ONNX
File size: 1,843 Bytes
23d2d79
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
---
license: gpl-3.0
datasets:
- detection-datasets/coco
---

# Introduction

This repository stores the model for YOLOv5, compatible with Kalray's neural network API. </br>
Please see www.github.com/kalray/kann-model-zoo for details and proper usage [WIKI](https://github.com/kalray/kann-model-zoo/wiki). </br>

# 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]>