Fashion-MNIST
A compact fashion-image classification benchmark suitable for learning and quick model comparisons.
What the resource contains
- 60,000 training and 10,000 test images
- 28-by-28 grayscale images
- Ten clothing and accessory classes
Potential uses
- Teach image classification
- Benchmark lightweight models
- Test preprocessing pipelines
Access & formats
IDX / image arrays. Repository / linked release. Follow the project’s instructions on GitHub; some resources require external files, registration or approval.
License & reuse
The dataset project declares MIT licensing. Review and retain the repository license notice.
Important limitations
Low-resolution grayscale classes are not detailed SKU metadata, product listings, or a visual shopping catalog.
Source & attribution
zalandoresearch — original GitHub project ↗
Documentation reviewed 2026-09-24. Review scope and licensing notes. Suggest a correction.
Related resources
FEIDEGGER
zalandoresearch
Fashion photographs paired with multiple German descriptions for multimodal retrieval research.
Images / text annotations · Repository / linked release
DeepFashion2
switchablenorms
Clothing images with detailed annotations for detection, segmentation and consumer-to-shop retrieval.
Images / JSON annotations · External dependencies
DeepFashion MultiModal
yumingj
Fashion images with rich visual and language annotations for controlled multimodal research.
Images / annotation files · External dependencies