Skip to content

Latest commit

 

History

History

cifair

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 

ciFAIR-10/100 Dataset

ciFAIR is a variant of the popular CIFAR dataset, which uses a slightly modified test set avoiding near-duplicates between training and test data. It comprises RGB images of size 32x32 spanning 10 and 100 classes of everyday objects for ciFAIR-10 and ciFAIR-100, respectively.

CIFAR homepage: https://www.cs.toronto.edu/~kriz/cifar.html
CIFAR paper: https://www.cs.toronto.edu/~kriz/learning-features-2009-TR.pdf

ciFAIR homepage: https://cvjena.github.io/cifair/
ciFAIR Paper: https://arxiv.org/abs/1902.00423

Example images from ciFAIR-10

Splits

We use the following splits of the ciFAIR-10/100 dataset for testing small-data performance:

Split Total Images Images / Class
train 300 / 3,000 30
val 200 / 2,000 20
trainval 500 / 5,000 50
test 10,000 1,000 / 100

train comprises the first 30 training images from each class and val the following 20. trainval is a combination of both. test is the full original ciFAIR-10 test set.

Baseline Performance

We achieved the following baseline performance using a Wide ResNet 16-8 trained on the trainval split and averaged over 10 runs.

Dataset Variant Accuracy
ciFAIR-10 58.22%
ciFAIR-100 53.42%