基于U
num_classes = 4 network = PetNet(num_classes) model = paddle.Model(network) model.summary( ( -1, 3, ) + IMAGE_SIZE )
W0509 15:39:03.322153 249 gpu_context.cc:278] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 10.1, Runtime API Version: 10.1 W0509 15:39:03.327960 249 gpu_context.cc:306] device: 0, cuDNN Version: 7.6. ----------------------------------------------------------------------------- Layer (type) Input Shape Output Shape Param ============================================================================= Conv2D-1 [[1, 3, 160, 160]] [1, 32, 80, 80] 896 BatchNorm2D-1 [[1, 32, 80, 80]] [1, 32, 80, 80] 128 ReLU-1 [[1, 32, 80, 80]] [1, 32, 80, 80] 0 ReLU-2 [[1, 32, 80, 80]] [1, 32, 80, 80] 0 SeparableConv2D-1 [[1, 32, 80, 80]] [1, 64, 80, 80] 2,400 BatchNorm2D-2 [[1, 64, 80, 80]] [1, 64, 80, 80] 256 ReLU-3 [[1, 64, 80, 80]] [1, 64, 80, 80] 0 SeparableConv2D-2 [[1, 64, 80, 80]] [1, 64, 80, 80] 4,736 BatchNorm2D-3 [[1, 64, 80, 80]] [1, 64, 80, 80] 256 MaxPool2D-1 [[1, 64, 80, 80]] [1, 64, 40, 40] 0 Conv2D-2 [[1, 32, 80, 80]] [1, 64, 40, 40] 2,112 Encoder-1 [[1, 32, 80, 80]] [1, 64, 40, 40] 0 ReLU-4 [[1, 64, 40, 40]] [1, 64, 40, 40] 0 SeparableConv2D-3 [[1, 64, 40, 40]] [1, 128, 40, 40] 8,896 BatchNorm2D-4 [[1, 128, 40, 40]] [1, 128, 40, 40] 512 ReLU-5 [[1, 128, 40, 40]] [1, 128, 40, 40] 0 SeparableConv2D-4 [[1, 128, 40, 40]] [1, 128, 40, 40] 17,664 BatchNorm2D-5 [[1, 128, 40, 40]] [1, 128, 40, 40] 512 MaxPool2D-2 [[1, 128, 40, 40]] [1, 128, 20, 20] 0 Conv2D-3 [[1, 64, 40, 40]] [1, 128, 20, 20] 8,320 Encoder-2 [[1, 64, 40, 40]] [1, 128, 20, 20] 0 ReLU-6 [[1, 128, 20, 20]] [1, 128, 20, 20] 0 SeparableConv2D-5 [[1, 128, 20, 20]] [1, 256, 20, 20] 34,176 BatchNorm2D-6 [[1, 256, 20, 20]] [1, 256, 20, 20] 1,024 ReLU-7 [[1, 256, 20, 20]] [1, 256, 20, 20] 0 SeparableConv2D-6 [[1, 256, 20, 20]] [1, 256, 20, 20] 68,096 BatchNorm2D-7 [[1, 256, 20, 20]] [1, 256, 20, 20] 1,024 MaxPool2D-3 [[1, 256, 20, 20]] [1, 256, 10, 10] 0 Conv2D-4 [[1, 128, 20, 20]] [1, 256, 10, 10] 33,024 Encoder-3 [[1, 128, 20, 20]] [1, 256, 10, 10] 0 ReLU-8 [[1, 256, 10, 10]] [1, 256, 10, 10] 0 Conv2DTranspose-1 [[1, 256, 10, 10]] [1, 256, 10, 10] 590,080 BatchNorm2D-8 [[1, 256, 10, 10]] [1, 256, 10, 10] 1,024 ReLU-9 [[1, 256, 10, 10]] [1, 256, 10, 10] 0 Conv2DTranspose-2 [[1, 256, 10, 10]] [1, 256, 10, 10] 590,080 BatchNorm2D-9 [[1, 256, 10, 10]] [1, 256, 10, 10] 1,024 Upsample-1 [[1, 256, 10, 10]] [1, 256, 20, 20] 0 Upsample-2 [[1, 256, 10, 10]] [1, 256, 20, 20] 0 Conv2D-5 [[1, 256, 20, 20]] [1, 256, 20, 20] 65,792 Decoder-1 [[1, 256, 10, 10]] [1, 256, 20, 20] 0 ReLU-10 [[1, 256, 20, 20]] [1, 256, 20, 20] 0 Conv2DTranspose-3 [[1, 256, 20, 20]] [1, 128, 20, 20] 295,040 BatchNorm2D-10 [[1, 128, 20, 20]] [1, 128, 20, 20] 512 ReLU-11 [[1, 128, 20, 20]] [1, 128, 20, 20] 0 Conv2DTranspose-4 [[1, 128, 20, 20]] [1, 128, 20, 20] 147,584 BatchNorm2D-11 [[1, 128, 20, 20]] [1, 128, 20, 20] 512 Upsample-3 [[1, 128, 20, 20]] [1, 128, 40, 40] 0 Upsample-4 [[1, 256, 20, 20]] [1, 256, 40, 40] 0 Conv2D-6 [[1, 256, 40, 40]] [1, 128, 40, 40] 32,896 Decoder-2 [[1, 256, 20, 20]] [1, 128, 40, 40] 0 ReLU-12 [[1, 128, 40, 40]] [1, 128, 40, 40] 0 Conv2DTranspose-5 [[1, 128, 40, 40]] [1, 64, 40, 40] 73,792 BatchNorm2D-12 [[1, 64, 40, 40]] [1, 64, 40, 40] 256 ReLU-13 [[1, 64, 40, 40]] [1, 64, 40, 40] 0 Conv2DTranspose-6 [[1, 64, 40, 40]] [1, 64, 40, 40] 36,928 BatchNorm2D-13 [[1, 64, 40, 40]] [1, 64, 40, 40] 256 Upsample-5 [[1, 64, 40, 40]] [1, 64, 80, 80] 0 Upsample-6 [[1, 128, 40, 40]] [1, 128, 80, 80] 0 Conv2D-7 [[1, 128, 80, 80]] [1, 64, 80, 80] 8,256 Decoder-3 [[1, 128, 40, 40]] [1, 64, 80, 80] 0 ReLU-14 [[1, 64, 80, 80]] [1, 64, 80, 80] 0 Conv2DTranspose-7 [[1, 64, 80, 80]] [1, 32, 80, 80] 18,464 BatchNorm2D-14 [[1, 32, 80, 80]] [1, 32, 80, 80] 128 ReLU-15 [[1, 32, 80, 80]] [1, 32, 80, 80] 0 Conv2DTranspose-8 [[1, 32, 80, 80]] [1, 32, 80, 80] 9,248 BatchNorm2D-15 [[1, 32, 80, 80]] [1, 32, 80, 80] 128 Upsample-7 [[1, 32, 80, 80]] [1, 32, 160, 160] 0 Upsample-8 [[1, 64, 80, 80]] [1, 64, 160, 160] 0 Conv2D-8 [[1, 64, 160, 160]] [1, 32, 160, 160] 2,080 Decoder-4 [[1, 64, 80, 80]] [1, 32, 160, 160] 0 Conv2D-9 [[1, 32, 160, 160]] [1, 4, 160, 160] 1,156 ============================================================================= Total params: 2,059,268 Trainable params: 2,051,716 Non-trainable params: 7,552 ----------------------------------------------------------------------------- Input size (MB): 0.29 Forward/backward pass size (MB): 117.77 Params size (MB): 7.86 Estimated Total Size (MB): 125.92 ----------------------------------------------------------------------------- {'total_params': 2059268, 'trainable_params': 2051716}
相关知识
HIV病毒 U=U测不出=不具传染力 在2017年起飞
根据U=U的意思是不是可以不用戴安全套了？
Bhiva强调U＝U在性传播中为“零风险”或“无风险”
基于U
一种基于智能家居的宠物状态远程监测设备的制作方法
基于springboot+mybatis+mysql+html实现宠物医院管理系统（包含实验报告）资源
基于Attention U
基于社会认知的儿童心理弹性研究
【Proteus仿真】基于51单片机的宠物喂食系统设计
2023北京U熊评判标准出炉
网址: 基于U https://m.mcbbbk.com/newsview668755.html
| 上一篇: 家鼠的毛色由一对常染色体上的基因 |
下一篇: 【美国可卡犬】美国可卡犬多少钱一 |
