标题: ---rknn 生成识别模型我遇到这样的问题, 麻烦问一下 [打印本页] 作者: linuxdog 时间: 2022-1-20 13:14 标题: ---rknn 生成识别模型我遇到这样的问题, 麻烦问一下 通过: cd ../rknn-toolkit2-1.0.0_no_docker/examples/tflite/mobilenet_v1,
python3 test.py命令生成模型成功没有问题, 现在是分别用dog_224x224.jpg和
cat_224x224.jpg生成模型, 为啥回逆的Top5打印不同啊！ 是否说太低表示失败了, 有了解帮助一下，刚刚起步了
1. dog图片的情况:
--> config model
done
--> Loading model
done
--> Building model
Analysing : 100%|██████████████████████████████████████████████████| 58/58 [00:00<00:00, 196.76it/s]
Quantizating : 100%|███████████████████████████████████████████████| 58/58 [00:00<00:00, 164.75it/s]
I RKNN: librknnc version: 1.0.0 (454a7218@2021-04-30T16:32:25)
I RKNN: set log level to 0
done
--> Export RKNN model
done
--> Init runtime environment
W init_runtime: target is None, use simulator!
done
--> Running model
I RKNN: librknnc version: 1.0.0 (454a7218@2021-04-30T16:32:25)
I RKNN: set log level to 0
mobilenet_v1
-----TOP 5-----
[156]: 0.84228515625
[155]: 0.088134765625
[205]: 0.01418304443359375
[284]: 0.0082855224609375
[194 260]: 0.0028285980224609375
2. cat图片的情况
--> config model
done
--> Loading model
done
--> Building model
Analysing : 100%|██████████████████████████████████████████████████| 58/58 [00:00<00:00, 196.25it/s]
Quantizating : 100%|████████████████████████████████████████████████| 58/58 [00:00<00:00, 60.71it/s]
I RKNN: librknnc version: 1.0.0 (454a7218@2021-04-30T16:32:25)
I RKNN: set log level to 0
done
--> Export RKNN model
done
--> Init runtime environment
W init_runtime: target is None, use simulator!
done
--> Running model
I RKNN: librknnc version: 1.0.0 (454a7218@2021-04-30T16:32:25)
I RKNN: set log level to 0
mobilenet_v1
-----TOP 5-----
[282]: 0.224365234375
[464]: 0.20849609375
[286]: 0.1802978515625
[283]: 0.1448974609375
[357]: 0.0419921875