|
|
板凳
楼主 |
发表于 2023-2-9 20:19:41
|
只看该作者
附上一份转换yolov3-tiny的log
- D Save log info to: log/best_rv1126_sample_asymmetric_affine-u8_True_2023-02-09_02:43:42.log
- D Using CPPUTILS: True
- I Start importing onnx...
- W Call onnx.optimizer.optimize fail, skip optimize
- I Current ONNX Model use ir_version 6 opset_version 11
- I Call RKNN onnx optimize fail, skip optimize
- D Calc tensor Constant_66 (0,)
- D Calc tensor Initializer_model.8.conv.weight (256, 128, 3, 3)
- D Calc tensor Initializer_model.8.conv.bias (256,)
- D Calc tensor Initializer_model.6.conv.weight (128, 64, 3, 3)
- D Calc tensor Initializer_model.6.conv.bias (128,)
- D Calc tensor Initializer_model.4.conv.weight (64, 32, 3, 3)
- D Calc tensor Initializer_model.4.conv.bias (64,)
- D Calc tensor Initializer_model.2.conv.weight (32, 16, 3, 3)
- D Calc tensor Initializer_model.2.conv.bias (32,)
- D Calc tensor Initializer_model.19.conv.weight (256, 384, 3, 3)
- D Calc tensor Initializer_model.19.conv.bias (256,)
- D Calc tensor Initializer_model.16.conv.weight (128, 256, 1, 1)
- D Calc tensor Initializer_model.16.conv.bias (128,)
- D Calc tensor Initializer_model.15.conv.weight (512, 256, 3, 3)
- D Calc tensor Initializer_model.15.conv.bias (512,)
- D Calc tensor Initializer_model.14.conv.weight (256, 1024, 1, 1)
- D Calc tensor Initializer_model.14.conv.bias (256,)
- D Calc tensor Initializer_model.13.conv.weight (1024, 512, 3, 3)
- D Calc tensor Initializer_model.13.conv.bias (1024,)
- D Calc tensor Initializer_model.11.conv.weight (512, 512, 2, 2)
- D Calc tensor Initializer_model.11.conv.bias (512,)
- D Calc tensor Initializer_model.10.conv.weight (512, 256, 3, 3)
- D Calc tensor Initializer_model.10.conv.bias (512,)
- D Calc tensor Initializer_model.0.conv.weight (16, 3, 3, 3)
- D Calc tensor Initializer_model.0.conv.bias (16,)
- D Calc tensor Conv_30 (1, 16, 640, 640)
- D Calc tensor Relu_31 (1, 16, 640, 640)
- D Calc tensor MaxPool_32 (1, 16, 320, 320)
- D Calc tensor Conv_33 (1, 32, 320, 320)
- D Calc tensor Relu_34 (1, 32, 320, 320)
- D Calc tensor MaxPool_35 (1, 32, 160, 160)
- D Calc tensor Conv_36 (1, 64, 160, 160)
- D Calc tensor Relu_37 (1, 64, 160, 160)
- D Calc tensor MaxPool_38 (1, 64, 80, 80)
- D Calc tensor Conv_39 (1, 128, 80, 80)
- D Calc tensor Relu_40 (1, 128, 80, 80)
- D Calc tensor MaxPool_41 (1, 128, 40, 40)
- D Calc tensor Conv_42 (1, 256, 40, 40)
- D Calc tensor Relu_43 (1, 256, 40, 40)
- D Calc tensor MaxPool_44 (1, 256, 20, 20)
- D Calc tensor Conv_45 (1, 512, 20, 20)
- D Calc tensor Relu_46 (1, 512, 20, 20)
- D Calc tensor Conv_47 (1, 512, 21, 21)
- D Calc tensor Relu_48 (1, 512, 21, 21)
- D Calc tensor MaxPool_49 (1, 512, 20, 20)
- D Calc tensor Conv_50 (1, 1024, 20, 20)
- D Calc tensor Relu_51 (1, 1024, 20, 20)
- D Calc tensor Conv_52 (1, 256, 20, 20)
- D Calc tensor Relu_53 (1, 256, 20, 20)
- D Calc tensor Conv_54 (1, 512, 20, 20)
- D Calc tensor Relu_55 (1, 512, 20, 20)
- D Calc tensor Conv_56 (1, 128, 20, 20)
- D Calc tensor Relu_57 (1, 128, 20, 20)
- D Calc tensor Initializer_75 (4,)
- D Calc tensor Resize_67 (1, 128, 40, 40)
- D Calc tensor Concat_68 (1, 384, 40, 40)
- D Calc tensor Conv_69 (1, 256, 40, 40)
- D Calc tensor Relu_output (1, 256, 40, 40)
- D import clients finished
- I build output layer attach_Relu_Relu_32:out0
- I build output layer attach_Relu_Relu_25:out0
- I Try match Relu_Relu_25:out0
- I Match r_relu [['Relu_Relu_25']] [['Relu']] to [['relu']]
- I Try match Relu_Relu_32:out0
- I Match r_relu [['Relu_Relu_32']] [['Relu']] to [['relu']]
- I Try match Conv_Conv_24:out0
- I Match r_conv [['Conv_Conv_24', 'Initializer_model.15.conv.weight', 'Initializer_model.15.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Conv_Conv_31:out0
- I Match r_conv [['Conv_Conv_31', 'Initializer_model.19.conv.weight', 'Initializer_model.19.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Relu_Relu_23:out0
- I Match r_relu [['Relu_Relu_23']] [['Relu']] to [['relu']]
- I Try match Concat_Concat_30:out0
- I Match concat_2 [['Concat_Concat_30']] [['Concat']] to [['concat']]
- I Try match Conv_Conv_22:out0
- I Match r_conv [['Conv_Conv_22', 'Initializer_model.14.conv.weight', 'Initializer_model.14.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Resize_Resize_29:out0
- I Match r_resize [['Resize_Resize_29', 'Constant_Constant_28', 'Initializer_75']] [['Resize', 'Constant', 'Constant_1']] to [['image_resize']]
- I Try match Relu_Relu_13:out0
- I Match r_relu [['Relu_Relu_13']] [['Relu']] to [['relu']]
- I Try match Relu_Relu_21:out0
- I Match r_relu [['Relu_Relu_21']] [['Relu']] to [['relu']]
- I Try match Relu_Relu_27:out0
- I Match r_relu [['Relu_Relu_27']] [['Relu']] to [['relu']]
- I Try match Conv_Conv_12:out0
- I Match r_conv [['Conv_Conv_12', 'Initializer_model.8.conv.weight', 'Initializer_model.8.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Conv_Conv_20:out0
- I Match r_conv [['Conv_Conv_20', 'Initializer_model.13.conv.weight', 'Initializer_model.13.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Conv_Conv_26:out0
- I Match r_conv [['Conv_Conv_26', 'Initializer_model.16.conv.weight', 'Initializer_model.16.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match MaxPool_MaxPool_11:out0
- I Match r_maxpool [['MaxPool_MaxPool_11']] [['MaxPool']] to [['pooling']]
- I Try match MaxPool_MaxPool_19:out0
- I Match r_maxpool [['MaxPool_MaxPool_19']] [['MaxPool']] to [['pooling']]
- I Try match Relu_Relu_18:out0
- I Match r_relu [['Relu_Relu_18']] [['Relu']] to [['relu']]
- I Try match Relu_Relu_10:out0
- I Match r_relu [['Relu_Relu_10']] [['Relu']] to [['relu']]
- I Try match Conv_Conv_17:out0
- I Match r_conv [['Conv_Conv_17', 'Initializer_model.11.conv.weight', 'Initializer_model.11.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Conv_Conv_9:out0
- I Match r_conv [['Conv_Conv_9', 'Initializer_model.6.conv.weight', 'Initializer_model.6.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Relu_Relu_16:out0
- I Match r_relu [['Relu_Relu_16']] [['Relu']] to [['relu']]
- I Try match MaxPool_MaxPool_8:out0
- I Match r_maxpool [['MaxPool_MaxPool_8']] [['MaxPool']] to [['pooling']]
- I Try match Conv_Conv_15:out0
- I Match r_conv [['Conv_Conv_15', 'Initializer_model.10.conv.weight', 'Initializer_model.10.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match Relu_Relu_7:out0
- I Match r_relu [['Relu_Relu_7']] [['Relu']] to [['relu']]
- I Try match MaxPool_MaxPool_14:out0
- I Match r_maxpool [['MaxPool_MaxPool_14']] [['MaxPool']] to [['pooling']]
- I Try match Conv_Conv_6:out0
- I Match r_conv [['Conv_Conv_6', 'Initializer_model.4.conv.weight', 'Initializer_model.4.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match MaxPool_MaxPool_5:out0
- I Match r_maxpool [['MaxPool_MaxPool_5']] [['MaxPool']] to [['pooling']]
- I Try match Relu_Relu_4:out0
- I Match r_relu [['Relu_Relu_4']] [['Relu']] to [['relu']]
- I Try match Conv_Conv_3:out0
- I Match r_conv [['Conv_Conv_3', 'Initializer_model.2.conv.weight', 'Initializer_model.2.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I Try match MaxPool_MaxPool_2:out0
- I Match r_maxpool [['MaxPool_MaxPool_2']] [['MaxPool']] to [['pooling']]
- I Try match Relu_Relu_1:out0
- I Match r_relu [['Relu_Relu_1']] [['Relu']] to [['relu']]
- I Try match Conv_Conv_0:out0
- I Match r_conv [['Conv_Conv_0', 'Initializer_model.0.conv.weight', 'Initializer_model.0.conv.bias']] [['Conv', 'Constant_0', 'Constant_1']] to [['convolution']]
- I build input layer images:out0
- D connect Conv_Conv_24_4 0 ~ Relu_Relu_25_2 0
- D connect Conv_Conv_31_5 0 ~ Relu_Relu_32_3 0
- D connect Relu_Relu_23_6 0 ~ Conv_Conv_24_4 0
- D connect Concat_Concat_30_7 0 ~ Conv_Conv_31_5 0
- D connect Conv_Conv_22_8 0 ~ Relu_Relu_23_6 0
- D connect Resize_Resize_29_9 0 ~ Concat_Concat_30_7 0
- D connect Relu_Relu_13_10 0 ~ Concat_Concat_30_7 1
- D connect Relu_Relu_21_11 0 ~ Conv_Conv_22_8 0
- D connect Relu_Relu_27_12 0 ~ Resize_Resize_29_9 0
- D connect Conv_Conv_12_13 0 ~ Relu_Relu_13_10 0
- D connect Conv_Conv_20_14 0 ~ Relu_Relu_21_11 0
- D connect Conv_Conv_26_15 0 ~ Relu_Relu_27_12 0
- D connect MaxPool_MaxPool_11_16 0 ~ Conv_Conv_12_13 0
- D connect MaxPool_MaxPool_19_17 0 ~ Conv_Conv_20_14 0
- D connect Relu_Relu_23_6 0 ~ Conv_Conv_26_15 0
- D connect Relu_Relu_10_19 0 ~ MaxPool_MaxPool_11_16 0
- D connect Relu_Relu_18_18 0 ~ MaxPool_MaxPool_19_17 0
- D connect Conv_Conv_17_20 0 ~ Relu_Relu_18_18 0
- D connect Conv_Conv_9_21 0 ~ Relu_Relu_10_19 0
- D connect Relu_Relu_16_22 0 ~ Conv_Conv_17_20 0
- D connect MaxPool_MaxPool_8_23 0 ~ Conv_Conv_9_21 0
- D connect Conv_Conv_15_24 0 ~ Relu_Relu_16_22 0
- D connect Relu_Relu_7_25 0 ~ MaxPool_MaxPool_8_23 0
- D connect MaxPool_MaxPool_14_26 0 ~ Conv_Conv_15_24 0
- D connect Conv_Conv_6_27 0 ~ Relu_Relu_7_25 0
- D connect Relu_Relu_13_10 0 ~ MaxPool_MaxPool_14_26 0
- D connect MaxPool_MaxPool_5_28 0 ~ Conv_Conv_6_27 0
- D connect Relu_Relu_4_29 0 ~ MaxPool_MaxPool_5_28 0
- D connect Conv_Conv_3_30 0 ~ Relu_Relu_4_29 0
- D connect MaxPool_MaxPool_2_31 0 ~ Conv_Conv_3_30 0
- D connect Relu_Relu_1_32 0 ~ MaxPool_MaxPool_2_31 0
- D connect Conv_Conv_0_33 0 ~ Relu_Relu_1_32 0
- D connect images_34 0 ~ Conv_Conv_0_33 0
- D connect Relu_Relu_32_3 0 ~ attach_Relu_Relu_32/out0_0 0
- D connect Relu_Relu_25_2 0 ~ attach_Relu_Relu_25/out0_1 0
- D Process images_34 ...
- D RKNN output shape(input): (1 640 640 3)
- D Tensor @images_34:out0 type: float32
- D Process Conv_Conv_0_33 ...
- D RKNN output shape(convolution): (1 640 640 16)
- D Tensor @Conv_Conv_0_33:out0 type: float32
- D Process Relu_Relu_1_32 ...
- D RKNN output shape(relu): (1 640 640 16)
- D Tensor @Relu_Relu_1_32:out0 type: float32
- D Process MaxPool_MaxPool_2_31 ...
- D RKNN output shape(pooling): (1 320 320 16)
- D Tensor @MaxPool_MaxPool_2_31:out0 type: float32
- D Process Conv_Conv_3_30 ...
- D RKNN output shape(convolution): (1 320 320 32)
- D Tensor @Conv_Conv_3_30:out0 type: float32
- D Process Relu_Relu_4_29 ...
- D RKNN output shape(relu): (1 320 320 32)
- D Tensor @Relu_Relu_4_29:out0 type: float32
- D Process MaxPool_MaxPool_5_28 ...
- D RKNN output shape(pooling): (1 160 160 32)
- D Tensor @MaxPool_MaxPool_5_28:out0 type: float32
- D Process Conv_Conv_6_27 ...
- D RKNN output shape(convolution): (1 160 160 64)
- D Tensor @Conv_Conv_6_27:out0 type: float32
- D Process Relu_Relu_7_25 ...
- D RKNN output shape(relu): (1 160 160 64)
- D Tensor @Relu_Relu_7_25:out0 type: float32
- D Process MaxPool_MaxPool_8_23 ...
- D RKNN output shape(pooling): (1 80 80 64)
- D Tensor @MaxPool_MaxPool_8_23:out0 type: float32
- D Process Conv_Conv_9_21 ...
- D RKNN output shape(convolution): (1 80 80 128)
- D Tensor @Conv_Conv_9_21:out0 type: float32
- D Process Relu_Relu_10_19 ...
- D RKNN output shape(relu): (1 80 80 128)
- D Tensor @Relu_Relu_10_19:out0 type: float32
- D Process MaxPool_MaxPool_11_16 ...
- D RKNN output shape(pooling): (1 40 40 128)
- D Tensor @MaxPool_MaxPool_11_16:out0 type: float32
- D Process Conv_Conv_12_13 ...
- D RKNN output shape(convolution): (1 40 40 256)
- D Tensor @Conv_Conv_12_13:out0 type: float32
- D Process Relu_Relu_13_10 ...
- D RKNN output shape(relu): (1 40 40 256)
- D Tensor @Relu_Relu_13_10:out0 type: float32
- D Process MaxPool_MaxPool_14_26 ...
- D RKNN output shape(pooling): (1 20 20 256)
- D Tensor @MaxPool_MaxPool_14_26:out0 type: float32
- D Process Conv_Conv_15_24 ...
- D RKNN output shape(convolution): (1 20 20 512)
- D Tensor @Conv_Conv_15_24:out0 type: float32
- D Process Relu_Relu_16_22 ...
- D RKNN output shape(relu): (1 20 20 512)
- D Tensor @Relu_Relu_16_22:out0 type: float32
- D Process Conv_Conv_17_20 ...
- D RKNN output shape(convolution): (1 21 21 512)
- D Tensor @Conv_Conv_17_20:out0 type: float32
- D Process Relu_Relu_18_18 ...
- D RKNN output shape(relu): (1 21 21 512)
- D Tensor @Relu_Relu_18_18:out0 type: float32
- D Process MaxPool_MaxPool_19_17 ...
- D RKNN output shape(pooling): (1 20 20 512)
- D Tensor @MaxPool_MaxPool_19_17:out0 type: float32
- D Process Conv_Conv_20_14 ...
- D RKNN output shape(convolution): (1 20 20 1024)
- D Tensor @Conv_Conv_20_14:out0 type: float32
- D Process Relu_Relu_21_11 ...
- D RKNN output shape(relu): (1 20 20 1024)
- D Tensor @Relu_Relu_21_11:out0 type: float32
- D Process Conv_Conv_22_8 ...
- D RKNN output shape(convolution): (1 20 20 256)
- D Tensor @Conv_Conv_22_8:out0 type: float32
- D Process Relu_Relu_23_6 ...
- D RKNN output shape(relu): (1 20 20 256)
- D Tensor @Relu_Relu_23_6:out0 type: float32
- D Process Conv_Conv_26_15 ...
- D RKNN output shape(convolution): (1 20 20 128)
- D Tensor @Conv_Conv_26_15:out0 type: float32
- D Process Relu_Relu_27_12 ...
- D RKNN output shape(relu): (1 20 20 128)
- D Tensor @Relu_Relu_27_12:out0 type: float32
- D Process Resize_Resize_29_9 ...
- D RKNN output shape(image_resize): (1 40 40 128)
- D Tensor @Resize_Resize_29_9:out0 type: float32
- D Process Concat_Concat_30_7 ...
- D RKNN output shape(concat): (1 40 40 384)
- D Tensor @Concat_Concat_30_7:out0 type: float32
- D Process Conv_Conv_31_5 ...
- D RKNN output shape(convolution): (1 40 40 256)
- D Tensor @Conv_Conv_31_5:out0 type: float32
- D Process Relu_Relu_32_3 ...
- D RKNN output shape(relu): (1 40 40 256)
- D Tensor @Relu_Relu_32_3:out0 type: float32
- D Process attach_Relu_Relu_32/out0_0 ...
- D RKNN output shape(output): (1 40 40 256)
- D Tensor @attach_Relu_Relu_32/out0_0:out0 type: float32
- D Process Conv_Conv_24_4 ...
- D RKNN output shape(convolution): (1 20 20 512)
- D Tensor @Conv_Conv_24_4:out0 type: float32
- D Process Relu_Relu_25_2 ...
- D RKNN output shape(relu): (1 20 20 512)
- D Tensor @Relu_Relu_25_2:out0 type: float32
- D Process attach_Relu_Relu_25/out0_1 ...
- D RKNN output shape(output): (1 20 20 512)
- D Tensor @attach_Relu_Relu_25/out0_1:out0 type: float32
- I Build torch-jit-export complete.
- I Start C2T Switcher...
- D Optimizing network with broadcast_op
- D convert Concat_Concat_30_7(concat) axis 1 to 3
- I End C2T Switcher...
- D Optimizing network with force_1d_tensor, swapper, merge_duplicate_quantize_dequantize, merge_layer, auto_fill_bn, auto_fill_l2normalizescale, auto_fill_instancenormalize, resize_nearest_transformer, eltwise_transform, auto_fill_multiply, auto_fill_zero_bias, proposal_opt_import, special_add_to_conv2d
- I End importing onnx...
- W The RKNN Model generated can not run on simulator when pre_compile is True.
- I Generate input meta ...
- I Load input meta
- I Generate input meta ...
- D import clients finished
- I Load net...
- I Load data...
- I Load input meta
- I Start quantization...
- D import clients finished
- D iterations: 7, batch_size: 16
- I Quantization start...
- D set up a quantize net
- D *********** Setup input meta ***********
- D import clients finished
- D *********** Setup database (1) ***********
- D Setup provider layer "text_input_layer":
- D Lids: ['images_34']
- D Shapes: [[16, 640, 640, 3]]
- D Data types: ['float32']
- D Sparse tensors: []
- D Tensor names(H5FS only): []
- D Add preprocess "[('reverse_channel', False), ('mean', [0, 0, 0]), ('scale', [0.00392156862745098, 0.00392156862745098, 0.00392156862745098]), ('preproc_node_params', ordereddict([('add_preproc_node', False), ('preproc_type', 'IMAGE_RGB'), ('preproc_perm', [0, 1, 2, 3])]))]" for "images_34"
- D *********** Setup input meta complete ***********
- D Process images_34 ...
- D RKNN output shape(input): (16 640 640 3)
- D Tensor @images_34:out0 type: asymmetric_affine
- D Real output shape: (16, 640, 640, 3)
- D Process Conv_Conv_0_33 ...
- D RKNN output shape(convolution): (16 640 640 16)
- D Tensor @Conv_Conv_0_33:out0 type: asymmetric_affine
- D Real output shape: (16, 640, 640, 16)
- D Process Relu_Relu_1_32 ...
- D RKNN output shape(relu): (16 640 640 16)
- D Tensor @Relu_Relu_1_32:out0 type: asymmetric_affine
- D Real output shape: (16, 640, 640, 16)
- D Process MaxPool_MaxPool_2_31 ...
- D RKNN output shape(pooling): (16 320 320 16)
- D Tensor @MaxPool_MaxPool_2_31:out0 type: asymmetric_affine
- D Real output shape: (16, 320, 320, 16)
- D Process Conv_Conv_3_30 ...
- D RKNN output shape(convolution): (16 320 320 32)
- D Tensor @Conv_Conv_3_30:out0 type: asymmetric_affine
- D Real output shape: (16, 320, 320, 32)
- D Process Relu_Relu_4_29 ...
- D RKNN output shape(relu): (16 320 320 32)
- D Tensor @Relu_Relu_4_29:out0 type: asymmetric_affine
- D Real output shape: (16, 320, 320, 32)
- D Process MaxPool_MaxPool_5_28 ...
- D RKNN output shape(pooling): (16 160 160 32)
- D Tensor @MaxPool_MaxPool_5_28:out0 type: asymmetric_affine
- D Real output shape: (16, 160, 160, 32)
- D Process Conv_Conv_6_27 ...
- D RKNN output shape(convolution): (16 160 160 64)
- D Tensor @Conv_Conv_6_27:out0 type: asymmetric_affine
- D Real output shape: (16, 160, 160, 64)
- D Process Relu_Relu_7_25 ...
- D RKNN output shape(relu): (16 160 160 64)
- D Tensor @Relu_Relu_7_25:out0 type: asymmetric_affine
- D Real output shape: (16, 160, 160, 64)
- D Process MaxPool_MaxPool_8_23 ...
- D RKNN output shape(pooling): (16 80 80 64)
- D Tensor @MaxPool_MaxPool_8_23:out0 type: asymmetric_affine
- D Real output shape: (16, 80, 80, 64)
- D Process Conv_Conv_9_21 ...
- D RKNN output shape(convolution): (16 80 80 128)
- D Tensor @Conv_Conv_9_21:out0 type: asymmetric_affine
- D Real output shape: (16, 80, 80, 128)
- D Process Relu_Relu_10_19 ...
- D RKNN output shape(relu): (16 80 80 128)
- D Tensor @Relu_Relu_10_19:out0 type: asymmetric_affine
- D Real output shape: (16, 80, 80, 128)
- D Process MaxPool_MaxPool_11_16 ...
- D RKNN output shape(pooling): (16 40 40 128)
- D Tensor @MaxPool_MaxPool_11_16:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 128)
- D Process Conv_Conv_12_13 ...
- D RKNN output shape(convolution): (16 40 40 256)
- D Tensor @Conv_Conv_12_13:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 256)
- D Process Relu_Relu_13_10 ...
- D RKNN output shape(relu): (16 40 40 256)
- D Tensor @Relu_Relu_13_10:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 256)
- D Process MaxPool_MaxPool_14_26 ...
- D RKNN output shape(pooling): (16 20 20 256)
- D Tensor @MaxPool_MaxPool_14_26:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 256)
- D Process Conv_Conv_15_24 ...
- D RKNN output shape(convolution): (16 20 20 512)
- D Tensor @Conv_Conv_15_24:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 512)
- D Process Relu_Relu_16_22 ...
- D RKNN output shape(relu): (16 20 20 512)
- D Tensor @Relu_Relu_16_22:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 512)
- D Process Conv_Conv_17_20 ...
- D RKNN output shape(convolution): (16 21 21 512)
- D Tensor @Conv_Conv_17_20:out0 type: asymmetric_affine
- D Real output shape: (16, 21, 21, 512)
- D Process Relu_Relu_18_18 ...
- D RKNN output shape(relu): (16 21 21 512)
- D Tensor @Relu_Relu_18_18:out0 type: asymmetric_affine
- D Real output shape: (16, 21, 21, 512)
- D Process MaxPool_MaxPool_19_17 ...
- D RKNN output shape(pooling): (16 20 20 512)
- D Tensor @MaxPool_MaxPool_19_17:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 512)
- D Process Conv_Conv_20_14 ...
- D RKNN output shape(convolution): (16 20 20 1024)
- D Tensor @Conv_Conv_20_14:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 1024)
- D Process Relu_Relu_21_11 ...
- D RKNN output shape(relu): (16 20 20 1024)
- D Tensor @Relu_Relu_21_11:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 1024)
- D Process Conv_Conv_22_8 ...
- D RKNN output shape(convolution): (16 20 20 256)
- D Tensor @Conv_Conv_22_8:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 256)
- D Process Relu_Relu_23_6 ...
- D RKNN output shape(relu): (16 20 20 256)
- D Tensor @Relu_Relu_23_6:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 256)
- D Process Conv_Conv_26_15 ...
- D RKNN output shape(convolution): (16 20 20 128)
- D Tensor @Conv_Conv_26_15:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 128)
- D Process Relu_Relu_27_12 ...
- D RKNN output shape(relu): (16 20 20 128)
- D Tensor @Relu_Relu_27_12:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 128)
- D Process Resize_Resize_29_9 ...
- D RKNN output shape(image_resize): (16 40 40 128)
- D Tensor @Resize_Resize_29_9:out0 type: float32
- D Real output shape: (16, 40, 40, 128)
- D Process Concat_Concat_30_7 ...
- D RKNN output shape(concat): (16 40 40 384)
- D Tensor @Concat_Concat_30_7:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 384)
- D Process Conv_Conv_31_5 ...
- D RKNN output shape(convolution): (16 40 40 256)
- D Tensor @Conv_Conv_31_5:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 256)
- D Process Relu_Relu_32_3 ...
- D RKNN output shape(relu): (16 40 40 256)
- D Tensor @Relu_Relu_32_3:out0 type: asymmetric_affine
- D Real output shape: (16, 40, 40, 256)
- D Process attach_Relu_Relu_32/out0_0 ...
- D RKNN output shape(output): (16 40 40 256)
- D Tensor @attach_Relu_Relu_32/out0_0:out0 type: float32
- D Real output shape: (16, 40, 40, 256)
- D Process Conv_Conv_24_4 ...
- D RKNN output shape(convolution): (16 20 20 512)
- D Tensor @Conv_Conv_24_4:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 512)
- D Process Relu_Relu_25_2 ...
- D RKNN output shape(relu): (16 20 20 512)
- D Tensor @Relu_Relu_25_2:out0 type: asymmetric_affine
- D Real output shape: (16, 20, 20, 512)
- D Process attach_Relu_Relu_25/out0_1 ...
- D RKNN output shape(output): (16 20 20 512)
- D Tensor @attach_Relu_Relu_25/out0_1:out0 type: float32
- D Real output shape: (16, 20, 20, 512)
- I Build torch-jit-export complete.
- I Analyze activation range of layer: Conv_Conv_0_33
- I Analyze activation range of layer: Relu_Relu_1_32
- I Analyze activation range of layer: MaxPool_MaxPool_2_31
- I Analyze activation range of layer: Conv_Conv_3_30
- I Analyze activation range of layer: Relu_Relu_4_29
- I Analyze activation range of layer: MaxPool_MaxPool_5_28
- I Analyze activation range of layer: Conv_Conv_6_27
- I Analyze activation range of layer: Relu_Relu_7_25
- I Analyze activation range of layer: MaxPool_MaxPool_8_23
- I Analyze activation range of layer: Conv_Conv_9_21
- I Analyze activation range of layer: Relu_Relu_10_19
- I Analyze activation range of layer: MaxPool_MaxPool_11_16
- I Analyze activation range of layer: Conv_Conv_12_13
- I Analyze activation range of layer: Relu_Relu_13_10
- I Analyze activation range of layer: MaxPool_MaxPool_14_26
- I Analyze activation range of layer: Conv_Conv_15_24
- I Analyze activation range of layer: Relu_Relu_16_22
- I Analyze activation range of layer: Conv_Conv_17_20
- I Analyze activation range of layer: Relu_Relu_18_18
- I Analyze activation range of layer: MaxPool_MaxPool_19_17
- I Analyze activation range of layer: Conv_Conv_20_14
- I Analyze activation range of layer: Relu_Relu_21_11
- I Analyze activation range of layer: Conv_Conv_22_8
- I Analyze activation range of layer: Relu_Relu_23_6
- I Analyze activation range of layer: Conv_Conv_24_4
- I Analyze activation range of layer: Conv_Conv_26_15
- I Analyze activation range of layer: Relu_Relu_25_2
- I Analyze activation range of layer: Relu_Relu_27_12
- I Analyze activation range of layer: attach_Relu_Relu_25/out0_1
- I Analyze activation range of layer: Resize_Resize_29_9
- I Analyze activation range of layer: Concat_Concat_30_7
- I Analyze activation range of layer: Conv_Conv_31_5
- I Analyze activation range of layer: Relu_Relu_32_3
- I Analyze activation range of layer: attach_Relu_Relu_32/out0_0
- I Running 7 iterations
- D 0(14.28%), Queue size 16
- D 1(28.57%), Queue size 16
- D 2(42.85%), Queue size 16
- D 3(57.14%), Queue size 16
- D 4(71.42%), Queue size 16
- D 5(85.71%), Queue size 16
- D 6(100.00%), Queue size 16
- I Queue cancelled.
- D Quantize tensor @Relu_Relu_25_2:out0.
- D Quantize tensor @Relu_Relu_32_3:out0.
- D Quantize tensor @Conv_Conv_24_4:out0.
- D Quantize tensor @Conv_Conv_31_5:out0.
- D Quantize tensor @Relu_Relu_23_6:out0.
- D Quantize tensor @Concat_Concat_30_7:out0.
- D Quantize tensor @Conv_Conv_22_8:out0.
- D Quantize tensor @Relu_Relu_13_10:out0.
- D Quantize tensor @Relu_Relu_21_11:out0.
- D Quantize tensor @Relu_Relu_27_12:out0.
- D Quantize tensor @Conv_Conv_12_13:out0.
- D Quantize tensor @Conv_Conv_20_14:out0.
- D Quantize tensor @Conv_Conv_26_15:out0.
- D Quantize tensor @MaxPool_MaxPool_11_16:out0.
- D Quantize tensor @MaxPool_MaxPool_19_17:out0.
- D Quantize tensor @Relu_Relu_18_18:out0.
- D Quantize tensor @Relu_Relu_10_19:out0.
- D Quantize tensor @Conv_Conv_17_20:out0.
- D Quantize tensor @Conv_Conv_9_21:out0.
- D Quantize tensor @Relu_Relu_16_22:out0.
- D Quantize tensor @MaxPool_MaxPool_8_23:out0.
- D Quantize tensor @Conv_Conv_15_24:out0.
- D Quantize tensor @Relu_Relu_7_25:out0.
- D Quantize tensor @MaxPool_MaxPool_14_26:out0.
- D Quantize tensor @Conv_Conv_6_27:out0.
- D Quantize tensor @MaxPool_MaxPool_5_28:out0.
- D Quantize tensor @Relu_Relu_4_29:out0.
- D Quantize tensor @Conv_Conv_3_30:out0.
- D Quantize tensor @MaxPool_MaxPool_2_31:out0.
- D Quantize tensor @Relu_Relu_1_32:out0.
- D Quantize tensor @Conv_Conv_0_33:out0.
- D Quantize tensor @images_34:out0.
- D Quantize tensor @Conv_Conv_24_4:weight.
- D Quantize tensor @Conv_Conv_31_5:weight.
- D Quantize tensor @Conv_Conv_22_8:weight.
- D Quantize tensor @Conv_Conv_12_13:weight.
- D Quantize tensor @Conv_Conv_20_14:weight.
- D Quantize tensor @Conv_Conv_26_15:weight.
- D Quantize tensor @Conv_Conv_17_20:weight.
- D Quantize tensor @Conv_Conv_9_21:weight.
- D Quantize tensor @Conv_Conv_15_24:weight.
- D Quantize tensor @Conv_Conv_6_27:weight.
- D Quantize tensor @Conv_Conv_3_30:weight.
- D Quantize tensor @Conv_Conv_0_33:weight.
- D Quantize tensor @Conv_Conv_24_4:bias.
- D Quantize tensor @Conv_Conv_31_5:bias.
- D Quantize tensor @Conv_Conv_22_8:bias.
- D Quantize tensor @Conv_Conv_12_13:bias.
- D Quantize tensor @Conv_Conv_20_14:bias.
- D Quantize tensor @Conv_Conv_26_15:bias.
- D Quantize tensor @Conv_Conv_17_20:bias.
- D Quantize tensor @Conv_Conv_9_21:bias.
- D Quantize tensor @Conv_Conv_15_24:bias.
- D Quantize tensor @Conv_Conv_6_27:bias.
- D Quantize tensor @Conv_Conv_3_30:bias.
- D Quantize tensor @Conv_Conv_0_33:bias.
- I Clean.
- D Optimizing network with align_quantize, broadcast_quantize, qnt_adjust_coef, qnt_adjust_param
- D Align @Relu_Relu_27_12:out0 scale to [0.05264517]
- D Align @Concat_Concat_30_7:out0 scale to [0.05264517]
- D Align @Relu_Relu_13_10:out0 scale to [0.05264517]
- D Quantize tensor(@attach_Relu_Relu_25/out0_1:out0) with tensor(@Relu_Relu_25_2:out0)
- D Quantize tensor(@Resize_Resize_29_9:out0) with tensor(@Relu_Relu_27_12:out0)
- D Quantize tensor(@attach_Relu_Relu_32/out0_0:out0) with tensor(@Relu_Relu_32_3:out0)
- I Conv_Conv_0_33 merged relu nexted.
- I Conv_Conv_3_30 merged relu nexted.
- I Conv_Conv_6_27 merged relu nexted.
- I Conv_Conv_9_21 merged relu nexted.
- I Conv_Conv_15_24 merged relu nexted.
- I Conv_Conv_17_20 merged relu nexted.
- I Conv_Conv_20_14 merged relu nexted.
- I Conv_Conv_22_8 merged relu nexted.
- I Conv_Conv_24_4 merged relu nexted.
- I Conv_Conv_26_15 merged relu nexted.
- I Conv_Conv_31_5 merged relu nexted.
- I Quantization complete.
- I End quantization...
- D import clients finished
- I Load net...
- I Load data...
- I Load quantization tensor table
- I Load input meta
- D Process images_34 ...
- D RKNN output shape(input): (1 640 640 3)
- D Tensor @images_34:out0 type: asymmetric_affine
- D Process Conv_Conv_0_33 ...
- D RKNN output shape(convolution): (1 640 640 16)
- D Tensor @Conv_Conv_0_33:out0 type: asymmetric_affine
- D Process Relu_Relu_1_32 ...
- D RKNN output shape(relu): (1 640 640 16)
- D Tensor @Relu_Relu_1_32:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_2_31 ...
- D RKNN output shape(pooling): (1 320 320 16)
- D Tensor @MaxPool_MaxPool_2_31:out0 type: asymmetric_affine
- D Process Conv_Conv_3_30 ...
- D RKNN output shape(convolution): (1 320 320 32)
- D Tensor @Conv_Conv_3_30:out0 type: asymmetric_affine
- D Process Relu_Relu_4_29 ...
- D RKNN output shape(relu): (1 320 320 32)
- D Tensor @Relu_Relu_4_29:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_5_28 ...
- D RKNN output shape(pooling): (1 160 160 32)
- D Tensor @MaxPool_MaxPool_5_28:out0 type: asymmetric_affine
- D Process Conv_Conv_6_27 ...
- D RKNN output shape(convolution): (1 160 160 64)
- D Tensor @Conv_Conv_6_27:out0 type: asymmetric_affine
- D Process Relu_Relu_7_25 ...
- D RKNN output shape(relu): (1 160 160 64)
- D Tensor @Relu_Relu_7_25:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_8_23 ...
- D RKNN output shape(pooling): (1 80 80 64)
- D Tensor @MaxPool_MaxPool_8_23:out0 type: asymmetric_affine
- D Process Conv_Conv_9_21 ...
- D RKNN output shape(convolution): (1 80 80 128)
- D Tensor @Conv_Conv_9_21:out0 type: asymmetric_affine
- D Process Relu_Relu_10_19 ...
- D RKNN output shape(relu): (1 80 80 128)
- D Tensor @Relu_Relu_10_19:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_11_16 ...
- D RKNN output shape(pooling): (1 40 40 128)
- D Tensor @MaxPool_MaxPool_11_16:out0 type: asymmetric_affine
- D Process Conv_Conv_12_13 ...
- D RKNN output shape(convolution): (1 40 40 256)
- D Tensor @Conv_Conv_12_13:out0 type: asymmetric_affine
- D Process Relu_Relu_13_10 ...
- D RKNN output shape(relu): (1 40 40 256)
- D Tensor @Relu_Relu_13_10:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_14_26 ...
- D RKNN output shape(pooling): (1 20 20 256)
- D Tensor @MaxPool_MaxPool_14_26:out0 type: asymmetric_affine
- D Process Conv_Conv_15_24 ...
- D RKNN output shape(convolution): (1 20 20 512)
- D Tensor @Conv_Conv_15_24:out0 type: asymmetric_affine
- D Process Relu_Relu_16_22 ...
- D RKNN output shape(relu): (1 20 20 512)
- D Tensor @Relu_Relu_16_22:out0 type: asymmetric_affine
- D Process Conv_Conv_17_20 ...
- D RKNN output shape(convolution): (1 21 21 512)
- D Tensor @Conv_Conv_17_20:out0 type: asymmetric_affine
- D Process Relu_Relu_18_18 ...
- D RKNN output shape(relu): (1 21 21 512)
- D Tensor @Relu_Relu_18_18:out0 type: asymmetric_affine
- D Process MaxPool_MaxPool_19_17 ...
- D RKNN output shape(pooling): (1 20 20 512)
- D Tensor @MaxPool_MaxPool_19_17:out0 type: asymmetric_affine
- D Process Conv_Conv_20_14 ...
- D RKNN output shape(convolution): (1 20 20 1024)
- D Tensor @Conv_Conv_20_14:out0 type: asymmetric_affine
- D Process Relu_Relu_21_11 ...
- D RKNN output shape(relu): (1 20 20 1024)
- D Tensor @Relu_Relu_21_11:out0 type: asymmetric_affine
- D Process Conv_Conv_22_8 ...
- D RKNN output shape(convolution): (1 20 20 256)
- D Tensor @Conv_Conv_22_8:out0 type: asymmetric_affine
- D Process Relu_Relu_23_6 ...
- D RKNN output shape(relu): (1 20 20 256)
- D Tensor @Relu_Relu_23_6:out0 type: asymmetric_affine
- D Process Conv_Conv_26_15 ...
- D RKNN output shape(convolution): (1 20 20 128)
- D Tensor @Conv_Conv_26_15:out0 type: asymmetric_affine
- D Process Relu_Relu_27_12 ...
- D RKNN output shape(relu): (1 20 20 128)
- D Tensor @Relu_Relu_27_12:out0 type: asymmetric_affine
- D Process Resize_Resize_29_9 ...
- D RKNN output shape(image_resize): (1 40 40 128)
- D Tensor @Resize_Resize_29_9:out0 type: asymmetric_affine
- D Process Concat_Concat_30_7 ...
- D RKNN output shape(concat): (1 40 40 384)
- D Tensor @Concat_Concat_30_7:out0 type: asymmetric_affine
- D Process Conv_Conv_31_5 ...
- D RKNN output shape(convolution): (1 40 40 256)
- D Tensor @Conv_Conv_31_5:out0 type: asymmetric_affine
- D Process Relu_Relu_32_3 ...
- D RKNN output shape(relu): (1 40 40 256)
- D Tensor @Relu_Relu_32_3:out0 type: asymmetric_affine
- D Process attach_Relu_Relu_32/out0_0 ...
- D RKNN output shape(output): (1 40 40 256)
- D Tensor @attach_Relu_Relu_32/out0_0:out0 type: asymmetric_affine
- D Process Conv_Conv_24_4 ...
- D RKNN output shape(convolution): (1 20 20 512)
- D Tensor @Conv_Conv_24_4:out0 type: asymmetric_affine
- D Process Relu_Relu_25_2 ...
- D RKNN output shape(relu): (1 20 20 512)
- D Tensor @Relu_Relu_25_2:out0 type: asymmetric_affine
- D Process attach_Relu_Relu_25/out0_1 ...
- D RKNN output shape(output): (1 20 20 512)
- D Tensor @attach_Relu_Relu_25/out0_1:out0 type: asymmetric_affine
- I Build torch-jit-export complete.
- I Initialzing network optimizer by /home/victor/.local/lib/python3.6/site-packages/rknn/config/npu_config/PUMA ...
- D Optimizing network with merge_ximum, qnt_adjust_coef, multiply_transform, add_extra_io, format_input_ops, auto_fill_zero_bias, conv_kernel_transform, twod_op_transform, strip_op, extend_unstack_split, merge_layer, transform_layer, broadcast_op, strip_op, auto_fill_reshape_zero, adjust_output_attrs, insert_dtype_converter
- I Conv_Conv_0_33 merged relu nexted.
- I Conv_Conv_3_30 merged relu nexted.
- I Conv_Conv_6_27 merged relu nexted.
- I Conv_Conv_9_21 merged relu nexted.
- I Conv_Conv_15_24 merged relu nexted.
- I Conv_Conv_17_20 merged relu nexted.
- I Conv_Conv_20_14 merged relu nexted.
- I Conv_Conv_22_8 merged relu nexted.
- I Conv_Conv_24_4 merged relu nexted.
- I Conv_Conv_26_15 merged relu nexted.
- I Conv_Conv_31_5 merged relu nexted.
- I Start T2C Switcher...
- D Optimizing network with broadcast_op, t2c_fc
- D convert Concat_Concat_30_7(concat) axis 3 to 1
- I End T2C Switcher...
- D Optimizing network with merge_ximum, qnt_adjust_coef, multiply_transform, add_extra_io, format_input_ops, auto_fill_zero_bias, conv_kernel_transform, twod_op_transform, strip_op, extend_unstack_split, merge_layer, transform_layer, broadcast_op, strip_op, auto_fill_reshape_zero, adjust_output_attrs, insert_dtype_converter
- I Conv_Conv_0_33 merged relu nexted.
- I Conv_Conv_3_30 merged relu nexted.
- I Conv_Conv_6_27 merged relu nexted.
- I Conv_Conv_9_21 merged relu nexted.
- I Conv_Conv_15_24 merged relu nexted.
- I Conv_Conv_17_20 merged relu nexted.
- I Conv_Conv_20_14 merged relu nexted.
- I Conv_Conv_22_8 merged relu nexted.
- I Conv_Conv_24_4 merged relu nexted.
- I Conv_Conv_26_15 merged relu nexted.
- I Conv_Conv_31_5 merged relu nexted.
- D Optimizing network with conv_1xn_transform, proposal_opt, c2drv_convert_axis, c2drv_convert_shape, c2drv_convert_array, c2drv_cast_dtype, c2drv_trans_data
- I Building data ...
- D /home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION
- D /home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/x64_linux:/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/x64_linux/RKm
- I Packing data ...
- D Packing Conv_Conv_0_33 ...
- D Quantize @Conv_Conv_0_33:bias to asymmetric_affine.
- D Quantize @Conv_Conv_0_33:weight to asymmetric_affine.
- D Packing Conv_Conv_12_13 ...
- D Quantize @Conv_Conv_12_13:bias to asymmetric_affine.
- D Quantize @Conv_Conv_12_13:weight to asymmetric_affine.
- D Packing Conv_Conv_15_24 ...
- D Quantize @Conv_Conv_15_24:bias to asymmetric_affine.
- D Quantize @Conv_Conv_15_24:weight to asymmetric_affine.
- D Packing Conv_Conv_17_20 ...
- D Quantize @Conv_Conv_17_20:bias to asymmetric_affine.
- D Quantize @Conv_Conv_17_20:weight to asymmetric_affine.
- D Packing Conv_Conv_20_14 ...
- D Quantize @Conv_Conv_20_14:bias to asymmetric_affine.
- D Quantize @Conv_Conv_20_14:weight to asymmetric_affine.
- D Packing Conv_Conv_22_8 ...
- D Quantize @Conv_Conv_22_8:bias to asymmetric_affine.
- D Quantize @Conv_Conv_22_8:weight to asymmetric_affine.
- D Packing Conv_Conv_24_4 ...
- D Quantize @Conv_Conv_24_4:bias to asymmetric_affine.
- D Quantize @Conv_Conv_24_4:weight to asymmetric_affine.
- D Packing Conv_Conv_26_15 ...
- D Quantize @Conv_Conv_26_15:bias to asymmetric_affine.
- D Quantize @Conv_Conv_26_15:weight to asymmetric_affine.
- D Packing Conv_Conv_31_5 ...
- D Quantize @Conv_Conv_31_5:bias to asymmetric_affine.
- D Quantize @Conv_Conv_31_5:weight to asymmetric_affine.
- D Packing Conv_Conv_3_30 ...
- D Quantize @Conv_Conv_3_30:bias to asymmetric_affine.
- D Quantize @Conv_Conv_3_30:weight to asymmetric_affine.
- D Packing Conv_Conv_6_27 ...
- D Quantize @Conv_Conv_6_27:bias to asymmetric_affine.
- D Quantize @Conv_Conv_6_27:weight to asymmetric_affine.
- D Packing Conv_Conv_9_21 ...
- D Quantize @Conv_Conv_9_21:bias to asymmetric_affine.
- D Quantize @Conv_Conv_9_21:weight to asymmetric_affine.
- D Generate fake input
- D Compiling start...
- D gcc -Wall -std=c++0x -I. -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/CL -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/VX -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ovxlib -D__linux__ -DLINUX -O3 -c rknn_torchjitexport.c
- D gcc -Wall -std=c++0x -I. -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/CL -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/VX -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ovxlib -D__linux__ -DLINUX -O3 -c rknn_pre_process.c
- D gcc -Wall -std=c++0x -I. -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/CL -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/VX -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ovxlib -D__linux__ -DLINUX -O3 -c main.c
- D gcc -Wall -std=c++0x -I. -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/CL -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/VX -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ovxlib -D__linux__ -DLINUX -O3 -c rknn_post_process.c
- D gcc -Wall -std=c++0x -I. -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/CL -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/VX -I/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/include/ovxlib -D__linux__ -DLINUX -O3 -O3 rknn_torchjitexport.o rknn_pre_process.o main.o rknn_post_process.o -L/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib -lOpenVX -lOpenVXU -lCLC -lVSC -lGAL -lovxlib -lEmulator -lvdtproxy -L/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/RKm -lOpenVX -lOpenVXU -lCLC -lVSC -lGAL -lovxlib -lEmulator -lvdtproxy -L/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/x64_linux -lOpenVX -lOpenVXU -lCLC -lVSC -lGAL -lovxlib -lEmulator -lvdtproxy -L/home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/x64_linux/RKm -lOpenVX -lOpenVXU -lCLC -lVSC -lGAL -lovxlib -lEmulator -lvdtproxy /home/victor/.local/lib/python3.6/site-packages/rknn/simulator/LION/lib/x64_linux/libjpeg.a -o gen_nbg
- D Compiling finished.
- D Generate pre-compile model data start...
- D Create Neural Network: 91ms or 91039us
- D Verify...
- D Verify Graph: 2590ms or 2590676us
- D Start run graph [1] times...
- D Run the 1 time: 1.00ms or 1461.00us
- D vxProcessGraph execution time:
- D Total 1.00ms or 1466.00us
- D Average 1.47ms or 1466.00us
- D --- Top5 ---
- D 0: 0.000000
- D 1: 0.000000
- D 2: 0.000000
- D 3: 0.000000
- D 4: 0.000000
- D Warning: NN VipSram is NULL, but SRAM is enabled!
- D Generate pre-compile model data finished.
- I Dump nbg input meta to /tmp/tmpqtwxevhd/nbg_unify/nbg_meta.json
- D output tensor id = 0, name = Relu_Relu_32/out0_0
- D output tensor id = 1, name = Relu_Relu_25/out0_1
- D input tensor id = 2, name = images_34
- I Build config finished.
|
|
收藏
闽公网安备 35010202001061号