Improve model card and surface related exports
Browse files
README.md
CHANGED
|
@@ -10,19 +10,24 @@ tags:
|
|
| 10 |
|
| 11 |
# YOLO26-RKNN2
|
| 12 |
|
| 13 |
-
|
| 14 |
|
| 15 |
-
Available tasks:
|
| 16 |
|
| 17 |
-
|
| 18 |
|
| 19 |
-
|
| 20 |
|
| 21 |
-
|
|
|
|
|
|
|
| 22 |
|
|
|
|
| 23 |
|
| 24 |
-
|
| 25 |
|
| 26 |
-
The
|
| 27 |
|
| 28 |
-
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
# YOLO26-RKNN2
|
| 12 |
|
| 13 |
+
> **Looking for the Ultralytics RKNN exports?** See [YOLO26-RKNN2-Ultralytics](https://huggingface.co/HanzoHuang/YOLO26-RKNN2-Ultralytics).
|
| 14 |
|
| 15 |
+
YOLO26 RKNN exports for **RK3576** and **RK3588**, compiled with RKNN Toolkit2 2.3.2. Available tasks: detection, classification, depth, oriented bounding boxes, pose, instance segmentation, and semantic segmentation.
|
| 16 |
|
| 17 |
+
## Choose a model
|
| 18 |
|
| 19 |
+
Models are organized by toolkit version, task, variant, target chip, and precision:
|
| 20 |
|
| 21 |
+
```text
|
| 22 |
+
models/v<toolkit-version>/<task>/<variant>/<chip>/<precision>/
|
| 23 |
+
```
|
| 24 |
|
| 25 |
+
Download `model.rknn` and `metadata.yaml` from the same directory. The [manifest](manifest.json) lists the available models and checksums.
|
| 26 |
|
| 27 |
+
## Conversion
|
| 28 |
|
| 29 |
+
Detection models were converted on Seeed-Dev with RKNN Toolkit2 2.3.2 using ONNX inputs supplied by the Rockchip Model Zoo. The [manifest](manifest.json) records input ONNX and output RKNN checksums. The ONNX graphs and Model Zoo download URLs are in [source/onnx](tree/main/source/onnx); the graphs declare their Ultralytics exporter versions and AGPL-3.0 license.
|
| 30 |
|
| 31 |
+
## Source and license
|
| 32 |
+
|
| 33 |
+
Original checkpoints, conversion scripts, and provenance are in [source](tree/main/source). The models and checkpoints are distributed under AGPL-3.0; see [LICENSE](LICENSE) and [checkpoint details](source/weights.json).
|