Package ai.djl.training.dataset

package ai.djl.training.dataset

Contains classes to download and prepare training and testing data.

The central class to work with in this package is the Dataset. In practice, most of the implementations of Dataset will actually extend RandomAccessDataset instead.

Related Packages

Package	Description
ai.djl.training	Contains classes and implementations to train a neural network.

Class	Description
ArrayDataset	ArrayDataset is an implementation of RandomAccessDataset that consist entirely of large NDArrays.
ArrayDataset.Builder	The Builder to construct an ArrayDataset.
Batch	A Batch is used to hold multiple items (data and label pairs) from a Dataset.
BatchSampler	BatchSampler is a Sampler that returns a single epoch over the data.
BulkDataIterable	BulkDataIterable specializes DataIterable in using ArrayDataset.getByRange(NDManager, long, long) or ArrayDataset.getByIndices(NDManager, long...) to create Batch instances more efficiently.
DataIterable	DataIterable is a data loader that combines Dataset, Batchifier, Pipeline, and Sampler to provide an iterable over the given RandomAccessDataset.
Dataset	An interface to represent a set of sample data/label pairs to train a model.
Dataset.Usage	An enum that indicates the mode - training, test or validation.
RandomAccessDataset	RandomAccessDataset represent the dataset that support random access reads.
RandomAccessDataset.BaseBuilder<T extends RandomAccessDataset.BaseBuilder<T>>	The Builder to construct a RandomAccessDataset.
RandomSampler	RandomSampler is an implementation of the Sampler.SubSampler interface.
RawDataset<T>	An interface can read a plain java object from dataset.
Record	Record represents a single element of data and labels from Dataset.
Sampler	An interface for sampling data items from a RandomAccessDataset.
Sampler.SubSampler	An interface that samples a single data item at a time.
SequenceSampler	SequenceSampler is an implementation of the Sampler.SubSampler interface.