LuckfoxPicoMax运行RKNN-Toolkit2中的Yolov5adbUSB仿真

1：下载rknn-toolkit2

git clone github /rockchip-linux/rknn-toolkit2

2：修改onnx目录下的yolov5的test.py的代码

 # pre-process config     print('--> Config model')     rknn.config(mean_values=[[0, 0, 0]], std_values=[[255, 255, 255]], target_platform='rv1106')  #target_platform='rk3566')

# Init runtime environment     print('--> Init runtime environment')     # ret = rknn.init_runtime()     ret = rknn.init_runtime(target='rv1106', device_id= 'bd547ee6900c058b')

3：adb push rknn_sever和依赖库

RV1103/RV1106上使用的RKNPU Runtime库是librknnmrt.so，使用32-bit的rknn_server,启动步骤如下：（armhf-uclibc）

adb push RV1106/Linux/rknn_server/armhf-uclibc/usr/bin下的所有文件到/oem/usr/bin目录adb push RV1106/Linux/librknn_api/armhf-uclibc/librknnmrt.so到/oem/usr/lib目录进入板子的串口终端，执行： chmod +x /oem/usr/bin/rknn_server chmod +x /oem/usr/bin/start_rknn.sh chmod +x /oem/usr/bin/restart_rknn.sh restart_rknn.sh

4：运行yolov5的python代码进行adb连接仿真

(RKNN-Toolkit2) ubuntu@ubuntu:~/Downloads/rknn-toolkit2-master/rknn-toolkit2/examples/onnx/yolov5$ python3 test.py W __init__: rknn-toolkit2 version: 1.6.0+81f21f4d --> Config model done --> Loading model W load_onnx: It is recommended onnx opset 19, but your onnx model opset is 12! W load_onnx: Model converted from pytorch, 'opset_version' should be set 19 in torch.onnx.export for successful convert! Loading : 100%|█████████████████████████████████████████████████| 125/125 [00:00<00:00, 6242.27it/s] done --> Building model I base_optimize ... I base_optimize done. I I fold_constant ... I fold_constant done. I I correct_ops ... I correct_ops done. I I fuse_ops ... I fuse_ops done. I I sparse_weight ... I sparse_weight done. I GraphPreparing : 100%|██████████████████████████████████████████| 149/149 [00:00<00:00, 1538.03it/s] Quantizating : 100%|██████████████████████████████████████████████| 149/149 [00:01<00:00, 85.94it/s] I I quant_optimizer ... I quant_optimizer results: I adjust_tanh_sigmoid: ['Sigmoid_146', 'Sigmoid_148', 'Sigmoid_150'] I adjust_relu: ['Relu_144', 'Relu_141', 'Relu_139', 'Relu_137', 'Relu_135', 'Relu_132', 'Relu_130', 'Relu_127', 'Relu_125', 'Relu_123', 'Relu_121', 'Relu_118', 'Relu_116', 'Relu_113', 'Relu_111', 'Relu_109', 'Relu_107', 'Relu_102', 'Relu_100', 'Relu_97', 'Relu_95', 'Relu_93', 'Relu_91', 'Relu_86', 'Relu_84', 'Relu_75', 'Relu_73', 'Relu_70', 'Relu_67', 'Relu_65', 'Relu_63', 'Relu_61', 'Relu_59', 'Relu_56', 'Relu_53', 'Relu_51', 'Relu_48', 'Relu_46', 'Relu_43', 'Relu_41', 'Relu_39', 'Relu_37', 'Relu_35', 'Relu_32', 'Relu_29', 'Relu_27', 'Relu_24', 'Relu_22', 'Relu_20', 'Relu_18', 'Relu_16', 'Relu_13', 'Relu_10', 'Relu_8', 'Relu_6', 'Relu_4', 'Relu_2'] I adjust_no_change_node: ['MaxPool_81', 'MaxPool_80'] I quant_optimizer done. I W build: The default input dtype of 'images' is changed from 'float32' to 'int8' in rknn model for performance! Please take care of this change when deploy rknn model with Runtime API! W build: The default output dtype of 'output' is changed from 'float32' to 'int8' in rknn model for performance! Please take care of this change when deploy rknn model with Runtime API! W build: The default output dtype of '283' is changed from 'float32' to 'int8' in rknn model for performance! Please take care of this change when deploy rknn model with Runtime API! W build: The default output dtype of '285' is changed from 'float32' to 'int8' in rknn model for performance! Please take care of this change when deploy rknn model with Runtime API! I rknn building ... I RKNN: [11:24:23.838] compress = 0, conv_eltwise_activation_fuse = 1, global_fuse = 1, multi-core-model-mode = 7, output_optimize = 1,enable_argb_group=0 ,layout_match = 1, pipeline_fuse = 0 I RKNN: librknnc version: 1.6.0 (585b3edcf@2023-12-11T07:42:56) D RKNN: [11:24:24.052] RKNN is invoked W RKNN: [11:24:24.721] Model initializer tensor data is empty, name: 219 W RKNN: [11:24:24.721] Model initializer tensor data is empty, name: 238 D RKNN: [11:24:24.748] >>>>>> start: rknn::RKNNExtractCustomOpAttrs D RKNN: [11:24:24.749] <<<<<<<< end: rknn::RKNNExtractCustomOpAttrs D RKNN: [11:24:24.749] >>>>>> start: rknn::RKNNSetOpTargetPass D RKNN: [11:24:24.749] <<<<<<<< end: rknn::RKNNSetOpTargetPass D RKNN: [11:24:24.749] >>>>>> start: rknn::RKNNBindNorm D RKNN: [11:24:24.750] <<<<<<<< end: rknn::RKNNBindNorm D RKNN: [11:24:24.750] >>>>>> start: rknn::RKNNAddFirstConv D RKNN: [11:24:24.751] <<<<<<<< end: rknn::RKNNAddFirstConv D RKNN: [11:24:24.751] >>>>>> start: rknn::RKNNEliminateQATDataConvert D RKNN: [11:24:24.752] <<<<<<<< end: rknn::RKNNEliminateQATDataConvert D RKNN: [11:24:24.752] >>>>>> start: rknn::RKNNTileGroupConv D RKNN: [11:24:24.752] <<<<<<<< end: rknn::RKNNTileGroupConv D RKNN: [11:24:24.752] >>>>>> start: rknn::RKNNTileFcBatchFuse D RKNN: [11:24:24.752] <<<<<<<< end: rknn::RKNNTileFcBatchFuse D RKNN: [11:24:24.752] >>>>>> start: rknn::RKNNAddConvBias D RKNN: [11:24:24.754] <<<<<<<< end: rknn::RKNNAddConvBias D RKNN: [11:24:24.754] >>>>>> start: rknn::RKNNTileChannel D RKNN: [11:24:24.754] <<<<<<<< end: rknn::RKNNTileChannel D RKNN: [11:24:24.754] >>>>>> start: rknn::RKNNPerChannelPrep D RKNN: [11:24:24.754] <<<<<<<< end: rknn::RKNNPerChannelPrep D RKNN: [11:24:24.754] >>>>>> start: rknn::RKNNBnQuant D RKNN: [11:24:24.754] <<<<<<<< end: rknn::RKNNBnQuant D RKNN: [11:24:24.754] >>>>>> start: rknn::RKNNFuseOptimizerPass D RKNN: [11:24:24.866] <<<<<<<< end: rknn::RKNNFuseOptimizerPass D RKNN: [11:24:24.866] >>>>>> start: rknn::RKNNTurnAutoPad D RKNN: [11:24:24.866] <<<<<<<< end: rknn::RKNNTurnAutoPad D RKNN: [11:24:24.866] >>>>>> start: rknn::RKNNInitRNNConst D RKNN: [11:24:24.866] <<<<<<<< end: rknn::RKNNInitRNNConst D RKNN: [11:24:24.866] >>>>>> start: rknn::RKNNInitCastConst D RKNN: [11:24:24.866] <<<<<<<< end: rknn::RKNNInitCastConst D RKNN: [11:24:24.866] >>>>>> start: rknn::RKNNMultiSurfacePass D RKNN: [11:24:24.866] <<<<<<<< end: rknn::RKNNMultiSurfacePass D RKNN: [11:24:24.866] >>>>>> start: rknn::RKNNReplaceConstantTensorPass D RKNN: [11:24:24.867] <<<<<<<< end: rknn::RKNNReplaceConstantTensorPass D RKNN: [11:24:24.867] >>>>>> start: OpEmit D RKNN: [11:24:24.867] <<<<<<<< end: OpEmit D RKNN: [11:24:24.867] >>>>>> start: rknn::RKNNLayoutMatchPass I RKNN: [11:24:24.867] AppointLayout: t->setNativeLayout(64), tname:[128] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[131] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[133] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[142] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[137] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[140] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[143] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[145] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[147] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[161] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[151] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[156] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[159] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[162] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[166] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[185] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[170] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[175] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[180] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[183] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[186] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[190] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[199] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[194] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[197] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[200] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[202] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[204] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[205] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[206] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[207] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[208] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[209] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[210] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[211] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[213] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[221] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[223] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[229] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[225] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[227] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[230] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[232] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[240] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[242] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[248] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[244] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[246] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[249] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[251] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[253] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[254] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[256] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[262] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[258] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[260] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[263] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[265] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[267] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[268] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[270] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[276] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[272] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[274] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[277] I RKNN: [11:24:24.868] AppointLayout: t->setNativeLayout(64), tname:[279] D RKNN: [11:24:24.868] <<<<<<<< end: rknn::RKNNLayoutMatchPass D RKNN: [11:24:24.868] >>>>>> start: rknn::RKNNAddSecondaryNode D RKNN: [11:24:24.869] <<<<<<<< end: rknn::RKNNAddSecondaryNode D RKNN: [11:24:24.869] >>>>>> start: OpEmit D RKNN: [11:24:24.883] <<<<<<<< end: OpEmit D RKNN: [11:24:24.883] >>>>>> start: rknn::RKNNProfileAnalysisPass D RKNN: [11:24:24.884] <<<<<<<< end: rknn::RKNNProfileAnalysisPass D RKNN: [11:24:24.888] >>>>>> start: rknn::RKNNOperatorIdGenPass D RKNN: [11:24:24.888] <<<<<<<< end: rknn::RKNNOperatorIdGenPass D RKNN: [11:24:24.888] >>>>>> start: rknn::RKNNWeightTransposePass W RKNN: [11:24:25.219] Warning: Tensor 289 need paramter qtype, type is set to float16 by default! W RKNN: [11:24:25.219] Warning: Tensor 219 need paramter qtype, type is set to float16 by default! W RKNN: [11:24:25.220] Warning: Tensor 290 need paramter qtype, type is set to float16 by default! W RKNN: [11:24:25.220] Warning: Tensor 238 need paramter qtype, type is set to float16 by default! D RKNN: [11:24:25.220] <<<<<<<< end: rknn::RKNNWeightTransposePass D RKNN: [11:24:25.220] >>>>>> start: rknn::RKNNCPUWeightTransposePass D RKNN: [11:24:25.220] <<<<<<<< end: rknn::RKNNCPUWeightTransposePass D RKNN: [11:24:25.220] >>>>>> start: rknn::RKNNModelBuildPass D RKNN: [11:24:25.844] RKNNModelBuildPass: [Statistics] D RKNN: [11:24:25.844] total_regcfg_size : 266816 D RKNN: [11:24:25.844] total_diff_regcfg_size: 164280 D RKNN: [11:24:25.844] <<<<<<<< end: rknn::RKNNModelBuildPass D RKNN: [11:24:25.844] >>>>>> start: rknn::RKNNModelRegCmdbuildPass D RKNN: [11:24:25.846] ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ D RKNN: [11:24:25.846] Network Layer Information Table D RKNN: [11:24:25.846] ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ D RKNN: [11:24:25.846] ID OpType DataType Target InputShape OutputShape DDRCycles NPUCycles MaxCycles TaskNumber RW(KB) FullName D RKNN: [11:24:25.846] ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ D RKNN: [11:24:25.846] 0 InputOperator INT8 CPU \ (1,3,640,640) 0 0 0 0/0 0 InputOperator:images D RKNN: [11:24:25.846] 1 Conv INT8 NPU (1,3,640,640),(12,3,2,2),(12) (1,12,320,320) 465543 409600 465543 8/0 1200 Conv:Conv_0 D RKNN: [11:24:25.846] 2 ConvRelu INT8 NPU (1,12,320,320),(32,12,3,3),(32) (1,32,320,320) 798774 1843200 1843200 8/0 1604 Conv:Conv_1 D RKNN: [11:24:25.846] 3 ConvRelu INT8 NPU (1,32,320,320),(64,32,3,3),(64) (1,64,160,160) 801060 921600 921600 16/0 3218 Conv:Conv_3 D RKNN: [11:24:25.846] 4 ConvRelu INT8 NPU (1,64,160,160),(32,64,1,1),(32) (1,32,160,160) 399366 204800 399366 8/0 1602 Conv:Conv_5 D RKNN: [11:24:25.846] 5 ConvRelu INT8 NPU (1,64,160,160),(32,64,1,1),(32) (1,32,160,160) 399366 204800 399366 8/0 1602 Conv:Conv_12 D RKNN: [11:24:25.846] 6 ConvRelu INT8 NPU (1,32,160,160),(32,32,1,1),(32) (1,32,160,160) 266203 204800 266203 4/0 801 Conv:Conv_7 D RKNN: [11:24:25.846] 7 ConvReluAdd INT8 NPU (1,32,160,160),(32,32,3,3),(32),... (1,32,160,160) 400530 460800 460800 4/0 1609 Conv:Conv_9 D RKNN: [11:24:25.846] 8 Concat INT8 NPU (1,32,160,160),(1,32,160,160) (1,64,160,160) 531990 0 531990 2/0 1600 Concat:Concat_14 D RKNN: [11:24:25.846] 9 ConvRelu INT8 NPU (1,64,160,160),(64,64,1,1),(64) (1,64,160,160) 532738 409600 532738 8/0 1604 Conv:Conv_15 D RKNN: [11:24:25.846] 10 ConvRelu INT8 NPU (1,64,160,160),(128,64,3,3),(128) (1,128,80,80) 411128 921600 921600 9/0 1673 Conv:Conv_17 D RKNN: [11:24:25.846] 11 ConvRelu INT8 NPU (1,128,80,80),(64,128,1,1),(64) (1,64,80,80) 200910 102400 200910 4/0 808 Conv:Conv_19 D RKNN: [11:24:25.846] 12 ConvRelu INT8 NPU (1,128,80,80),(64,128,1,1),(64) (1,64,80,80) 200910 102400 200910 4/0 808 Conv:Conv_31 D RKNN: [11:24:25.846] 13 ConvRelu INT8 NPU (1,64,80,80),(64,64,1,1),(64) (1,64,80,80) 133746 102400 133746 2/0 404 Conv:Conv_21 D RKNN: [11:24:25.846] 14 ConvReluAdd INT8 NPU (1,64,80,80),(64,64,3,3),(64),... (1,64,80,80) 205564 460800 460800 3/0 836 Conv:Conv_23 D RKNN: [11:24:25.846] 15 ConvRelu INT8 NPU (1,64,80,80),(64,64,1,1),(64) (1,64,80,80) 133746 102400 133746 2/0 404 Conv:Conv_26 D RKNN: [11:24:25.846] 16 ConvReluAdd INT8 NPU (1,64,80,80),(64,64,3,3),(64),... (1,64,80,80) 205564 460800 460800 3/0 836 Conv:Conv_28 D RKNN: [11:24:25.846] 17 Concat INT8 NPU (1,64,80,80),(1,64,80,80) (1,128,80,80) 265995 0 265995 2/0 800 Concat:Concat_33 D RKNN: [11:24:25.846] 18 ConvRelu INT8 NPU (1,128,80,80),(128,128,1,1),(128) (1,128,80,80) 268821 204800 268821 4/0 817 Conv:Conv_34 D RKNN: [11:24:25.846] 19 ConvRelu INT8 NPU (1,128,80,80),(256,128,3,3),(256) (1,256,40,40) 247708 921600 921600 5/0 1090 Conv:Conv_36 D RKNN: [11:24:25.846] 20 ConvRelu INT8 NPU (1,256,40,40),(128,256,1,1),(128) (1,128,40,40) 105235 102400 105235 2/0 433 Conv:Conv_38 D RKNN: [11:24:25.846] 21 ConvRelu INT8 NPU (1,256,40,40),(128,256,1,1),(128) (1,128,40,40) 105235 102400 105235 2/0 433 Conv:Conv_55 D RKNN: [11:24:25.846] 22 ConvRelu INT8 NPU (1,128,40,40),(128,128,1,1),(128) (1,128,40,40) 69325 51200 69325 1/0 217 Conv:Conv_40 D RKNN: [11:24:25.846] 23 ConvReluAdd INT8 NPU (1,128,40,40),(128,128,3,3),(128),... (1,128,40,40) 123854 460800 460800 2/0 545 Conv:Conv_42 D RKNN: [11:24:25.846] 24 ConvRelu INT8 NPU (1,128,40,40),(128,128,1,1),(128) (1,128,40,40) 69325 51200 69325 1/0 217 Conv:Conv_45 D RKNN: [11:24:25.846] 25 ConvReluAdd INT8 NPU (1,128,40,40),(128,128,3,3),(128),... (1,128,40,40) 123854 460800 460800 2/0 545 Conv:Conv_47 D RKNN: [11:24:25.846] 26 ConvRelu INT8 NPU (1,128,40,40),(128,128,1,1),(128) (1,128,40,40) 69325 51200 69325 1/0 217 Conv:Conv_50 D RKNN: [11:24:25.846] 27 ConvReluAdd INT8 NPU (1,128,40,40),(128,128,3,3),(128),... (1,128,40,40) 123854 460800 460800 2/0 545 Conv:Conv_52 D RKNN: [11:24:25.846] 28 Concat INT8 NPU (1,128,40,40),(1,128,40,40) (1,256,40,40) 132998 0 132998 2/0 400 Concat:Concat_57 D RKNN: [11:24:25.846] 29 ConvRelu INT8 NPU (1,256,40,40),(256,256,1,1),(256) (1,256,40,40) 143970 204800 204800 3/0 466 Conv:Conv_58 D RKNN: [11:24:25.846] 30 ConvRelu INT8 NPU (1,256,40,40),(512,256,3,3),(512) (1,512,20,20) 291930 921600 921600 3/0 1556 Conv:Conv_60 D RKNN: [11:24:25.846] 31 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_62 D RKNN: [11:24:25.846] 32 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_69 D RKNN: [11:24:25.846] 33 ConvRelu INT8 NPU (1,256,20,20),(256,256,1,1),(256) (1,256,20,20) 44222 51200 51200 1/0 166 Conv:Conv_64 D RKNN: [11:24:25.846] 34 ConvReluAdd INT8 NPU (1,256,20,20),(256,256,3,3),(256),... (1,256,20,20) 145965 460800 460800 1/0 778 Conv:Conv_66 D RKNN: [11:24:25.846] 35 Concat INT8 NPU (1,256,20,20),(1,256,20,20) (1,512,20,20) 66499 0 66499 2/0 200 Concat:Concat_71 D RKNN: [11:24:25.846] 36 ConvRelu INT8 NPU (1,512,20,20),(512,512,1,1),(512) (1,512,20,20) 109723 204800 204800 2/0 460 Conv:Conv_72 D RKNN: [11:24:25.846] 37 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_74 D RKNN: [11:24:25.846] 38 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_76 D RKNN: [11:24:25.846] 39 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_77 D RKNN: [11:24:25.846] 40 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_78 D RKNN: [11:24:25.846] 41 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_79 D RKNN: [11:24:25.846] 42 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_80 D RKNN: [11:24:25.846] 43 MaxPool INT8 NPU (1,256,20,20) (1,256,20,20) 33250 0 33250 1/0 100 MaxPool:MaxPool_81 D RKNN: [11:24:25.846] 44 Concat INT8 NPU (1,256,20,20),(1,256,20,20),... (1,1024,20,20) 132998 0 132998 4/0 400 Concat:Concat_82 D RKNN: [11:24:25.846] 45 ConvRelu INT8 NPU (1,1024,20,20),(512,1024,1,1),(512) (1,512,20,20) 185532 409600 409600 3/0 916 Conv:Conv_83 D RKNN: [11:24:25.846] 46 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_85 D RKNN: [11:24:25.846] 47 Resize INT8 NPU (1,256,20,20),(0),(4) (1,256,40,40) 83126 0 83126 16/0 100 Resize:Resize_88 D RKNN: [11:24:25.846] 48 Concat INT8 NPU (1,256,40,40),(1,256,40,40) (1,512,40,40) 265995 0 265995 2/0 800 Concat:Concat_89 D RKNN: [11:24:25.846] 49 ConvRelu INT8 NPU (1,512,40,40),(128,512,1,1),(128) (1,128,40,40) 177053 204800 204800 5/0 865 Conv:Conv_90 D RKNN: [11:24:25.846] 50 ConvRelu INT8 NPU (1,512,40,40),(128,512,1,1),(128) (1,128,40,40) 177053 204800 204800 5/0 865 Conv:Conv_96 D RKNN: [11:24:25.846] 51 ConvRelu INT8 NPU (1,128,40,40),(128,128,1,1),(128) (1,128,40,40) 69325 51200 69325 1/0 217 Conv:Conv_92 D RKNN: [11:24:25.846] 52 ConvRelu INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 90605 460800 460800 2/0 345 Conv:Conv_94 D RKNN: [11:24:25.846] 53 Concat INT8 NPU (1,128,40,40),(1,128,40,40) (1,256,40,40) 132998 0 132998 2/0 400 Concat:Concat_98 D RKNN: [11:24:25.846] 54 ConvRelu INT8 NPU (1,256,40,40),(256,256,1,1),(256) (1,256,40,40) 143970 204800 204800 3/0 466 Conv:Conv_99 D RKNN: [11:24:25.846] 55 ConvRelu INT8 NPU (1,256,40,40),(128,256,1,1),(128) (1,128,40,40) 105235 102400 105235 2/0 433 Conv:Conv_101 D RKNN: [11:24:25.846] 56 Resize INT8 NPU (1,128,40,40),(0),(4) (1,128,80,80) 166250 0 166250 8/0 200 Resize:Resize_104 D RKNN: [11:24:25.846] 57 Concat INT8 NPU (1,128,80,80),(1,128,80,80) (1,256,80,80) 531990 0 531990 2/0 1600 Concat:Concat_105 D RKNN: [11:24:25.846] 58 ConvRelu INT8 NPU (1,256,80,80),(64,256,1,1),(64) (1,64,80,80) 335237 204800 335237 8/0 1616 Conv:Conv_106 D RKNN: [11:24:25.846] 59 ConvRelu INT8 NPU (1,256,80,80),(64,256,1,1),(64) (1,64,80,80) 335237 204800 335237 8/0 1616 Conv:Conv_112 D RKNN: [11:24:25.846] 60 ConvRelu INT8 NPU (1,64,80,80),(64,64,1,1),(64) (1,64,80,80) 133746 102400 133746 2/0 404 Conv:Conv_108 D RKNN: [11:24:25.846] 61 ConvRelu INT8 NPU (1,64,80,80),(64,64,3,3),(64) (1,64,80,80) 139066 460800 460800 3/0 436 Conv:Conv_110 D RKNN: [11:24:25.846] 62 Concat INT8 NPU (1,64,80,80),(1,64,80,80) (1,128,80,80) 265995 0 265995 2/0 800 Concat:Concat_114 D RKNN: [11:24:25.846] 63 ConvRelu INT8 NPU (1,128,80,80),(128,128,1,1),(128) (1,128,80,80) 268821 204800 268821 4/0 817 Conv:Conv_115 D RKNN: [11:24:25.846] 64 ConvRelu INT8 NPU (1,128,80,80),(128,128,3,3),(128) (1,128,40,40) 190353 460800 460800 5/0 945 Conv:Conv_117 D RKNN: [11:24:25.846] 65 Concat INT8 NPU (1,128,40,40),(1,128,40,40) (1,256,40,40) 132998 0 132998 2/0 400 Concat:Concat_119 D RKNN: [11:24:25.846] 66 ConvSigmoid INT8 NPU (1,128,80,80),(255,128,1,1),(255) (1,255,80,80) 404624 409600 409600 5/1 833 Conv:Conv_145 D RKNN: [11:24:25.846] 67 ConvRelu INT8 NPU (1,256,40,40),(128,256,1,1),(128) (1,128,40,40) 105235 102400 105235 2/0 433 Conv:Conv_120 D RKNN: [11:24:25.846] 68 ConvRelu INT8 NPU (1,256,40,40),(128,256,1,1),(128) (1,128,40,40) 105235 102400 105235 2/0 433 Conv:Conv_126 D RKNN: [11:24:25.846] 69 ConvRelu INT8 NPU (1,128,40,40),(128,128,1,1),(128) (1,128,40,40) 69325 51200 69325 1/0 217 Conv:Conv_122 D RKNN: [11:24:25.846] 70 ConvRelu INT8 NPU (1,128,40,40),(128,128,3,3),(128) (1,128,40,40) 90605 460800 460800 2/0 345 Conv:Conv_124 D RKNN: [11:24:25.846] 71 Concat INT8 NPU (1,128,40,40),(1,128,40,40) (1,256,40,40) 132998 0 132998 2/0 400 Concat:Concat_128 D RKNN: [11:24:25.846] 72 ConvRelu INT8 NPU (1,256,40,40),(256,256,1,1),(256) (1,256,40,40) 143970 204800 204800 3/0 466 Conv:Conv_129 D RKNN: [11:24:25.846] 73 ConvRelu INT8 NPU (1,256,40,40),(256,256,3,3),(256) (1,256,20,20) 179214 460800 460800 3/0 978 Conv:Conv_131 D RKNN: [11:24:25.846] 74 Concat INT8 NPU (1,256,20,20),(1,256,20,20) (1,512,20,20) 66499 0 66499 2/0 200 Concat:Concat_133 D RKNN: [11:24:25.846] 75 ConvSigmoid INT8 NPU (1,256,40,40),(255,256,1,1),(255) (1,255,40,40) 143929 204800 204800 4/1 465 Conv:Conv_147 D RKNN: [11:24:25.846] 76 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_134 D RKNN: [11:24:25.846] 77 ConvRelu INT8 NPU (1,512,20,20),(256,512,1,1),(256) (1,256,20,20) 71487 102400 102400 2/0 330 Conv:Conv_140 D RKNN: [11:24:25.846] 78 ConvRelu INT8 NPU (1,256,20,20),(256,256,1,1),(256) (1,256,20,20) 44222 51200 51200 1/0 166 Conv:Conv_136 D RKNN: [11:24:25.846] 79 ConvRelu INT8 NPU (1,256,20,20),(256,256,3,3),(256) (1,256,20,20) 129340 460800 460800 1/0 678 Conv:Conv_138 D RKNN: [11:24:25.846] 80 Concat INT8 NPU (1,256,20,20),(1,256,20,20) (1,512,20,20) 66499 0 66499 2/0 200 Concat:Concat_142 D RKNN: [11:24:25.846] 81 ConvRelu INT8 NPU (1,512,20,20),(512,512,1,1),(512) (1,512,20,20) 109723 204800 204800 2/0 460 Conv:Conv_143 D RKNN: [11:24:25.846] 82 ConvSigmoid INT8 NPU (1,512,20,20),(255,512,1,1),(255) (1,255,20,20) 71403 102400 102400 3/1 329 Conv:Conv_149 D RKNN: [11:24:25.846] 83 OutputOperator INT8 NPU (1,255,80,80),(1,80,80,256) \ 531990 0 531990 21/0 3200 OutputOperator:output D RKNN: [11:24:25.846] 84 OutputOperator INT8 NPU (1,255,40,40),(1,40,40,256) \ 146298 0 146298 8/0 880 OutputOperator:283 D RKNN: [11:24:25.846] 85 OutputOperator INT8 NPU (1,255,20,20),(1,20,20,256) \ 43225 0 43225 4/0 260 OutputOperator:285 D RKNN: [11:24:25.846] ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ D RKNN: [11:24:25.847] <<<<<<<< end: rknn::RKNNModelRegCmdbuildPass D RKNN: [11:24:25.848] >>>>>> start: rknn::RKNNFlatcModelBuildPass D RKNN: [11:24:25.868] Export Mini RKNN model to /tmp/tmpmzghee_t/dumps/torch-jit-export.mini.rknn D RKNN: [11:24:25.871] >>>>>> end: rknn::RKNNFlatcModelBuildPass D RKNN: [11:24:25.872] >>>>>> start: rknn::RKNNMemStatisticsPass D RKNN: [11:24:25.873] -------------------------------------------------------------------------------------------------------------------------------- D RKNN: [11:24:25.873] Feature Tensor Information Table D RKNN: [11:24:25.873] ----------------------------------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.873] ID User Tensor DataType DataFormat OrigShape NativeShape | [Start End) Size D RKNN: [11:24:25.873] ----------------------------------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.873] 1 Conv images INT8 NC1HWC2 (1,3,640,640) (1,1,640,640,3) | 0x007297c0 0x008557c0 0x0012c000 D RKNN: [11:24:25.873] 2 ConvRelu 128 INT8 NC1HWC2 (1,12,320,320) (1,1,320,320,16) | 0x008557c0 0x009e57c0 0x00190000 D RKNN: [11:24:25.873] 3 ConvRelu 131 INT8 NC1HWC2 (1,32,320,320) (1,2,320,320,16) | 0x009e57c0 0x00d057c0 0x00320000 D RKNN: [11:24:25.873] 4 ConvRelu 133 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x007297c0 0x008b97c0 0x00190000 D RKNN: [11:24:25.873] 5 ConvRelu 133 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x007297c0 0x008b97c0 0x00190000 D RKNN: [11:24:25.873] 6 ConvRelu 135 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x008b97c0 0x009817c0 0x000c8000 D RKNN: [11:24:25.873] 7 ConvReluAdd 137 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x007297c0 0x007f17c0 0x000c8000 D RKNN: [11:24:25.873] 7 ConvReluAdd 135 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x008b97c0 0x009817c0 0x000c8000 D RKNN: [11:24:25.873] 8 Concat 140 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x007f17c0 0x008b97c0 0x000c8000 D RKNN: [11:24:25.873] 8 Concat 142 INT8 NC1HWC2 (1,32,160,160) (1,2,160,160,16) | 0x009817c0 0x00a497c0 0x000c8000 D RKNN: [11:24:25.873] 9 ConvRelu 143 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x00a497c0 0x00bd97c0 0x00190000 D RKNN: [11:24:25.873] 10 ConvRelu 145 INT8 NC1HWC2 (1,64,160,160) (1,4,160,160,16) | 0x007297c0 0x008b97c0 0x00190000 D RKNN: [11:24:25.873] 11 ConvRelu 147 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x008b97c0 0x009817c0 0x000c8000 D RKNN: [11:24:25.873] 12 ConvRelu 147 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x008b97c0 0x009817c0 0x000c8000 D RKNN: [11:24:25.873] 13 ConvRelu 149 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 14 ConvReluAdd 151 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 14 ConvReluAdd 149 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 15 ConvRelu 154 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x008557c0 0x008b97c0 0x00064000 D RKNN: [11:24:25.873] 16 ConvReluAdd 156 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 16 ConvReluAdd 154 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x008557c0 0x008b97c0 0x00064000 D RKNN: [11:24:25.873] 17 Concat 159 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 17 Concat 161 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x0078d7c0 0x007f17c0 0x00064000 D RKNN: [11:24:25.873] 18 ConvRelu 162 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x008557c0 0x0091d7c0 0x000c8000 D RKNN: [11:24:25.873] 19 ConvRelu 164 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x007297c0 0x007f17c0 0x000c8000 D RKNN: [11:24:25.873] 20 ConvRelu 166 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 21 ConvRelu 166 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 22 ConvRelu 168 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 23 ConvReluAdd 170 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 23 ConvReluAdd 168 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 24 ConvRelu 173 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008237c0 0x008557c0 0x00032000 D RKNN: [11:24:25.873] 25 ConvReluAdd 175 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 25 ConvReluAdd 173 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008237c0 0x008557c0 0x00032000 D RKNN: [11:24:25.873] 26 ConvRelu 178 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 27 ConvReluAdd 180 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 27 ConvReluAdd 178 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 28 Concat 183 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008237c0 0x008557c0 0x00032000 D RKNN: [11:24:25.873] 28 Concat 185 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008877c0 0x008b97c0 0x00032000 D RKNN: [11:24:25.873] 29 ConvRelu 186 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008b97c0 0x0091d7c0 0x00064000 D RKNN: [11:24:25.873] 30 ConvRelu 188 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 31 ConvRelu 190 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 32 ConvRelu 190 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 33 ConvRelu 192 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 34 ConvReluAdd 194 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008557c0 0x0086e7c0 0x00019000 D RKNN: [11:24:25.873] 34 ConvReluAdd 192 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 35 Concat 197 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0086e7c0 0x008877c0 0x00019000 D RKNN: [11:24:25.873] 35 Concat 199 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008a07c0 0x008b97c0 0x00019000 D RKNN: [11:24:25.873] 36 ConvRelu 200 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008b97c0 0x008eb7c0 0x00032000 D RKNN: [11:24:25.873] 37 ConvRelu 202 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 38 MaxPool 204 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 39 MaxPool 205 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008557c0 0x0086e7c0 0x00019000 D RKNN: [11:24:25.873] 40 MaxPool 206 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0086e7c0 0x008877c0 0x00019000 D RKNN: [11:24:25.873] 41 MaxPool 207 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008557c0 0x0086e7c0 0x00019000 D RKNN: [11:24:25.873] 42 MaxPool 208 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008a07c0 0x008b97c0 0x00019000 D RKNN: [11:24:25.873] 43 MaxPool 209 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008557c0 0x0086e7c0 0x00019000 D RKNN: [11:24:25.873] 44 Concat 204 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 44 Concat 206 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0086e7c0 0x008877c0 0x00019000 D RKNN: [11:24:25.873] 44 Concat 208 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008a07c0 0x008b97c0 0x00019000 D RKNN: [11:24:25.873] 44 Concat 210 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008b97c0 0x008d27c0 0x00019000 D RKNN: [11:24:25.873] 45 ConvRelu 211 INT8 NC1HWC2 (1,1024,20,20) (1,64,20,20,16) | 0x008d27c0 0x009367c0 0x00064000 D RKNN: [11:24:25.873] 46 ConvRelu 213 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008557c0 0x008877c0 0x00032000 D RKNN: [11:24:25.873] 47 Resize 215 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 48 Concat 220 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008a07c0 0x009047c0 0x00064000 D RKNN: [11:24:25.873] 48 Concat 188 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 49 ConvRelu 221 INT8 NC1HWC2 (1,512,40,40) (1,32,40,40,16) | 0x009047c0 0x009cc7c0 0x000c8000 D RKNN: [11:24:25.873] 50 ConvRelu 221 INT8 NC1HWC2 (1,512,40,40) (1,32,40,40,16) | 0x009047c0 0x009cc7c0 0x000c8000 D RKNN: [11:24:25.873] 51 ConvRelu 223 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008a07c0 0x008d27c0 0x00032000 D RKNN: [11:24:25.873] 52 ConvRelu 225 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 53 Concat 227 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008a07c0 0x008d27c0 0x00032000 D RKNN: [11:24:25.873] 53 Concat 229 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008d27c0 0x009047c0 0x00032000 D RKNN: [11:24:25.873] 54 ConvRelu 230 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 55 ConvRelu 232 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008a07c0 0x009047c0 0x00064000 D RKNN: [11:24:25.873] 56 Resize 234 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 57 Concat 239 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x008a07c0 0x009687c0 0x000c8000 D RKNN: [11:24:25.873] 57 Concat 164 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x007297c0 0x007f17c0 0x000c8000 D RKNN: [11:24:25.873] 58 ConvRelu 240 INT8 NC1HWC2 (1,256,80,80) (1,16,80,80,16) | 0x009687c0 0x00af87c0 0x00190000 D RKNN: [11:24:25.873] 59 ConvRelu 240 INT8 NC1HWC2 (1,256,80,80) (1,16,80,80,16) | 0x009687c0 0x00af87c0 0x00190000 D RKNN: [11:24:25.873] 60 ConvRelu 242 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x008237c0 0x008877c0 0x00064000 D RKNN: [11:24:25.873] 61 ConvRelu 244 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x0078d7c0 0x007f17c0 0x00064000 D RKNN: [11:24:25.873] 62 Concat 246 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x008237c0 0x008877c0 0x00064000 D RKNN: [11:24:25.873] 62 Concat 248 INT8 NC1HWC2 (1,64,80,80) (1,4,80,80,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 63 ConvRelu 249 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x008a07c0 0x009687c0 0x000c8000 D RKNN: [11:24:25.873] 64 ConvRelu 251 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x007297c0 0x007f17c0 0x000c8000 D RKNN: [11:24:25.873] 65 Concat 253 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008237c0 0x008557c0 0x00032000 D RKNN: [11:24:25.873] 65 Concat 234 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 66 ConvSigmoid 251 INT8 NC1HWC2 (1,128,80,80) (1,8,80,80,16) | 0x007297c0 0x007f17c0 0x000c8000 D RKNN: [11:24:25.873] 67 ConvRelu 254 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008a07c0 0x009047c0 0x00064000 D RKNN: [11:24:25.873] 68 ConvRelu 254 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008a07c0 0x009047c0 0x00064000 D RKNN: [11:24:25.873] 69 ConvRelu 256 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007297c0 0x0075b7c0 0x00032000 D RKNN: [11:24:25.873] 70 ConvRelu 258 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x008a07c0 0x008d27c0 0x00032000 D RKNN: [11:24:25.873] 71 Concat 260 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x007297c0 0x0075b7c0 0x00032000 D RKNN: [11:24:25.873] 71 Concat 262 INT8 NC1HWC2 (1,128,40,40) (1,8,40,40,16) | 0x0075b7c0 0x0078d7c0 0x00032000 D RKNN: [11:24:25.873] 72 ConvRelu 263 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x008a07c0 0x009047c0 0x00064000 D RKNN: [11:24:25.873] 73 ConvRelu 265 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 74 Concat 267 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008a07c0 0x008b97c0 0x00019000 D RKNN: [11:24:25.873] 74 Concat 215 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008877c0 0x008a07c0 0x00019000 D RKNN: [11:24:25.873] 75 ConvSigmoid 265 INT8 NC1HWC2 (1,256,40,40) (1,16,40,40,16) | 0x007297c0 0x0078d7c0 0x00064000 D RKNN: [11:24:25.873] 76 ConvRelu 268 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008b97c0 0x008eb7c0 0x00032000 D RKNN: [11:24:25.873] 77 ConvRelu 268 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x008b97c0 0x008eb7c0 0x00032000 D RKNN: [11:24:25.873] 78 ConvRelu 270 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x008eb7c0 0x009047c0 0x00019000 D RKNN: [11:24:25.873] 79 ConvRelu 272 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x007427c0 0x0075b7c0 0x00019000 D RKNN: [11:24:25.873] 80 Concat 274 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x0075b7c0 0x007747c0 0x00019000 D RKNN: [11:24:25.873] 80 Concat 276 INT8 NC1HWC2 (1,256,20,20) (1,16,20,20,16) | 0x007297c0 0x007427c0 0x00019000 D RKNN: [11:24:25.873] 81 ConvRelu 277 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x007f17c0 0x008237c0 0x00032000 D RKNN: [11:24:25.873] 82 ConvSigmoid 279 INT8 NC1HWC2 (1,512,20,20) (1,32,20,20,16) | 0x007297c0 0x0075b7c0 0x00032000 D RKNN: [11:24:25.873] 83 OutputOperator output INT8 NC1HWC2 (1,255,80,80) (1,16,80,80,16) | 0x009047c0 0x00a947c0 0x00190000 D RKNN: [11:24:25.873] 83 OutputOperator output_exSecondary0 INT8 NC1HWC2 (1,80,80,256) (1,5,80,256,16) | 0x00a947c0 0x00c247c0 0x00190000 D RKNN: [11:24:25.873] 83 OutputOperator output_exSecondary INT8 NHWC (1,80,80,255) (1,80,80,255) | 0x00c247c0 0x00db2ec0 0x0018e700 D RKNN: [11:24:25.873] 84 OutputOperator 283 INT8 NC1HWC2 (1,255,40,40) (1,16,40,40,16) | 0x0078d7c0 0x007f17c0 0x00064000 D RKNN: [11:24:25.873] 84 OutputOperator 283_exSecondary0 INT8 NC1HWC2 (1,40,40,256) (1,2,40,256,16) | 0x007f17c0 0x008557c0 0x00064000 D RKNN: [11:24:25.873] 84 OutputOperator 283_exSecondary INT8 NHWC (1,40,40,255) (1,40,40,255) | 0x008557c0 0x008b9180 0x000639c0 D RKNN: [11:24:25.873] 85 OutputOperator 285 INT8 NC1HWC2 (1,255,20,20) (1,16,20,20,16) | 0x0075b7c0 0x007747c0 0x00019000 D RKNN: [11:24:25.873] 85 OutputOperator 285_exSecondary0 INT8 NC1HWC2 (1,20,20,256) (1,1,20,256,16) | 0x007747c0 0x0078d7c0 0x00019000 D RKNN: [11:24:25.873] 85 OutputOperator 285_exSecondary INT8 NHWC (1,20,20,255) (1,20,20,255) | 0x007297c0 0x00742630 0x00018e70 D RKNN: [11:24:25.873] ----------------------------------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.873] -------------------------------------------------------------------------------------------------------- D RKNN: [11:24:25.873] Const Tensor Information Table D RKNN: [11:24:25.873] ----------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.873] ID User Tensor DataType OrigShape | [Start End) Size D RKNN: [11:24:25.873] ----------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.873] 1 Conv model.0.convsp.weight INT8 (12,3,2,2) | 0x00000000 0x000000c0 0x000000c0 D RKNN: [11:24:25.873] 1 Conv model.0.convsp.weight_bias_0 INT32 (12) | 0x006f6140*0x006f61c0 0x00000080 D RKNN: [11:24:25.873] 2 ConvRelu 287 INT8 (32,12,3,3) | 0x006f4d40 0x006f5f40 0x00001200 D RKNN: [11:24:25.873] 2 ConvRelu 288 INT32 (32) | 0x006f5f40 0x006f6040 0x00000100 D RKNN: [11:24:25.873] 3 ConvRelu model.1.conv.weight INT8 (64,32,3,3) | 0x000000c0 0x000048c0 0x00004800 D RKNN: [11:24:25.873] 3 ConvRelu model.1.conv.bias INT32 (64) | 0x000048c0 0x00004ac0 0x00000200 D RKNN: [11:24:25.873] 4 ConvRelu model.2.cv1.conv.weight INT8 (32,64,1,1) | 0x00004ac0 0x000052c0 0x00000800 D RKNN: [11:24:25.873] 4 ConvRelu model.2.cv1.conv.bias INT32 (32) | 0x000052c0 0x000053c0 0x00000100 D RKNN: [11:24:25.873] 5 ConvRelu model.2.cv2.conv.weight INT8 (32,64,1,1) | 0x000053c0 0x00005bc0 0x00000800 D RKNN: [11:24:25.873] 5 ConvRelu model.2.cv2.conv.bias INT32 (32) | 0x00005bc0 0x00005cc0 0x00000100 D RKNN: [11:24:25.873] 6 ConvRelu model.2.m.0.cv1.conv.weight INT8 (32,32,1,1) | 0x00006ec0 0x000072c0 0x00000400 D RKNN: [11:24:25.873] 6 ConvRelu model.2.m.0.cv1.conv.bias INT32 (32) | 0x000072c0 0x000073c0 0x00000100 D RKNN: [11:24:25.873] 7 ConvReluAdd model.2.m.0.cv2.conv.weight INT8 (32,32,3,3) | 0x000073c0 0x000097c0 0x00002400 D RKNN: [11:24:25.873] 7 ConvReluAdd model.2.m.0.cv2.conv.bias INT32 (32) | 0x000097c0 0x000098c0 0x00000100 D RKNN: [11:24:25.873] 9 ConvRelu model.2.cv3.conv.weight INT8 (64,64,1,1) | 0x00005cc0 0x00006cc0 0x00001000 D RKNN: [11:24:25.873] 9 ConvRelu model.2.cv3.conv.bias INT32 (64) | 0x00006cc0 0x00006ec0 0x00000200 D RKNN: [11:24:25.873] 10 ConvRelu model.3.conv.weight INT8 (128,64,3,3) | 0x000098c0 0x0001b8c0 0x00012000 D RKNN: [11:24:25.873] 10 ConvRelu model.3.conv.bias INT32 (128) | 0x0001b8c0 0x0001bcc0 0x00000400 D RKNN: [11:24:25.873] 11 ConvRelu model.4.cv1.conv.weight INT8 (64,128,1,1) | 0x0001bcc0 0x0001dcc0 0x00002000 D RKNN: [11:24:25.873] 11 ConvRelu model.4.cv1.conv.bias INT32 (64) | 0x0001dcc0 0x0001dec0 0x00000200 D RKNN: [11:24:25.873] 12 ConvRelu model.4.cv2.conv.weight INT8 (64,128,1,1) | 0x0001dec0 0x0001fec0 0x00002000 D RKNN: [11:24:25.873] 12 ConvRelu model.4.cv2.conv.bias INT32 (64) | 0x0001fec0 0x000200c0 0x00000200 D RKNN: [11:24:25.873] 13 ConvRelu model.4.m.0.cv1.conv.weight INT8 (64,64,1,1) | 0x000244c0 0x000254c0 0x00001000 D RKNN: [11:24:25.873] 13 ConvRelu model.4.m.0.cv1.conv.bias INT32 (64) | 0x000254c0 0x000256c0 0x00000200 D RKNN: [11:24:25.873] 14 ConvReluAdd model.4.m.0.cv2.conv.weight INT8 (64,64,3,3) | 0x000256c0 0x0002e6c0 0x00009000 D RKNN: [11:24:25.873] 14 ConvReluAdd model.4.m.0.cv2.conv.bias INT32 (64) | 0x0002e6c0 0x0002e8c0 0x00000200 D RKNN: [11:24:25.873] 15 ConvRelu model.4.m.1.cv1.conv.weight INT8 (64,64,1,1) | 0x0002e8c0 0x0002f8c0 0x00001000 D RKNN: [11:24:25.873] 15 ConvRelu model.4.m.1.cv1.conv.bias INT32 (64) | 0x0002f8c0 0x0002fac0 0x00000200 D RKNN: [11:24:25.873] 16 ConvReluAdd model.4.m.1.cv2.conv.weight INT8 (64,64,3,3) | 0x0002fac0 0x00038ac0 0x00009000 D RKNN: [11:24:25.873] 16 ConvReluAdd model.4.m.1.cv2.conv.bias INT32 (64) | 0x00038ac0 0x00038cc0 0x00000200 D RKNN: [11:24:25.873] 18 ConvRelu model.4.cv3.conv.weight INT8 (128,128,1,1) | 0x000200c0 0x000240c0 0x00004000 D RKNN: [11:24:25.873] 18 ConvRelu model.4.cv3.conv.bias INT32 (128) | 0x000240c0 0x000244c0 0x00000400 D RKNN: [11:24:25.873] 19 ConvRelu model.5.conv.weight INT8 (256,128,3,3) | 0x00038cc0 0x00080cc0 0x00048000 D RKNN: [11:24:25.873] 19 ConvRelu model.5.conv.bias INT32 (256) | 0x00080cc0 0x000814c0 0x00000800 D RKNN: [11:24:25.873] 20 ConvRelu model.6.cv1.conv.weight INT8 (128,256,1,1) | 0x000814c0 0x000894c0 0x00008000 D RKNN: [11:24:25.873] 20 ConvRelu model.6.cv1.conv.bias INT32 (128) | 0x000894c0 0x000898c0 0x00000400 D RKNN: [11:24:25.873] 21 ConvRelu model.6.cv2.conv.weight INT8 (128,256,1,1) | 0x000898c0 0x000918c0 0x00008000 D RKNN: [11:24:25.873] 21 ConvRelu model.6.cv2.conv.bias INT32 (128) | 0x000918c0 0x00091cc0 0x00000400 D RKNN: [11:24:25.873] 22 ConvRelu model.6.m.0.cv1.conv.weight INT8 (128,128,1,1) | 0x000a24c0 0x000a64c0 0x00004000 D RKNN: [11:24:25.873] 22 ConvRelu model.6.m.0.cv1.conv.bias INT32 (128) | 0x000a64c0 0x000a68c0 0x00000400 D RKNN: [11:24:25.873] 23 ConvReluAdd model.6.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x000a68c0 0x000ca8c0 0x00024000 D RKNN: [11:24:25.873] 23 ConvReluAdd model.6.m.0.cv2.conv.bias INT32 (128) | 0x000ca8c0 0x000cacc0 0x00000400 D RKNN: [11:24:25.873] 24 ConvRelu model.6.m.1.cv1.conv.weight INT8 (128,128,1,1) | 0x000cacc0 0x000cecc0 0x00004000 D RKNN: [11:24:25.873] 24 ConvRelu model.6.m.1.cv1.conv.bias INT32 (128) | 0x000cecc0 0x000cf0c0 0x00000400 D RKNN: [11:24:25.873] 25 ConvReluAdd model.6.m.1.cv2.conv.weight INT8 (128,128,3,3) | 0x000cf0c0 0x000f30c0 0x00024000 D RKNN: [11:24:25.873] 25 ConvReluAdd model.6.m.1.cv2.conv.bias INT32 (128) | 0x000f30c0 0x000f34c0 0x00000400 D RKNN: [11:24:25.873] 26 ConvRelu model.6.m.2.cv1.conv.weight INT8 (128,128,1,1) | 0x000f34c0 0x000f74c0 0x00004000 D RKNN: [11:24:25.873] 26 ConvRelu model.6.m.2.cv1.conv.bias INT32 (128) | 0x000f74c0 0x000f78c0 0x00000400 D RKNN: [11:24:25.873] 27 ConvReluAdd model.6.m.2.cv2.conv.weight INT8 (128,128,3,3) | 0x000f78c0 0x0011b8c0 0x00024000 D RKNN: [11:24:25.873] 27 ConvReluAdd model.6.m.2.cv2.conv.bias INT32 (128) | 0x0011b8c0 0x0011bcc0 0x00000400 D RKNN: [11:24:25.873] 29 ConvRelu model.6.cv3.conv.weight INT8 (256,256,1,1) | 0x00091cc0 0x000a1cc0 0x00010000 D RKNN: [11:24:25.873] 29 ConvRelu model.6.cv3.conv.bias INT32 (256) | 0x000a1cc0 0x000a24c0 0x00000800 D RKNN: [11:24:25.873] 30 ConvRelu model.7.conv.weight INT8 (512,256,3,3) | 0x0011bcc0 0x0023bcc0 0x00120000 D RKNN: [11:24:25.873] 30 ConvRelu model.7.conv.bias INT32 (512) | 0x0023bcc0 0x0023ccc0 0x00001000 D RKNN: [11:24:25.873] 31 ConvRelu model.8.cv1.conv.weight INT8 (256,512,1,1) | 0x0023ccc0 0x0025ccc0 0x00020000 D RKNN: [11:24:25.873] 31 ConvRelu model.8.cv1.conv.bias INT32 (256) | 0x0025ccc0 0x0025d4c0 0x00000800 D RKNN: [11:24:25.873] 32 ConvRelu model.8.cv2.conv.weight INT8 (256,512,1,1) | 0x0025d4c0 0x0027d4c0 0x00020000 D RKNN: [11:24:25.873] 32 ConvRelu model.8.cv2.conv.bias INT32 (256) | 0x0027d4c0 0x0027dcc0 0x00000800 D RKNN: [11:24:25.873] 33 ConvRelu model.8.m.0.cv1.conv.weight INT8 (256,256,1,1) | 0x002becc0 0x002cecc0 0x00010000 D RKNN: [11:24:25.873] 33 ConvRelu model.8.m.0.cv1.conv.bias INT32 (256) | 0x002cecc0 0x002cf4c0 0x00000800 D RKNN: [11:24:25.873] 34 ConvReluAdd model.8.m.0.cv2.conv.weight INT8 (256,256,3,3) | 0x002cf4c0 0x0035f4c0 0x00090000 D RKNN: [11:24:25.873] 34 ConvReluAdd model.8.m.0.cv2.conv.bias INT32 (256) | 0x0035f4c0 0x0035fcc0 0x00000800 D RKNN: [11:24:25.873] 36 ConvRelu model.8.cv3.conv.weight INT8 (512,512,1,1) | 0x0027dcc0 0x002bdcc0 0x00040000 D RKNN: [11:24:25.873] 36 ConvRelu model.8.cv3.conv.bias INT32 (512) | 0x002bdcc0 0x002becc0 0x00001000 D RKNN: [11:24:25.873] 37 ConvRelu model.9.cv1.conv.weight INT8 (256,512,1,1) | 0x0035fcc0 0x0037fcc0 0x00020000 D RKNN: [11:24:25.873] 37 ConvRelu model.9.cv1.conv.bias INT32 (256) | 0x0037fcc0 0x003804c0 0x00000800 D RKNN: [11:24:25.873] 45 ConvRelu model.9.cv2.conv.weight INT8 (512,1024,1,1) | 0x003804c0 0x004004c0 0x00080000 D RKNN: [11:24:25.873] 45 ConvRelu model.9.cv2.conv.bias INT32 (512) | 0x004004c0 0x004014c0 0x00001000 D RKNN: [11:24:25.873] 46 ConvRelu model.10.conv.weight INT8 (256,512,1,1) | 0x004014c0 0x004214c0 0x00020000 D RKNN: [11:24:25.873] 46 ConvRelu model.10.conv.bias INT32 (256) | 0x004214c0 0x00421cc0 0x00000800 D RKNN: [11:24:25.873] 47 Resize 219 FLOAT (0) | 0x00000000 0x00000000 0x00000000 D RKNN: [11:24:25.873] 47 Resize 289 FLOAT (4) | 0x006f6040 0x006f60c0 0x00000080 D RKNN: [11:24:25.873] 49 ConvRelu model.13.cv1.conv.weight INT8 (128,512,1,1) | 0x00421cc0 0x00431cc0 0x00010000 D RKNN: [11:24:25.873] 49 ConvRelu model.13.cv1.conv.bias INT32 (128) | 0x00431cc0 0x004320c0 0x00000400 D RKNN: [11:24:25.873] 50 ConvRelu model.13.cv2.conv.weight INT8 (128,512,1,1) | 0x004320c0 0x004420c0 0x00010000 D RKNN: [11:24:25.873] 50 ConvRelu model.13.cv2.conv.bias INT32 (128) | 0x004420c0 0x004424c0 0x00000400 D RKNN: [11:24:25.873] 51 ConvRelu model.13.m.0.cv1.conv.weight INT8 (128,128,1,1) | 0x00452cc0 0x00456cc0 0x00004000 D RKNN: [11:24:25.873] 51 ConvRelu model.13.m.0.cv1.conv.bias INT32 (128) | 0x00456cc0 0x004570c0 0x00000400 D RKNN: [11:24:25.873] 52 ConvRelu model.13.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x004570c0 0x0047b0c0 0x00024000 D RKNN: [11:24:25.873] 52 ConvRelu model.13.m.0.cv2.conv.bias INT32 (128) | 0x0047b0c0 0x0047b4c0 0x00000400 D RKNN: [11:24:25.873] 54 ConvRelu model.13.cv3.conv.weight INT8 (256,256,1,1) | 0x004424c0 0x004524c0 0x00010000 D RKNN: [11:24:25.873] 54 ConvRelu model.13.cv3.conv.bias INT32 (256) | 0x004524c0 0x00452cc0 0x00000800 D RKNN: [11:24:25.873] 55 ConvRelu model.14.conv.weight INT8 (128,256,1,1) | 0x0047b4c0 0x004834c0 0x00008000 D RKNN: [11:24:25.873] 55 ConvRelu model.14.conv.bias INT32 (128) | 0x004834c0 0x004838c0 0x00000400 D RKNN: [11:24:25.873] 56 Resize 238 FLOAT (0) | 0x00000000 0x00000000 0x00000000 D RKNN: [11:24:25.873] 56 Resize 290 FLOAT (4) | 0x006f60c0 0x006f6140 0x00000080 D RKNN: [11:24:25.873] 58 ConvRelu model.17.cv1.conv.weight INT8 (64,256,1,1) | 0x004838c0 0x004878c0 0x00004000 D RKNN: [11:24:25.873] 58 ConvRelu model.17.cv1.conv.bias INT32 (64) | 0x004878c0 0x00487ac0 0x00000200 D RKNN: [11:24:25.873] 59 ConvRelu model.17.cv2.conv.weight INT8 (64,256,1,1) | 0x00487ac0 0x0048bac0 0x00004000 D RKNN: [11:24:25.873] 59 ConvRelu model.17.cv2.conv.bias INT32 (64) | 0x0048bac0 0x0048bcc0 0x00000200 D RKNN: [11:24:25.873] 60 ConvRelu model.17.m.0.cv1.conv.weight INT8 (64,64,1,1) | 0x004900c0 0x004910c0 0x00001000 D RKNN: [11:24:25.873] 60 ConvRelu model.17.m.0.cv1.conv.bias INT32 (64) | 0x004910c0 0x004912c0 0x00000200 D RKNN: [11:24:25.873] 61 ConvRelu model.17.m.0.cv2.conv.weight INT8 (64,64,3,3) | 0x004912c0 0x0049a2c0 0x00009000 D RKNN: [11:24:25.873] 61 ConvRelu model.17.m.0.cv2.conv.bias INT32 (64) | 0x0049a2c0 0x0049a4c0 0x00000200 D RKNN: [11:24:25.873] 63 ConvRelu model.17.cv3.conv.weight INT8 (128,128,1,1) | 0x0048bcc0 0x0048fcc0 0x00004000 D RKNN: [11:24:25.873] 63 ConvRelu model.17.cv3.conv.bias INT32 (128) | 0x0048fcc0 0x004900c0 0x00000400 D RKNN: [11:24:25.873] 64 ConvRelu model.18.conv.weight INT8 (128,128,3,3) | 0x0049a4c0 0x004be4c0 0x00024000 D RKNN: [11:24:25.873] 64 ConvRelu model.18.conv.bias INT32 (128) | 0x004be4c0 0x004be8c0 0x00000400 D RKNN: [11:24:25.873] 66 ConvSigmoid model.24.m.0.weight INT8 (255,128,1,1) | 0x006bb8c0 0x006c3840 0x00007f80 D RKNN: [11:24:25.873] 66 ConvSigmoid model.24.m.0.bias INT32 (255) | 0x006c3840 0x006c4040 0x00000800 D RKNN: [11:24:25.873] 67 ConvRelu model.20.cv1.conv.weight INT8 (128,256,1,1) | 0x004be8c0 0x004c68c0 0x00008000 D RKNN: [11:24:25.873] 67 ConvRelu model.20.cv1.conv.bias INT32 (128) | 0x004c68c0 0x004c6cc0 0x00000400 D RKNN: [11:24:25.873] 68 ConvRelu model.20.cv2.conv.weight INT8 (128,256,1,1) | 0x004c6cc0 0x004cecc0 0x00008000 D RKNN: [11:24:25.873] 68 ConvRelu model.20.cv2.conv.bias INT32 (128) | 0x004cecc0 0x004cf0c0 0x00000400 D RKNN: [11:24:25.873] 69 ConvRelu model.20.m.0.cv1.conv.weight INT8 (128,128,1,1) | 0x004df8c0 0x004e38c0 0x00004000 D RKNN: [11:24:25.873] 69 ConvRelu model.20.m.0.cv1.conv.bias INT32 (128) | 0x004e38c0 0x004e3cc0 0x00000400 D RKNN: [11:24:25.873] 70 ConvRelu model.20.m.0.cv2.conv.weight INT8 (128,128,3,3) | 0x004e3cc0 0x00507cc0 0x00024000 D RKNN: [11:24:25.873] 70 ConvRelu model.20.m.0.cv2.conv.bias INT32 (128) | 0x00507cc0 0x005080c0 0x00000400 D RKNN: [11:24:25.873] 72 ConvRelu model.20.cv3.conv.weight INT8 (256,256,1,1) | 0x004cf0c0 0x004df0c0 0x00010000 D RKNN: [11:24:25.873] 72 ConvRelu model.20.cv3.conv.bias INT32 (256) | 0x004df0c0 0x004df8c0 0x00000800 D RKNN: [11:24:25.873] 73 ConvRelu model.21.conv.weight INT8 (256,256,3,3) | 0x005080c0 0x005980c0 0x00090000 D RKNN: [11:24:25.873] 73 ConvRelu model.21.conv.bias INT32 (256) | 0x005980c0 0x005988c0 0x00000800 D RKNN: [11:24:25.873] 75 ConvSigmoid model.24.m.1.weight INT8 (255,256,1,1) | 0x006c4040 0x006d3f40 0x0000ff00 D RKNN: [11:24:25.873] 75 ConvSigmoid model.24.m.1.bias INT32 (255) | 0x006d3f40 0x006d4740 0x00000800 D RKNN: [11:24:25.873] 76 ConvRelu model.23.cv1.conv.weight INT8 (256,512,1,1) | 0x005988c0 0x005b88c0 0x00020000 D RKNN: [11:24:25.873] 76 ConvRelu model.23.cv1.conv.bias INT32 (256) | 0x005b88c0 0x005b90c0 0x00000800 D RKNN: [11:24:25.873] 77 ConvRelu model.23.cv2.conv.weight INT8 (256,512,1,1) | 0x005b90c0 0x005d90c0 0x00020000 D RKNN: [11:24:25.873] 77 ConvRelu model.23.cv2.conv.bias INT32 (256) | 0x005d90c0 0x005d98c0 0x00000800 D RKNN: [11:24:25.873] 78 ConvRelu model.23.m.0.cv1.conv.weight INT8 (256,256,1,1) | 0x0061a8c0 0x0062a8c0 0x00010000 D RKNN: [11:24:25.873] 78 ConvRelu model.23.m.0.cv1.conv.bias INT32 (256) | 0x0062a8c0 0x0062b0c0 0x00000800 D RKNN: [11:24:25.873] 79 ConvRelu model.23.m.0.cv2.conv.weight INT8 (256,256,3,3) | 0x0062b0c0 0x006bb0c0 0x00090000 D RKNN: [11:24:25.873] 79 ConvRelu model.23.m.0.cv2.conv.bias INT32 (256) | 0x006bb0c0 0x006bb8c0 0x00000800 D RKNN: [11:24:25.873] 81 ConvRelu model.23.cv3.conv.weight INT8 (512,512,1,1) | 0x005d98c0 0x006198c0 0x00040000 D RKNN: [11:24:25.873] 81 ConvRelu model.23.cv3.conv.bias INT32 (512) | 0x006198c0 0x0061a8c0 0x00001000 D RKNN: [11:24:25.873] 82 ConvSigmoid model.24.m.2.weight INT8 (255,512,1,1) | 0x006d4740 0x006f4540 0x0001fe00 D RKNN: [11:24:25.873] 82 ConvSigmoid model.24.m.2.bias INT32 (255) | 0x006f4540 0x006f4d40 0x00000800 D RKNN: [11:24:25.873] ----------------------------------------------------------------------+--------------------------------- D RKNN: [11:24:25.878] ---------------------------------------- D RKNN: [11:24:25.879] Total Internal Memory Size: 6693.75KB D RKNN: [11:24:25.879] Total Weight Memory Size: 7128.44KB D RKNN: [11:24:25.879] ---------------------------------------- D RKNN: [11:24:25.879] <<<<<<<< end: rknn::RKNNMemStatisticsPass I rknn buiding done. done --> Export rknn model done --> Init runtime environment I target set by user is: rv1106 I Check RV1106 board npu runtime version I Starting ntp or adb, target is RV1106 I Device [bd547ee6900c058b] not found in ntb device list. I Start adb... I Connect to Device success! I NPUTransfer: Starting NPU Transfer Client, Transfer version 2.1.0 (b5861e7@2020-11-23T11:50:36) D NPUTransfer: Transfer spec = local:transfer_proxy D NPUTransfer: Transfer interface successfully opened, fd = 3 D RKNNAPI: ============================================== D RKNNAPI: RKNN VERSION: D RKNNAPI: API: 1.6.0 (535b468 build@2023-12-11T09:05:46) D RKNNAPI: DRV: rknn_server: 1.6.0 (535b468 build@2023-12-11T17:05:28) D RKNNAPI: DRV: rknnrt: 1.6.0 (9a7b5d24c@2023-12-13T17:33:10) D RKNNAPI: ============================================== D RKNNAPI: Input tensors: D RKNNAPI: index=0, name=images, n_dims=4, dims=[1, 640, 640, 3], n_elems=1228800, size=1228800, w_stride = 0, size_with_stride = 0, fmt=NHWC, type=UINT8, qnt_type=AFFINE, zp=-128, scale=0.003922 D RKNNAPI: Output tensors: D RKNNAPI: index=0, name=output, n_dims=4, dims=[1, 255, 80, 80], n_elems=1632000, size=1632000, w_stride = 0, size_with_stride = 0, fmt=NCHW, type=INT8, qnt_type=AFFINE, zp=-128, scale=0.003860 D RKNNAPI: index=1, name=283, n_dims=4, dims=[1, 255, 40, 40], n_elems=408000, size=408000, w_stride = 0, size_with_stride = 0, fmt=NCHW, type=INT8, qnt_type=AFFINE, zp=-128, scale=0.003922 D RKNNAPI: index=2, name=285, n_dims=4, dims=[1, 255, 20, 20], n_elems=102000, size=102000, w_stride = 0, size_with_stride = 0, fmt=NCHW, type=INT8, qnt_type=AFFINE, zp=-128, scale=0.003915 done --> Running model done class score xmin, ymin, xmax, ymax -------------------------------------------------- person 0.884 [ 209, 244, 286, 506] person 0.868 [ 478, 238, 559, 526] person 0.825 [ 110, 238, 230, 534] person 0.339 [ 79, 354, 122, 516] bus 0.705 [ 94, 129, 553, 468] Save results to result.jpg! D NPUTransfer: Transfer client closed, fd = 3