Toybrick

标题: load模型报错 [打印本页]

作者: mogududu 时间: 2021-8-7 15:01
标题: load模型报错
D [set_chip_platform_env:208]evaluate model on RK1808
I [vnn_CreateRKNN:5718]rknn model optimization_level=3
I [setup_client_ops:5215]setup_client_ops num=0
I [setup_client_ops:5340]setup_client_ops finish
I [register_client_ops:5359]register_client_ops num=0
I [register_client_ops:5377]register_client_ops finish
I [setup_nodes:4413]name=nbg_0
I [vnn_CreateRKNN:5844]Norm tensor initialize
D [setup_norm_tensor:4672]tensor id=0, name=Conv_Conv_71/out0_0
D [setup_norm_tensor:4672]tensor id=1, name=Conv_Conv_74/out0_1
D [setup_norm_tensor:4672]tensor id=2, name=Conv_Conv_113/out0_2
D [setup_norm_tensor:4672]tensor id=3, name=Conv_Conv_116/out0_3
D [setup_norm_tensor:4672]tensor id=4, name=data_121
I [vnn_CreateRKNN:5859]Const tensor initialize
I [vnn_CreateRKNN:5874]Virtual tensor initialize
I [vnn_CreateRKNN:5886]Connection initialize
I [vnn_CreateRKNN:5896]graph io initialize
D [setup_node:434]Setup node id[0] uid[0] op[NBG]
D [print_tensor:146]in(0) : id[ 4] vtl[0] const[0] shape[ 456, 256, 3, 1 ] fmt[u8 ] qnt[ASM zp=128, scale=0.003922]
D [print_tensor:146]out(0): id[ 1] vtl[0] const[0] shape[ 57, 32, 38, 1 ] fmt[u8 ] qnt[ASM zp=125, scale=0.007891]
D [print_tensor:146]out(1): id[ 0] vtl[0] const[0] shape[ 57, 32, 19, 1 ] fmt[u8 ] qnt[ASM zp=125, scale=0.007891]
D [print_tensor:146]out(2): id[ 3] vtl[0] const[0] shape[ 57, 32, 38, 1 ] fmt[u8 ] qnt[ASM zp=122, scale=0.007543]
D [print_tensor:146]out(3): id[ 2] vtl[0] const[0] shape[ 57, 32, 19, 1 ] fmt[u8 ] qnt[ASM zp= 5, scale=0.004004]
D [optimize_node:378]Backward optimize neural network
D [optimize_node:385]Forward optimize neural network
I [compute_node:326]Create vx node
D [compute_node:349]Instance node[0] "NBG" ...
E [compute_node:353]Create node[0] NBG fail
I [remove_client_ops:5388]remove_client_ops num=0
I [remove_client_ops:5399]remove_client_ops finish
Create RKNN model fail, error=-6
E Catch exception when init runtime!
E Traceback (most recent call last):
E File "rknn/api/rknn_base.py", line 844, in rknn.api.rknn_base.RKNNBase.init_runtime
E File "rknn/api/rknn_runtime.py", line 340, in rknn.api.rknn_runtime.RKNNRuntime.build_graph
E Exception: RKNN init failed. error code: RKNN_ERR_MODEL_INVALID
Init runtime environment failed

这个报错事什么原因啊？

欢迎光临 Toybrick (https://t.rock-chips.com/)