---
# For reference on dataset card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/datasetcard.md?plain=1
# Doc / guide: https://huggingface.co/docs/hub/datasets-cards
{}
---

# Dataset Card for Imagenette

<!-- Provide a quick summary of the dataset. -->

## Dataset Details

### Dataset Description

<!-- Provide a longer summary of what this dataset is. -->
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

<!-- Provide the basic links for the dataset. -->

- **Homepage:** https://github.com/fastai/imagenette

## Dataset Structure

<!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->

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

<!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->

**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}
}