sthalles/SimCLR

Health v2 · maintenance only

Adoption not part of the score

2491 stars · 495 forks observed · 2026-08-28

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

A PyTorch reference implementation of SimCLR, a self-supervised contrastive learning framework for learning visual representations from unlabeled images. It provides configurable training scripts (including mixed-precision AMP support) and a linear-probe evaluation protocol for measuring representation quality on benchmarks like STL10 and CIFAR10.

Use cases

• train a SimCLR contrastive learning model in PyTorch
• self-supervised pretraining on unlabeled images
• learn visual representations without labeled data
• reproduce SimCLR paper results on STL10 and CIFAR10
• evaluate learned image features with a linear classifier
• pretrain a ResNet encoder for downstream transfer learning

When to choose

• You want a compact, readable PyTorch implementation of SimCLR to study, run, or extend
• You need to pretrain image encoders on unlabeled data using contrastive learning
• You want a reproducible linear-evaluation setup for comparing representation quality

When to avoid

• You need large-scale distributed training or support for many datasets and architectures beyond small benchmarks
• You want a broader, actively extended self-supervised learning framework covering newer methods
• You need production-ready supervised training pipelines or off-the-shelf pretrained models for deployment

Facets

library · maturity stable

deep-learning machine-learning computer-vision image-processing machine-learning deep-learning computer-vision image-processing python cli simclr contrastive-learning self-supervised-learning representation-learning unsupervised-learning pytorch torchvision resnet transfer-learning linear-evaluation gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="sthalles/SimCLR")