I was trying to follow this tutorial : rknn_model_zoo/examples/yolov10 at main · airockchip/rknn_model_zoo (github.com)
to convert a yolov10 model to rknn to then deploy it on a rock 5a.
first i converted the .pt model to .onnx running : !yolo export model=model format=onnx opset=13 simplify
then after trying to convert from onnx to rknn i git the following
W init : rknn-toolkit2 version: 1.5.0+1fa95b5c
–> Config model
done
–> Loading model
W load_onnx: If you don’t need to crop the model, don’t set ‘inputs’/‘input_size_list’/‘outputs’!
W load_onnx: Onnx opset13 is not fully supported, it may cause convert fail, it is recommended to use opset12!
Loading : 100%|█████████████████████████████████████████████████| 343/343 [00:00<00:00, 8585.19it/s]
done
–> Building model
I base_optimize …
I base_optimize done.
I
I fold_constant …
E build: The input 0 of TopK(’/model.23/TopK’) need to be constant! It will cause the graph to be a dynamic graph!
Remove TopK(’/model.23/TopK’) manually and try again!
W build: ===================== WARN(3) =====================
E rknn-toolkit2 version: 1.5.0+1fa95b5c
E build: Catch exception when building RKNN model!
E build: Traceback (most recent call last):
E build: File “rknn/api/rknn_base.py”, line 1789, in rknn.api.rknn_base.RKNNBase.build
E build: File “rknn/api/graph_optimizer.py”, line 727, in rknn.api.graph_optimizer.GraphOptimizer.fold_constant
E build: File “rknn/api/graph_optimizer.py”, line 331, in rknn.api.graph_optimizer._dynamic_check
E build: File “rknn/api/rknn_log.py”, line 112, in rknn.api.rknn_log.RKNNLog.e
E build: ValueError: The input 0 of TopK(’/model.23/TopK’) need to be constant! It will cause the graph to be a dynamic graph!
E build: Remove TopK(’/model.23/TopK’) manually and try again!
Build model failed!