MAC系统下，1808计算棒，按照quick start手册运行示例tflite出错

[复制链接] · 发表于 2021-1-6 17:09:12

按照手册一步步操作下来的，执行示例程序时出错，信息如下：

python3.6 test.py

--> config model

done

--> Loading model

done

--> Building model

W The target_platform is not set in config, using default target platform rk1808.

W The channel_mean_value filed will not be used in the future!

done

--> Export RKNN model

done

--> Init runtime environment

E NPUTransfer: Cannot connect to proxy: Permission denied

E RKNNAPI: rknn_init, driver open fail! ret = -4(ERROR_NO_DEVICE)!

E Catch exception when init runtime!

E Traceback (most recent call last):

E File "rknn/api/rknn_base.py", line 1154, in rknn.api.rknn_base.RKNNBase.init_runtime

E File "rknn/api/rknn_runtime.py", line 356, in rknn.api.rknn_runtime.RKNNRuntime.build_graph

E Exception: RKNN init failed. error code: RKNN_ERR_DEVICE_UNAVAILABLE

E Current device id is: None

E Devices connected:

E ['TS018083201101749']

Init runtime environment failed

只看该作者 · 发表于 2021-1-6 17:34:26

MAC不支持模拟器仿真，请填一个计算棒的device id在init runtime时候

只看该作者 · 发表于 2021-1-6 17:42:10

计算棒设备信息如下：
./npu_transfer_proxy devices
List of ntb devices attached
TS018083201101749 422a7a1a USB_DEVICE

我该怎么填？
ret = rknn.init_runtime(target='TS018083201101749') ?
ret = rknn.init_runtime('TS018083201101749') ?
ret = rknn.init_runtime(target='422a7a1a') ?
ret = rknn.init_runtime('422a7a1a')

只看该作者 · 发表于 2021-1-6 19:40:01

我这样修改了，还是报上面一样的错误

# init runtime environment
print('--> Init runtime environment')
ret = rknn.init_runtime(target='rk1808',device_id='TS018083201101749')
if ret != 0:
print('Init runtime environment failed')
exit(ret)
print('done')

只看该作者 · 发表于 2021-1-7 08:39:41

关键是这个错误，我的是rknn-toolkit-v1.4.0，直接从百度网盘下载的
E NPUTransfer: Cannot connect to proxy: Permission denied

只看该作者 · 发表于 2021-1-7 08:55:58

nj1tony 发表于 2021-1-7 08:39
关键是这个错误，我的是rknn-toolkit-v1.4.0，直接从百度网盘下载的
E NPUTransfer: Cannot connect to pro ...

sudo 运行试试，我也是mac，挺正常的跑的，还没遇到过你的问题。

只看该作者 · 发表于 2021-1-7 08:58:57

jefferyzhang 发表于 2021-1-7 08:55
sudo 运行试试，我也是mac，挺正常的跑的，还没遇到过你的问题。

谢谢版主！
sudo也试过，不行。
刚刚重启了一下电脑，可以执行了。应该是安装环境及插拔计算棒时，proxy的address被占用了

python3 test.py
--> config model
done
--> Loading model
done
--> Building model
W The channel_mean_value filed will not be used in the future!
done
--> Export RKNN model
done
--> Init runtime environment
D RKNNAPI: ==============================================
D RKNNAPI: RKNN VERSION:
D RKNNAPI: API: 1.4.0 (b4a8096 build: 2020-08-12 10:18:51)
D RKNNAPI: DRV: 1.4.0 (b4a8096 build: 2020-09-14 11:15:57)
D RKNNAPI: ==============================================
done
--> Running model
mobilenet_v1
-----TOP 5-----
[156]: 0.85107421875
[155]: 0.09173583984375
[205]: 0.01358795166015625
[284]: 0.006465911865234375
[194]: 0.002239227294921875

done
--> Begin evaluate model performance
W When performing performance evaluation, inputs can be set to None to use fake inputs.
========================================================================
Performance
========================================================================
Total Time(us): 6940
FPS: 144.09
========================================================================