Dataset Card for Imagenette

Dataset Details

Dataset Description

Imagenette is a subset of 10 easily classified classes from ImageNet (tench, English springer, cassette player, chain saw, church, French horn, garbage truck, gas pump, golf ball, parachute).

• License: Apache 2.0

Dataset Sources

• Homepage: https://github.com/fastai/imagenette

Dataset Structure

Total images: 13,394

Splits:

• Train: 9,469 images

• Test: 3,925 images

Classes (labels): 10 categories

Image specs: JPEG format, various resolutions, 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/imagenette", split="train", trust_remote_code=True)
# dataset = load_dataset("randall-lab/imagenette", split="test", trust_remote_code=True)

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

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

Citation

BibTeX:

@misc{imagenette, author = {FastAI}, title = {Imagenette: A smaller subset of ImageNet}, year = {2019}, url = {https://github.com/fastai/imagenette}, note = {Accessed: March 17, 2025} }