Dataset Card for Tiny ImageNet

Dataset Details

Dataset Description

In Tiny ImageNet, there are 100,000 images divided up into 200 classes.

• License: MIT License

Dataset Sources

• Homepage: https://www.kaggle.com/c/tiny-imagenet
• Paper: Le, Y., & Yang, X. (2015). Tiny imagenet visual recognition challenge. CS 231N, 7(7), 3.

Dataset Structure

Total images: 110,000

Classes: 200 categories

Splits:

• Train: 100,000 images

• Validation: 10,000 images

Image specs: JPEG format, 64×64 pixels, RGB

Example Usage

Below is a quick example of how to load this dataset via the Hugging Face Datasets library.

from datasets import load_dataset 

# Load the dataset 
dataset = load_dataset("randall-lab/tiny-imagenet", split="train", trust_remote_code=True)
# dataset = load_dataset("randall-lab/tiny-imagenet", split="validation", trust_remote_code=True) 

# Access a sample from the dataset 
example = dataset[0] 
image = example["image"] 
label = example["label"] 

image.show() # Display the image 
print(f"Label: {label}")

Citation

BibTeX:

@article{le2015tiny, title={Tiny imagenet visual recognition challenge}, author={Le, Yann and Yang, Xuan}, journal={CS 231N}, volume={7}, number={7}, pages={3}, year={2015} }