org.deeplearning4j.datasets.iterator.impl

Class MnistDataSetIterator

java.lang.Object

org.nd4j.linalg.dataset.api.iterator.BaseDatasetIterator

org.deeplearning4j.datasets.iterator.impl.MnistDataSetIterator

All Implemented Interfaces:

Serializable, Iterator<org.nd4j.linalg.dataset.DataSet>, org.nd4j.linalg.dataset.api.iterator.DataSetIterator

public class MnistDataSetIterator
extends org.nd4j.linalg.dataset.api.iterator.BaseDatasetIterator

Mnist data applyTransformToDestination iterator.

Author:

Adam Gibson

See Also:

Serialized Form

Field Summary

Fields inherited from class org.nd4j.linalg.dataset.api.iterator.BaseDatasetIterator

batch, fetcher, numExamples, preProcessor

Constructor Summary

Constructors

Constructor and Description
MnistDataSetIterator(int batchSize, boolean train, int seed) Constructor to get the full MNIST data set (either test or train sets) without binarization (i.e., just normalization into range of 0 to 1), with shuffling based on a random seed.
MnistDataSetIterator(int batch, int numExamples)
MnistDataSetIterator(int batch, int numExamples, boolean binarize) Get the specified number of examples for the MNIST training data set.
MnistDataSetIterator(int batch, int numExamples, boolean binarize, boolean train, boolean shuffle, long rngSeed) Get the specified number of MNIST examples (test or train set), with optional shuffling and binarization.

Method Summary

Methods inherited from class org.nd4j.linalg.dataset.api.iterator.BaseDatasetIterator

asyncSupported, batch, cursor, getLabels, getPreProcessor, hasNext, inputColumns, next, next, numExamples, remove, reset, resetSupported, setPreProcessor, totalExamples, totalOutcomes

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface java.util.Iterator

forEachRemaining

Constructor Detail

MnistDataSetIterator

public MnistDataSetIterator(int batch,
                            int numExamples)
                     throws IOException

Throws:

IOException

MnistDataSetIterator

public MnistDataSetIterator(int batch,
                            int numExamples,
                            boolean binarize)
                     throws IOException

Get the specified number of examples for the MNIST training data set.

Parameters:

batch - the batch size of the examples

numExamples - the overall number of examples

binarize - whether to binarize mnist or not

Throws:

IOException

MnistDataSetIterator

public MnistDataSetIterator(int batchSize,
                            boolean train,
                            int seed)
                     throws IOException

Constructor to get the full MNIST data set (either test or train sets) without binarization (i.e., just normalization into range of 0 to 1), with shuffling based on a random seed.

Parameters:

batchSize -

train -

Throws:

IOException

MnistDataSetIterator

public MnistDataSetIterator(int batch,
                            int numExamples,
                            boolean binarize,
                            boolean train,
                            boolean shuffle,
                            long rngSeed)
                     throws IOException

Get the specified number of MNIST examples (test or train set), with optional shuffling and binarization.

Parameters:

batch - Size of each patch

numExamples - total number of examples to load

binarize - whether to binarize the data or not (if false: normalize in range 0 to 1)

train - Train vs. test set

shuffle - whether to shuffle the examples

rngSeed - random number generator seed to use when shuffling examples

Throws:

IOException