Add YOLOv8m model card
Browse files
README.md
CHANGED
|
@@ -1,3 +1,35 @@
|
|
| 1 |
---
|
| 2 |
license: agpl-3.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: agpl-3.0
|
| 3 |
+
library_name: onnx
|
| 4 |
+
pipeline_tag: object-detection
|
| 5 |
+
tags:
|
| 6 |
+
- yolov8
|
| 7 |
+
- onnx
|
| 8 |
+
- object-detection
|
| 9 |
+
- coco
|
| 10 |
+
- aidge
|
| 11 |
---
|
| 12 |
+
|
| 13 |
+
# YOLOv8m ONNX
|
| 14 |
+
|
| 15 |
+
COCO-pretrained YOLOv8m model exported from Ultralytics to ONNX for Aidge validation.
|
| 16 |
+
|
| 17 |
+
## Model
|
| 18 |
+
|
| 19 |
+
- Source model: Ultralytics YOLOv8m
|
| 20 |
+
- Task: object detection
|
| 21 |
+
- Dataset: COCO
|
| 22 |
+
- Export format: ONNX
|
| 23 |
+
- ONNX opset: 17
|
| 24 |
+
- Input: `images`, shape `[1, 3, 640, 640]`
|
| 25 |
+
- Output: `output0`, shape `[1, 84, 8400]`
|
| 26 |
+
|
| 27 |
+
## Usage in Aidge
|
| 28 |
+
|
| 29 |
+
This model is used to validate YOLOv8 CPU import and inference in Aidge through the ONNX import path.
|
| 30 |
+
|
| 31 |
+
The static `640x640` export is used for model-zoo testing because it has fixed input and output shapes.
|
| 32 |
+
|
| 33 |
+
## License
|
| 34 |
+
|
| 35 |
+
The original YOLOv8 model is provided by Ultralytics under the AGPL-3.0 license.
|