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

A

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

Returns the absolute value of this NDArray element-wise.

AbstractAccuracy - Class in ai.djl.training.evaluator

Accuracy is an Evaluator that computes the accuracy score.

AbstractAccuracy() - Constructor for class ai.djl.training.evaluator.AbstractAccuracy

Creates an accuracy evaluator that computes accuracy across axis 1 along the 0th index.

AbstractAccuracy(String, int) - Constructor for class ai.djl.training.evaluator.AbstractAccuracy

Creates an accuracy evaluator that computes accuracy across axis 1 along given index.

AbstractAccuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.AbstractAccuracy

Creates an accuracy evaluator.

AbstractBlock - Class in ai.djl.nn

AbstractBlock is an abstract implementation of Block.

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

AbstractCompositeLoss - Class in ai.djl.training.loss

AbstractCompositeLoss is a Loss class that can combine other Losses together to make a larger loss.

AbstractCompositeLoss(String) - Constructor for class ai.djl.training.loss.AbstractCompositeLoss

Constructs a composite loss with the given name.

Accuracy - Class in ai.djl.training.evaluator

Accuracy is the AbstractAccuracy with multiple classes.

Accuracy() - Constructor for class ai.djl.training.evaluator.Accuracy

Creates a multiclass accuracy evaluator that computes accuracy across axis 1 along the 0th index.

Accuracy(String, int) - Constructor for class ai.djl.training.evaluator.Accuracy

Creates a multiclass accuracy evaluator that computes accuracy across axis 1 along given index.

Accuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.Accuracy

Creates a multiclass accuracy evaluator.

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

A helper for classes extending AbstractAccuracy.

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

A helper for classes extending AbstractAccuracy.

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

A helper for classes extending AbstractAccuracy.

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

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

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

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

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

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

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

Activation - Class in ai.djl.nn

Utility class that provides activation functions and blocks.

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

Adam - Class in ai.djl.training.optimizer

Adam is a generalization of the AdaGrad Optimizer.

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

Creates a new instance of Adam optimizer.

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

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

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

The Builder to construct an Adam object.

add(List<String>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Adds the given sentence to the Vocabulary.

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

Adds a number to this NDArray element-wise.

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

Adds other NDArrays to this NDArray element-wise.

add(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Adds a number to the NDArray element-wise.

add(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Adds a NDArray to a number element-wise.

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

Adds a NDArray to a NDArray element-wise.

add(Block) - Method in class ai.djl.nn.ParallelBlock

Adds the given Block to the block, which is one parallel branch.

add(Function<NDList, NDList>) - Method in class ai.djl.nn.ParallelBlock

Adds a LambdaBlock, that applies the given function, to the list of parallel branches.

add(Block) - Method in class ai.djl.nn.SequentialBlock

Adds the given Block to the block to be executed in order.

add(Function<NDList, NDList>) - Method in class ai.djl.nn.SequentialBlock

Adds a LambdaBlock that applies the given function to the sequence of blocks.

add(Hyperparameter<?>) - Method in class ai.djl.training.hyperparameter.param.HpSet

Adds a hyperparameter to the set.

add(Transform) - Method in class ai.djl.translate.Pipeline

Adds the given Transform to the list of transforms to be applied on the input when the transform method is called on this object.

add(int, Transform) - Method in class ai.djl.translate.Pipeline

Adds the given Transform to the list of transforms to be applied on the NDArray at the given index in the input NDList.

add(String, Transform) - Method in class ai.djl.translate.Pipeline

Adds the given Transform to the list of transforms to be applied on the NDArray with the given key as name in the input NDList.

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

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

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

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

addAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss

addAccumulator(String) - Method in class ai.djl.training.loss.Loss

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

addAll(List<List<String>>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Adds the given list of sentences to the Vocabulary.

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

Appends all of the NDArrays in the specified NDList to the end of this NDList, in the order that they are returned by the specified NDList's iterator.

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

Joins a this shape with specified other shape.

addAll(Block...) - Method in class ai.djl.nn.ParallelBlock

Adds an array of blocks, each of which is a parallel branch.

addAll(Collection<Block>) - Method in class ai.djl.nn.ParallelBlock

Adds a Collection of blocks, each of which is a parallel branch.

addAll(Block...) - Method in class ai.djl.nn.SequentialBlock

Adds an array of blocks to be executed in sequence, in order.

addAll(Collection<Block>) - Method in class ai.djl.nn.SequentialBlock

Adds a Collection of blocks to be executed in sequence, in order.

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

Updates the NDIndex by appending a boolean NDArray.

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

Adds an Evaluator that needs to be computed during training.

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

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

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

Adds other NDArrays to this NDArray element-wise in place.

addi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

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

addi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

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

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

Adds all of the NDArrays together element-wise in place.

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

Updates the NDIndex by appending indices to the array.

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

Updates the NDIndex by appending indices as specified values on the NDArray.

addLoss(Loss) - Method in class ai.djl.training.loss.SimpleCompositeLoss

Adds a Loss that applies to all labels and predictions to this composite loss.

addLoss(Loss, int) - Method in class ai.djl.training.loss.SimpleCompositeLoss

Adds a Loss that applies to a single index of the label and predictions to this composite loss.

addMetric(Metric) - Method in class ai.djl.metric.Metrics

Adds a Metric to the collection.

addMetric(String, Number) - Method in class ai.djl.metric.Metrics

Adds a Metric given the metric's name and value.

addMetric(String, Number, String) - Method in class ai.djl.metric.Metrics

Adds a Metric given the metric's name, value, and unit.

addSliceDim(long, long) - Method in class ai.djl.ndarray.index.NDIndex

Appends a new index to slice the dimension and returns a range of values.

addSliceDim(long, long, long) - Method in class ai.djl.ndarray.index.NDIndex

Appends a new index to slice the dimension and returns a range of values.

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

Adds TrainingListeners for training.

ai.djl - package ai.djl

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

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

Contains classes responsible for loading a deep learning framework.

ai.djl.inference - package ai.djl.inference

Contains classes to implement inference tasks.

ai.djl.metric - package ai.djl.metric

Contains classes to collect metrics information.

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

Contains utility classes for each of the predefined modalities.

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

Contains utility classes for computer vision tasks and image processing.

ai.djl.modality.cv.output - package ai.djl.modality.cv.output

Contains output types used in various computer vision applications.

ai.djl.modality.cv.transform - package ai.djl.modality.cv.transform

Contains Transforms for working with Images.

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

Contains translators used for computer vision applications.

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

Contains utility classes for image manipulation.

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

Contains utility classes for natural language processing tasks.

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

Contains utility classes for natural language pre-processing tasks.

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

Contains utility classes for question and answer processing.

ai.djl.ndarray - package ai.djl.ndarray

Contains classes and interfaces that define an n-dimensional array.

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

Contains classes that help access NDArray's indices.

ai.djl.ndarray.types - package ai.djl.ndarray.types

Contains classes that define n-dimensional array data types.

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

Contains classes to construct neural networks.

ai.djl.nn.convolutional - package ai.djl.nn.convolutional

Contains classes that define convolutional operations extending Convolution.

ai.djl.nn.core - package ai.djl.nn.core

Contains classes that define simple neural network operations.

ai.djl.nn.norm - package ai.djl.nn.norm

Contains classes that define normalizing neural network operations.

ai.djl.nn.pooling - package ai.djl.nn.pooling

Contains pooling neural network operations in Pool and helpers for it.

ai.djl.nn.recurrent - package ai.djl.nn.recurrent

Contains classes for recurrent neural network operations.

ai.djl.training - package ai.djl.training

Contains classes and implementations to train a neural network.

ai.djl.training.dataset - package ai.djl.training.dataset

Contains classes to download and prepare training and testing data.

ai.djl.training.evaluator - package ai.djl.training.evaluator

Contains classes for evaluating the effectiveness of models.

ai.djl.training.hyperparameter.optimizer - package ai.djl.training.hyperparameter.optimizer

Contains classes to optimize Hyperparameters.

ai.djl.training.hyperparameter.param - package ai.djl.training.hyperparameter.param

Contains different types of Hyperparameters.

ai.djl.training.initializer - package ai.djl.training.initializer

Contains classes for initializing a Block's Parameters.

ai.djl.training.listener - package ai.djl.training.listener

Contains classes providing functionality during training through TrainingListener.

ai.djl.training.loss - package ai.djl.training.loss

Contains classes for measuring the Loss of a model.

ai.djl.training.optimizer - package ai.djl.training.optimizer

Contains classes for optimizing a neural network Block.

ai.djl.training.optimizer.learningrate - package ai.djl.training.optimizer.learningrate

Contains classes for having a gradually changing learning rate.

ai.djl.training.util - package ai.djl.training.util

Contains utilities to use during training.

ai.djl.translate - package ai.djl.translate

Contains classes and interfaces that translate between java objects and NDArrays.

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

Returns true if all elements within this NDArray are non-zero or true.

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

Returns true if two NDArrays are element-wise equal within a tolerance.

allClose(NDArray, double, double, boolean) - Method in interface ai.djl.ndarray.NDArray

Returns true if two NDArray are element-wise equal within a tolerance.

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

Returns true if two NDArray are element-wise equal within a tolerance.

allClose(NDArray, NDArray, double, double, boolean) - Static method in class ai.djl.ndarray.NDArrays

Returns true if two NDArray are element-wise equal within a tolerance.

allocateDirect(int) - Method in interface ai.djl.ndarray.NDManager

Allocates a new engine specific direct byte buffer.

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

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

Application - Class in ai.djl

A class contains common deep learning applications.

Application.CV - Interface in ai.djl

The common set of applications for computer vision.

Application.NLP - Interface in ai.djl

The common set of applications for natural language processing.

arange(int) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced values starting from 0.

arange(float) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced values starting from 0.

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

Returns evenly spaced values within a given interval with step 1.

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

Returns evenly spaced values within a given interval with step 1.

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

Returns evenly spaced values within a given interval.

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

Returns evenly spaced values within a given interval.

arange(int, int, int, DataType, Device) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced values within a given interval.

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

Returns evenly spaced values within a given interval.

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

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

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

Returns the indices of the maximum values along given axis.

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

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

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

Returns the indices of the minimum values along given axis.

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

Returns the indices that would sort this NDArray.

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

Returns the indices that would sort this NDArray given the axis.

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

Returns the indices that would sort this NDArray given the axis.

ArrayDataset - Class in ai.djl.training.dataset

ArrayDataset is an implementation of RandomAccessDataset that consist entirely of large NDArrays.

ArrayDataset(RandomAccessDataset.BaseBuilder<?>) - Constructor for class ai.djl.training.dataset.ArrayDataset

Creates a new instance of ArrayDataset with the arguments in ArrayDataset.Builder.

ArrayDataset.Builder - Class in ai.djl.training.dataset

The Builder to construct an ArrayDataset.

artifacts - Variable in class ai.djl.BaseModel

asDataType(ByteBuffer) - Method in enum ai.djl.ndarray.types.DataType

Converts a ByteBuffer to a buffer for this data type.

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

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

asInDevice(Device, boolean) - Method in class ai.djl.ndarray.NDList

Converts all the NDArray in NDList to a different Device.

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

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

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

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

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

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

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

Attaches a NDArray or NDManager to this NDManager.

attach(NDManager) - Method in interface ai.djl.ndarray.NDArray

Attaches this NDArray to the specified NDManager.

attach(NDManager) - Method in class ai.djl.ndarray.NDList

Attaches each ndarray in this list to the specified manager.

attach(String, AutoCloseable) - Method in interface ai.djl.ndarray.NDManager

Attaches a NDArray or NDManager to this NDManager.

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

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

availableSize() - Method in class ai.djl.training.dataset.ArrayDataset

Returns the number of records available to be read in this Dataset.

availableSize() - Method in class ai.djl.training.dataset.RandomAccessDataset

Returns the number of records available to be read in this Dataset.

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

Performs Avg Pooling on the input.

avgPool1DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.

avgPool1DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.

avgPool1DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.

avgPool1DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.

avgPool1DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool1DBlock pooling function in its forward function.

avgPool2DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.

avgPool2DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.

avgPool2DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.

avgPool2DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.

avgPool2DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool2DBlock pooling function in its forward function.

avgPool3DBlock(Shape, Shape, Shape, PoolingConvention, boolean) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.

avgPool3DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.

avgPool3DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.

avgPool3DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.

avgPool3DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the avgPool3DBlock pooling function in its forward function.

axis - Variable in class ai.djl.training.evaluator.AbstractAccuracy

B

backward(NDArray) - Method in interface ai.djl.training.GradientCollector

Calculate the gradient w.r.t previously marked variable (head).

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

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

BaseBuilder() - Constructor for class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

BaseHpOptimizer - Class in ai.djl.training.hyperparameter.optimizer

A base containing shared implementations for HpOptimizers.

BaseHpOptimizer(HpSet) - Constructor for class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

Constructs a BaseHpOptimizer.

BaseModel - Class in ai.djl

BaseModel is the basic implementation of Model.

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

BaseNDManager - Class in ai.djl.ndarray

BaseNDManager is the default implementation of NDManager.

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

BasePredictor<I,O> - Class in ai.djl.inference

BasePredictor contains common code for implementations of the Predictor interface.

BasePredictor(Model, Translator<I, O>, boolean) - Constructor for class ai.djl.inference.BasePredictor

Creates a new instance of BasePredictor with the given Model and Translator.

basic() - Static method in interface ai.djl.training.listener.TrainingListener.Defaults

A basic TrainingListener set with minimal recommended functionality.

Batch - Class in ai.djl.training.dataset

A Batch is used to hold multiple items (data and label pairs) from a Dataset.

Batch(NDManager, NDList, NDList, int) - Constructor for class ai.djl.training.dataset.Batch

Creates a new instance of Batch with the given manager, data and labels.

Batch(NDManager, NDList, NDList, int, Batchifier) - Constructor for class ai.djl.training.dataset.Batch

Creates a new instance of Batch with the given manager, data and labels.

Batch(NDManager, NDList, NDList, int, Batchifier, long, long) - Constructor for class ai.djl.training.dataset.Batch

Creates a new instance of Batch with the given manager, data and labels.

BatchData(Batch, Map<Device, NDList>, Map<Device, NDList>) - Constructor for class ai.djl.training.listener.TrainingListener.BatchData

Constructs a new TrainingListener.BatchData.

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

Creates a Block whose forward function applies the batchFlatten method.

batchFlattenBlock(long) - Static method in class ai.djl.nn.Blocks

Creates a Block whose forward function applies the batchFlatten method.

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

Creates a Block whose forward function applies the batchFlatten method.

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

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

Batchifier - Interface in ai.djl.translate

An interface that provides methods to convert an un-batched NDList into a batched NDList and vice versa.

batchify(NDList[]) - Method in interface ai.djl.translate.Batchifier

Converts an array of NDList into an NDList.

batchify(NDList[]) - Method in class ai.djl.translate.StackBatchifier

Converts an array of NDList into an NDList.

BatchNorm - Class in ai.djl.nn.norm

Normalizes a data batch by mean and variance, and applies a scale gamma as well as offset beta.

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

The Builder to construct a BatchNorm.

batchPredict(List<I>) - Method in class ai.djl.inference.BasePredictor

Predicts a batch for inference.

batchPredict(List<I>) - Method in interface ai.djl.inference.Predictor

Predicts a batch for inference.

BatchSampler - Class in ai.djl.training.dataset

BatchSampler is a Sampler that returns a single epoch over the data.

BatchSampler(Sampler.SubSampler, int) - Constructor for class ai.djl.training.dataset.BatchSampler

Creates a new instance of BatchSampler that samples from the given Sampler.SubSampler, and yields a mini-batch of indices.

BatchSampler(Sampler.SubSampler, int, boolean) - Constructor for class ai.djl.training.dataset.BatchSampler

Creates a new instance of BatchSampler that samples from the given Sampler.SubSampler, and yields a mini-batch of indices.

beforeInitialize(Shape[]) - Method in class ai.djl.nn.AbstractBlock

Performs any action necessary before initialization.

beforeInitialize(Shape[]) - Method in class ai.djl.nn.convolutional.Convolution

Performs any action necessary before initialization.

beforeInitialize(Shape[]) - Method in class ai.djl.nn.core.Linear

Performs any action necessary before initialization.

beforeInitialize(Shape[]) - Method in class ai.djl.nn.norm.BatchNorm

Performs any action necessary before initialization.

beforeInitialize(Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell

Performs any action necessary before initialization.

best() - Method in class ai.djl.modality.Classifications

Returns the most likely class for the classification.

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

BinaryAccuracy - Class in ai.djl.training.evaluator

BinaryAccuracy is the AbstractAccuracy with two classes.

BinaryAccuracy(String, float, int, int) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy

Creates a binary (two class) accuracy evaluator.

BinaryAccuracy(String, float, int) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy

Creates a binary (two class) accuracy evaluator that computes accuracy across axis 1 along given index.

BinaryAccuracy(float) - Constructor for class ai.djl.training.evaluator.BinaryAccuracy

Creates a binary (two class) accuracy evaluator that computes accuracy across axis 1 along the 0th index.

BinaryAccuracy() - Constructor for class ai.djl.training.evaluator.BinaryAccuracy

Creates a binary (two class) accuracy evaluator with 0 threshold.

block - Variable in class ai.djl.BaseModel

Block - Interface in ai.djl.nn

A Block is a composable function that forms a neural network.

BlockList - Class in ai.djl.nn

Represents a set of names and Blocks.

BlockList() - Constructor for class ai.djl.nn.BlockList

Creates an empty BlockList.

BlockList(int) - Constructor for class ai.djl.nn.BlockList

Constructs an empty BlockList with the specified initial capacity.

BlockList(List<String>, List<Block>) - Constructor for class ai.djl.nn.BlockList

Constructs a BlockList containing the elements of the specified keys and values.

BlockList(List<Pair<String, Block>>) - Constructor for class ai.djl.nn.BlockList

Constructs a BlockList containing the elements of the specified list of Pairs.

BlockList(Map<String, Block>) - Constructor for class ai.djl.nn.BlockList

Constructs a BlockList containing the elements of the specified map.

Blocks - Class in ai.djl.nn

Utility class that provides some useful blocks.

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

Returns portion of this NDArray given the index boolean NDArray along first axis.

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

Returns portion of this NDArray given the index boolean NDArray along given axis.

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

Returns portion of the NDArray given the index boolean NDArray along first axis.

booleanMask(NDArray, NDArray, int) - Static method in class ai.djl.ndarray.NDArrays

Returns portion of the NDArray given the index boolean NDArray along given axis.

BoundingBox - Interface in ai.djl.modality.cv.output

An interface representing a bounding box for the detected object.

BoundingBoxError - Class in ai.djl.training.evaluator

BoundingBoxError is an Evaluator that computes the error in the prediction of bounding boxes in SingleShotDetection model.

BoundingBoxError(String) - Constructor for class ai.djl.training.evaluator.BoundingBoxError

Creates an BoundingBoxError evaluator.

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

Broadcasts this NDArray to be the given shape.

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

Broadcasts this NDArray to be the given shape.

BufferedImageUtils - Class in ai.djl.modality.cv.util

BufferedImageUtils is an image processing utility to load, reshape and convert images using BufferedImage.

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

Builds a MultiBoxDetection block.

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

Builds a MultiBoxPrior block.

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

Builds a MultiBoxTarget block.

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

Builds the ImageClassificationTranslator with the provided data.

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

Builds the translator.

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

Builds the translator.

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

Builds the translator.

build() - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Builds the Vocabulary object with the set arguments.

build() - Method in class ai.djl.nn.convolutional.Conv1D.Builder

Builds a Conv1D block.

build() - Method in class ai.djl.nn.convolutional.Conv2D.Builder

Builds a Conv2D block.

build() - Method in class ai.djl.nn.convolutional.Conv3D.Builder

Builds a Conv3D block.

build() - Method in class ai.djl.nn.core.Embedding.Builder

Builds the Embedding.

build() - Method in class ai.djl.nn.core.Linear.Builder

Returns the constructed Linear.

build() - Method in class ai.djl.nn.norm.BatchNorm.Builder

Builds a BatchNorm block.

build() - Method in class ai.djl.nn.norm.Dropout.Builder

Builds a Dropout block.

build() - Method in class ai.djl.nn.recurrent.GRU.Builder

Builds a GRU block.

build() - Method in class ai.djl.nn.recurrent.LSTM.Builder

Builds a LSTM block.

build() - Method in class ai.djl.nn.recurrent.RNN.Builder

Builds a RNN block.

build() - Method in class ai.djl.training.dataset.ArrayDataset.Builder

Builds a new instance of ArrayDataset with the specified data and labels.

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

Builds a Adam block.

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

Builds a FactorTracker block.

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

Builds a MultiFactorTracker block.

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

Builds a Nag block.

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

Builds a Sgd block.

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

Creates a builder to build a MultiBoxDetection.

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

Creates a builder to build a MultiBoxPrior.

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

Creates a builder to build a MultiBoxTarget.

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

Creates a builder to build a ImageClassificationTranslator.

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

Creates a builder to build a InstanceSegmentationTranslator.

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

Creates a builder to build a SimplePoseTranslator.

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

Creates a builder to build a SingleShotDetectionTranslator.

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

builder() - Static method in class ai.djl.nn.convolutional.Conv1D

Creates a builder to build a Conv1D.

builder() - Static method in class ai.djl.nn.convolutional.Conv2D

Creates a builder to build a Conv2D.

builder() - Static method in class ai.djl.nn.convolutional.Conv3D

Creates a builder to build a Conv3D.

builder() - Static method in class ai.djl.nn.core.Embedding

Creates a builder to build an Embedding.

builder() - Static method in class ai.djl.nn.core.Linear

Creates a builder to build a Linear.

builder() - Static method in class ai.djl.nn.norm.BatchNorm

Creates a builder to build a BatchNorm.

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

Creates a builder to build a Dropout.

builder() - Static method in class ai.djl.nn.recurrent.GRU

Creates a builder to build a GRU.

Builder() - Constructor for class ai.djl.nn.recurrent.GRU.Builder

builder() - Static method in class ai.djl.nn.recurrent.LSTM

Creates a builder to build a LSTM.

Builder() - Constructor for class ai.djl.nn.recurrent.LSTM.Builder

builder() - Static method in class ai.djl.nn.recurrent.RNN

Creates a builder to build a RNN.

Builder() - Constructor for class ai.djl.nn.recurrent.RNN.Builder

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

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

Creates a builder to build a Adam.

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

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

Combines a list of tokens to form a sentence.

C

cast(DataType) - Method in interface ai.djl.Model

Casts the model to support a different precision level.

cast(DataType) - Method in class ai.djl.nn.AbstractBlock

Guaranteed to throw an exception.

cast(DataType) - Method in interface ai.djl.nn.Block

Guaranteed to throw an exception.

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

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

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

Returns the ceiling of this NDArray element-wise.

CenterCrop - Class in ai.djl.modality.cv.transform

A Transform that crops the center of an image.

CenterCrop() - Constructor for class ai.djl.modality.cv.transform.CenterCrop

Creates a CenterCrop Transform that crops to size min(width, height).

CenterCrop(int, int) - Constructor for class ai.djl.modality.cv.transform.CenterCrop

Creates a CenterCrop Transform that crops the given width and height.

centerCrop(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Crops an image to a square of size min(width, height).

centerCrop(NDArray, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Crops an image to a given width and height from the center of the image.

checkConcatInput(NDList) - Static method in class ai.djl.ndarray.NDUtils

Check two criteria of concat input: 1.

checkLabelShapes(NDArray, NDArray, boolean) - Method in class ai.djl.training.evaluator.Evaluator

Checks if the two input NDArray have the same length or shape.

checkLabelShapes(NDArray, NDArray) - Method in class ai.djl.training.evaluator.Evaluator

Checks the length of NDArrays.

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

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

Constructs a single class result for a classification.

Classifications - Class in ai.djl.modality

Classifications is the container that stores the classification results for classification on a single input.

Classifications(List<String>, List<Double>) - Constructor for class ai.djl.modality.Classifications

Constructs a Classifications using a parallel list of classNames and probabilities.

Classifications(List<String>, NDArray) - Constructor for class ai.djl.modality.Classifications

Constructs a Classifications using list of classNames parallel to an NDArray of probabilities.

Classifications.Classification - Class in ai.djl.modality

A Classification stores the classification result for a single class on a single input.

classNames - Variable in class ai.djl.modality.Classifications

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

Closes all the parameters of the block.

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

Closes all the parameters of the block.

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

Clips (limit) the values in this NDArray.

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

clipLstmState - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

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

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

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

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

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

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

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

close() - Method in class ai.djl.nn.Parameter

close() - Method in class ai.djl.training.dataset.Batch

close() - Method in interface ai.djl.training.GradientCollector

close() - Method in class ai.djl.training.LocalParameterServer

close() - Method in interface ai.djl.training.ParameterServer

close() - Method in interface ai.djl.training.Trainer

close() - Method in interface ai.djl.translate.TranslatorContext

closed - Variable in class ai.djl.ndarray.BaseNDManager

collectMemoryInfo(Metrics) - Static method in class ai.djl.training.listener.MemoryTrainingListener

Collect memory information.

components - Variable in class ai.djl.training.loss.AbstractCompositeLoss

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

Joins a NDArray along the first axis.

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

Joins a NDArray along an existing axis.

concat(NDList) - Static method in class ai.djl.ndarray.NDArrays

Joins a NDList along the first axis.

concat(NDList, int) - Static method in class ai.djl.ndarray.NDArrays

Joins a NDList along an existing axis.

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

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

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

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

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

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

contentEquals(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns true if all elements in NDArray a are equal to NDArray b.

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

Returns true if all elements in NDArray a are equal to NDArray b.

Conv1D - Class in ai.djl.nn.convolutional

Computes 1-D convolution on 3-D input.

Conv1D.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv1D type of Block.

Conv2D - Class in ai.djl.nn.convolutional

Computes 2-D convolution on 4-D input.

Conv2D.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv2D type of Block.

Conv3D - Class in ai.djl.nn.convolutional

Computes 3-D convolution on 5-D input.

Conv3D.Builder - Class in ai.djl.nn.convolutional

The Builder to construct a Conv3D type of Block.

Convolution - Class in ai.djl.nn.convolutional

Computes N-D convolution on (N+2)-D input.

Convolution(Convolution.ConvolutionBuilder<?>) - Constructor for class ai.djl.nn.convolutional.Convolution

Creates a Convolution object.

Convolution.ConvolutionBuilder<T extends Convolution.ConvolutionBuilder> - Class in ai.djl.nn.convolutional

A builder that can build any Convolution block.

ConvolutionBuilder() - Constructor for class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

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

Deep-copies the current NDArray to the one passed in.

correctInstances - Variable in class ai.djl.training.evaluator.AbstractAccuracy

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

Returns the trigonometric cosine of this NDArray element-wise.

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

Returns the hyperbolic cosine of this NDArray element-wise.

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

Counts the number of non-zero values in this NDArray.

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

Counts the number of non-zero values in this NDArray along a given axis.

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

Returns the default CPU Device.

CPU - Static variable in interface ai.djl.Device.Type

create(Shape) - Method in interface ai.djl.ndarray.NDManager

Creates an uninitialized instance of DataType.FLOAT32 NDArray with specified Shape.

create(Number) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(float) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(int) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(double) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(long) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(byte) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(boolean) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a scalar NDArray.

create(float[]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 1D NDArray.

create(int[]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 1D NDArray.

create(double[]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 1D NDArray.

create(long[]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 1D NDArray.

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

Creates and initializes a 1D NDArray.

create(boolean[]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 1D NDArray.

create(float[][]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 2D NDArray.

create(int[][]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 2D NDArray.

create(double[][]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 2D NDArray.

create(long[][]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 2-D NDArray.

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

Creates and initializes a 2-D NDArray.

create(boolean[][]) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a 2-D NDArray.

create(Buffer, Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes a NDArray with specified Shape.

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

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

create(Buffer, Shape, DataType) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and DataType.

create(float[], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and float array.

create(int[], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and int array.

create(double[], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and double array.

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

Creates and initializes an instance of NDArray with specified Shape and long array.

create(byte[], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and byte array.

create(boolean[], Shape) - Method in interface ai.djl.ndarray.NDManager

Creates and initializes an instance of NDArray with specified Shape and boolean array.

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

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

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

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

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

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

createMask(NDIndex) - Method in interface ai.djl.ndarray.NDArray

Returns a mask on whether each element matches the given index.

createMask(Predicate<Number>) - Method in interface ai.djl.ndarray.NDArray

Returns a mask on whether each element matches the given condition.

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

Stores the matrix in row sparse format.

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

Stores the matrix in row sparse format.

Crop - Class in ai.djl.modality.cv.transform

A Transform that crops the image to a given location and size.

Crop(int, int, int, int) - Constructor for class ai.djl.modality.cv.transform.Crop

Creates a CenterCrop Transform.

crop(NDArray, int, int, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Crops an image with a given location and size.

CUDA - Static variable in class ai.djl.engine.StandardCapabilities

CUDNN - Static variable in class ai.djl.engine.StandardCapabilities

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

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

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

Return the cumulative sum of the elements along a given axis.

currentPoint() - Method in interface ai.djl.modality.cv.output.PathIterator

Returns the coordinates and type of the current path segment in the iteration.

D

data - Variable in class ai.djl.training.dataset.ArrayDataset

DataDesc - Class in ai.djl.ndarray.types

A data descriptor class that encapsulates information of a NDArray.

DataDesc(Shape) - Constructor for class ai.djl.ndarray.types.DataDesc

Constructs and initializes a DataDesc with specified Shape.

DataDesc(Shape, String) - Constructor for class ai.djl.ndarray.types.DataDesc

Constructs and initializes a DataDesc with specified Shape and name.

DataDesc(Shape, DataType) - Constructor for class ai.djl.ndarray.types.DataDesc

Constructs and initializes a DataDesc with specified Shape and DataType.

DataDesc(Shape, DataType, String) - Constructor for class ai.djl.ndarray.types.DataDesc

Constructs and initializes a DataDesc with specified Shape, DataType, name, Device and SparseFormat.

DataIterable - Class in ai.djl.training.dataset

DataIterable is a data loader that combines Dataset, Batchifier, Pipeline, and Sampler to provide an iterable over the given RandomAccessDataset.

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

Creates a new instance of DataIterable with the given parameters.

Dataset - Interface in ai.djl.training.dataset

An interface to represent a set of sample data/label pairs to train a model.

Dataset.Usage - Enum in ai.djl.training.dataset

An enum that indicates the mode - training, test or validation.

dataType - Variable in class ai.djl.BaseModel

DataType - Enum in ai.djl.ndarray.types

An enum representing the underlying NDArray's data type.

DataType.Format - Enum in ai.djl.ndarray.types

The general data type format categories.

debugDump(int) - Method in class ai.djl.ndarray.BaseNDManager

Prints information about this NDManager and all sub-managers to the console.

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

Logs debug information about the environment for use when debugging environment issues.

decode(DataInputStream) - Static method in class ai.djl.ndarray.types.Shape

Decodes the data in the given DataInputStream and converts it into the corresponding Shape object.

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

Returns the default context used in Engine.

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

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

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

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

DefaultTrainingConfig - Class in ai.djl.training

DefaultTrainingConfig is an implementation of the TrainingConfig interface.

DefaultTrainingConfig(Loss) - Constructor for class ai.djl.training.DefaultTrainingConfig

Creates an instance of DefaultTrainingConfig with the given Loss.

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

Returns the input descriptor of the model.

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

Returns the input descriptor of the model.

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

Returns a PairList of input names, and shapes.

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

Returns a PairList of input names, and shapes.

describeInput() - Method in class ai.djl.nn.core.Linear

Returns a PairList of input names, and shapes.

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

Returns the output descriptor of the model.

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

Returns the output descriptor of the model.

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

Detaches a NDArray from this NDManager's lifecycle.

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

Detaches the NDArray from current NDManager's lifecycle.

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

Detaches each ndarray in this list from their current managers.

detach(String) - Method in interface ai.djl.ndarray.NDManager

Detaches a NDArray from this NDManager's lifecycle.

DetectedObject(String, double, BoundingBox) - Constructor for class ai.djl.modality.cv.output.DetectedObjects.DetectedObject

Constructs a bounding box with the given data.

DetectedObjects - Class in ai.djl.modality.cv.output

A class representing the detected objects results for a single image in an object detection case.

DetectedObjects(List<String>, List<Double>, List<BoundingBox>) - Constructor for class ai.djl.modality.cv.output.DetectedObjects

Constructs a DetectedObjects, usually during post-processing.

DetectedObjects.DetectedObject - Class in ai.djl.modality.cv.output

A DetectedObject represents a single potential detected Object for an image.

detection(NDList) - Method in class ai.djl.modality.cv.MultiBoxDetection

Converts multi-box detection predictions.

Device - Class in ai.djl

The Device class provides the specified assignment for CPU/GPU processing on the NDArray.

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

Creates a Device with basic information.

Device(String, int) - Constructor for class ai.djl.Device

Creates a Device with basic information.

device - Variable in class ai.djl.ndarray.BaseNDManager

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

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

Device.Type - Interface in ai.djl

Contains device type string constants.

DeviceType - Interface in ai.djl

DeviceType is a class used to map the Device name to their corresponding type number.

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

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

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

Returns the number of dimensions of this Shape.

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

Divides this NDArray by a number element-wise.

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

Divides this NDArray by the other NDArray element-wise.

div(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Divides the NDArray by a number element-wise.

div(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Divides a number by a NDArray element-wise.

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

Divides a NDArray by a NDArray element-wise.

DivergenceCheckTrainingListener - Class in ai.djl.training.listener

TrainingListener that gives early warning if your training has failed by divergence.

DivergenceCheckTrainingListener() - Constructor for class ai.djl.training.listener.DivergenceCheckTrainingListener

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

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

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

Divides this NDArray by the other NDArray element-wise in place.

divi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

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

divi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Divides a number by a NDArray element-wise.

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

Divides a NDArray by a NDArray element-wise.

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

Dot product of this NDArray and the other NDArray.

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

Dot product of NDArray a and NDArray b.

download(String, String) - Static method in class ai.djl.training.util.DownloadUtils

Downloads a file from specified url.

download(String, String, Progress) - Static method in class ai.djl.training.util.DownloadUtils

Downloads a file from specified url.

download(URL, Path, Progress) - Static method in class ai.djl.training.util.DownloadUtils

Downloads a file from specified url.

DownloadUtils - Class in ai.djl.training.util

A utility class downloads the file from specified url.

draw(Graphics2D, int, int) - Method in interface ai.djl.modality.cv.output.BoundingBox

Draws the bounding box using the Graphics2D.

draw(Graphics2D, int, int) - Method in class ai.djl.modality.cv.output.Mask

Draws the bounding box using the Graphics2D.

draw(Graphics2D, int, int) - Method in class ai.djl.modality.cv.output.Rectangle

Draws the bounding box using the Graphics2D.

drawBoundingBoxes(BufferedImage, DetectedObjects) - Static method in class ai.djl.modality.cv.ImageVisualization

Draws the bounding boxes on an image.

drawJoints(BufferedImage, Joints) - Static method in class ai.djl.modality.cv.ImageVisualization

Draws all joints of a body on an image.

Dropout - Class in ai.djl.nn.norm

Applies dropout operation to an input array.

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

The Builder to construct a Dropout type of Block.

dropRate - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

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

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

Dump memory metrics into log directory.

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

Returns a copy of this NDArray.

E

elu(NDArray, float) - Static method in class ai.djl.nn.Activation

Applies ELU activation on the input NDArray.

elu(NDList, float) - Static method in class ai.djl.nn.Activation

Applies ELU(Exponential Linear Unit) activation on the input singleton NDList.

eluBlock(float) - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the ELU activation function in its forward function.

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

Embeds an array of items.

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

Embeds an item.

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

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

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

Constructs a pretrained embedding.

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

The Builder to construct a Embedding type of Block.

EmbeddingException - Exception in ai.djl.modality.nlp

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

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

Constructs a new exception with the specified detail message.

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

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

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

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.

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

Embeds a word.

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

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

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

endEpoch() - Method in interface ai.djl.training.Trainer

Runs the end epoch actions.

Engine - Class in ai.djl.engine

The Engine interface is the base of the provided implementation for DJL.

Engine() - Constructor for class ai.djl.engine.Engine

EngineException - Exception in ai.djl.engine

Thrown to indicate that a native error is raised from the underlying Engine.

EngineException(String) - Constructor for exception ai.djl.engine.EngineException

Constructs a new exception with the specified detail message.

EngineException(String, Throwable) - Constructor for exception ai.djl.engine.EngineException

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

EngineException(Throwable) - Constructor for exception ai.djl.engine.EngineException

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) (which typically contains the class and detail message of cause).

EngineProvider - Interface in ai.djl.engine

The EngineProvider instance manufactures an Engine instance, which is available in the system.

EpochTrainingListener - Class in ai.djl.training.listener

EpochTrainingListener that tracks epochs.

EpochTrainingListener() - Constructor for class ai.djl.training.listener.EpochTrainingListener

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

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

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

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

eq(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

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

eq(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

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

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

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

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

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

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

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

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

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.HingeLoss

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.L1Loss

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.L2Loss

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss

Calculates the evaluation between the labels and the predictions.

evaluate(NDList, NDList) - Method in class ai.djl.training.loss.SoftmaxCrossEntropyLoss

Calculates the evaluation between the labels and the predictions.

Evaluator - Class in ai.djl.training.evaluator

Base class for all Evaluators that can be used to evaluate the performance of a model.

Evaluator(String) - Constructor for class ai.djl.training.evaluator.Evaluator

Creates an evaluator with abstract update methods.

EvaluatorTrainingListener - Class in ai.djl.training.listener

TrainingListener that records evaluator results.

EvaluatorTrainingListener() - Constructor for class ai.djl.training.listener.EvaluatorTrainingListener

Constructs an EvaluatorTrainingListener that updates the training progress the default frequency.

EvaluatorTrainingListener(int) - Constructor for class ai.djl.training.listener.EvaluatorTrainingListener

Constructs an EvaluatorTrainingListener that updates the training progress the given frequency.

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

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

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

Returns the exponential value of this NDArray element-wise.

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

Expands the Shape of a NDArray.

eye(int) - Method in interface ai.djl.ndarray.NDManager

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

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

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

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

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

eye(int, int, int, DataType, Device) - Method in interface ai.djl.ndarray.NDManager

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

F

FactorTracker - Class in ai.djl.training.optimizer.learningrate

FactorTracker is an implementation of LearningRateTracker which is updated by a multiplicative factor, at a constant interval of steps, until it reaches a specified stop value.

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

Creates a new instance of FactorTracker.

factorTracker() - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker

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

FactorTracker.Builder - Class in ai.djl.training.optimizer.learningrate

The Builder to construct an FactorTracker object.

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

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

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

Creates a FileTranslator instance.

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

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

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

Creates a FileTranslatorFactory instance.

filterByLayoutType(Predicate<LayoutType>) - Method in class ai.djl.ndarray.types.Shape

Returns only the axes of the Shape whose layout types match the predicate.

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

fixedLearningRate(float) - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker

Returns a new instance of MultiFactorTracker.

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

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

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

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

Returns the floor of this NDArray element-wise.

forward(TranslatorContext, NDList) - Method in class ai.djl.inference.BasePredictor

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

Applies the operating function of the block once.

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

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

fromBuffer(Buffer) - Static method in enum ai.djl.ndarray.types.DataType

Returns the data type to use for a data buffer.

fromDeviceType(int) - Static method in interface ai.djl.DeviceType

Map device to its type number.

fromFile(Path) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Loads the image from the specified file.

fromUrl(URL) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Loads the image from the specified URL.

fromUrl(String) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Loads the image from the specified URL.

fromValue(char) - Static method in enum ai.djl.ndarray.types.LayoutType

Converts the character to the matching layout type.

fromValue(String) - Static method in enum ai.djl.ndarray.types.LayoutType

Converts each character to the matching layout type.

fromValue(int) - Static method in enum ai.djl.ndarray.types.SparseFormat

Gets the SparseFormat from it's integer value.

G

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

gelu(NDArray) - Static method in class ai.djl.nn.Activation

Applies GELU(Gaussian Error Linear Unit) activation on the input NDArray.

gelu(NDList) - Static method in class ai.djl.nn.Activation

Applies GELU(Gaussian Error Linear Unit) activation on the input singleton NDList.

geluBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the GELU activation function in its forward function.

generateAnchorBoxes(NDArray) - Method in class ai.djl.modality.cv.MultiBoxPrior

Generates the anchorBoxes array in the input's manager and device.

get(String) - Method in class ai.djl.modality.Classifications

Returns the result for a particular class name.

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

Returns the index affecting the given dimension.

get(NDIndex) - Method in interface ai.djl.ndarray.NDArray

Returns a partial NDArray.

get(String) - Method in interface ai.djl.ndarray.NDArray

Returns a partial NDArray.

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

Returns a partial NDArray.

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

Returns a partial NDArray.

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

Returns the shape in the given dimension.

get(NDManager, long) - Method in class ai.djl.training.dataset.ArrayDataset

Gets the Record for the given index from the dataset.

get(NDManager, long) - Method in class ai.djl.training.dataset.RandomAccessDataset

Gets the Record for the given index from the dataset.

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

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

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

Returns the accumulated evaluator value.

getAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss

Returns the accumulated evaluator value.

getAccumulator(String) - Method in class ai.djl.training.loss.Loss

Returns the accumulated evaluator value.

getArray() - Method in class ai.djl.nn.Parameter

Gets the values of this Parameter as an NDArray.

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

Attempts to load the artifact using the given function and cache it if the specified artifact is not already cached.

getArtifact(String) - Method in class ai.djl.BaseModel

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

getArtifact(String, Function<InputStream, T>) - Method in interface ai.djl.Model

Attempts to load the artifact using the given function and cache it if the specified artifact is not already cached.

getArtifact(String) - Method in interface ai.djl.Model

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

getArtifactAsStream(String) - Method in class ai.djl.BaseModel

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

getArtifactAsStream(String) - Method in interface ai.djl.Model

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

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

Returns the artifact names associated with the model.

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

Returns this index as a full slice if it can be represented as one.

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

Returns value of attached key-value pair to context.

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

Returns the original batch.

getBatchifier() - Method in class ai.djl.translate.NoopTranslator

Gets the Batchifier.

getBatchifier() - Method in interface ai.djl.translate.Translator

Gets the Batchifier.

getBatchSize() - Method in class ai.djl.training.dataset.BatchSampler

Returns the batch size of the Sampler.

getBatchSize() - Method in interface ai.djl.training.dataset.Sampler

Returns the batch size of the Sampler.

getBest() - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

Returns the best hyperparameters and loss.

getBest() - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer

Returns the best hyperparameters and loss.

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

Gets the block from the Model.

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

Gets the block from the Model.

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

Returns a boolean element from this NDArray.

getBoundingBox() - Method in class ai.djl.modality.cv.output.DetectedObjects.DetectedObject

Returns the BoundingBox of the detected object.

getBounds() - Method in interface ai.djl.modality.cv.output.BoundingBox

Returns the bounding Rectangle of this BoundingBox.

getBounds() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the bounding Rectangle of this BoundingBox.

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

Returns an byte element from this NDArray.

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

Returns a list of all the children of the block.

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

Returns a list of all the children of the block.

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

Returns a list of all the children of the block.

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

Returns a list of all the children of the block.

getClasses() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Get classes.

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

Returns the class name.

getComponents() - Method in class ai.djl.training.loss.AbstractCompositeLoss

Returns the component losses that make up the composite loss.

getConfidence() - Method in class ai.djl.modality.cv.output.Joints.Joint

Returns the confidence probability for the joint.

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

Gets the data of this Batch.

getData(NDManager) - Method in interface ai.djl.training.dataset.Dataset

Fetches an iterator that can iterate through the Dataset.

getData(NDManager) - Method in class ai.djl.training.dataset.RandomAccessDataset

Fetches an iterator that can iterate through the Dataset.

getData(NDManager, Sampler) - Method in class ai.djl.training.dataset.RandomAccessDataset

Fetches an iterator that can iterate through the Dataset with a custom sampler.

getData() - Method in class ai.djl.training.dataset.Record

Gets the data of this Record.

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

Returns the standard data type used within the model.

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

Returns the standard data type used within the model.

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

Returns the DataType of this NDArray.

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

Returns the DataType of the NDArray.

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

Returns the default Device of this NDManager.

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

Returns the Device of this NDArray.

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

Returns the default Device of this NDManager.

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

Returns the deviceId of the Device.

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

Returns an array of devices given the maximum number of GPUs to use.

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

Gets the Device that are available for computation.

getDevices() - Method in interface ai.djl.training.Trainer

Returns the devices used for training.

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

Gets the Device that are available for computation.

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

Returns the device type of the Device.

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

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

getDirectParameters() - Method in class ai.djl.nn.convolutional.Convolution

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

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

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

getDirectParameters() - Method in class ai.djl.nn.core.Linear

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

getDirectParameters() - Method in class ai.djl.nn.core.Prelu

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

getDirectParameters() - Method in class ai.djl.nn.LambdaBlock

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

getDirectParameters() - Method in class ai.djl.nn.norm.BatchNorm

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

getDirectParameters() - Method in class ai.djl.nn.norm.Dropout

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

getDirectParameters() - Method in class ai.djl.nn.ParallelBlock

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

getDirectParameters() - Method in class ai.djl.nn.recurrent.RecurrentCell

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

getDirectParameters() - Method in class ai.djl.nn.SequentialBlock

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

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

Returns a double element from this NDArray.

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

Returns the embedded type.

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

Gets the byte array representation of this Shape for serialization.

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

Returns the Engine with the given name.

getEngine() - Method in interface ai.djl.engine.EngineProvider

Returns the instance of the Engine class EngineProvider should bind to.

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

Returns the Engine associated with this manager.

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

Returns the name of the Engine.

getEvaluator(Class<T>) - Method in interface ai.djl.training.Trainer

Gets the Evaluator that is an instance of the given Class.

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

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

getEvaluators() - Method in interface ai.djl.training.Trainer

Gets all Evaluators.

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

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

getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv1D

Returns the expected layout of the input.

getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv2D

Returns the expected layout of the input.

getExpectedLayout() - Method in class ai.djl.nn.convolutional.Conv3D

Returns the expected layout of the input.

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

Returns the expected layout of the input.

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

Returns a float element from this NDArray.

getFormat() - Method in enum ai.djl.ndarray.types.DataType

Returns the format of the data type.

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

Returns the number of GPUs available in the system.

getGpuCount(String) - Static method in class ai.djl.Device

Returns the number of GPUs available with specified framework.

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

Returns the gradient NDArray attached to this NDArray.

getHeight() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the height of the Rectangle.

getHParam(String) - Method in class ai.djl.training.hyperparameter.param.HpSet

Returns the hyperparameter in the set with the given name.

getId() - Method in class ai.djl.nn.Parameter

Gets the ID of this Parameter.

getImageHeight() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Get resized image height.

getImageWidth() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Get resized image width.

getIndex(String) - Method in class ai.djl.modality.nlp.Vocabulary

Returns the index of the given token.

getIndex() - Method in class ai.djl.ndarray.index.NDIndexBooleans

Returns the mask binary NDArray.

getIndex() - Method in class ai.djl.ndarray.index.NDIndexFixed

Returns the dimension of the index.

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

Returns the indices.

getInitializer() - Method in enum ai.djl.nn.ParameterType

Gets the Initializer of this ParameterType.

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

Gets the Initializer to initialize the parameters of the model.

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

Gets the Initializer to initialize the parameters of the model.

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

Returns the default Engine.

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

Returns an int element from this NDArray.

getIoU(BoundingBox) - Method in interface ai.djl.modality.cv.output.BoundingBox

Gets the Intersection over Union (IoU) value between bounding boxes.

getIoU(BoundingBox) - Method in class ai.djl.modality.cv.output.Rectangle

Gets the Intersection over Union (IoU) value between bounding boxes.

getJoints() - Method in class ai.djl.modality.cv.output.Joints

Gets the joints for the image.

getLabels() - Method in class ai.djl.training.dataset.Batch

Gets the labels corresponding to the data of this Batch.

getLabels() - Method in class ai.djl.training.dataset.Record

Gets the labels that correspond to the data of this Record.

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

Returns the labels for each device.

getLastValidLength() - Method in class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer

Returns the valid length of the sentence that was last served as input to SentenceLengthNormalizer.preprocess(List).

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

Returns the layout type for each axis in this shape.

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

Returns the layout type in the given dimension.

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

Returns the number of leading ones in the array shape.

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

Returns a long element from this NDArray.

getLoss(HpSet) - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

Returns the recorded loss.

getLoss(HpSet) - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer

Returns the recorded loss.

getLoss() - Method in interface ai.djl.training.Trainer

Gets the training Loss function of the trainer.

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

Gets the Loss function to compute the loss against.

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

Gets the Loss function to compute the loss against.

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

Returns the NDManager used to create this NDArray.

getManager() - Method in class ai.djl.training.dataset.Batch

Gets the NDManager that is attached to this Batch.

getManager() - Method in interface ai.djl.training.Trainer

Gets the NDManager from the model.

getMax() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the slice max for each axis.

getMax() - Method in class ai.djl.ndarray.index.NDIndexSlice

Returns the end of the range.

getMetric(String) - Method in class ai.djl.metric.Metrics

Returns all Metrics with the specified metric name.

getMetricName() - Method in class ai.djl.metric.Metric

Returns the name of the Metric.

getMetrics() - Method in interface ai.djl.training.Trainer

Returns the Metrics param used for benchmarking.

getMetrics() - Method in interface ai.djl.translate.TranslatorContext

Returns the Metric tool to do benchmark.

getMin() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the slice min for each axis.

getMin() - Method in class ai.djl.ndarray.index.NDIndexSlice

Returns the start of the range.

getModel() - Method in interface ai.djl.training.Trainer

Returns the model used to create this trainer.

getModel() - Method in interface ai.djl.translate.TranslatorContext

Returns the Model object to understand the input/output.

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

Gets the model name.

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

Gets the model name.

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

Returns the name of this NDArray.

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

Returns the name of the NDArray.

getName() - Method in class ai.djl.nn.Parameter

Gets the name of this Parameter.

getName() - Method in class ai.djl.training.evaluator.Evaluator

Returns the name of this Evaluator.

getName() - Method in class ai.djl.training.hyperparameter.param.Hyperparameter

Returns the name of the hyperparameter.

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

Returns an internal representative of Native NDArray.

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

Gets the NDManager from the model.

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

Gets the NDManager from the model.

getNDManager() - Method in interface ai.djl.translate.TranslatorContext

Returns the NDManager to create NDArray.

getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker

Fetches the value of the learning rate after the given number of steps/updates.

getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker

Fetches the value of the learning rate after the given number of steps/updates.

getNewLearningRate(int) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker

Fetches the value of the learning rate after the given number of steps/updates.

getNumberOfObjects() - Method in class ai.djl.modality.cv.output.DetectedObjects

Returns the number of objects found in an image.

getNumOfBytes() - Method in enum ai.djl.ndarray.types.DataType

Returns the number of bytes for each element.

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

Gets the Optimizer to use during training.

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

Gets the Optimizer to use during training.

getOutput(long, long, long, long) - Method in enum ai.djl.nn.pooling.PoolingConvention

Computes the size of the output in the given input dimension for this PoolingConvention.

getOutputShapes(NDManager, Shape[]) - Method in interface ai.djl.nn.Block

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.convolutional.Convolution

Returns the expected output shapes of the block for the specified input shapes.

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

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.Linear

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.core.Prelu

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.LambdaBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.norm.BatchNorm

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.norm.Dropout

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.ParallelBlock

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell

Returns the expected output shapes of the block for the specified input shapes.

getOutputShapes(NDManager, Shape[]) - Method in class ai.djl.nn.SequentialBlock

Returns the expected output shapes of the block for the specified input shapes.

getParagraph() - Method in class ai.djl.modality.nlp.qa.QAInput

Gets the resource document that contains the answer.

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

Returns a list of all the parameters of the block, including the parameters of its children fetched recursively.

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

Returns a list of all the parameters of the block, including the parameters of its children fetched recursively.

getParameterShape(String, Shape[]) - Method in interface ai.djl.nn.Block

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.convolutional.Convolution

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Embedding

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Linear

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.core.Prelu

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.LambdaBlock

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.norm.BatchNorm

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.norm.Dropout

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.ParallelBlock

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.recurrent.RecurrentCell

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

getParameterShape(String, Shape[]) - Method in class ai.djl.nn.SequentialBlock

Returns the shape of the specified direct parameter of this block given the shapes of the input to the block.

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

Returns the parent NDManager.

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

Returns the parent NDManager.

getPath() - Method in class ai.djl.Application

Returns the repository path of the application.

getPath() - Method in interface ai.djl.modality.cv.output.BoundingBox

Returns an iterator object that iterates along the BoundingBox boundary and provides access to the geometry of the BoundingBox outline.

getPath() - Method in class ai.djl.modality.cv.output.Rectangle

Returns an iterator object that iterates along the BoundingBox boundary and provides access to the geometry of the BoundingBox outline.

getPipeline() - Method in class ai.djl.modality.cv.translator.ImageTranslator

Gets the Pipeline applied to the input.

getPipeline() - Method in interface ai.djl.translate.PreProcessor

Gets the Pipeline applied to the input.

getPoint() - Method in interface ai.djl.modality.cv.output.BoundingBox

Returns the top left point of the bounding box.

getPoint() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the top left point of the bounding box.

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

Returns the predictions for each device.

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

Returns the probability.

getProbDist() - Method in class ai.djl.modality.cv.output.Mask

Returns the probability for each pixel.

getProgress() - Method in class ai.djl.training.dataset.Batch

Returns the progress of the batch if it is part of some kind of iteration like a dataset iteration.

getProgressTotal() - Method in class ai.djl.training.dataset.Batch

Returns the total or end value for the progress of the batch if it is part of some kind of iteration like a dataset iteration.

getProperty(String) - Method in class ai.djl.BaseModel

Gets the property of the model based on property name.

getProperty(String) - Method in interface ai.djl.Model

Gets the property of the model based on property name.

getQuestion() - Method in class ai.djl.modality.nlp.qa.QAInput

Gets the question for the model.

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

Returns the number of dimensions specified in the Index.

getRank() - Method in class ai.djl.ndarray.index.NDIndexAll

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.NDIndexBooleans

Returns the number of dimensions occupied by this index element.

getRank() - Method in interface ai.djl.ndarray.index.NDIndexElement

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.NDIndexFixed

Returns the number of dimensions occupied by this index element.

getRank() - Method in class ai.djl.ndarray.index.NDIndexSlice

Returns the number of dimensions occupied by this index element.

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

Gets the Sampler for the dataset.

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

Returns a scalar NDArray corresponding to a single element.

getSeqLength() - Method in class ai.djl.modality.nlp.qa.QAInput

Gets the sequence length.

getShape() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the slice shape without squeezing.

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

Returns the Shape of this NDArray.

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

Returns the Shape of the NDArray.

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

Returns the dimensions of the Shape.

getShapeFromEmptyNDArrayForReductionOp(Shape, int) - Static method in class ai.djl.ndarray.NDUtils

Get Shape of the empty NDArray after applying reduction operations.

getSize() - Method in class ai.djl.training.dataset.Batch

Returns the batchSize.

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

Returns the SparseFormat of this NDArray.

getSqueezedShape() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the slice shape with squeezing.

getStep() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the slice step for each axis.

getStep() - Method in class ai.djl.ndarray.index.NDIndexSlice

Returns the step between each slice.

getStringLayout() - Method in class ai.djl.nn.convolutional.Conv1D

Returns the string representing the layout of the input.

getStringLayout() - Method in class ai.djl.nn.convolutional.Conv2D

Returns the string representing the layout of the input.

getStringLayout() - Method in class ai.djl.nn.convolutional.Conv3D

Returns the string representing the layout of the input.

getStringLayout() - Method in class ai.djl.nn.convolutional.Convolution

Returns the string representing the layout of the input.

getSynsetArtifactName() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Get symset artifact name.

getThreshold() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Get threshold.

getTimestamp() - Method in class ai.djl.metric.Metric

Returns the timestamp of the Metric.

getToken(int) - Method in class ai.djl.modality.nlp.Vocabulary

Returns the token corresponding to the given index.

getToSqueeze() - Method in class ai.djl.ndarray.index.NDIndexFullSlice

Returns the list of axis to squeeze.

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

Returns the number of trailing ones in the array shape.

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

Returns the list of TrainingListeners that should be used during training.

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

Returns the list of TrainingListeners that should be used during training.

getType() - Method in enum ai.djl.ndarray.types.SparseFormat

Returns the SparseFormat name.

getType() - Method in class ai.djl.nn.Parameter

Gets the type of this Parameter.

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

Returns unique identifier of this NDArray.

getUint8(long...) - Method in interface ai.djl.ndarray.NDArray

Returns an integer element from this NDArray that represent unsigned integer with 8 bits.

getUnit() - Method in class ai.djl.metric.Metric

Returns the unit of the Metric.

getUnknownToken() - Method in class ai.djl.modality.nlp.Vocabulary

Return a String used for unseen or rarely-seen tokens.

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

Return the count of unknown value in this Shape.

getValue() - Method in class ai.djl.metric.Metric

Returns the value of the Metric.

getValue() - Method in enum ai.djl.ndarray.types.LayoutType

Returns the character representation of the layout type.

getValue() - Method in enum ai.djl.ndarray.types.SparseFormat

Returns the integer value of this SparseFormat.

getValue(Parameter, Device) - Method in class ai.djl.training.ParameterStore

Returns the value of a mirrored parameter on a device.

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

Returns the version of the deep learning framework.

getWeightDecay() - Method in class ai.djl.training.optimizer.Optimizer

Gets the value of weight decay.

getWidth() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the width of the Rectangle.

getX() - Method in class ai.djl.modality.cv.output.Point

Returns the X coordinate of this Point in double precision.

getX() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the left x-coordinate of the Rectangle.

getY() - Method in class ai.djl.modality.cv.output.Point

Returns the Y coordinate of this Point in double precision.

getY() - Method in class ai.djl.modality.cv.output.Rectangle

Returns the top y-coordinate of the Rectangle.

globalAvgPool1DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool1D pooling function.

globalAvgPool2DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool2D pooling function.

globalAvgPool3DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalAvgPool3D pooling function.

globalLpPool1DBlock(int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool1D pooling function.

globalLpPool2DBlock(int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool2D pooling function.

globalLpPool3DBlock(int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalLpPool3D pooling function.

globalMaxPool1DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalmaxPool1DBlock pooling function.

globalMaxPool2DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalmaxPool2DBlock pooling function.

globalMaxPool3DBlock() - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the globalmaxPool3DBlock pooling function.

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

Returns the default GPU Device.

gpu(int) - Static method in class ai.djl.Device

Returns a new instance of GPU Device with the specified deviceId.

GPU - Static variable in interface ai.djl.Device.Type

GradientCollector - Interface in ai.djl.training

An interface that provides a mechanism to collect gradients during training.

GRU - Class in ai.djl.nn.recurrent

Applies GRU (Gated Recurrent Unit) recurrent layer to input.

GRU.Builder - Class in ai.djl.nn.recurrent

The Builder to construct a GRU type of Block.

gt(Number) - Method in interface ai.djl.ndarray.NDArray

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

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

Returns the boolean NDArray for element-wise "Greater Than" comparison.

gt(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Greater Than" comparison.

gt(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Greater Than" comparison.

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

Returns the boolean NDArray for element-wise "Greater Than" comparison.

gte(Number) - Method in interface ai.djl.ndarray.NDArray

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

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

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

gte(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

gte(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

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

Returns the boolean NDArray for element-wise "Greater or equals" comparison.

H

hasCapability(String) - Method in class ai.djl.engine.Engine

Returns whether the engine has the specified capability.

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

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

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

Returns whether an item is in the embedding.

hasMetric(String) - Method in class ai.djl.metric.Metrics

Returns true if the metrics object has a metric with the given name.

hasNext() - Method in interface ai.djl.modality.cv.output.PathIterator

Tests if the iteration is complete.

hasNext() - Method in class ai.djl.training.dataset.DataIterable

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

Returns the head index of the NDList.

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

Returns the head index of the shape.

HingeLoss - Class in ai.djl.training.loss

HingeLoss is a type of Loss.

HingeLoss() - Constructor for class ai.djl.training.loss.HingeLoss

Calculates Hinge loss.

HingeLoss(String) - Constructor for class ai.djl.training.loss.HingeLoss

Calculates Hinge loss.

HingeLoss(String, int, float) - Constructor for class ai.djl.training.loss.HingeLoss

Calculates Hinge loss.

hingeLoss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of HingeLoss with default arguments.

hingeLoss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of HingeLoss with default arguments.

hingeLoss(String, int, float) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of HingeLoss with the given arguments.

HpBool - Class in ai.djl.training.hyperparameter.param

A Hyperparameter for a boolean option.

HpBool(String) - Constructor for class ai.djl.training.hyperparameter.param.HpBool

Constructs a HpBool.

HpCategorical<T> - Class in ai.djl.training.hyperparameter.param

A Hyperparameter which is one of a fixed number of options (similar to an enum).

HpCategorical(String, List<T>) - Constructor for class ai.djl.training.hyperparameter.param.HpCategorical

Constructs a HpCategorical.

HpFloat - Class in ai.djl.training.hyperparameter.param

A Hyperparameter for a float.

HpFloat(String, float, float, boolean) - Constructor for class ai.djl.training.hyperparameter.param.HpFloat

Constructs a HpFloat.

HpInt - Class in ai.djl.training.hyperparameter.param

A Hyperparameter for an integer.

HpInt(String, int, int) - Constructor for class ai.djl.training.hyperparameter.param.HpInt

Constructs a HpInt.

HpOptimizer - Interface in ai.djl.training.hyperparameter.optimizer

An optimizer for Hyperparameters.

HpORandom - Class in ai.djl.training.hyperparameter.optimizer

A simple HpOptimizer that tries random hyperparameter choices within the range.

HpORandom(HpSet) - Constructor for class ai.djl.training.hyperparameter.optimizer.HpORandom

Constructs a new HpORandom.

HpSet - Class in ai.djl.training.hyperparameter.param

A nestable set of Hyperparameters.

HpSet(String, List<Hyperparameter<?>>) - Constructor for class ai.djl.training.hyperparameter.param.HpSet

Cosntructs a new HpSet.

HpSet(String) - Constructor for class ai.djl.training.hyperparameter.param.HpSet

Cosntructs a new empty HpSet.

HpVal<T> - Class in ai.djl.training.hyperparameter.param

A Hyperparameter with a known value instead of a range of possible values.

HpVal(String, T) - Constructor for class ai.djl.training.hyperparameter.param.HpVal

Cosntructs a new HpVal.

Hyperparameter<T> - Class in ai.djl.training.hyperparameter.param

A class representing an input to the network that can't be differentiated.

Hyperparameter(String) - Constructor for class ai.djl.training.hyperparameter.param.Hyperparameter

Constructs a hyperparameter with the given name.

hyperParams - Variable in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

I

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

Creates a LambdaBlock that performs the identity function.

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

ImageClassificationTranslator - Class in ai.djl.modality.cv.translator

A generic Translator for Image Classification tasks.

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

Constructs an Image Classification using ImageClassificationTranslator.Builder.

ImageClassificationTranslator.Builder - Class in ai.djl.modality.cv.translator

A Builder to construct a ImageClassificationTranslator.

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

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

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

Constructs an ImageTranslator with the provided builder.

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

A builder to extend for all classes extending the ImageTranslator.

ImageVisualization - Class in ai.djl.modality.cv

A Collection of utilities for visualizing the results of Computer Vision tasks.

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

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

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

index - Variable in class ai.djl.training.evaluator.AbstractAccuracy

init(String, NDArray[]) - Method in class ai.djl.training.LocalParameterServer

Initializes the ParameterStore for the given parameter.

init(String, NDArray[]) - Method in interface ai.djl.training.ParameterServer

Initializes the ParameterStore for the given parameter.

initialize(NDManager, DataType, Shape...) - Method in interface ai.djl.nn.Block

Initializes the parameters of the block.

initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.ParallelBlock

Initializes the parameters of the block.

initialize(NDManager, DataType, Shape[]) - Method in class ai.djl.nn.Parameter

Initializes the parameter with the given NDManager, with given DataType for the given expected input shapes.

initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.ParameterBlock

Initializes the parameters of the block.

initialize(NDManager, DataType, Shape...) - Method in class ai.djl.nn.SequentialBlock

Initializes the parameters of the block.

initialize(NDManager, Shape, DataType) - Method in interface ai.djl.training.initializer.Initializer

Initializes a single NDArray.

initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.NormalInitializer

Initializes a single NDArray.

initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.UniformInitializer

Initializes a single NDArray.

initialize(NDManager, Shape, DataType) - Method in class ai.djl.training.initializer.XavierInitializer

Initializes a single NDArray.

initialize(Shape...) - Method in interface ai.djl.training.Trainer

Initializes the Model that the Trainer is going to train.

Initializer - Interface in ai.djl.training.initializer

An interface representing an initialization method.

inputData - Variable in class ai.djl.BaseModel

inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss

Returns the inputs to computing the loss for a component loss.

inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.SimpleCompositeLoss

inputForComponent(int, NDList, NDList) - Method in class ai.djl.training.loss.SingleShotDetectionLoss

Calculate loss between label and prediction.

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

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

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

Built-in Translator that provides image pre-processing from InputStream.

InputStreamTranslator(Translator<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.InputStreamTranslator

Creates a UrlTranslator instance.

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

A factory class creates Translator that can process image from an InputStream.

InputStreamTranslatorFactory(TranslatorFactory<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.InputStreamTranslatorFactory

Creates a FileTranslatorFactory instance.

insert(int, Transform) - Method in class ai.djl.translate.Pipeline

Inserts the given Transform to the list of transforms at the given position.

insert(int, int, Transform) - Method in class ai.djl.translate.Pipeline

Inserts the given Transform to the list of transforms at the given position to be applied on the NDArray at the given index in the input NDList.

insert(int, String, Transform) - Method in class ai.djl.translate.Pipeline

Inserts the given Transform to the list of transforms at the given position to be applied on the NDArray with the given name in the input NDList.

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

InstanceSegmentationTranslator - Class in ai.djl.modality.cv.translator

A ImageTranslator that post-process the NDArray into DetectedObjects with boundaries at the detailed pixel level.

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

Creates the Instance Segmentation translator from the given builder.

InstanceSegmentationTranslator.Builder - Class in ai.djl.modality.cv.translator

The builder for Instance Segmentation translator.

invoke(String, NDArray[], NDArray[], PairList<String, ?>) - Method in interface ai.djl.ndarray.NDManager

An engine specific generic invocation to native operation.

invoke(String, NDList, PairList<String, ?>) - Method in interface ai.djl.ndarray.NDManager

An engine specific generic invocation to native operation.

isBidirectional() - Method in class ai.djl.nn.recurrent.RecurrentCell

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

Returns true if this NDArray is special case: no-value NDArray.

isFloating() - Method in enum ai.djl.ndarray.types.DataType

Checks whether it is a floating data type.

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

Returns the boolean NDArray with value true where this NDArray's entries are infinite, or false where they are not infinite.

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

Returns a boolean whether the block is initialized.

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

Returns a boolean whether the block is initialized.

isInitialized() - Method in class ai.djl.nn.Parameter

Checks if this Parameter is initialized.

isInteger() - Method in enum ai.djl.ndarray.types.DataType

Checks whether it is an integer data type.

isKnownToken(String) - Method in class ai.djl.modality.nlp.Vocabulary

Returns whether the given token is a known word.

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

Returns true if a layout is set.

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

Returns the boolean NDArray with value true where this NDArray's entries are NaN, or false where they are not NaN.

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

Check if the manager is still valid.

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

Check if the manager is still valid.

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

Returns true if this NDArray is a scalar NDArray with empty Shape.

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

Returns true if the NDArray is a scalar.

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

Returns true if this NDArray is a SparseNDArray.

item(int) - Method in class ai.djl.modality.Classifications

Returns the item at a given index based on the order used to construct the Classifications.

item(int) - Method in class ai.djl.modality.cv.output.DetectedObjects

Returns the item at a given index based on the order used to construct the Classifications.

items() - Method in class ai.djl.modality.Classifications

Returns a classification item for each potential class for the input.

iterateDataset(Dataset) - Method in interface ai.djl.training.Trainer

Fetches an iterator that can iterate through the given Dataset.

iterator() - Method in class ai.djl.training.dataset.DataIterable

J

Joint(double, double, double) - Constructor for class ai.djl.modality.cv.output.Joints.Joint

Constructs a Joint with given data.

Joints - Class in ai.djl.modality.cv.output

A result of all joints found during Human Pose Estimation on a single image.

Joints(List<Joints.Joint>) - Constructor for class ai.djl.modality.cv.output.Joints

Constructs the Joints with the provided joints.

Joints.Joint - Class in ai.djl.modality.cv.output

A joint that was detected using Human Pose Estimation on an image.

K

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

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

L

L1Loss - Class in ai.djl.training.loss

L1Loss calculates L1 loss between label and prediction.

L1Loss() - Constructor for class ai.djl.training.loss.L1Loss

Calculates L1 Loss between the label and prediction, a.k.a.

L1Loss(String) - Constructor for class ai.djl.training.loss.L1Loss

Calculates L1 Loss between the label and prediction, a.k.a.

L1Loss(String, float) - Constructor for class ai.djl.training.loss.L1Loss

Calculates L1 Loss between the label and prediction, a.k.a.

l1Loss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L1Loss with default weight and batch axis.

l1Loss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L1Loss with default weight and batch axis.

l1Loss(String, float) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L1Loss with given weight and batch axis.

L2Loss - Class in ai.djl.training.loss

Calculates L2Loss between label and prediction, a.k.a.

L2Loss() - Constructor for class ai.djl.training.loss.L2Loss

Calculate L2Loss between the label and prediction, a.k.a.

L2Loss(String) - Constructor for class ai.djl.training.loss.L2Loss

Calculate L2Loss between the label and prediction, a.k.a.

L2Loss(String, float) - Constructor for class ai.djl.training.loss.L2Loss

Calculates L2Loss between the label and prediction, a.k.a.

l2Loss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L2Loss with default weight and batch axis.

l2Loss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L2Loss with default weight and batch axis.

l2Loss(String, float) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of L2Loss with given weight and batch axis.

labels - Variable in class ai.djl.training.dataset.ArrayDataset

LambdaBlock - Class in ai.djl.nn

LambdaBlock is a Block with no parameters or children.

LambdaBlock(Function<NDList, NDList>) - Constructor for class ai.djl.nn.LambdaBlock

Creates a LambdaBlock that can apply the specified function.

latestMetric(String) - Method in class ai.djl.metric.Metrics

Returns the latest Metric with the specified metric name.

LayoutType - Enum in ai.djl.ndarray.types

An enum to represent the meaning of a particular axis in an NDArray.

leakyRelu(NDArray, float) - Static method in class ai.djl.nn.Activation

Applies Leaky ReLU activation on the input NDArray.

leakyRelu(NDList, float) - Static method in class ai.djl.nn.Activation

Applies Leaky ReLU activation on the input singleton NDList.

leakyReluBlock(float) - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the LeakyReLU activation function in its forward function.

LearningRateTracker - Class in ai.djl.training.optimizer.learningrate

A LearningRateTracker tracks the evolution of the learning rate through the training process.

LearningRateTracker.LrBaseBuilder<T extends LearningRateTracker.LrBaseBuilder> - Class in ai.djl.training.optimizer.learningrate

The Builder to construct a LearningRateTracker.

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

Returns an uninitialized NDArray with the same Shape, DataType and SparseFormat as the input NDArray.

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

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

Linear - Class in ai.djl.nn.core

A Linear block applies a linear transformation \(Y = XW^T + b\).

Linear.Builder - Class in ai.djl.nn.core

The Builder to construct a Linear type of Block.

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

Returns evenly spaced numbers over a specified interval.

linspace(float, float, int) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced numbers over a specified interval.

linspace(int, int, int, boolean, Device) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced numbers over a specified interval.

linspace(float, float, int, boolean, Device) - Method in interface ai.djl.ndarray.NDManager

Returns evenly spaced numbers over a specified interval.

load(Path) - Method in interface ai.djl.Model

Loads the model from the modelPath.

load(Path, String) - Method in interface ai.djl.Model

Loads the model from the modelPath and the given name.

load(Path, String, Map<String, String>) - Method in interface ai.djl.Model

Loads the model from the modelPath with the name and options provided.

load(Path) - Method in interface ai.djl.ndarray.NDManager

Loads the NDArrays saved to a file.

load(Path, Device) - Method in interface ai.djl.ndarray.NDManager

Loads the NDArrays saved to a file.

load(NDManager, DataInputStream) - Method in class ai.djl.nn.Parameter

Loads parameter NDArrays from InputStream.

loadParameters(NDManager, DataInputStream) - Method in interface ai.djl.nn.Block

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.convolutional.Convolution

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Embedding

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Linear

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.core.Prelu

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.LambdaBlock

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.norm.BatchNorm

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.norm.Dropout

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.ParallelBlock

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.recurrent.LSTM

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.recurrent.RecurrentCell

Loads the parameters from the given input stream.

loadParameters(NDManager, DataInputStream) - Method in class ai.djl.nn.SequentialBlock

Loads the parameters from the given input stream.

LocalParameterServer - Class in ai.djl.training

LocalParameterServer is an implementation of the ParameterServer interface.

LocalParameterServer(Optimizer) - Constructor for class ai.djl.training.LocalParameterServer

Create a new instance of LocalParameterServer for the given optimizer.

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

Returns the natural logarithmic value of this NDArray element-wise.

log10() - Method in interface ai.djl.ndarray.NDArray

Returns the base 10 logarithm of this NDArray element-wise.

log2() - Method in interface ai.djl.ndarray.NDArray

Returns the base 2 logarithm of this NDArray element-wise.

logging() - Static method in interface ai.djl.training.listener.TrainingListener.Defaults

A default TrainingListener set including batch output logging.

logging(String) - Static method in interface ai.djl.training.listener.TrainingListener.Defaults

A default TrainingListener set including batch output logging and output directory.

LoggingTrainingListener - Class in ai.djl.training.listener

TrainingListener that outputs the progress of training each batch and epoch into logs.

LoggingTrainingListener() - Constructor for class ai.djl.training.listener.LoggingTrainingListener

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

Returns the truth value of this NDArray AND the other NDArray element-wise.

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

Returns the truth value of NDArray a AND NDArray b element-wise.

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

Computes the truth value of NOT this NDArray element-wise.

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

Computes the truth value of this NDArray OR the other NDArray element-wise.

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

Computes the truth value of NDArray a AND NDArray b element-wise.

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

Computes the truth value of this NDArray XOR the other NDArray element-wise.

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

Computes the truth value of NDArray a AND NDArray b element-wise.

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

Applies the softmax function followed by a logarithm.

logSoftmax(int, float) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function followed by a logarithm.

logSoftmax(int[]) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function followed by a logarithm.

logSoftmax(int[], float) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function followed by a logarithm.

Loss - Class in ai.djl.training.loss

Loss functions (or Cost functions) are used to evaluate the model predictions against true labels for optimization.

Loss(String) - Constructor for class ai.djl.training.loss.Loss

Base class for metric with abstract update methods.

LowerCaseConvertor - Class in ai.djl.modality.nlp.preprocess

LowerCaseConvertor converts every character of the input tokens to it's respective lower case character.

LowerCaseConvertor(Locale) - Constructor for class ai.djl.modality.nlp.preprocess.LowerCaseConvertor

Creates a TextProcessor that converts input text into lower case character given the Locale.

LowerCaseConvertor() - Constructor for class ai.djl.modality.nlp.preprocess.LowerCaseConvertor

Creates a TextProcessor that converts input text into lower case character with the default english Locale.

lpPool1DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.

lpPool1DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.

lpPool1DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.

lpPool1DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool1DBlock pooling function in its forward function.

lpPool2DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.

lpPool2DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.

lpPool2DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.

lpPool2DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool2DBlock pooling function in its forward function.

lpPool3DBlock(Shape, Shape, Shape, PoolingConvention, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.

lpPool3DBlock(Shape, Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.

lpPool3DBlock(Shape, Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the LpPoo3D pooling function in its forward function.

lpPool3DBlock(Shape, int) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the lpPool3DBlock pooling function in its forward function.

LrBaseBuilder() - Constructor for class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

LSTM - Class in ai.djl.nn.recurrent

Applies Long Short-Term Memory recurrent layer to input.

LSTM.Builder - Class in ai.djl.nn.recurrent

The Builder to construct a LSTM type of Block.

lstmStateClipMax - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

lstmStateClipMin - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

lt(Number) - Method in interface ai.djl.ndarray.NDArray

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

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

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

lt(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

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

lt(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

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

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

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

lte(Number) - Method in interface ai.djl.ndarray.NDArray

Returns the boolean NDArray for element-wise "Less or equals" comparison.

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

Returns the boolean NDArray for element-wise "Less or equals" comparison.

lte(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Less or equals" comparison.

lte(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Less or equals" comparison.

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

Returns the boolean NDArray for element-wise "Less or equals" comparison.

M

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

MalformedModelException - Exception in ai.djl

Thrown to indicate Model parameters are not in expected format or are malformed.

MalformedModelException(String) - Constructor for exception ai.djl.MalformedModelException

Constructs a new exception with the specified detail message.

MalformedModelException(String, Throwable) - Constructor for exception ai.djl.MalformedModelException

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

MalformedModelException(Throwable) - Constructor for exception ai.djl.MalformedModelException

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.

manager - Variable in class ai.djl.BaseModel

manager - Variable in class ai.djl.inference.BasePredictor

map(Function<Pair<Long, LayoutType>, Pair<Long, LayoutType>>) - Method in class ai.djl.ndarray.types.Shape

Returns a mapped shape.

Mask - Class in ai.djl.modality.cv.output

A mask with a probability for each pixel within a bounding rectangle.

Mask(double, double, double, double, float[][]) - Constructor for class ai.djl.modality.cv.output.Mask

Constructs a Mask with the given data.

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

Returns the maximum of this NDArray.

max(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the maximum of this NDArray along given axes.

max(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the maximum of this NDArray along given axes.

maximum(Number) - Method in interface ai.djl.ndarray.NDArray

Returns the maximum of this NDArray and a number element-wise.

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

Returns the maximum of this NDArray and the other NDArray element-wise.

maximum(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the maximum of a NDArray and a number element-wise.

maximum(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the maximum of a number and a NDArray element-wise.

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

Returns the maximum of NDArray a and NDArray b element-wise.

maxPool1DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool1D pooling function in its forward function.

maxPool1DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.

maxPool1DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.

maxPool1DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool1DBlock pooling function in its forward function.

maxPool2DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.

maxPool2DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.

maxPool2DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.

maxPool2DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool2DBlock pooling function in its forward function.

maxPool3DBlock(Shape, Shape, Shape, PoolingConvention) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.

maxPool3DBlock(Shape, Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.

maxPool3DBlock(Shape, Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.

maxPool3DBlock(Shape) - Static method in class ai.djl.nn.pooling.Pool

Creates a LambdaBlock that applies the maxPool3DBlock pooling function in its forward function.

mean(String) - Method in class ai.djl.metric.Metrics

Returns the average value of the specified metric.

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

Returns the average of this NDArray.

mean(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the average of this NDArray along given axes.

mean(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the average of this NDArray along given axes.

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

Returns median value for this NDArray.

median(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns median value along given axes.

MemoryTrainingListener - Class in ai.djl.training.listener

TrainingListener that collects the memory usage information.

MemoryTrainingListener() - Constructor for class ai.djl.training.listener.MemoryTrainingListener

Constructs a MemoryTrainingListener that does not output data to a file.

MemoryTrainingListener(String) - Constructor for class ai.djl.training.listener.MemoryTrainingListener

Constructs a MemoryTrainingListener that outputs data in the given directory.

Metric - Class in ai.djl.metric

A class representing a single recorded Metric value.

Metric(String, Number) - Constructor for class ai.djl.metric.Metric

Constructs a Metric instance with the specified metricName and value.

Metric(String, Number, String) - Constructor for class ai.djl.metric.Metric

Constructs a Metric instance with the specified metricName, value , and unit.

metricName(Evaluator, String) - Static method in class ai.djl.training.listener.EvaluatorTrainingListener

Returns the metric created with the evaluator for the given stage.

Metrics - Class in ai.djl.metric

A collection of Metric objects organized by metric name.

Metrics() - Constructor for class ai.djl.metric.Metrics

Constructs an empty Metrics instance.

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

Returns the minimum of this NDArray.

min(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the minimum of this NDArray along given axes.

min(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the minimum of this NDArray along given axes.

minFrequency - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

minimum(Number) - Method in interface ai.djl.ndarray.NDArray

Returns the minimum of this NDArray and a number element-wise.

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

Returns the maximum of this NDArray and the other NDArray element-wise.

minimum(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the minimum of a NDArray and a number element-wise.

minimum(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the minimum of a number and a NDArray element-wise.

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

Returns the minimum of NDArray a and NDArray b element-wise.

mish(NDArray) - Static method in class ai.djl.nn.Activation

Applies Mish activation on the input NDArray.

mish(NDList) - Static method in class ai.djl.nn.Activation

Applies Mish activation on the input singleton NDList.

mishBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the Mish activation function in its forward function.

MKL - Static variable in class ai.djl.engine.StandardCapabilities

MKLDNN - Static variable in class ai.djl.engine.StandardCapabilities

mod(Number) - Method in interface ai.djl.ndarray.NDArray

Returns element-wise remainder of division.

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

Returns element-wise remainder of division.

mod(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns element-wise remainder of division.

mod(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns element-wise remainder of division.

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

Returns element-wise remainder of division.

mode - Variable in class ai.djl.nn.recurrent.RecurrentCell

model - Variable in class ai.djl.inference.BasePredictor

Model - Interface in ai.djl

A model is a collection of artifacts that is created by the training process.

modelDir - Variable in class ai.djl.BaseModel

ModelException - Exception in ai.djl

Thrown to indicate Model parameter or load exceptions parent to Model Exceptions.

ModelException(String) - Constructor for exception ai.djl.ModelException

Constructs a new exception with the specified detail message.

ModelException(String, Throwable) - Constructor for exception ai.djl.ModelException

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

ModelException(Throwable) - Constructor for exception ai.djl.ModelException

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.

modelName - Variable in class ai.djl.BaseModel

modi(Number) - Method in interface ai.djl.ndarray.NDArray

Returns element-wise remainder of division in place.

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

Returns in place element-wise remainder of division in place.

modi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns element-wise remainder of division in place.

modi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns element-wise remainder of division in place.

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

Returns element-wise remainder of division.

mul(Number) - Method in interface ai.djl.ndarray.NDArray

Multiplies this NDArray by a number element-wise.

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

Multiplies this NDArray by other NDArrays element-wise.

mul(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Multiplies the NDArray by a number element-wise.

mul(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Multiplies a number by a NDArray element-wise.

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

Multiplies all of the NDArrays together element-wise.

muli(Number) - Method in interface ai.djl.ndarray.NDArray

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

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

Multiplies this NDArray by other NDArray element-wise in place.

muli(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Multiplies the NDArray by a number element-wise in place.

muli(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Multiplies a number by a NDArray element-wise.

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

Multiplies all of the NDArrays together element-wise in place.

MultiBoxDetection - Class in ai.djl.modality.cv

MultiBoxDetection is the class that takes the output of a multi-box detection model, and converts it into an NDList that contains the object detections.

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

Creates a new instance of MultiBoxDetection with the arguments from the given MultiBoxDetection.Builder.

MultiBoxDetection.Builder - Class in ai.djl.modality.cv

The Builder to construct a MultiBoxDetection object.

MultiBoxPrior - Class in ai.djl.modality.cv

MultiBoxPrior is the class that generates anchor boxes that act as priors for object detection.

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

Creates a new instance of MultiBoxPrior with the arguments from the given MultiBoxPrior.Builder.

MultiBoxPrior.Builder - Class in ai.djl.modality.cv

The Builder to construct a MultiBoxPrior object.

MultiBoxTarget - Class in ai.djl.modality.cv

MultiBoxTarget is the class that computes the training targets for training a Single Shot Detection (SSD) models.

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

Creates a new instance of MultiBoxTarget with the arguments from the given MultiBoxTarget.Builder.

MultiBoxTarget.Builder - Class in ai.djl.modality.cv

The Builder to construct a MultiBoxTarget object.

multiFactorTracker() - Static method in class ai.djl.training.optimizer.learningrate.LearningRateTracker

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

MultiFactorTracker - Class in ai.djl.training.optimizer.learningrate

MultiFactorTracker is an implementation of LearningRateTracker which is updated by a multiplicative factor, at an uneven interval of steps, until it reaches a specified stop value.

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

Creates a new instance of MultiFactorTracker.

MultiFactorTracker.Builder - Class in ai.djl.training.optimizer.learningrate

The Builder to construct an MultiFactorTracker object.

N

Nag - Class in ai.djl.training.optimizer

Nag is a Nesterov accelerated gradient optimizer.

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

Creates a new instance of Nag optimizer.

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

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

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

The Builder to construct an Nag object.

name - Variable in class ai.djl.training.hyperparameter.param.Hyperparameter

NDArray - Interface in ai.djl.ndarray

An interface representing an n-dimensional array.

NDArrays - Class in ai.djl.ndarray

This class contains various methods for manipulating NDArrays.

NDImageUtils - Class in ai.djl.modality.cv.util

NDImageUtils is an image processing utility to load, reshape, and convert images using NDArray images.

NDImageUtils.Flag - Enum in ai.djl.modality.cv.util

Flag indicates the color channel options for images.

NDIndex - Class in ai.djl.ndarray.index

The NDIndex allows you to specify a subset of an NDArray that can be used for fetching or updating.

NDIndex() - Constructor for class ai.djl.ndarray.index.NDIndex

Creates an empty NDIndex to append values to.

NDIndex(String) - Constructor for class ai.djl.ndarray.index.NDIndex

Creates a NDIndex given the index values.

NDIndex(long...) - Constructor for class ai.djl.ndarray.index.NDIndex

Creates an NDIndex with the given indices as specified values on the NDArray.

NDIndexAll - Class in ai.djl.ndarray.index

An NDIndexElement to return all values in a particular dimension.

NDIndexAll() - Constructor for class ai.djl.ndarray.index.NDIndexAll

NDIndexBooleans - Class in ai.djl.ndarray.index

An NDIndexElement to return values based on a mask binary NDArray.

NDIndexBooleans(NDArray) - Constructor for class ai.djl.ndarray.index.NDIndexBooleans

Constructs a NDIndexBooleans instance with specified mask binary NDArray.

NDIndexElement - Interface in ai.djl.ndarray.index

An index for particular dimensions created by NDIndex.

NDIndexFixed - Class in ai.djl.ndarray.index

An NDIndexElement that returns only a specific value in the corresponding dimension.

NDIndexFixed(long) - Constructor for class ai.djl.ndarray.index.NDIndexFixed

Constructs a NDIndexFixed instance with specified dimension.

NDIndexFullSlice - Class in ai.djl.ndarray.index

An index as a slice on all dimensions where some dimensions can be squeezed.

NDIndexSlice - Class in ai.djl.ndarray.index

An NDIndexElement that returns a range of values in the specified dimension.

NDIndexSlice(Long, Long, Long) - Constructor for class ai.djl.ndarray.index.NDIndexSlice

Constructs a NDIndexSlice instance with specified range and step.

NDList - Class in ai.djl.ndarray

An NDList represents a sequence of NDArrays with names.

NDList() - Constructor for class ai.djl.ndarray.NDList

Constructs an empty NDList.

NDList(int) - Constructor for class ai.djl.ndarray.NDList

Constructs an empty NDList with the specified initial capacity.

NDList(NDArray...) - Constructor for class ai.djl.ndarray.NDList

Constructs and initiates an NDList with the specified NDArrays.

NDList(Collection<NDArray>) - Constructor for class ai.djl.ndarray.NDList

Constructs and initiates an NDList with the specified NDArrays.

NDManager - Interface in ai.djl.ndarray

NDArray managers are used to create NDArrays (n-dimensional array on native engine).

NDUtils - Class in ai.djl.ndarray

A class containing utility methods for NDArray operations.

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

Returns the numerical negative NDArray element-wise.

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

Returns the numerical negative NDArray element-wise in place.

neq(Number) - Method in interface ai.djl.ndarray.NDArray

Returns the boolean NDArray for element-wise "Not equals" comparison.

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

Returns the boolean NDArray for element-wise "Not equals" comparison.

neq(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Not equals" comparison.

neq(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Returns the boolean NDArray for element-wise "Not equals" comparison.

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

Returns the boolean NDArray for element-wise "Not equals" comparison.

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

Creates a new top-level NDManager.

newBaseManager(Device) - Method in class ai.djl.engine.Engine

Creates a new top-level NDManager with specified Device.

newBaseManager() - Static method in interface ai.djl.ndarray.NDManager

Creates a new top-level NDManager.

newBaseManager(Device) - Static method in interface ai.djl.ndarray.NDManager

Creates a new top-level NDManager with specified Device.

newBaseManager(Device, String) - Static method in interface ai.djl.ndarray.NDManager

Creates a new top-level NDManager with specified Device and engine.

newGradientCollector() - Method in interface ai.djl.training.Trainer

Returns a new instance of GradientCollector.

newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.FileTranslatorFactory

Returns a new instance of the Translator class.

newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.InputStreamTranslatorFactory

Returns a new instance of the Translator class.

newInstance(Map<String, Object>) - Method in class ai.djl.modality.cv.translator.UrlTranslatorFactory

Returns a new instance of the Translator class.

newInstance() - Static method in interface ai.djl.Model

Creates an empty model instance.

newInstance(Device) - Static method in interface ai.djl.Model

Creates an empty model instance on the specified Device.

newInstance(Device, String) - Static method in interface ai.djl.Model

Creates an empty model instance on the specified Device and engine.

newInstance(Map<String, Object>) - Method in interface ai.djl.translate.TranslatorFactory

Returns a new instance of the Translator class.

newModel(Device) - Method in class ai.djl.engine.Engine

Constructs a new model.

newPredictor(Translator<I, O>) - Method in interface ai.djl.Model

Creates a new Predictor based on the model.

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

Creates a child NDManager.

newSubManager(Device) - Method in interface ai.djl.ndarray.NDManager

Creates a child NDManager with specified default Device.

newTrainer(TrainingConfig) - Method in interface ai.djl.Model

Creates a new Trainer instance for a Model.

next() - Method in interface ai.djl.modality.cv.output.PathIterator

Moves the iterator to the next segment of the path forwards along the primary direction of traversal as long as there are more points in that direction.

next() - Method in class ai.djl.training.dataset.DataIterable

nextConfig() - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer

Returns the next hyperparameters to test.

nextConfig() - Method in class ai.djl.training.hyperparameter.optimizer.HpORandom

Returns the next hyperparameters to test.

none() - Method in interface ai.djl.ndarray.NDArray

Returns true if none of the elements within this NDArray are non-zero or true.

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

Returns the indices of elements that are non-zero.

NoopTranslator - Class in ai.djl.translate

A no operational Translator implementation.

NoopTranslator() - Constructor for class ai.djl.translate.NoopTranslator

NormalInitializer - Class in ai.djl.training.initializer

NormalInitializer initializes weights with random values sampled from a normal distribution with a mean of zero and standard deviation of sigma.

NormalInitializer() - Constructor for class ai.djl.training.initializer.NormalInitializer

Creates an instance of NormalInitializer with a default sigma of 0.01.

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

Creates a Normal initializer.

Normalize - Class in ai.djl.modality.cv.transform

A Transform that normalizes an image NDArray of shape CHW or NCHW.

Normalize(float[], float[]) - Constructor for class ai.djl.modality.cv.transform.Normalize

Creates a Normalize Transform that normalizes.

normalize(NDArray, float, float) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Normalizes an image NDArray of shape CHW or NCHW with a single mean and standard deviation to apply to all channels.

normalize(NDArray, float[], float[]) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Normalizes an image NDArray of shape CHW or NCHW with mean and standard deviation.

numDimensions() - Method in class ai.djl.nn.convolutional.Conv1D

Returns the number of dimensions of the input.

numDimensions() - Method in class ai.djl.nn.convolutional.Conv2D

Returns the number of dimensions of the input.

numDimensions() - Method in class ai.djl.nn.convolutional.Conv3D

Returns the number of dimensions of the input.

numDimensions() - Method in class ai.djl.nn.convolutional.Convolution

Returns the number of dimensions of the input.

numDirections - Variable in class ai.djl.nn.recurrent.RecurrentCell

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

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

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

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

numStackedLayers - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

numStackedLayers - Variable in class ai.djl.nn.recurrent.RecurrentCell

O

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

onEpoch(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener

Listens to the end of an epoch during training.

onEpoch(Trainer) - Method in interface ai.djl.training.listener.TrainingListener

Listens to the end of an epoch during training.

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

Creates an instance of NDArray with specified Shape filled with ones.

ones(Shape) - Method in interface ai.djl.ndarray.NDManager

Creates an instance of NDArray with specified Shape filled with ones.

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

Creates an instance of NDArray with specified Device, Shape, and DataType filled with ones.

ONES - Static variable in interface ai.djl.training.initializer.Initializer

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

Returns an NDArray of ones with the same Shape, DataType and SparseFormat as the input NDArray.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EpochTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.LoggingTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.MemoryTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener

Listens to the end of training one batch of data during training.

onTrainingBatch(Trainer, TrainingListener.BatchData) - Method in interface ai.djl.training.listener.TrainingListener

Listens to the end of training one batch of data during training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener

Listens to the beginning of training.

onTrainingBegin(Trainer) - Method in interface ai.djl.training.listener.TrainingListener

Listens to the beginning of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.EpochTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.LoggingTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.MemoryTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener

Listens to the end of training.

onTrainingEnd(Trainer) - Method in interface ai.djl.training.listener.TrainingListener

Listens to the end of training.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.DivergenceCheckTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EpochTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.EvaluatorTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.LoggingTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.MemoryTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in class ai.djl.training.listener.TimeMeasureTrainingListener

Listens to the end of validating one batch of data during validation.

onValidationBatch(Trainer, TrainingListener.BatchData) - Method in interface ai.djl.training.listener.TrainingListener

Listens to the end of validating one batch of data during validation.

OPENMP - Static variable in class ai.djl.engine.StandardCapabilities

opInputs(ParameterStore, NDList) - Method in class ai.djl.nn.recurrent.LSTM

opInputs(ParameterStore, NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell

optApplySoftmax(boolean) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

Sets whether to apply softmax when processing output.

optAxis(int) - Method in class ai.djl.nn.norm.BatchNorm.Builder

Set the axis in which channel is specified.

optBackgroundId(int) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the class ID for the background.

optBaseLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

Sets the base learning rate.

optBatchier(Batchifier) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Batchifier for the dataset.

optBeginNumUpdate(int) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the initial value of the number of updates.

optBeta1(float) - Method in class ai.djl.training.optimizer.Adam.Builder

Sets the decay rate for the first moment estimates.

optBeta2(float) - Method in class ai.djl.training.optimizer.Adam.Builder

Sets the decay rate for the second moment estimates.

optBias(boolean) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the optional parameter of whether to include a bias vector.

optBias(boolean) - Method in class ai.djl.nn.core.Linear.Builder

Sets the optional parameter that indicates whether to include a bias vector with default value of true.

optBidrectional(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the optional parameter that indicates whether to use bidirectional recurrent layers.

optCenter(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder

If True, add offset of `beta` to normalized tensor.

optClip(boolean) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.

optClip(boolean) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.

optClipGrad(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the value of the \(clipGrad\).

optDataType(DataType) - Method in class ai.djl.nn.core.Embedding.Builder

Sets the data type of the embedding arrays (default is Float32).

optDevice(Device) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Device.

optDevices(Device[]) - Method in class ai.djl.training.DefaultTrainingConfig

Sets the array of Device available for training.

optDilate(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the dilation along each dimension.

optDropRate(float) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the drop rate of the dropout on the outputs of each RNN layer, except the last layer.

optEpsilon(float) - Method in class ai.djl.nn.norm.BatchNorm.Builder

Sets the epsilon value to prevent division by 0.

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

Sets \(epsilon\) - a small quantity for numerical stability.

optExecutor(ExecutorService, int) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the ExecutorService to spawn threads to fetch data.

optFactor(float) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder

Sets the value of the multiplicative factor.

optFactor(float) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder

Set the value of the multiplicative factor.

optFlag(NDImageUtils.Flag) - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder

Sets the optional NDImageUtils.Flag (default is NDImageUtils.Flag#COLOR).

optForceSuppress(boolean) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the boolean parameter that indicates whether to suppress all detections regardless of class_id.

optIgnoreLabel(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the label for ignored anchors.

Optimizer - Class in ai.djl.training.optimizer

An Optimizer updates the weight parameters to minimize the loss function.

Optimizer(Optimizer.OptimizerBuilder<?>) - Constructor for class ai.djl.training.optimizer.Optimizer

Creates a new instance of Optimizer.

Optimizer.OptimizerBuilder<T extends Optimizer.OptimizerBuilder> - Class in ai.djl.training.optimizer

The Builder to construct an Optimizer.

OptimizerBuilder() - Constructor for class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

optInitializer(Initializer) - Method in class ai.djl.training.DefaultTrainingConfig

Sets the Initializer to use for the parameters (default from paper).

optIouThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the anchor-GroundTruth overlap threshold to be regarded as a positive match.

optLabels(NDArray...) - Method in class ai.djl.training.dataset.ArrayDataset.Builder

Sets the labels for the data in the ArrayDataset.

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

Sets the LearningRateTracker for this optimizer.

optLimit(long) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets this dataset's limit.

optLstmStateClipMin(float, float) - Method in class ai.djl.nn.recurrent.LSTM.Builder

Sets the minimum and maximum clip value of LSTM states.

optMaxEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the maximum edge length of the rescaled image.

optMinFrequency(int) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Sets the optional parameter that specifies the minimum frequency to consider a token to be part of the Vocabulary.

optMinNegativeSamples(int) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the minimum number of negative samples.

optMomentum(float) - Method in class ai.djl.nn.norm.BatchNorm.Builder

Set the momentum for moving average.

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

Sets the momentum for Sgd.

optNegativeMinigRatio(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the max negative to positive samples ratio.

optNegativeMiningThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxTarget.Builder

Sets the threshold used for negative mining.

optNmsThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the non-maximum suppression(NMS) threshold.

optNmsTopK(int) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the boolean parameter that indicates whether to clip out-of-boundary boxes.

optNumGroups(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the number of group partitions.

optOffsets(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the value of the center-box offsets across \(x\) and \(y\) dimensions.

optOptimizer(Optimizer) - Method in class ai.djl.training.DefaultTrainingConfig

Sets the Optimizer used during training (default Adam).

optPad(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the padding along each dimension.

optPipeline(Pipeline) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Pipeline of Transform to be applied on the data.

optProbability(float) - Method in class ai.djl.nn.norm.Dropout.Builder

Sets the probability or the fraction of the input that gets dropped out during training time.

optRescaleSize(double, double) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Sets the optional rescale size.

optReservedTokens(Collection<String>) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Sets the optional parameter that sets the list of reserved tokens.

optScale(boolean) - Method in class ai.djl.nn.norm.BatchNorm.Builder

If True, multiply result by `gamma`.

optSharedAxes(int[]) - Method in class ai.djl.nn.norm.Dropout.Builder

Sets the axes for variational dropout kernel.

optShortEdge(int) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the shorter edge length of the rescaled image.

optStateOutput(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the optional parameter that indicates whether to have the states as symbol outputs.

optSteps(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the step across \(x\) and \(y\) dimensions.

optStopFactorLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder

Sets the stop value after which the learning rate should remain constant.

optStride(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the stride of the convolution.

optTargetPipeline(Pipeline) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Pipeline of Transform to be applied on the labels.

optThreshold(float) - Method in class ai.djl.modality.cv.MultiBoxDetection.Builder

Sets the threshold score for a detection to be a positive prediction.

optThreshold(float) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the threshold for prediction accuracy.

optThreshold(float) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder

Sets the threshold for prediction accuracy.

optThreshold(float) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Sets the threshold for prediction accuracy.

optUnknownToken(String) - Method in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

Sets the optional parameter that specifies the unknown token's string value.

optUseDefault(boolean) - Method in class ai.djl.nn.core.Embedding.Builder

Sets whether to use a default embedding for undefined items (default true).

optWarmUpBeginLearningRate(float) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

Sets the value of the learning rate at the beginning of warm-up mode.

optWarmUpMode(WarmUpMode) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

Sets the WarmUpMode for the LearningRateTracker.

optWarmUpSteps(int) - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

Sets the number of steps until the point the learning rate is updated in warm-up mode.

optWeightDecays(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the value of weight decay.

P

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

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

ParallelBlock - Class in ai.djl.nn

ParallelBlock is a Block whose children form a parallel branch in the network and are combined to produce a single output.

ParallelBlock(Function<List<NDList>, NDList>) - Constructor for class ai.djl.nn.ParallelBlock

Creates a parallel block whose branches are combined to form a single output by the given function.

ParallelBlock(Function<List<NDList>, NDList>, List<Block>) - Constructor for class ai.djl.nn.ParallelBlock

Creates a parallel block whose branches are formed by each block in the list of blocks, and are combined to form a single output by the given function.

Parameter - Class in ai.djl.nn

Parameter is a container class that holds a learnable parameter of a model.

Parameter(String, Block, ParameterType) - Constructor for class ai.djl.nn.Parameter

Creates a Parameter with the given name, and parameter type, and associated with the given Block.

Parameter(String, Block, ParameterType, boolean) - Constructor for class ai.djl.nn.Parameter

Creates a Parameter with the given name, and parameter type, and associated with the given Block.

ParameterBlock - Class in ai.djl.nn

ParameterBlock is an abstract implementation of Block.

ParameterBlock() - Constructor for class ai.djl.nn.ParameterBlock

ParameterList - Class in ai.djl.nn

Represents a set of names and Parameters.

ParameterList() - Constructor for class ai.djl.nn.ParameterList

Create an empty ParameterList.

ParameterList(int) - Constructor for class ai.djl.nn.ParameterList

Constructs an empty ParameterList with the specified initial capacity.

ParameterList(List<String>, List<Parameter>) - Constructor for class ai.djl.nn.ParameterList

Constructs a ParameterList containing the elements of the specified keys and values.

ParameterList(List<Pair<String, Parameter>>) - Constructor for class ai.djl.nn.ParameterList

Constructs a ParameterList containing the elements of the specified list of Pairs.

ParameterList(Map<String, Parameter>) - Constructor for class ai.djl.nn.ParameterList

Constructs a ParameterList containing the elements of the specified map.

parameters - Variable in class ai.djl.nn.recurrent.RecurrentCell

ParameterServer - Interface in ai.djl.training

An interface for a key-value store to store parameters, and their corresponding gradients.

ParameterStore - Class in ai.djl.training

The ParameterStore contains a map from a parameter to the mirrors of it on other devices.

ParameterStore(NDManager, boolean) - Constructor for class ai.djl.training.ParameterStore

Constructs an empty ParameterStore.

ParameterType - Enum in ai.djl.nn

Enumerates the types of Parameter.

paramPathResolver(Map<String, String>) - Method in class ai.djl.BaseModel

parent - Variable in class ai.djl.ndarray.BaseNDManager

PathIterator - Interface in ai.djl.modality.cv.output

A sequence of points used to outline an object in an image.

percentile(String, int) - Method in class ai.djl.metric.Metrics

Returns a percentile Metric object for the specified metric name.

percentile(Number) - Method in interface ai.djl.ndarray.NDArray

Returns percentile for this NDArray.

percentile(Number, int[]) - Method in interface ai.djl.ndarray.NDArray

Returns median along given dimension(s).

pipeline - Variable in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder

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

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

Pipeline - Class in ai.djl.translate

Pipeline allows applying multiple transforms on an input NDList.

Pipeline() - Constructor for class ai.djl.translate.Pipeline

Creates a new instance of Pipeline that has no Transform defined yet.

Pipeline(Transform...) - Constructor for class ai.djl.translate.Pipeline

Creates a new instance of Pipeline that can apply the given transforms on its input.

Point - Class in ai.djl.modality.cv.output

A point representing a location in (x,y) coordinate space, specified in double precision.

Point(double, double) - Constructor for class ai.djl.modality.cv.output.Point

Constructs and initializes a point at the specified (x,y) location in the coordinate space.

Pool - Class in ai.djl.nn.pooling

Utility class that provides Block and methods for different pooling functions.

PoolingConvention - Enum in ai.djl.nn.pooling

An enum that enumerates the different pooling conventions.

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

PostProcessor<O> - Interface in ai.djl.translate

An interface that provides post-processing functionality.

pow(Number) - Method in interface ai.djl.ndarray.NDArray

Takes the power of this NDArray with a number element-wise.

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

Takes the power of this NDArray with the other NDArray element-wise.

pow(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Takes the power of the NDArray with a number element-wise.

pow(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Takes the power of a number with a NDArray element-wise.

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

Takes the power of a NDArray with a NDArray element-wise.

powi(Number) - Method in interface ai.djl.ndarray.NDArray

Takes the power of this NDArray with a number element-wise in place.

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

Takes the power of this NDArray with the other NDArray element-wise in place.

powi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Takes the power of the NDArray with a number element-wise in place.

powi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Takes the power of a number with a NDArray element-wise in place.

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

Takes the power of a NDArray with a NDArray element-wise.

predict(I) - Method in class ai.djl.inference.BasePredictor

Predicts an item for inference.

predict(I) - Method in interface ai.djl.inference.Predictor

Predicts an item for inference.

Predictor<I,O> - Interface in ai.djl.inference

The Predictor interface provides a session for model inference.

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

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

Prelu - Class in ai.djl.nn.core

Applies Leaky Parametric ReLU activation element-wise to the input.

Prelu() - Constructor for class ai.djl.nn.core.Prelu

Creates a Parametric ReLU Block.

preluBlock() - Static method in class ai.djl.nn.Activation

Returns a Prelu block.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator

Prepares the translator with the manager and model to use.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Prepares the translator with the manager and model to use.

prepare(NDManager, Model) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

prepare(NDManager, Model) - Method in interface ai.djl.translate.Translator

Prepares the translator with the manager and model to use.

prepared - Variable in class ai.djl.inference.BasePredictor

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

Applies the preprocessing defined to the given input tokens.

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

Applies the preprocessing defined to the given input tokens.

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

Applies the preprocessing defined to the given input tokens.

preprocess(List<String>) - Method in interface ai.djl.modality.nlp.preprocess.TextProcessor

Applies the preprocessing defined to the given input tokens.

PreProcessor<I> - Interface in ai.djl.translate

An interface that provides pre-processing functionality.

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

Preprocesses the word to embed into an NDArray to pass into the model.

probabilities - Variable in class ai.djl.modality.Classifications

processInput(TranslatorContext, Path) - Method in class ai.djl.modality.cv.translator.FileTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, BufferedImage) - Method in class ai.djl.modality.cv.translator.ImageTranslator

Processes the BufferedImage input and converts it to NDList.

processInput(TranslatorContext, InputStream) - Method in class ai.djl.modality.cv.translator.InputStreamTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, BufferedImage) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Processes the BufferedImage input and converts it to NDList.

processInput(TranslatorContext, URL) - Method in class ai.djl.modality.cv.translator.UrlTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, NDList) - Method in class ai.djl.translate.NoopTranslator

Processes the input and converts it to NDList.

processInput(TranslatorContext, I) - Method in interface ai.djl.translate.PreProcessor

Processes the input and converts it to NDList.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.FileTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.InputStreamTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.modality.cv.translator.UrlTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in class ai.djl.translate.NoopTranslator

Processes the output NDList to the corresponding output object.

processOutput(TranslatorContext, NDList) - Method in interface ai.djl.translate.PostProcessor

Processes the output NDList to the corresponding output object.

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

Returns the product of this NDArray.

prod(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the product of this NDArray elements over the given axes.

prod(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the product of this NDArray elements over the given axes.

ProgressBar - Class in ai.djl.training.util

ProgressBar is an implementation of Progress.

ProgressBar() - Constructor for class ai.djl.training.util.ProgressBar

Creates an instance of ProgressBar with a maximum value of 1.

ProgressBar(String, long) - Constructor for class ai.djl.training.util.ProgressBar

Creates an instance of ProgressBar with the given maximum value, and displays the given message.

ProgressBar(String, long, String) - Constructor for class ai.djl.training.util.ProgressBar

Creates an instance of ProgressBar with the given maximum value, and displays the given message.

properties - Variable in class ai.djl.BaseModel

pull(String, NDArray[], int) - Method in class ai.djl.training.LocalParameterServer

Pulls the value of a key from Parameter Server to NDArrays.

pull(String, NDArray[], int) - Method in interface ai.djl.training.ParameterServer

Pulls the value of a key from Parameter Server to NDArrays.

PunctuationSeparator - Class in ai.djl.modality.nlp.preprocess

PunctuationSeparator converts every character of the input tokens to it's respective lower case character.

PunctuationSeparator(String) - Constructor for class ai.djl.modality.nlp.preprocess.PunctuationSeparator

Creates a TextProcessor that separates the given punctuations into distinct tokens in the text.

PunctuationSeparator() - Constructor for class ai.djl.modality.nlp.preprocess.PunctuationSeparator

Creates a TextProcessor that separates the given punctuations into distinct tokens in the text.

push(String, NDArray[], int) - Method in class ai.djl.training.LocalParameterServer

Updates values of a key in Parameter Server.

push(String, NDArray[], int) - Method in interface ai.djl.training.ParameterServer

Updates values of a key in Parameter Server.

Q

QAInput - Class in ai.djl.modality.nlp.qa

The input container for BERT QA model.

QAInput(String, String, int) - Constructor for class ai.djl.modality.nlp.qa.QAInput

Creates the BERT QA model.

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

Converts the model to use a lower precision quantized network.

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

R

random() - Method in class ai.djl.training.hyperparameter.param.HpCategorical

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

random() - Method in class ai.djl.training.hyperparameter.param.HpFloat

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

random() - Method in class ai.djl.training.hyperparameter.param.HpInt

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

random() - Method in class ai.djl.training.hyperparameter.param.HpSet

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

random() - Method in class ai.djl.training.hyperparameter.param.HpVal

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

random() - Method in class ai.djl.training.hyperparameter.param.Hyperparameter

Returns a random value for the hyperparameter for a range of a fixed value if it is a HpVal.

RandomAccessDataset - Class in ai.djl.training.dataset

RandomAccessDataset represent the dataset that support random access reads.

RandomAccessDataset(RandomAccessDataset.BaseBuilder<?>) - Constructor for class ai.djl.training.dataset.RandomAccessDataset

Creates a new instance of RandomAccessDataset with the given necessary configurations.

RandomAccessDataset.BaseBuilder<T extends RandomAccessDataset.BaseBuilder> - Class in ai.djl.training.dataset

The Builder to construct a RandomAccessDataset.

randomColor() - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Returns a random color.

randomMultinomial(int, NDArray) - Method in interface ai.djl.ndarray.NDManager

Draw samples from a multinomial distribution.

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

Draw samples from a multinomial distribution.

randomNormal(Shape) - Method in interface ai.djl.ndarray.NDManager

Draws random samples from a normal (Gaussian) distribution with mean 0 and standard deviation 1.

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

Draws random samples from a normal (Gaussian) distribution with mean 0 and standard deviation 1.

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

Draws random samples from a normal (Gaussian) distribution.

RandomSampler - Class in ai.djl.training.dataset

RandomSampler is an implementation of the Sampler.SubSampler interface.

RandomSampler() - Constructor for class ai.djl.training.dataset.RandomSampler

Creates a new instance of RandomSampler.

RandomSampler(int) - Constructor for class ai.djl.training.dataset.RandomSampler

Creates a new instance of RandomSampler with the given seed.

randomSplit(int...) - Method in class ai.djl.training.dataset.RandomAccessDataset

Splits the dataset set into multiple portions.

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

Draws samples from a uniform distribution.

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

Draws samples from a uniform distribution.

readFileToArray(NDManager, Path) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Reads an image file and converts to NDArray.

readFileToArray(NDManager, Path, NDImageUtils.Flag) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Reads an image file and converts to NDArray.

readInputShapes(DataInputStream) - Method in class ai.djl.nn.AbstractBlock

readParameters(Map<String, String>) - Method in class ai.djl.BaseModel

Record - Class in ai.djl.training.dataset

Record represents a single element of data and labels from Dataset.

Record(NDList, NDList) - Constructor for class ai.djl.training.dataset.Record

Creates a new instance of Record with a single element of data and its corresponding labels.

Rectangle - Class in ai.djl.modality.cv.output

A Rectangle specifies an area in a coordinate space that is enclosed by the Rectangle object's upper-left point Point in the coordinate space, its width, and its height.

Rectangle(double, double, double, double) - Constructor for class ai.djl.modality.cv.output.Rectangle

Constructs a new Rectangle whose upper-left corner is specified as (x,y) and whose width and height are specified by the arguments of the same name.

Rectangle(Point, double, double) - Constructor for class ai.djl.modality.cv.output.Rectangle

Constructs a new Rectangle whose upper-left corner is specified as coordinate point and whose width and height are specified by the arguments of the same name.

RecurrentCell - Class in ai.djl.nn.recurrent

Applies recurrent layers to input data.

RecurrentCell(RecurrentCell.BaseBuilder<?>) - Constructor for class ai.djl.nn.recurrent.RecurrentCell

Creates a RecurrentCell object.

RecurrentCell.BaseBuilder<T extends RecurrentCell.BaseBuilder> - Class in ai.djl.nn.recurrent

The Builder to construct a RecurrentCell type of Block.

relu(NDArray) - Static method in class ai.djl.nn.Activation

Applies ReLU activation on the input NDArray.

relu(NDList) - Static method in class ai.djl.nn.Activation

Applies ReLU activation on the input singleton NDList.

reluBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the ReLU activation function in its forward function.

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

Removes the first occurrence of the specified element from this NDList if it is present.

removeLastBlock() - Method in class ai.djl.nn.SequentialBlock

Removes the Block added last from the sequence of blocks.

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

Removes the last block in the symbolic graph.

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

Repeats element of this NDArray the number of times given repeats.

repeat(int, long) - Method in interface ai.djl.ndarray.NDArray

Repeats element of this NDArray the number of times given repeats along given axis.

repeat(long[]) - Method in interface ai.djl.ndarray.NDArray

Repeats element of this NDArray the number of times given repeats along each axis.

repeat(Shape) - Method in interface ai.djl.ndarray.NDArray

Repeats element of this NDArray to match the desired shape.

replaceLastBlock(Block) - Method in class ai.djl.nn.SequentialBlock

Replaces the Block last added from the sequence of blocks, and adds the given block.

requireGradient() - Method in class ai.djl.nn.Parameter

Returns whether this parameter needs gradients to be computed.

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

reservedTokens - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

reset(String, long, String) - Method in class ai.djl.training.util.ProgressBar

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

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

resetAccumulator(String) - Method in class ai.djl.training.evaluator.Evaluator

Resets the evaluator value with the given key.

resetAccumulator(String) - Method in class ai.djl.training.loss.AbstractCompositeLoss

Resets the evaluator value with the given key.

resetAccumulator(String) - Method in class ai.djl.training.loss.Loss

Resets the evaluator value with the given key.

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

Reshapes this NDArray to the given Shape.

reshape(Shape) - Method in interface ai.djl.ndarray.NDArray

Reshapes this NDArray to the given Shape.

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

Reshapes this NDArray to the match the Shape of the given NDArray.

Resize - Class in ai.djl.modality.cv.transform

A Transform that resizes the image.

Resize(int) - Constructor for class ai.djl.modality.cv.transform.Resize

Creates a Resize Transform that resizes to the given size.

Resize(int, int) - Constructor for class ai.djl.modality.cv.transform.Resize

Creates a Resize Transform that resizes to the given width and height.

resize(NDArray, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Resizes an image to the given size.

resize(NDArray, int, int) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Resizes an image to the given width and height.

resources - Variable in class ai.djl.ndarray.BaseNDManager

results - Variable in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

RNN - Class in ai.djl.nn.recurrent

Applies a single-gate recurrent layer to input.

RNN.Activation - Enum in ai.djl.nn.recurrent

An enum that enumerates the type of activation.

RNN.Builder - Class in ai.djl.nn.recurrent

The Builder to construct a RNN type of Block.

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

Returns the round of this NDArray element-wise.

S

sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.BatchSampler

Fetches an iterator that iterates through the given RandomAccessDataset in mini-batches of indices.

sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.RandomSampler

Fetches an iterator that iterates through the indices of the given RandomAccessDataset.

sample(RandomAccessDataset) - Method in interface ai.djl.training.dataset.Sampler

Fetches an iterator that iterates through the given RandomAccessDataset in mini-batches of indices.

sample(RandomAccessDataset) - Method in interface ai.djl.training.dataset.Sampler.SubSampler

Fetches an iterator that iterates through the indices of the given RandomAccessDataset.

sample(RandomAccessDataset) - Method in class ai.djl.training.dataset.SequenceSampler

Fetches an iterator that iterates through the indices of the given RandomAccessDataset.

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

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

Sampler - Interface in ai.djl.training.dataset

An interface for sampling data items from a RandomAccessDataset.

Sampler.SubSampler - Interface in ai.djl.training.dataset

An interface that samples a single data item at a time.

save(Path, String) - Method in class ai.djl.BaseModel

Saves the model to the specified modelPath with the name provided.

save(Path, String) - Method in interface ai.djl.Model

Saves the model to the specified modelPath with the name provided.

save(DataOutputStream) - Method in class ai.djl.nn.Parameter

Writes the parameter NDArrays to the given output stream.

saveInputShapes(DataOutputStream) - Method in class ai.djl.nn.AbstractBlock

saveParameters(DataOutputStream) - Method in interface ai.djl.nn.Block

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.convolutional.Convolution

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Embedding

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Linear

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.core.Prelu

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.LambdaBlock

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.norm.BatchNorm

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.norm.Dropout

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.ParallelBlock

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.recurrent.LSTM

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.recurrent.RecurrentCell

Writes the parameters of the block to the given outputStream.

saveParameters(DataOutputStream) - Method in class ai.djl.nn.SequentialBlock

Writes the parameters of the block to the given outputStream.

self() - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

self() - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder

self() - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

self() - Method in class ai.djl.modality.cv.translator.SimplePoseTranslator.Builder

self() - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

self() - Method in class ai.djl.nn.convolutional.Conv1D.Builder

self() - Method in class ai.djl.nn.convolutional.Conv2D.Builder

self() - Method in class ai.djl.nn.convolutional.Conv3D.Builder

self() - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

self() - Method in class ai.djl.nn.recurrent.GRU.Builder

self() - Method in class ai.djl.nn.recurrent.LSTM.Builder

self() - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

self() - Method in class ai.djl.nn.recurrent.RNN.Builder

self() - Method in class ai.djl.training.dataset.ArrayDataset.Builder

Returns this {code Builder} object.

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

Returns this {code Builder} object.

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

self() - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder

self() - Method in class ai.djl.training.optimizer.learningrate.LearningRateTracker.LrBaseBuilder

self() - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder

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

self() - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

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

selu(NDArray) - Static method in class ai.djl.nn.Activation

Applies Scaled ELU activation on the input NDArray.

selu(NDList) - Static method in class ai.djl.nn.Activation

Applies Scaled ELU activation on the input singleton NDList.

seluBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the SELU activation function in its forward function.

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

SentenceLengthNormalizer - Class in ai.djl.modality.nlp.preprocess

SentenceLengthNormalizer normalizes the length of all the input sentences to the specified number of tokens.

SentenceLengthNormalizer() - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer

Creates a TextProcessor that normalizes the length of the input.

SentenceLengthNormalizer(int, boolean) - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer

Creates a TextProcessor that normalizes the length of the input to the given sentence length.

SentenceLengthNormalizer(int, boolean, String, String, String) - Constructor for class ai.djl.modality.nlp.preprocess.SentenceLengthNormalizer

Creates a TextProcessor that normalizes the length of the input to the given sentence length.

sentences - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

SequenceSampler - Class in ai.djl.training.dataset

SequenceSampler is an implementation of the Sampler.SubSampler interface.

SequenceSampler() - Constructor for class ai.djl.training.dataset.SequenceSampler

SequentialBlock - Class in ai.djl.nn

SequentialBlock is a Block whose children form a chain of blocks with each child block feeding its output to the next.

SequentialBlock() - Constructor for class ai.djl.nn.SequentialBlock

set(Buffer) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from Buffer.

set(float[]) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from an array of floats.

set(int[]) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from an array of ints.

set(double[]) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from an array of doubles.

set(long[]) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from an array of longs.

set(byte[]) - Method in interface ai.djl.ndarray.NDArray

Sets this NDArray value from an array of bytes.

set(NDIndex, NDArray) - Method in interface ai.djl.ndarray.NDArray

Sets the specified index in this NDArray with the given values.

set(NDIndex, Number) - Method in interface ai.djl.ndarray.NDArray

Sets the specified index in this NDArray with the given value.

setActivation(RNN.Activation) - Method in class ai.djl.nn.recurrent.RNN.Builder

Sets the activation for the RNN - ReLu or Tanh.

setArray(NDArray) - Method in class ai.djl.nn.Parameter

Sets the values of this Parameter.

setAttachment(String, Object) - Method in interface ai.djl.translate.TranslatorContext

Set a key-value pair of attachments.

setBlock(Block) - Method in class ai.djl.BaseModel

Sets the block for the Model for training and inference.

setBlock(Block) - Method in interface ai.djl.Model

Sets the block for the Model for training and inference.

setClasses(List<String>) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Sets the class list.

setData(NDArray...) - Method in class ai.djl.training.dataset.ArrayDataset.Builder

Sets the data as an NDArray for the ArrayDataset.

setDataType(DataType) - Method in class ai.djl.BaseModel

Sets the standard data type used within the model.

setDataType(DataType) - Method in interface ai.djl.Model

Sets the standard data type used within the model.

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

Sets the DataType of the NDArray.

setEmbeddingSize(int) - Method in class ai.djl.nn.core.Embedding.Builder

Sets the size of the embeddings.

setInitializer(Initializer) - Method in class ai.djl.nn.AbstractBlock

Sets an Initializer to the block.

setInitializer(Initializer, String) - Method in class ai.djl.nn.AbstractBlock

Sets an Initializer to the specified direct parameter of the block, overriding the initializer of the parameter, if already set.

setInitializer(Initializer) - Method in interface ai.djl.nn.Block

Sets an Initializer to the block.

setInitializer(Initializer, String) - Method in interface ai.djl.nn.Block

Sets an Initializer to the specified direct parameter of the block, overriding the initializer of the parameter, if already set.

setInitializer(Initializer, boolean) - Method in class ai.djl.nn.Parameter

Sets the Initializer for this Parameter, if not already set.

setItems(Collection<T>) - Method in class ai.djl.nn.core.Embedding.Builder

Sets the collection of items that should feature embeddings.

setKernel(Shape) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the shape of the kernel.

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

Sets the LearningRateTracker for this optimizer.

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

Sets the LearningRateTracker for this optimizer.

setMandatoryDataType(DataType) - Method in class ai.djl.nn.Parameter

Sets the mandatory data type for this Parameter.

setMetrics(Metrics) - Method in class ai.djl.inference.BasePredictor

Attaches a Metrics param to use for benchmark.

setMetrics(Metrics) - Method in interface ai.djl.inference.Predictor

Attaches a Metrics param to use for benchmark.

setMetrics(Metrics) - Method in interface ai.djl.training.Trainer

Attaches a Metrics param to use for benchmarking.

setModelDir(Path) - Method in class ai.djl.BaseModel

setModelName(String) - Method in class ai.djl.BaseModel

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

Sets the momentum for Nag.

setName(String) - Method in interface ai.djl.ndarray.NDArray

Sets name of this NDArray.

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

Sets the name of the NDArray.

setNumFilters(int) - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Sets the Required number of filters.

setNumStackedLayers(int) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the Required number of stacked layers.

setOutChannels(long) - Method in class ai.djl.nn.core.Linear.Builder

Sets the number of output channels.

setParameterServer(ParameterServer, Device[]) - Method in class ai.djl.training.ParameterStore

Sets the parameterServer used to apply updates to the parameters.

setPipeline(Pipeline) - Method in class ai.djl.modality.cv.translator.ImageTranslator.BaseBuilder

Sets the Pipeline to use for pre-processing the image.

setProperty(String, String) - Method in class ai.djl.BaseModel

Sets a property to the model.

setProperty(String, String) - Method in interface ai.djl.Model

Sets a property to the model.

setRandomSeed(int) - Method in class ai.djl.engine.Engine

Seeds the random number generator in DJL Engine.

setRatios(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the aspect ratios of the anchor boxes to be generated around each pixel.

setRescaleGrad(float) - Method in class ai.djl.training.optimizer.Optimizer.OptimizerBuilder

Sets the value used to rescale the gradient.

setSampling(int, boolean) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Sampler with the given batch size.

setSampling(int, boolean, boolean) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Sampler with the given batch size.

setSampling(Sampler) - Method in class ai.djl.training.dataset.RandomAccessDataset.BaseBuilder

Sets the Sampler for the dataset.

setScalar(NDIndex, Number) - Method in interface ai.djl.ndarray.NDArray

Sets the specified index in this NDArray with the given value.

setSequenceLength(boolean) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the optional parameter that indicates whether to include an extra input parameter sequence_length to specify variable length sequence.

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

Sets the Shape of the NDArray.

setSizes(List<Float>) - Method in class ai.djl.modality.cv.MultiBoxPrior.Builder

Sets the sizes of the anchor boxes to be generated around each pixel.

setStateSize(int) - Method in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

Sets the Required size of the state for each layer.

setStep(int) - Method in class ai.djl.training.optimizer.learningrate.FactorTracker.Builder

Sets the number of steps after which the multiplicative factor must be applied once.

setSteps(int[]) - Method in class ai.djl.training.optimizer.learningrate.MultiFactorTracker.Builder

Sets an array of integers indicating when the learning rate should be changed, usually in an uneven interval of steps.

setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.ImageClassificationTranslator.Builder

Sets the name of the synset file listing the potential classes for an image.

setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator.Builder

Sets the name for the synset.

setSynsetArtifactName(String) - Method in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator.Builder

Sets the name for the synset.

setType(Class<T>) - Method in class ai.djl.nn.core.Embedding.Builder

Creates a new Conv2D.Builder with the specified embedding type.

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

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

Sgd - Class in ai.djl.training.optimizer

Sgd is a Stochastic Gradient Descent (SDG) optimizer.

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

Creates a new instance of Sgd.

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

The Builder to construct an Sgd object.

Shape - Class in ai.djl.ndarray.types

A class that presents the NDArray's shape information.

Shape(long...) - Constructor for class ai.djl.ndarray.types.Shape

Constructs and initializes a Shape with specified dimension as {@code (long...

Shape(List<Long>) - Constructor for class ai.djl.ndarray.types.Shape

Constructs and initializes a Shape with specified dimension.

Shape(PairList<Long, LayoutType>) - Constructor for class ai.djl.ndarray.types.Shape

Constructs and initializes a Shape with specified shape and layout pairList.

Shape(long[], String) - Constructor for class ai.djl.ndarray.types.Shape

Constructs and initializes a Shape with specified dimension and layout.

Shape(long[], LayoutType[]) - Constructor for class ai.djl.ndarray.types.Shape

Constructs and initializes a Shape with specified dimension and layout.

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

Checks 2 NDArrays for equal shapes.

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

Checks 2 NDArrays for equal shapes.

sigmoid(NDArray) - Static method in class ai.djl.nn.Activation

Applies Sigmoid activation on the input NDArray.

sigmoid(NDList) - Static method in class ai.djl.nn.Activation

Applies Sigmoid activation on the input singleton NDList.

sigmoidBinaryCrossEntropyLoss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SigmoidBinaryCrossEntropyLoss with default arguments.

sigmoidBinaryCrossEntropyLoss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SigmoidBinaryCrossEntropyLoss with default arguments.

sigmoidBinaryCrossEntropyLoss(String, float, boolean) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SigmoidBinaryCrossEntropyLoss with the given arguments.

SigmoidBinaryCrossEntropyLoss - Class in ai.djl.training.loss

SigmoidBinaryCrossEntropyLoss is a type of Loss.

SigmoidBinaryCrossEntropyLoss() - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss

Performs Sigmoid cross-entropy loss for binary classification.

SigmoidBinaryCrossEntropyLoss(String) - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss

Performs Sigmoid cross-entropy loss for binary classification.

SigmoidBinaryCrossEntropyLoss(String, float, boolean) - Constructor for class ai.djl.training.loss.SigmoidBinaryCrossEntropyLoss

Performs Sigmoid cross-entropy loss for binary classification.

sigmoidBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the Sigmoid activation function in its forward function.

SimpleCompositeLoss - Class in ai.djl.training.loss

SimpleCompositeLoss is an implementation of the Loss abstract class that can combine different Loss functions by adding the individual losses together.

SimpleCompositeLoss() - Constructor for class ai.djl.training.loss.SimpleCompositeLoss

Creates a new empty instance of CompositeLoss that can combine the given Loss components.

SimpleCompositeLoss(String) - Constructor for class ai.djl.training.loss.SimpleCompositeLoss

Creates a new empty instance of CompositeLoss that can combine the given Loss components.

SimplePoseTranslator - Class in ai.djl.modality.cv.translator

A ImageTranslator that post-process the NDArray into human Joints.

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

Creates the Pose Estimation translator from the given builder.

SimplePoseTranslator.Builder - Class in ai.djl.modality.cv.translator

The builder for Pose Estimation translator.

SimpleTokenizer - Class in ai.djl.modality.nlp.preprocess

SimpleTokenizer is an implementation of the Tokenizer interface that converts sentences into token by splitting them by a given delimiter.

SimpleTokenizer(String) - Constructor for class ai.djl.modality.nlp.preprocess.SimpleTokenizer

Creates an instance of SimpleTokenizer with the given delimiter.

SimpleTokenizer() - Constructor for class ai.djl.modality.nlp.preprocess.SimpleTokenizer

Creates an instance of SimpleTokenizer with the default delimiter.

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

Returns the trigonometric sine of this NDArray element-wise.

SingleShotDetectionAccuracy - Class in ai.djl.training.evaluator

SingleShotDetectionAccuracy is an implementation of AbstractAccuracy.

SingleShotDetectionAccuracy(String) - Constructor for class ai.djl.training.evaluator.SingleShotDetectionAccuracy

Creates a new instance of SingleShotDetectionAccuracy with the given name.

SingleShotDetectionLoss - Class in ai.djl.training.loss

SingleShotDetectionLoss is an implementation of Loss.

SingleShotDetectionLoss() - Constructor for class ai.djl.training.loss.SingleShotDetectionLoss

Base class for metric with abstract update methods.

SingleShotDetectionTranslator - Class in ai.djl.modality.cv.translator

A ImageTranslator that post-process the NDArray into DetectedObjects with boundaries.

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

Creates the SSD translator from the given builder.

SingleShotDetectionTranslator.Builder - Class in ai.djl.modality.cv.translator

The builder for SSD translator.

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

Returns the only element if this is a singleton NDList or throws an exception if multiple elements.

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

Returns the hyperbolic sine of this NDArray element-wise.

size() - Method in class ai.djl.modality.nlp.Vocabulary

Returns the size of the Vocabulary.

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

Returns the size of this NDArray along a given axis.

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

Returns the total number of elements in this NDArray.

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

Returns the size of a specific dimension or several specific dimensions.

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

Returns the total size.

size() - Method in class ai.djl.training.dataset.RandomAccessDataset

Returns the size of this Dataset.

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

Creates a new Shape whose content is a slice of this shape.

slice(int, int) - Method in class ai.djl.ndarray.types.Shape

Creates a new Shape whose content is a slice of this shape.

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

Applies the softmax function along the given axis.

softmax(int, float) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function along the given axis.

softmax(int[]) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function along the given axes.

softmax(int[], float) - Method in interface ai.djl.ndarray.NDArray

Applies the softmax function along the given axes.

softmaxCrossEntropyLoss() - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SoftmaxCrossEntropyLoss with default arguments.

softmaxCrossEntropyLoss(String) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SoftmaxCrossEntropyLoss with default arguments.

softmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Static method in class ai.djl.training.loss.Loss

Returns a new instance of SoftmaxCrossEntropyLoss with the given arguments.

SoftmaxCrossEntropyLoss - Class in ai.djl.training.loss

SoftmaxCrossEntropyLoss is a type of Loss that calculates the softmax cross entropy loss.

SoftmaxCrossEntropyLoss() - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss

Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.

SoftmaxCrossEntropyLoss(String) - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss

Creates a new instance of SoftmaxCrossEntropyLoss with default parameters.

SoftmaxCrossEntropyLoss(String, float, int, boolean, boolean) - Constructor for class ai.djl.training.loss.SoftmaxCrossEntropyLoss

Creates a new instance of SoftmaxCrossEntropyLoss with the given parameters.

softrelu(NDArray) - Static method in class ai.djl.nn.Activation

Applies soft ReLU activation on the input NDArray.

softrelu(NDList) - Static method in class ai.djl.nn.Activation

Applies soft ReLU activation on the input singleton NDList.

softreluBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the SoftReLU activation function in its forward function.

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

Sorts the flattened NDArray.

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

Sorts the flattened NDArray.

SparseFormat - Enum in ai.djl.ndarray.types

An enum representing Sparse matrix storage formats.

SparseNDArray - Interface in ai.djl.ndarray

An interface representing a Sparse NDArray.

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

Splits this NDArray into multiple subNDArrays given sections along first axis.

split(long[]) - Method in interface ai.djl.ndarray.NDArray

Splits this NDArray into multiple sub-NDArrays given indices along first axis.

split(long, int) - Method in interface ai.djl.ndarray.NDArray

Splits this NDArray into multiple subNDArrays given sections along the given axis.

split(long[], int) - Method in interface ai.djl.ndarray.NDArray

Splits this NDArray into multiple sub-NDArrays given indices along given axis.

split(Device[], boolean) - Method in class ai.djl.training.dataset.Batch

Splits the data and labels in the Batch across the given devices.

split(NDList, int, boolean) - Method in interface ai.djl.translate.Batchifier

Splits the given NDList into the given number of slices.

split(NDList, int, boolean) - Method in class ai.djl.translate.StackBatchifier

Splits the given NDList into the given number of slices.

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

Returns the square root of this NDArray element-wise.

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

Returns the square of this NDArray element-wise.

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

Removes all singleton dimensions from this NDArray Shape.

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

Removes a singleton dimension at the given axis.

squeeze(int[]) - Method in interface ai.djl.ndarray.NDArray

Removes singleton dimensions at the given axes.

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

Joins a NDArray along the first axis.

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

Joins a NDArray along a new axis.

stack(NDList) - Static method in class ai.djl.ndarray.NDArrays

Joins a sequence of NDArrays in NDList along the first axis.

stack(NDList, int) - Static method in class ai.djl.ndarray.NDArrays

Joins a sequence of NDArrays in NDList along a new axis.

STACK - Static variable in interface ai.djl.translate.Batchifier

StackBatchifier - Class in ai.djl.translate

StackBatchifier is used to merge a list of samples to form a mini-batch of NDArray(s).

StackBatchifier() - Constructor for class ai.djl.translate.StackBatchifier

StandardCapabilities - Class in ai.djl.engine

Constant definitions for the standard capability.

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

stateOutputs - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

stateOutputs - Variable in class ai.djl.nn.recurrent.RecurrentCell

stateShape - Variable in class ai.djl.nn.recurrent.RecurrentCell

stateSize - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

stateSize - Variable in class ai.djl.nn.recurrent.RecurrentCell

step() - Method in interface ai.djl.training.Trainer

Updates all of the parameters of the model once.

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

Returns a stream of the NDIndexElements.

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

Returns a stream of the Shape.

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

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

sub(Number) - Method in interface ai.djl.ndarray.NDArray

Subtracts a number from this NDArray element-wise.

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

Subtracts the other NDArray from this NDArray element-wise.

sub(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Subtracts a number from the NDArray element-wise.

sub(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Subtracts a NDArray from a number element-wise.

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

Subtracts a NDArray from a NDArray element-wise.

subi(Number) - Method in interface ai.djl.ndarray.NDArray

Subtracts a number from this NDArray element-wise in place.

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

Subtracts the other NDArray from this NDArray element-wise in place.

subi(NDArray, Number) - Static method in class ai.djl.ndarray.NDArrays

Subtracts a number from the NDArray element-wise in place.

subi(Number, NDArray) - Static method in class ai.djl.ndarray.NDArrays

Subtracts a NDArray from a number element-wise in place.

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

Subtracts a NDArray from a NDArray element-wise in place.

subNDList(int) - Method in class ai.djl.ndarray.NDList

Returns a view of the portion of this NDList between the specified fromIndex, inclusive, and to the end.

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

Returns the sum of this NDArray.

sum(int[]) - Method in interface ai.djl.ndarray.NDArray

Returns the minimum of this NDArray along given axes.

sum(int[], boolean) - Method in interface ai.djl.ndarray.NDArray

Returns the minimum of this NDArray along given axes.

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

Interchanges two axes of this NDArray.

swish(NDArray, float) - Static method in class ai.djl.nn.Activation

Applies Swish activation on the input NDArray.

swish(NDList, float) - Static method in class ai.djl.nn.Activation

Applies SWish activation on the input singleton NDList.

swishBlock(float) - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the Swish activation function in its forward function.

SymbolBlock - Interface in ai.djl.nn

SymbolBlock is a Block is used to load models that were exported directly from the engine in its native format.

sync() - Method in class ai.djl.training.ParameterStore

Synchronizes the values on all mirrors with the main parameter.

synsetArtifactName - Variable in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

T

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

Returns the trigonometric tangent of this NDArray element-wise.

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

Returns the hyperbolic tangent of this NDArray element-wise.

tanh(NDArray) - Static method in class ai.djl.nn.Activation

Applies Tanh activation on the input NDArray.

tanh(NDList) - Static method in class ai.djl.nn.Activation

Applies Tanh activation on the input singleton NDList.

tanhBlock() - Static method in class ai.djl.nn.Activation

Creates a LambdaBlock that applies the Tanh activation function in its forward function.

target(NDList) - Method in class ai.djl.modality.cv.MultiBoxTarget

Computes multi-box training targets.

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

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

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

TextProcessor - Interface in ai.djl.modality.nlp.preprocess

TextProcessor allows applying pre-processing to input tokens for natural language applications.

threshold - Variable in class ai.djl.modality.cv.translator.SingleShotDetectionTranslator

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

Constructs a NDArray by repeating this NDArray the number of times given repeats.

tile(int, long) - Method in interface ai.djl.ndarray.NDArray

Constructs a NDArray by repeating this NDArray the number of times given by repeats along given axis.

tile(long[]) - Method in interface ai.djl.ndarray.NDArray

Constructs a NDArray by repeating this NDArray the number of times given by repeats.

tile(Shape) - Method in interface ai.djl.ndarray.NDArray

Constructs a NDArray by repeating this NDArray the number of times to match the desired shape.

TimeMeasureTrainingListener - Class in ai.djl.training.listener

TrainingListener that outputs the training time metrics after training is done.

TimeMeasureTrainingListener(String) - Constructor for class ai.djl.training.listener.TimeMeasureTrainingListener

Constructs a TimeMeasureTrainingListener.

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

Converts this NDArray to a Number array based on its DataType.

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

Converts this NDArray to a boolean array.

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

Converts this NDArray to a byte array.

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

Converts this NDArray to a ByteBuffer.

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

Runs the debug string representation of this NDArray.

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

Converts this NDArray from radians to degrees element-wise.

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

Returns a dense representation of the sparse NDArray.

toDevice(Device, boolean) - Method in interface ai.djl.ndarray.NDArray

Moves this NDArray to a different Device.

toDeviceType(Device) - Static method in interface ai.djl.DeviceType

Map device to its type number.

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

Converts this NDArray to a double array.

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

Converts this NDArray to a float array.

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

Converts this NDArray to an int array.

tokenize(String) - Method in class ai.djl.modality.nlp.preprocess.SimpleTokenizer

tokenize(String) - Method in interface ai.djl.modality.nlp.preprocess.Tokenizer

Breaks down the given sentence into a list of tokens that can be represented by embeddings.

Tokenizer - Interface in ai.djl.modality.nlp.preprocess

Tokenizer interface provides the ability to break-down sentences into embeddable tokens.

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

Returns the string layout type for each axis in this shape.

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

Converts this NDArray to a long array.

toNDArray(NDManager, BufferedImage) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Converts a BufferedImage to an RGB NDArray.

toNDArray(NDManager, BufferedImage, NDImageUtils.Flag) - Static method in class ai.djl.modality.cv.util.BufferedImageUtils

Converts a BufferedImage to an NDArray with designated color mode.

topK(int) - Method in class ai.djl.modality.Classifications

Returns a list of the top k best classes.

TopKAccuracy - Class in ai.djl.training.evaluator

TopKAccuracy is an Evaluator that computes the accuracy of the top k predictions.

TopKAccuracy(String, int, int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy

Creates a TopKAccuracy instance.

TopKAccuracy(int, int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy

Creates an instance of TopKAccuracy evaluator that computes topK accuracy across axis 1 along the given index.

TopKAccuracy(int) - Constructor for class ai.djl.training.evaluator.TopKAccuracy

Creates an instance of TopKAccuracy evaluator that computes topK accuracy across axis 1 along the 0th index.

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

Converts this NDArray from degrees to radians element-wise.

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

Returns a sparse representation of NDArray.

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

toString() - Method in class ai.djl.metric.Metric

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

toString() - Method in class ai.djl.modality.Classifications

toString() - Method in class ai.djl.modality.cv.output.DetectedObjects.DetectedObject

toString() - Method in class ai.djl.modality.cv.output.Joints.Joint

toString() - Method in class ai.djl.modality.cv.output.Joints

toString() - Method in class ai.djl.modality.cv.output.Rectangle

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

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

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

toString() - Method in enum ai.djl.ndarray.types.DataType

toString(LayoutType[]) - Static method in enum ai.djl.ndarray.types.LayoutType

Converts a layout type array to a string of the character representations.

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

toString() - Method in class ai.djl.nn.LambdaBlock

toString() - Method in class ai.djl.nn.ParallelBlock

toString() - Method in class ai.djl.nn.ParameterBlock

toString() - Method in class ai.djl.nn.SequentialBlock

toString() - Method in class ai.djl.training.hyperparameter.param.HpCategorical

toString() - Method in class ai.djl.training.hyperparameter.param.HpFloat

toString() - Method in class ai.djl.training.hyperparameter.param.HpInt

toString() - Method in class ai.djl.training.hyperparameter.param.HpSet

toString() - Method in class ai.djl.training.hyperparameter.param.HpVal

totalInstances - Variable in class ai.djl.training.evaluator.Evaluator

ToTensor - Class in ai.djl.modality.cv.transform

A Transform that converts an image NDArray from preprocessing format to Neural Network format.

ToTensor() - Constructor for class ai.djl.modality.cv.transform.ToTensor

toTensor(NDArray) - Static method in class ai.djl.modality.cv.util.NDImageUtils

Converts an image NDArray from preprocessing format to Neural Network format.

toType(DataType, boolean) - Method in interface ai.djl.ndarray.NDArray

Converts this NDArray to a different DataType.

toUint8Array() - Method in interface ai.djl.ndarray.NDArray

Converts this NDArray to a uint8 array.

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

Returns the sum along diagonals of this NDArray.

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

Returns the sum along diagonals of this NDArray.

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

Returns the sum along diagonals of this NDArray.

TRAIN_ALL - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener

TRAIN_EPOCH - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener

TRAIN_PROGRESS - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener

trainBatch(Batch) - Method in interface ai.djl.training.Trainer

Trains the model with one iteration of the given Batch of data.

Trainer - Interface in ai.djl.training

The Trainer interface provides a session for model training.

TrainingConfig - Interface in ai.djl.training

An interface that is responsible for holding the configuration required by Trainer.

TrainingDivergedException - Exception in ai.djl

Thrown to indicate when there is a divergence during Training.

TrainingDivergedException(String) - Constructor for exception ai.djl.TrainingDivergedException

Constructs a new exception with the specified detail message.

TrainingDivergedException(String, Throwable) - Constructor for exception ai.djl.TrainingDivergedException

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

TrainingDivergedException(Throwable) - Constructor for exception ai.djl.TrainingDivergedException

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.

TrainingListener - Interface in ai.djl.training.listener

TrainingListener offers an interface that allows performing some actions when certain events have occurred in the Trainer.

TrainingListener.BatchData - Class in ai.djl.training.listener

A class to pass data from the batch into the training listeners.

TrainingListener.Defaults - Interface in ai.djl.training.listener

Contains default TrainingListener sets.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.CenterCrop

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.Crop

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.Normalize

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.Resize

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.transform.ToTensor

Applies the Transform to the given NDArray.

transform(NDArray) - Method in class ai.djl.modality.cv.translator.InstanceSegmentationTranslator

Applies the Transform to the given NDArray.

transform(NDList) - Method in class ai.djl.translate.Pipeline

Applies the transforms configured in this object on the input NDList.

Transform - Interface in ai.djl.translate

An interface to apply various transforms to the input.

transform(NDArray) - Method in interface ai.djl.translate.Transform

Applies the Transform to the given NDArray.

TranslateException - Exception in ai.djl.translate

Thrown to indicate that an error is raised during processing of the input or output.

TranslateException(String) - Constructor for exception ai.djl.translate.TranslateException

Constructs a new exception with the specified detail message.

TranslateException(String, Throwable) - Constructor for exception ai.djl.translate.TranslateException

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

TranslateException(Throwable) - Constructor for exception ai.djl.translate.TranslateException

Constructs a new exception with the specified cause and a detail message of (cause==null ? null : cause.toString()) which typically contains the class and detail message of cause.

Translator<I,O> - Interface in ai.djl.translate

The Translator interface provides model pre-processing and postprocessing functionality.

TranslatorContext - Interface in ai.djl.translate

The TranslatorContext interface provides a toolkit for pre-processing and postprocessing functionality.

TranslatorFactory<I,O> - Interface in ai.djl.translate

A utility class creates Translator instances.

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

Returns this NDArray with axes transposed.

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

Returns this NDArray with given axes transposed.

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

Returns the truncated value of this NDArray element-wise.

U

uid - Variable in class ai.djl.ndarray.BaseNDManager

unbatchify(NDList) - Method in interface ai.djl.translate.Batchifier

Reverses the batchify operation.

unbatchify(NDList) - Method in class ai.djl.translate.StackBatchifier

Reverses the batchify operation.

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

Returns the closest matching word for a given embedding.

UniformInitializer - Class in ai.djl.training.initializer

UniformInitializer initializes weights with random values uniformly sampled from a given range.

UniformInitializer() - Constructor for class ai.djl.training.initializer.UniformInitializer

Creates an instance of UniformInitializer with a default scale of 0.07.

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

Initializes a uniform initializer.

unknownToken - Variable in class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

update(HpSet, float) - Method in class ai.djl.training.hyperparameter.optimizer.BaseHpOptimizer

Updates the optimizer with the results of a hyperparameter test.

update(HpSet, float) - Method in interface ai.djl.training.hyperparameter.optimizer.HpOptimizer

Updates the optimizer with the results of a hyperparameter test.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Adam

Updates the parameters according to the gradients.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Nag

Updates the parameters according to the gradients.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Optimizer

Updates the parameters according to the gradients.

update(String, NDArray, NDArray) - Method in class ai.djl.training.optimizer.Sgd

Updates the parameters according to the gradients.

update(long, String) - Method in class ai.djl.training.util.ProgressBar

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

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

updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.evaluator.Evaluator

Updates the evaluator with the given key based on a NDList of labels and predictions.

updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.loss.AbstractCompositeLoss

Updates the evaluator with the given key based on a NDList of labels and predictions.

updateAccumulator(String, NDList, NDList) - Method in class ai.djl.training.loss.Loss

Updates the evaluator with the given key based on a NDList of labels and predictions.

updateAllParameters() - Method in class ai.djl.training.ParameterStore

Updates all the mirrored parameters.

updateCount(String) - Method in class ai.djl.training.optimizer.Optimizer

updateInputLayoutToTNC(NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell

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

Built-in Translator that provides image pre-processing from URL.

UrlTranslator(Translator<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.UrlTranslator

Creates a UrlTranslator instance.

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

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

UrlTranslatorFactory(TranslatorFactory<BufferedImage, T>) - Constructor for class ai.djl.modality.cv.translator.UrlTranslatorFactory

Creates a FileTranslatorFactory instance.

useBidirectional - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

useSequenceLength - Variable in class ai.djl.nn.recurrent.RecurrentCell.BaseBuilder

useSequenceLength - Variable in class ai.djl.nn.recurrent.RecurrentCell

V

validate() - Method in class ai.djl.nn.convolutional.Convolution.ConvolutionBuilder

Validates that the required arguments are set.

VALIDATE_EPOCH - Static variable in class ai.djl.training.listener.EvaluatorTrainingListener

validateBatch(Batch) - Method in interface ai.djl.training.Trainer

Validates the given batch of data.

validateInputSize(NDList) - Method in class ai.djl.nn.recurrent.RecurrentCell

validateLayout(LayoutType[], LayoutType[]) - Static method in interface ai.djl.nn.Block

Validates that actual layout matches the expected layout.

valueOf(String) - Static method in enum ai.djl.modality.cv.util.NDImageUtils.Flag

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.ndarray.types.DataType.Format

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.ndarray.types.DataType

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.ndarray.types.LayoutType

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.ndarray.types.SparseFormat

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.nn.ParameterType

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.nn.pooling.PoolingConvention

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.nn.recurrent.RNN.Activation

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.training.dataset.Dataset.Usage

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.training.initializer.XavierInitializer.FactorType

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.training.initializer.XavierInitializer.RandomType

Returns the enum constant of this type with the specified name.

valueOf(String) - Static method in enum ai.djl.training.optimizer.learningrate.WarmUpMode

Returns the enum constant of this type with the specified name.

values() - Static method in enum ai.djl.modality.cv.util.NDImageUtils.Flag

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.ndarray.types.DataType.Format

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.ndarray.types.DataType

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.ndarray.types.LayoutType

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.ndarray.types.SparseFormat

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.nn.ParameterType

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.nn.pooling.PoolingConvention

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.nn.recurrent.RNN.Activation

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.training.dataset.Dataset.Usage

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.training.initializer.XavierInitializer.FactorType

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.training.initializer.XavierInitializer.RandomType

Returns an array containing the constants of this enum type, in the order they are declared.

values() - Static method in enum ai.djl.training.optimizer.learningrate.WarmUpMode

Returns an array containing the constants of this enum type, in the order they are declared.

Vocabulary - Class in ai.djl.modality.nlp

Vocabulary is a collection of tokens.

Vocabulary(Vocabulary.VocabularyBuilder) - Constructor for class ai.djl.modality.nlp.Vocabulary

Create a Vocabulary object with ta Vocabulary.VocabularyBuilder.

Vocabulary.VocabularyBuilder - Class in ai.djl.modality.nlp

Builder class that is used to build the Vocabulary.

VocabularyBuilder() - Constructor for class ai.djl.modality.nlp.Vocabulary.VocabularyBuilder

vocabularyContains(String) - Method in interface ai.djl.modality.nlp.WordEmbedding

Returns whether an embedding exists for a word.

W

waitToRead(NDList) - Method in class ai.djl.inference.BasePredictor

WarmUpMode - Enum in ai.djl.training.optimizer.learningrate

An enum that enumerates the types of warm-up modes for a LearningRateTracker.

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

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

Returns elements chosen from the NDArray or the other NDArray depending on condition.

withDefaultState(Map<String, Map<Device, NDArray>>, String, Device, Function<String, NDArray>) - Method in class ai.djl.training.optimizer.Optimizer

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

WordEmbedding - Interface in ai.djl.modality.nlp

A class to manage 1-D NDArray representations of words.

X

XavierInitializer - Class in ai.djl.training.initializer

XavierInitializer is an Initializer that performs "Xavier" initialization for parameters.

XavierInitializer(XavierInitializer.RandomType, XavierInitializer.FactorType, float) - Constructor for class ai.djl.training.initializer.XavierInitializer

Initializes a Xavier initializer.

XavierInitializer() - Constructor for class ai.djl.training.initializer.XavierInitializer

Creates a new instance of XavierInitializer.

XavierInitializer.FactorType - Enum in ai.djl.training.initializer

Enum for different types of factor type.

XavierInitializer.RandomType - Enum in ai.djl.training.initializer

Enum for different types of random distributions.

Z

zeros(Shape) - Method in interface ai.djl.ndarray.NDManager

Creates an instance of NDArray with specified Shape filled with zeros.

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

Creates an instance of NDArray with specified Shape filled with zeros.

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

Creates an instance of NDArray with specified Device, Shape, and DataType filled with zeros.

ZEROS - Static variable in interface ai.djl.training.initializer.Initializer

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

Returns an NDArray of zeros with the same Shape, DataType and SparseFormat as the input NDArray.

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