org.openimaj.image.annotation.evaluation.datasets

Class CIFAR100Dataset

java.lang.Object

org.openimaj.image.annotation.evaluation.datasets.CIFARDataset

org.openimaj.image.annotation.evaluation.datasets.CIFAR100Dataset

@Reference(type=Article,
           author={"Krizhevsky, A.","Hinton, G."},
           title="Learning multiple layers of features from tiny images",
           year="2009",
           journal="Master\'s thesis, Department of Computer Science, University of Toronto",
           publisher="Citeseer")
 @DatasetDescription(name="CIFAR-100",
                    description="This dataset is just like CIFAR-10, except it has 100 classes containing 600 images each. There are 500 training images and 100 testing images per class. The 100 classes in the CIFAR-100 are grouped into 20 superclasses. Each image comes with a \"fine\" label (the class to which it belongs) and a \"coarse\" label (the superclass to which it belongs).",
                    creator="Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton",
                    url="http://www.cs.toronto.edu/~kriz/cifar.html",
                    downloadUrls="http://datasets.openimaj.org/cifar/cifar-100-binary.tar.gz")
public class CIFAR100Dataset
extends CIFARDataset

CIFAR-100 Dataset. Contains 60000 tiny images in 100 classes (600 per class). There are 500 training images/class and 100 test. Each image is 32x32 pixels.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Field Summary

Fields inherited from class org.openimaj.image.annotation.evaluation.datasets.CIFARDataset

FIMAGE_READER, HEIGHT, MBFIMAGE_READER, WIDTH

Method Summary

Modifier and Type	Method and Description
static <IMAGE> GroupedDataset<String,ListDataset<IMAGE>,IMAGE>	getTestImages(BinaryReader<IMAGE> reader, boolean fineLabels) Load the test images using the given reader.
static <IMAGE> GroupedDataset<String,ListDataset<IMAGE>,IMAGE>	getTrainingImages(BinaryReader<IMAGE> reader, boolean fineLabels) Load the training images using the given reader.

Methods inherited from class java.lang.Object

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

Method Detail

getTrainingImages

public static <IMAGE> GroupedDataset<String,ListDataset<IMAGE>,IMAGE> getTrainingImages(BinaryReader<IMAGE> reader,
                                                                                        boolean fineLabels)
                                                                                 throws IOException

Load the training images using the given reader. To load the images as MBFImages, you would do the following: CIFAR100Dataset.getTrainingImages(CIFAR100Dataset.MBFIMAGE_READER);

Parameters:

reader - the reader

fineLabels - if true, then the fine labels will be used; otherwise the coarse superclass labels will be used.

Returns:

the training image dataset

Throws:

IOException

getTestImages

public static <IMAGE> GroupedDataset<String,ListDataset<IMAGE>,IMAGE> getTestImages(BinaryReader<IMAGE> reader,
                                                                                    boolean fineLabels)
                                                                             throws IOException

Load the test images using the given reader. To load the images as MBFImages, you would do the following: CIFAR100Dataset.getTestImages(CIFAR100Dataset.MBFIMAGE_READER);

Parameters:

reader - the reader

fineLabels - if true, then the fine labels will be used; otherwise the coarse superclass labels will be used.

Returns:

the test image dataset

Throws:

IOException