Fafa-DL/Awesome-Backbones

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1953 stars · 270 forks observed · 2026-08-28

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

A PyTorch-based framework that integrates many deep learning backbone models (CNNs and vision transformers like ResNet, EfficientNet, Swin Transformer) for image classification. It is designed for learning, comparing, and easily modifying backbone architectures with config-driven training and evaluation pipelines.

Use cases

• train an image classifier with a pretrained backbone
• compare CNN vs transformer backbones on my dataset
• fine-tune resnet or swin transformer on custom images
• export a classification model to ONNX
• learn how vision transformer backbones are implemented
• generate class activation maps for a classifier
• train a classifier on a small dataset with configurable augmentation

When to choose

• you want a lightweight, config-driven PyTorch toolbox focused on classification backbones
• you want to experiment with or modify many backbone architectures in one codebase
• you are learning deep learning and want readable model implementations with tutorials

When to avoid

• you need detection, segmentation, or other vision tasks beyond classification
• you need a production-grade framework with long-term commercial support
• you require a permissively licensed project - it currently has no license file

Facets

framework · maturity active

deep-learning machine-learning image-processing deep-learning computer-vision image-processing machine-learning python cross-platform pytorch image-classification backbones cnn vision-transformer resnet swin-transformer model-zoo onnx-export gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="Fafa-DL/Awesome-Backbones")