wvangansbeke/Unsupervised-Classification

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1455 stars · 271 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

PyTorch implementation of SCAN (ECCV 2020), a two-step method for unsupervised image classification that combines self-supervised representation learning (e.g., SimCLR) with learnable clustering. It achieved state-of-the-art results on CIFAR-10, CIFAR-100-20, STL-10, and ImageNet image clustering benchmarks.

Use cases

• cluster images into semantic groups without labels
• perform unsupervised image classification on CIFAR or ImageNet
• learn self-supervised visual representations with SimCLR
• reproduce ECCV 2020 SCAN paper results
• benchmark unsupervised clustering on image datasets
• explore representation learning prior to clustering

When to choose

• you need to classify or cluster images without ground-truth labels
• you want a research-grade baseline for unsupervised image classification
• you need self-supervised feature learning combined with clustering

When to avoid

• you need a production-ready, actively maintained library
• you require a permissive license (license is non-standard)
• you need supervised classification with labeled data

Facets

library · maturity maintenance

machine-learning computer-vision image-processing machine-learning computer-vision image-processing deep-learning python unsupervised-learning self-supervised-learning contrastive-learning simclr clustering image-classification eccv-2020 pytorch research-code gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="wvangansbeke/Unsupervised-Classification")