argusswift/YOLOv4-pytorch

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1676 stars · 323 forks observed · 2026-08-28

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

A PyTorch re-implementation of YOLOv4 object detection with variants including attentive YOLOv4 (SEnet, CBAM, CoordAttention) and MobileNetv2/v3 backbones. It supports training and evaluation on PASCAL VOC, COCO, and custom datasets.

Use cases

• train a YOLOv4 object detector in PyTorch
• train YOLOv4 on a custom dataset
• compare attention modules like SEnet and CBAM in a detection backbone
• build a lightweight mobile-friendly detector with MobileNetv3 YOLOv4
• evaluate object detection mAP on PASCAL VOC or COCO
• experiment with focal loss, CIOU, mixup, and multi-scale training

When to choose

• you want a readable, hackable PyTorch YOLOv4 codebase for research or learning
• you need attention-augmented or MobileNet backbone variants of YOLOv4
• you are training on PASCAL VOC, COCO, or small custom datasets

When to avoid

• you need the latest YOLO versions (YOLOv7/v8/v10) or highest accuracy
• you need production-grade inference serving or long-term maintenance
• you require a permissively licensed project, since it uses the Anti-996 license

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision image-processing computer-vision deep-learning machine-learning image-processing python windows yolov4 object-detection pytorch attention-mechanisms mobilenet pascal-voc coco model-training linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="argusswift/YOLOv4-pytorch")