The CIFAR-10 data consists of 60,000 32x32 color images in 10 classes, with 6000 images per class. There are 50,000 training images and 10,000 test images in the official data.
The label classes in the dataset are:
- airplane
- automobile
- bird
- cat
- deer
- dog
- frog
- horse
- ship
- truck
The classes are completely mutually exclusive. There is no overlap between automobiles and trucks. "Automobile" includes sedans, SUVs, things of that sort. "Truck" includes only big trucks. Neither includes pickup trucks.
External Dataset Link -https://www.kaggle.com/c/cifar-10/data