c++ version of yolov7-mask with ncnn
mkdir build
cd build
cmake ..
make
./yolov7_mask_ncnn ../imgs/horses.jpg
https://github.com/Tencent/ncnn
https://github.com/WongKinYiu/yolov7/tree/mask
https://github.com/pytorch/vision
大佬有个问题我自己训练的yolov7 人体检测模型 在用ncnn推理时 输出层的维度 h,w,c都为0是啥情况.
可以分享下您这里的onnx模型吗 看下结构
how to export the yolov7_mask to onnx or ncnn ?
1、模型转换，怎么pt转ncnn模型
2、后处理，各层都是什么意思，怎么解析的
请问可以给出opencv和protobuf的安装和版本细节吗，谢谢！
can you give me a way to convert model .pt to ncnn format ? thanks
I had
#link_directories(${CMAKE_CURRENT_SOURCE_DIR}/lib)
add:
include_directories(~~/CLionProjects/ncnn/build/include)
link_directories(~/CLionProjects/ncnn/build/src)
in CMakeLists.txt & build complited
Convolution Conv_0 18.39ms | [640, 448, 3 *1] -> [640, 448, 4 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_2 5.30ms | [640, 448, 4 *8] -> [640, 448, 4 *8]
Convolution Conv_3 42.14ms | [640, 448, 4 *8] -> [320, 224, 8 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_5 1.51ms | [320, 224, 8 *8] -> [320, 224, 8 *8]
Convolution Conv_6 21.13ms | [320, 224, 8 *8] -> [320, 224, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_8 2.12ms | [320, 224, 8 *8] -> [320, 224, 8 *8]
Convolution Conv_9 23.97ms | [320, 224, 8 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_11 5.94ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Split splitncnn_0 0.01ms |
Convolution Conv_15 2.83ms | [160, 112, 16 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_17 0.39ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Split splitncnn_1 0.00ms |
Convolution Conv_18 4.14ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_20 0.40ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_21 26.67ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_23 5.35ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Split splitncnn_2 0.00ms |
Convolution Conv_24 4.72ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_26 0.45ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_27 4.63ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_29 0.98ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_12 2.20ms | [160, 112, 16 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_14 0.35ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Concat Concat_30 1.30ms |
Convolution Conv_31 11.11ms | [160, 112, 32 *8] -> [160, 112, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_33 1.57ms | [160, 112, 32 *8] -> [160, 112, 32 *8]
Split splitncnn_3 0.00ms |
Convolution Conv_38 5.98ms | [160, 112, 32 *8] -> [160, 112, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_40 4.50ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_41 10.63ms | [160, 112, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_43 0.17ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Pooling MaxPool_34 0.97ms | [160, 112, 32 *8] -> [ 80, 56, 32 *8]
Convolution Conv_35 1.28ms | [ 80, 56, 32 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_37 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Concat Concat_44 0.27ms |
Split splitncnn_4 0.00ms |
Convolution Conv_48 1.48ms | [ 80, 56, 32 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_50 3.60ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Split splitncnn_5 0.00ms |
Convolution Conv_51 3.44ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_53 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_54 2.56ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_56 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Split splitncnn_6 0.00ms |
Convolution Conv_57 2.63ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_59 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_60 2.66ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_62 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_45 1.59ms | [ 80, 56, 32 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_47 0.21ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Concat Concat_63 0.62ms |
Convolution Conv_64 11.37ms | [ 80, 56, 64 *8] -> [ 80, 56, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_66 0.55ms | [ 80, 56, 64 *8] -> [ 80, 56, 64 *8]
Split splitncnn_7 0.00ms |
Convolution Conv_195 3.01ms | [ 80, 56, 64 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_197 0.35ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_71 3.65ms | [ 80, 56, 64 *8] -> [ 80, 56, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_73 0.28ms | [ 80, 56, 32 *8] -> [ 80, 56, 32 *8]
Convolution Conv_74 10.07ms | [ 80, 56, 32 *8] -> [ 40, 28, 32 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_76 0.08ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Pooling MaxPool_67 0.53ms | [ 80, 56, 64 *8] -> [ 40, 28, 64 *8]
Convolution Conv_68 0.79ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_70 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Concat Concat_77 0.14ms |
Split splitncnn_8 0.00ms |
Convolution Conv_81 1.33ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_83 0.11ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_9 0.00ms |
Convolution Conv_84 1.31ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_86 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Convolution Conv_87 0.96ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_89 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_10 0.00ms |
Convolution Conv_90 104.40ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_92 19.90ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Convolution Conv_93 30.59ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_95 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Convolution Conv_78 0.81ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_80 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Concat Concat_96 0.31ms |
Convolution Conv_97 5.38ms | [ 40, 28, 128 *8] -> [ 40, 28, 128 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_99 0.26ms | [ 40, 28, 128 *8] -> [ 40, 28, 128 *8]
Split splitncnn_11 0.00ms |
Convolution Conv_164 1.63ms | [ 40, 28, 128 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_166 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Convolution Conv_104 3.19ms | [ 40, 28, 128 *8] -> [ 40, 28, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_106 0.14ms | [ 40, 28, 64 *8] -> [ 40, 28, 64 *8]
Convolution Conv_107 4.88ms | [ 40, 28, 64 *8] -> [ 20, 14, 64 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_109 1.56ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Pooling MaxPool_100 1.86ms | [ 40, 28, 128 *8] -> [ 20, 14, 128 *8]
Convolution Conv_101 0.82ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_103 0.06ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Concat Concat_110 0.10ms |
Split splitncnn_12 0.00ms |
Convolution Conv_114 0.47ms | [ 20, 14, 128 *8] -> [ 20, 14, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_116 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Split splitncnn_13 0.00ms |
Convolution Conv_117 0.58ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_119 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Convolution Conv_120 0.46ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_122 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Split splitncnn_14 0.00ms |
Convolution Conv_123 0.44ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_125 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Convolution Conv_126 0.42ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_128 0.51ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Convolution Conv_111 0.40ms | [ 20, 14, 128 *8] -> [ 20, 14, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_113 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Concat Concat_129 0.08ms |
Convolution Conv_130 1.21ms | [ 20, 14, 128 *8] -> [ 20, 14, 128 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_132 0.06ms | [ 20, 14, 128 *8] -> [ 20, 14, 128 *8]
Split splitncnn_15 0.00ms |
Convolution Conv_133 0.66ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_135 0.04ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Convolution Conv_136 3.70ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_138 11.37ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Convolution Conv_139 4.22ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_141 0.05ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Split splitncnn_16 0.00ms |
Pooling MaxPool_142 0.12ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Pooling MaxPool_143 0.28ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Pooling MaxPool_144 0.44ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Concat Concat_145 0.15ms |
Convolution Conv_146 1.53ms | [ 20, 14, 256 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_148 0.04ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Convolution Conv_149 1.70ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_151 0.04ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Convolution Conv_152 0.71ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_154 0.04ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Concat Concat_155 0.09ms |
Convolution Conv_156 0.72ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_158 0.03ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Split splitncnn_17 0.00ms |
Convolution Conv_159 0.20ms | [ 20, 14, 64 *8] -> [ 20, 14, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_161 0.02ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Interp Resize_163 0.05ms | [ 20, 14, 32 *8] -> [ 40, 28, 32 *8]
Concat Concat_167 0.16ms |
Split splitncnn_18 0.00ms |
Convolution Conv_171 0.91ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_173 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_19 0.00ms |
Convolution Conv_174 3.30ms | [ 40, 28, 32 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_176 0.60ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_20 0.00ms |
Convolution Conv_177 0.48ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_179 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_21 0.00ms |
Convolution Conv_180 0.36ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_182 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_22 0.00ms |
Convolution Conv_183 0.35ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_185 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Convolution Conv_168 0.93ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_170 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Concat Concat_186 0.34ms |
Convolution Conv_187 1.86ms | [ 40, 28, 128 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_189 0.07ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_23 0.00ms |
Convolution Conv_190 0.27ms | [ 40, 28, 32 *8] -> [ 40, 28, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_192 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Interp Resize_194 0.14ms | [ 40, 28, 16 *8] -> [ 80, 56, 16 *8]
Concat Concat_198 0.35ms |
Split splitncnn_24 0.00ms |
Convolution Conv_202 1.42ms | [ 80, 56, 32 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_204 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Split splitncnn_25 0.00ms |
Convolution Conv_205 1.70ms | [ 80, 56, 16 *8] -> [ 80, 56, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_207 0.09ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8]
Split splitncnn_26 0.00ms |
Convolution Conv_208 0.87ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_210 0.09ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8]
Split splitncnn_27 0.00ms |
Convolution Conv_211 0.62ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_213 0.09ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8]
Split splitncnn_28 0.00ms |
Convolution Conv_214 0.51ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_216 0.09ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8]
Convolution Conv_199 1.56ms | [ 80, 56, 32 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_201 1.72ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Concat Concat_217 0.70ms |
Convolution Conv_218 3.82ms | [ 80, 56, 64 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_220 0.20ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Split splitncnn_29 0.00ms |
Convolution Conv_287 3.45ms | [ 80, 56, 16 *8] -> [ 80, 56, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_289 0.33ms | [ 80, 56, 32 *8] -> [ 80, 56, 32 *8]
Split splitncnn_42 0.00ms |
Convolution Conv_476 2.03ms | [ 80, 56, 32 *8] -> [ 80, 56, 255 *1] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_490 1.34ms | [ 80, 56, 255 *1] -> [4480, 85, 3 *1]
Permute Transpose_491 0.38ms | [4480, 85, 3 *1] -> [ 85, 4480, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_460 19.95ms | [ 80, 56, 32 *8] -> [ 80, 56, 735 *4] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_474 15.56ms | [ 80, 56, 735 *4] -> [4480, 980, 3 *1]
Permute Transpose_475 10.96ms | [4480, 980, 3 *1] -> [980, 4480, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_225 26.35ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_227 0.34ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_228 3.50ms | [ 80, 56, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_230 0.12ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Pooling MaxPool_221 0.17ms | [ 80, 56, 16 *8] -> [ 40, 28, 16 *8]
Convolution Conv_222 0.19ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_224 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Concat Concat_231 0.17ms |
Split splitncnn_30 0.00ms |
Convolution Conv_235 0.85ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_237 1.67ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_31 0.00ms |
Convolution Conv_238 1.98ms | [ 40, 28, 32 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_240 0.08ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_32 0.00ms |
Convolution Conv_241 0.46ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_243 0.06ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_33 0.00ms |
Convolution Conv_244 0.39ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_246 0.06ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Split splitncnn_34 0.00ms |
Convolution Conv_247 0.41ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_249 0.06ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Convolution Conv_232 0.91ms | [ 40, 28, 64 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_234 0.11ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Concat Concat_250 0.41ms |
Convolution Conv_251 2.03ms | [ 40, 28, 128 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_253 0.12ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Split splitncnn_35 0.00ms |
Convolution Conv_290 2.60ms | [ 40, 28, 32 *8] -> [ 40, 28, 64 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_292 0.19ms | [ 40, 28, 64 *8] -> [ 40, 28, 64 *8]
Split splitncnn_43 0.00ms |
Convolution Conv_508 1.99ms | [ 40, 28, 64 *8] -> [ 40, 28, 255 *1] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_522 0.23ms | [ 40, 28, 255 *1] -> [1120, 85, 3 *1]
Permute Transpose_523 0.08ms | [1120, 85, 3 *1] -> [ 85, 1120, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_492 8.82ms | [ 40, 28, 64 *8] -> [ 40, 28, 735 *4] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_506 3.44ms | [ 40, 28, 735 *4] -> [1120, 980, 3 *1]
Permute Transpose_507 1.49ms | [1120, 980, 3 *1] -> [980, 1120, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_258 0.48ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_260 0.13ms | [ 40, 28, 32 *8] -> [ 40, 28, 32 *8]
Convolution Conv_261 1.26ms | [ 40, 28, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 2 x 2
Swish Mul_263 0.07ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Pooling MaxPool_254 0.08ms | [ 40, 28, 32 *8] -> [ 20, 14, 32 *8]
Convolution Conv_255 0.14ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_257 0.05ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Concat Concat_264 0.09ms |
Split splitncnn_36 0.00ms |
Convolution Conv_268 93.32ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_270 22.64ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Split splitncnn_37 0.00ms |
Convolution Conv_271 96.77ms | [ 20, 14, 64 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_273 0.19ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Split splitncnn_38 0.00ms |
Convolution Conv_274 0.65ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_276 0.05ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Split splitncnn_39 0.00ms |
Convolution Conv_277 0.49ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_279 0.05ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Split splitncnn_40 0.00ms |
Convolution Conv_280 0.46ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_282 0.05ms | [ 20, 14, 32 *8] -> [ 20, 14, 32 *8]
Convolution Conv_265 0.69ms | [ 20, 14, 128 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_267 0.07ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Concat Concat_283 0.17ms |
Convolution Conv_284 2.82ms | [ 20, 14, 256 *8] -> [ 20, 14, 64 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_286 0.08ms | [ 20, 14, 64 *8] -> [ 20, 14, 64 *8]
Split splitncnn_41 0.00ms |
Convolution Conv_293 3.13ms | [ 20, 14, 64 *8] -> [ 20, 14, 128 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_295 0.11ms | [ 20, 14, 128 *8] -> [ 20, 14, 128 *8]
Split splitncnn_44 0.00ms |
Convolution Conv_540 0.44ms | [ 20, 14, 128 *8] -> [ 20, 14, 255 *1] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_554 0.05ms | [ 20, 14, 255 *1] -> [280, 85, 3 *1]
Permute Transpose_555 0.05ms | [280, 85, 3 *1] -> [ 85, 280, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_524 6.07ms | [ 20, 14, 128 *8] -> [ 20, 14, 735 *4] kernel: 1 x 1 stride: 1 x 1
Reshape Reshape_538 0.63ms | [ 20, 14, 735 *4] -> [280, 980, 3 *1]
Permute Transpose_539 0.30ms | [280, 980, 3 *1] -> [980, 280, 3 *1]
denormals_zero 998277249 not supported
Convolution Conv_301 4.11ms | [160, 112, 32 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_303 0.42ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_296 0.48ms | [ 80, 56, 16 *8] -> [ 80, 56, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_298 0.11ms | [ 80, 56, 8 *8] -> [ 80, 56, 8 *8]
Interp Resize_300 0.26ms | [ 80, 56, 8 *8] -> [160, 112, 8 *8]
Concat Concat_304 0.63ms |
Convolution Conv_305 2.10ms | [160, 112, 16 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_307 0.45ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Split splitncnn_45 0.00ms |
Convolution Conv_308 0.95ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_310 0.46ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_311 8.74ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_313 0.39ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_314 1.07ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_316 0.39ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_317 4.67ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_319 0.50ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_320 1.82ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_322 0.41ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Concat Concat_323 0.97ms |
Convolution Conv_324 2.17ms | [160, 112, 16 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_326 0.40ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Convolution Conv_327 0.81ms | [160, 112, 8 *8] -> [160, 112, 8 *8] kernel: 1 x 1 stride: 1 x 1
Swish Mul_329 0.44ms | [160, 112, 8 *8] -> [160, 112, 8 *8]
Split splitncnn_46 0.00ms |
Convolution Conv_387 18.89ms | [160, 112, 8 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_389 0.87ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_330 6.39ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_332 0.25ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_333 1.04ms | [ 40, 28, 32 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_335 0.09ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Interp Resize_354 13.58ms | [ 40, 28, 16 *8] -> [ 80, 56, 16 *8]
BinaryOp Add_355 16.68ms |
Convolution Conv_356 1.82ms | [ 20, 14, 64 *8] -> [ 20, 14, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_358 0.04ms | [ 20, 14, 16 *8] -> [ 20, 14, 16 *8]
Interp Resize_377 0.13ms | [ 20, 14, 16 *8] -> [ 80, 56, 16 *8]
BinaryOp Add_378 0.33ms |
Convolution Conv_379 7.39ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_381 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Interp Resize_383 0.59ms | [ 80, 56, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_384 12.35ms | [160, 112, 16 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_386 4.52ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
BinaryOp Add_390 1.46ms |
Convolution Conv_391 12.85ms | [160, 112, 16 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_393 0.81ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_394 5.21ms | [160, 112, 16 *8] -> [160, 112, 1 *4] kernel: 1 x 1 stride: 1 x 1
Convolution Conv_452 7.90ms | [160, 112, 8 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_454 0.77ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_395 2.53ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_397 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Convolution Conv_398 1.03ms | [ 40, 28, 32 *8] -> [ 40, 28, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_400 0.05ms | [ 40, 28, 16 *8] -> [ 40, 28, 16 *8]
Interp Resize_419 0.15ms | [ 40, 28, 16 *8] -> [ 80, 56, 16 *8]
BinaryOp Add_420 0.30ms |
Convolution Conv_421 0.78ms | [ 20, 14, 64 *8] -> [ 20, 14, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_423 0.01ms | [ 20, 14, 16 *8] -> [ 20, 14, 16 *8]
Interp Resize_442 0.16ms | [ 20, 14, 16 *8] -> [ 80, 56, 16 *8]
BinaryOp Add_443 4.72ms |
Convolution Conv_444 3.17ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_446 0.18ms | [ 80, 56, 16 *8] -> [ 80, 56, 16 *8]
Interp Resize_448 0.62ms | [ 80, 56, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_449 13.86ms | [160, 112, 16 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_451 1.54ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
BinaryOp Add_455 1.53ms |
Convolution Conv_456 12.57ms | [160, 112, 16 *8] -> [160, 112, 16 *8] kernel: 3 x 3 stride: 1 x 1
Swish Mul_458 0.81ms | [160, 112, 16 *8] -> [160, 112, 16 *8]
Convolution Conv_459 4.81ms | [160, 112, 16 *8] -> [160, 112, 1 *1] kernel: 1 x 1 stride: 1 x 1
Concat Concat_10000 0.04ms |
Segmentation fault (core dumped)
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.