PolarSeg源码解析

模型打印理解
my_model：
ptBEVnet(
(PPmodel): Sequential(
(0): BatchNorm1d(9, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(1): Linear(in_features=9, out_features=64, bias=True)
(2): BatchNorm1d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(3): ReLU(inplace=True)
(4): Linear(in_features=64, out_features=128, bias=True)
(5): BatchNorm1d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(6): ReLU(inplace=True)
(7): Linear(in_features=128, out_features=256, bias=True)
(8): BatchNorm1d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(9): ReLU(inplace=True)
(10): Linear(in_features=256, out_features=512, bias=True)
)
(BEV_model): BEV_Unet(
(network): UNet(
(inc): inconv(
(conv): Sequential(
(0): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(1): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
)
)
(down1): down(
(mpconv): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(1): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
)
)
(down2): down(
(mpconv): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(1): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
)
)
(down3): down(
(mpconv): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(1): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
)
)
(down4): down(
(mpconv): Sequential(
(0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
(1): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
)
)
(up1): up(
(up): Upsample(scale_factor=2.0, mode=bilinear)
(conv): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(1024, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
(dropblock): DropBlock2D()
)
(up2): up(
(up): Upsample(scale_factor=2.0, mode=bilinear)
(conv): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(512, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
(dropblock): DropBlock2D()
)
(up3): up(
(up): Upsample(scale_factor=2.0, mode=bilinear)
(conv): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(256, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
(dropblock): DropBlock2D()
)
(up4): up(
(up): Upsample(scale_factor=2.0, mode=bilinear)
(conv): double_conv_circular(
(conv1): Sequential(
(0): Conv2d(128, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
(conv2): Sequential(
(0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 0))
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): LeakyReLU(negative_slope=0.01, inplace=True)
)
)
(dropblock): DropBlock2D()
)
(dropout): Dropout(p=0.0, inplace=False)
(outc): outconv(
(conv): Conv2d(64, 608, kernel_size=(1, 1), stride=(1, 1))
)
)
)
(fea_compression): Sequential(
(0): Linear(in_features=512, out_features=32, bias=True)
(1): ReLU()
)
)