Kalray/yolov5

Introduction

This repository stores the model for YOLOv5, compatible with Kalray's neural network API.
Please see www.github.com/kalray/kann-model-zoo for details and proper usage WIKI.

Contents

ONNX model are described below:

Model	FLOPs	Params	mAP-50	mAP-50/95
yolov5nu.onnx	8.0 G	2.6 M	48.9 %	34.4 %*
yolov5su.onnx	24.4 G	9.1 M	58.9 %	43.1 %*
yolov5su_relu.onnx	24.4 G	9.1 M	54.7 %	39.0 %*
yolov5mu.onnx	64.9 G	25.0 M	65.0 %	49.3 %*
yolov5lu.onnx	136.1 G	53.1 M	68.2%	52.3 %*
yolov5xu.onnx	248.0 G	97.1 M	69.3 %	53.7 %*

Lecture note reference

Repository or links references

• repository: https://github.com/ultralytics/yolov5
• configuration : https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/models/v5/yolov5.yaml
• weights:
  • https://github.com/ultralytics/assets/releases/download/v8.3.0/yolov5nu.pt
  • https://github.com/ultralytics/assets/releases/download/v8.3.0/yolov5su.pt
  • https://github.com/ultralytics/assets/releases/download/v8.3.0/yolov5mu.pt
  • https://github.com/ultralytics/assets/releases/download/v8.3.0/yolov5lu.pt
  • https://github.com/ultralytics/assets/releases/download/v8.3.0/yolov5xu.pt

BibTeX entry and citation info

@software{yolov5,
  title = {Ultralytics YOLOv5},
  author = {Glenn Jocher},
  year = {2020},
  version = {7.0},
  license = {AGPL-3.0},
  url = {https://github.com/ultralytics/yolov5},
  doi = {10.5281/zenodo.3908559},
  orcid = {0000-0001-5950-6979}
}

Authors: Quentin Muller, [email protected]