Package org.deeplearning4j.iterator

Interface Summary

Interface	Description
LabeledSentenceProvider	LabeledSentenceProvider: a simple iterator interface over sentences/documents that have a label. This is intended for use with CnnSentenceDataSetIterator

Class Summary

Class	Description
BertIterator	BertIterator is a MultiDataSetIterator for training BERT (Transformer) models in the following way: (a) Unsupervised - Masked language model task (no sentence matching task is implemented thus far) (b) Supervised - For sequence classification (i.e., 1 label per sequence, typically used for fine tuning) The task can be specified using BertIterator.Task.
BertIterator.Builder
CnnSentenceDataSetIterator	A DataSetIterator that provides data for training a CNN sentence classification models (though can of course be used for general documents, not just sentences.
CnnSentenceDataSetIterator.Builder

Enum Summary

Enum	Description
BertIterator.FeatureArrays
BertIterator.LengthHandling
BertIterator.Task
BertIterator.UnsupervisedLabelFormat
CnnSentenceDataSetIterator.Format	Format of features: CNN1D: For use with 1d convolution layers: Shape [minibatch, vectorSize, sentenceLength] CNN2D: For use with 2d convolution layers: Shape [minibatch, 1, vectorSize, sentenceLength] or [minibatch, 1, sentenceLength, vectorSize], depending on the setting for 'sentencesAlongHeight' configuration.
CnnSentenceDataSetIterator.UnknownWordHandling