org.deeplearning4j.datasets.iterator.impl

Class TinyImageNetDataSetIterator

java.lang.Object

org.deeplearning4j.datasets.datavec.RecordReaderDataSetIterator

org.deeplearning4j.datasets.iterator.impl.TinyImageNetDataSetIterator

All Implemented Interfaces:

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

public class TinyImageNetDataSetIterator
extends RecordReaderDataSetIterator

Tiny ImageNet is a subset of the ImageNet database. TinyImageNet is the default course challenge for CS321n at Stanford University. Tiny ImageNet has 200 classes, each consisting of 500 training images. See: http://cs231n.stanford.edu/ and https://tiny-imagenet.herokuapp.com/

Author:

Justin Long (crockpotveggies)

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class org.deeplearning4j.datasets.datavec.RecordReaderDataSetIterator

RecordReaderDataSetIterator.Builder

Field Summary

Fields

Modifier and Type	Field and Description
protected org.nd4j.linalg.dataset.api.DataSetPreProcessor	preProcessor

Fields inherited from class org.deeplearning4j.datasets.datavec.RecordReaderDataSetIterator

batchNum, batchSize, converter, labelIndex, labelIndexTo, last, maxNumBatches, numPossibleLabels, recordReader, regression, sequenceIter, useCurrent

Constructor Summary

Constructors

Constructor and Description
TinyImageNetDataSetIterator(int batchSize)
TinyImageNetDataSetIterator(int batchSize, DataSetType set)
TinyImageNetDataSetIterator(int batchSize, int[] imgDim, DataSetType set, org.datavec.image.transform.ImageTransform imageTransform, long rngSeed) Get the Tiny ImageNet iterator with specified train/test set and custom transform.

Method Summary

Methods inherited from class org.deeplearning4j.datasets.datavec.RecordReaderDataSetIterator

asyncSupported, batch, cursor, getLabels, getPreProcessor, hasNext, inputColumns, isCollectMetaData, loadFromMetaData, loadFromMetaData, next, next, numExamples, remove, reset, resetSupported, setCollectMetaData, 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

Field Detail

preProcessor

protected org.nd4j.linalg.dataset.api.DataSetPreProcessor preProcessor

Constructor Detail

TinyImageNetDataSetIterator

public TinyImageNetDataSetIterator(int batchSize)

TinyImageNetDataSetIterator

public TinyImageNetDataSetIterator(int batchSize,
                                   DataSetType set)

TinyImageNetDataSetIterator

public TinyImageNetDataSetIterator(int batchSize,
                                   int[] imgDim,
                                   DataSetType set,
                                   org.datavec.image.transform.ImageTransform imageTransform,
                                   long rngSeed)

Get the Tiny ImageNet iterator with specified train/test set and custom transform.

Parameters:

batchSize - Size of each patch

imgDim - Dimensions of desired output

set - Train, test, or validation

imageTransform - Additional image transform for output

rngSeed - random number generator seed to use when shuffling examples