rknn模型在香橙派5B运行报错

前文帖子：https://t.rock-chips.com/forum.p ... tid=4677&extra=
根据上贴中网友的回复，①更新rknn为最新版本；②测试仿真结果是否正常推理；

在虚拟机中，重建环境，使用rknn-toolkit 1.6版本；


在虚拟机中运行仿真测试程序(simulator_LDA_RKNN_Lee20240126.py)，程序连接：https://gitee.com/collisionandconflict/rknn-communication
测试仿真结果输出，如下图：


1、将生成的rknn模型：LDA_rknn1.6_Lee20240126.rknn
导入香橙派。
2、更新香橙派中的rknn版本为：rknn-toolkit-lite2 1.60：
3、运行rknn调用程序(inference_LDA_rknn_Lee20240124.py)，程序连接：https://gitee.com/collisionandconflict/rknn-communication

运行报错：
  warnings.warn(
--> Init runtime environment
W Verbose file path is invalid, debug info will not dump to file.
D target set by user is: None
D Starting ntp or adb, target soc is RK3588, device id is: None
I RKNN: [16:31:12.501] RKNN Runtime Information: librknnrt version: 1.4.0 (a10f100eb@2022-09-09T09:07:14)
I RKNN: [16:31:12.501] RKNN Driver Information: version: 0.8.2
Init runtime environment failed
E RKNN: [16:31:12.501] 6, 1
E RKNN: [16:31:12.501] Invalid RKNN model version 6
E RKNN: [16:31:12.501] rknn_init, load model failed!
E Catch exception when init runtime!
E Traceback (most recent call last):
  File "/home/orangepi/anaconda3/envs/LDA_rknn_ENV/lib/python3.8/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 919, in rknnlite.api.rknn_runtime.RKNNRuntime.build_graph
Exception: RKNN init failed. error code: RKNN_ERR_FAIL




求帮助，如何解决。

按照下述博客操作，Bug解决。
https://blog.csdn.net/qq_36497369/article/details/134883399