MultiLayerNetwork.fit(DataSetIterator)
This is mainly used for debugging purposes; generally an iterator that isn't safe to asynchronously prefetch from
should simply return asyncSupported() == falseDataSetCallbackLoader<DataSet> such as SerializedDataSetLoader.Iterator<String>) the DataSetIterator cannot be reset.DefaultCallbackIterable<Pair<double[], double[]>>.Iterable<DataSet> and Iterator<DataSet>.DataSet objects that have
previously been saved to files with DataSet.save(File).Random instance is providedMultiDataSet objects that have
previously been saved to files with MultiDataSet.save(File).Random instance is providedIterable<Pair<float[], float[]>>
First value in pair is the features vector, second value in pair is the labels.MultiDataSetPreProcessor, if one has previously been set.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.true if the iteration has more elements.Iterable<Pair<INDArray,INDArray>>.Iterator<DataSet>, combining and splitting the input DataSet objects as
required to get the specified batch size.Iterator<MultiDataSet>, combining and splitting the input DataSet objects as
required to get a specified batch size.Collection<DataSet>.Loader<MultiDataSet> such as SerializedMultiDataSetLoader.Iterator<String>) the MultiDataSetIterator cannot be reset.RandomDataSetIterator.Values enumeration.RandomMultiDataSetIterator.Values enumeration.Copyright © 2019. All rights reserved.