galilai-group/

cifar100

	Dataset Card for CIFAR-100






	
	Dataset Details




	
	Dataset Description

The CIFAR-100 dataset consists of 60,000 color images of size 32×32 pixels, divided into 100 different classes. Each class contains 600 images. The dataset is commonly used for image classification tasks in deep learning research. The 100 classes are grouped into 20 superclasses, each containing 5 related classes.

Homepage: https://www.cs.toronto.edu/~kriz/cifar.html… See the full description on the dataset page: https://huggingface.co/datasets/galilai-group/cifar100.