kuangliu/pytorch-cifar resource

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

6423 stars · 2167 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A PyTorch reference repository for training image classification models on the CIFAR10 dataset, with implementations of many popular architectures like ResNet, DenseNet, and DLA. It serves as a well-known baseline benchmark achieving up to 95.47% accuracy.

Use cases

• train image classifiers on cifar10 with pytorch
• compare accuracy of different cnn architectures
• learn how to implement resnet in pytorch
• get a baseline for image classification experiments
• find reference pytorch model implementations

When to choose

• you want clean, minimal PyTorch implementations of common CNN architectures
• you need CIFAR10 accuracy baselines for research comparisons
• you are learning PyTorch image classification

When to avoid

• you need production-grade training pipelines with distributed training or experiment tracking
• you need datasets other than CIFAR10
• you need the latest training techniques like mixup or modern optimizers

Facets

learning-resource · maturity maintenance

machine-learning deep-learning benchmarking deep-learning computer-vision image-processing tutorials python cross-platform pytorch cifar10 image-classification model-zoo reference-implementation gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="kuangliu/pytorch-cifar")