Yolov3 on rk3566

#1

hii,

i am using rk3566 and while running yolov3 i am getting following error

yolov3.weight exist.
–> Load RKNN model
done

–> Init runtime environment
I RKNN: [08:52:37.638] RKNN Runtime Information: librknnrt version: 1.5.0 (e6fe0c678@2023-05-25T08:09:20)
I RKNN: [08:52:37.638] RKNN Driver Information: version: 0.4.2
I RKNN: [08:52:37.641] RKNN Model Information: version: 2, toolkit version: 1.4.0-22dcfef4(compiler version: 1.4.0 (3b4520e4f@2022-09-05T20:52:35)), target: RKNPU v2, target platform: rk3588, framework name: DarkNet, framework layout: NCHW, model inference type: static_shape
E RKNN: [08:52:37.641] failed to check rknpu hardware version: 0
E RKNN: [08:52:37.641] This rknn model is for RK3588, but current platform is RK3566/RK3568
E RKNN: [08:52:37.641] rknn_init, load model failed!
E Catch exception when init runtime!
E Traceback (most recent call last):
File “/home/rock/.local/lib/python3.9/site-packages/rknnlite/api/rknn_lite.py”, line 148, in init_runtime
self.rknn_runtime.build_graph(self.rknn_data, self.load_model_in_npu)
File “rknnlite/api/rknn_runtime.py”, line 875, in rknnlite.api.rknn_runtime.RKNNRuntime.build_graph
Exception: RKNN init failed. error code: RKNN_ERR_MODEL_INVALID

failed to initialise runtime environment

does it mean we cannot run yolov3 on rk3566 or do i need to change anything in code?

#2

You are using a model for the RK3588 and this will not work on RK3566. You need the RK3566/RK3568!

#3

how to get the yolov3 for rk3566

#4

When you convert your onnx to rknn you choose rk3568 instead of rk3588.

#5

This is a doc for convert yolo onnx model to rknn model for rk3588, even your target is rk3566, but the steps is same, just change the target platform to rk3566