Datasets:
annotations_creators:
- expert-generated
language:
- en
license: mit
multilinguality:
- monolingual
pretty_name: CIFAR-10
size_categories:
- 100K<n<1M
source_datasets:
- original
task_categories:
- image-classification
task_ids:
- multi-class-image-classification
tags:
- computer-vision
- image-classification
- benchmark
- cifar
- object-detection
CIFAR-10 - Object Recognition in Images
Benchmark dataset for object classification.
🖼️ 60,000 32x32 color images
🏷️ 10 classes
📁 Format: PNG, CSV
📦 Files: 4
🧪 Subset of the 80 million tiny images dataset
Dataset Summary
CIFAR-10 is a widely used computer vision dataset consisting of 60,000 32x32 color images in 10 mutually exclusive classes. It was created by Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton. The dataset is a labeled subset of the 80 million tiny images dataset and is often used as a benchmark for image classification tasks.
This Hugging Face version mirrors the original Kaggle competition structure, including additional junk test images to discourage cheating.
Dataset Structure
Files Included
| File | Description |
|---|---|
train.7z |
Training images in PNG format (50,000 images) |
test.7z |
Test images in PNG format (300,000 images incl. junk) |
trainLabels.csv |
Training image labels |
sampleSubmission.csv |
Sample format for submission predictions |
Label Classes
Each image is labeled with one of the following 10 classes:
- airplane
- automobile
- bird
- cat
- deer
- dog
- frog
- horse
- ship
- truck
Note: "automobile" includes sedans and SUVs; "truck" includes large trucks only (not pickups).
Data Splits
| Split | Number of Images |
|---|---|
| Train | 50,000 |
| Test | 10,000 (scored) + 290,000 (junk) |
Total: 300,000 test image predictions are required, though only 10,000 are scored.
Usage Example
from torchvision.datasets import CIFAR10
import torchvision.transforms as transforms
transform = transforms.Compose([
transforms.ToTensor()
])
trainset = CIFAR10(root='./data', train=True, download=True, transform=transform)
testset = CIFAR10(root='./data', train=False, download=True, transform=transform)
Citation
If you use this dataset, please cite the original technical report:
@techreport{Krizhevsky2009LearningML,
title={Learning Multiple Layers of Features from Tiny Images},
author={Alex Krizhevsky},
year={2009},
institution={University of Toronto},
url={https://www.cs.toronto.edu/~kriz/learning-features-2009-TR.pdf}
}