HanzoHuang commited on
Commit
9b33ad5
·
verified ·
1 Parent(s): 9921898

Improve model card and surface related exports

Browse files
Files changed (1) hide show
  1. README.md +13 -8
README.md CHANGED
@@ -10,19 +10,24 @@ tags:
10
 
11
  # YOLO26-RKNN2
12
 
13
- YOLO exports for RK3576 and RK3588. All models were compiled with RKNN Toolkit2 2.3.2.
14
 
15
- Available tasks: detect, classify, depth, obb, pose, segment, semantic.
16
 
17
- Download `model.rknn` and `metadata.yaml` from the same folder under `models/<toolkit-version>/<task>/<variant>/<chip>/<precision>/`.
18
 
19
- The manifest records model and metadata checksums. Original checkpoints, conversion scripts, and their provenance are in `source/`.
20
 
21
- The models and checkpoints are distributed under AGPL-3.0. See `LICENSE` and `source/weights.json`.
 
 
22
 
 
23
 
24
- We converted the detection models on Seeed-Dev using RKNN Toolkit2 2.3.2. Rockchip Model Zoo supplied the ONNX inputs. Each detection entry in the manifest records the input ONNX and output RKNN checksums.
25
 
26
- The detection ONNX graphs and their Model Zoo download URLs are recorded in `source/onnx/` and `manifest.json`. These graphs declare the AGPL-3.0 license and their Ultralytics exporter versions.
27
 
28
- Models are organized by RKNN Toolkit2 version under `models/v<version>/`. Each model and its metadata share a folder.
 
 
 
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).