Bobo-y/flexible-yolov5
A modular YOLOv5 object detection framework supporting swappable backbones and deployment optimization.
Not currently ranked — collecting fresh signals.
star history
This repository reorganizes YOLOv5 into backbone, neck, and head components to enable flexible experimentation with various architectures including ResNet, HRNet, Swin-Transformer, and MobileNet. It supports plug-in modules like SE, DCN, CBAM, and DropBlock. The project provides TensorRT and Triton inference serving, post-training quantization, and quantization-aware training for production deployment.
Frequently asked
- What is Bobo-y/flexible-yolov5?
- A modular YOLOv5 object detection framework supporting swappable backbones and deployment optimization.
- Is flexible-yolov5 open source?
- Yes — Bobo-y/flexible-yolov5 is open source, released under the GPL-3.0 license.
- What language is flexible-yolov5 written in?
- Bobo-y/flexible-yolov5 is primarily written in Python.
- How popular is flexible-yolov5?
- Bobo-y/flexible-yolov5 has 692 stars on GitHub.
- Where can I find flexible-yolov5?
- Bobo-y/flexible-yolov5 is on GitHub at https://github.com/Bobo-y/flexible-yolov5.