmUpTracker.Builder - Class in ai.djl.training.tracker

The Builder to construct a WarmUpTracker.

WarmUpTracker.Mode - Enum in ai.djl.training.tracker

An enum that enumerates the types of warm-up modes for a WarmUpTracker.

weight - Variable in class ai.djl.nn.convolutional.Convolution

weight - Variable in class ai.djl.nn.convolutional.Deconvolution

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.

width - Variable in class ai.djl.modality.cv.translator.BaseImageTranslator.BaseBuilder

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

An application that takes a word and returns a feature vector that represents the word.

WORD_RECOGNITION - Static variable in interface ai.djl.Application.CV

An application that accepts an image of a single word and returns the String text of the word.

WordEmbedding - Interface in ai.djl.modality.nlp.embedding

A class to manage 1-D NDArray representations of words.

WordpieceTokenizer - Class in ai.djl.modality.nlp.bert

WordpieceTokenizer tokenizes a piece of text into its word pieces.

WordpieceTokenizer(SimpleVocabulary, String, int) - Constructor for class ai.djl.modality.nlp.bert.WordpieceTokenizer

Creates an instance of WordpieceTokenizer.

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.

Y

YoloModelLoader - Class in ai.djl.modality.cv.zoo

A ModelLoader for YOLO models.

YoloModelLoader(Repository, String, String, String, ModelZoo) - Constructor for class ai.djl.modality.cv.zoo.YoloModelLoader

Creates the Model loader from the given repository.

YoloTranslator - Class in ai.djl.modality.cv.translator

A translator for yolo models.

YoloTranslator(YoloTranslator.Builder) - Constructor for class ai.djl.modality.cv.translator.YoloTranslator

Constructs an ImageTranslator with the provided builder.

YoloTranslator.Builder - Class in ai.djl.modality.cv.translator

The builder for YoloTranslator.

YoloV5Translator - Class in ai.djl.modality.cv.translator

A translator for YoloV5 models.

YoloV5Translator(YoloV5Translator.Builder) - Constructor for class ai.djl.modality.cv.translator.YoloV5Translator

Constructs an ImageTranslator with the provided builder.

YoloV5Translator.Builder - Class in ai.djl.modality.cv.translator

The builder for YoloV5Translator.

YoloV5Translator.YoloOutputType - Enum in ai.djl.modality.cv.translator

A enum represents the Yolo output type.

Z

zeros(Shape, DataType) - Method in class ai.djl.ndarray.BaseNDManager

Creates an instance of NDArray with specified Shape filled with zeros.

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.

zerosLike() - Method in interface ai.djl.ndarray.NDArrayAdapter

Returns an NDArray of zeros with the same Shape, DataType and SparseFormat as the input NDArray.

ZooModel<I,O> - Class in ai.djl.repository.zoo

A ZooModel is a Model loaded from a model zoo and includes a default Translator.

ZooModel(Model, Translator<I, O>) - Constructor for class ai.djl.repository.zoo.ZooModel

Constructs a ZooModel given the model and translator.

ZooProvider - Interface in ai.djl.repository.zoo

The ZooProvider is a service provider that enables ServiceLoader to locate and load at the run time.

ZooProviderNotFoundException - Exception in ai.djl.repository.zoo

Runtime exception thrown when a provider of the required type cannot be found.

ZooProviderNotFoundException(String) - Constructor for exception ai.djl.repository.zoo.ZooProviderNotFoundException

Constructs a new exception with the specified detail message.

ZooProviderNotFoundException(String, Throwable) - Constructor for exception ai.djl.repository.zoo.ZooProviderNotFoundException

Constructs a new exception with the specified detail message and cause.

ZooProviderNotFoundException(Throwable) - Constructor for exception ai.djl.repository.zoo.ZooProviderNotFoundException

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).

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