YoloBody(
(backbone): CSPDarkNet(
(conv1): BasicConv(
(conv): Conv2d(3, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(stages): ModuleList(
(0): Resblock_body(
(downsample_conv): BasicConv(
(conv): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv0): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv1): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(blocks_conv): Sequential(
(0): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(1): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(concat_conv): BasicConv(
(conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(1): Resblock_body(
(downsample_conv): BasicConv(
(conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv0): BasicConv(
(conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv1): BasicConv(
(conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(blocks_conv): Sequential(
(0): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(1): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(2): BasicConv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(concat_conv): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(2): Resblock_body(
(downsample_conv): BasicConv(
(conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv0): BasicConv(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv1): BasicConv(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(blocks_conv): Sequential(
(0): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(1): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(2): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(3): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(4): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(5): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(6): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(7): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(8): BasicConv(
(conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(concat_conv): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(3): Resblock_body(
(downsample_conv): BasicConv(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv0): BasicConv(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv1): BasicConv(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(blocks_conv): Sequential(
(0): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(1): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(2): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(3): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(4): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(5): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(6): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(7): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(8): BasicConv(
(conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(concat_conv): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(4): Resblock_body(
(downsample_conv): BasicConv(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv0): BasicConv(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(split_conv1): BasicConv(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(blocks_conv): Sequential(
(0): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(1): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(2): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(3): Resblock(
(block): Sequential(
(0): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
(1): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
(4): BasicConv(
(conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
(concat_conv): BasicConv(
(conv): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(activation): Mish()
)
)
)
)
(conv1): Sequential(
(0): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(SPP): SpatialPyramidPooling(
(maxpools): ModuleList(
(0): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False)
(1): MaxPool2d(kernel_size=9, stride=1, padding=4, dilation=1, ceil_mode=False)
(2): MaxPool2d(kernel_size=13, stride=1, padding=6, dilation=1, ceil_mode=False)
)
)
(conv2): Sequential(
(0): Sequential(
(conv): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(upsample1): Upsample(
(upsample): Sequential(
(0): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Upsample(scale_factor=2.0, mode=nearest)
)
)
(conv_for_P4): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(make_five_conv1): Sequential(
(0): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(3): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(4): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(upsample2): Upsample(
(upsample): Sequential(
(0): Sequential(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Upsample(scale_factor=2.0, mode=nearest)
)
)
(conv_for_P3): Sequential(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(make_five_conv2): Sequential(
(0): Sequential(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(3): Sequential(
(conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(4): Sequential(
(conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(yolo_head3): Sequential(
(0): Sequential(
(conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Conv2d(256, 75, kernel_size=(1, 1), stride=(1, 1))
)
(down_sample1): Sequential(
(conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(make_five_conv3): Sequential(
(0): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(3): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(4): Sequential(
(conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(yolo_head2): Sequential(
(0): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Conv2d(512, 75, kernel_size=(1, 1), stride=(1, 1))
)
(down_sample2): Sequential(
(conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(make_five_conv4): Sequential(
(0): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Sequential(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(2): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(3): Sequential(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(4): Sequential(
(conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
)
(yolo_head1): Sequential(
(0): Sequential(
(conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): LeakyReLU(negative_slope=0.1)
)
(1): Conv2d(1024, 75, kernel_size=(1, 1), stride=(1, 1))
)
)
ModuleList(
(0): Sequential(
(conv1): Conv2d(3, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish1): Mish()
)
(1): Sequential(
(conv2): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish2): Mish()
)
(2): Sequential(
(conv3): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish3): Mish()
)
(3): EmptyModule()
(4): Sequential(
(conv4): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn4): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish4): Mish()
)
(5): Sequential(
(conv5): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn5): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish5): Mish()
)
(6): Sequential(
(conv6): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn6): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish6): Mish()
)
(7): EmptyModule()
(8): Sequential(
(conv7): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn7): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish7): Mish()
)
(9): EmptyModule()
(10): Sequential(
(conv8): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn8): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish8): Mish()
)
(11): Sequential(
(conv9): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn9): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish9): Mish()
)
(12): Sequential(
(conv10): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn10): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish10): Mish()
)
(13): EmptyModule()
(14): Sequential(
(conv11): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn11): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish11): Mish()
)
(15): Sequential(
(conv12): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn12): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish12): Mish()
)
(16): Sequential(
(conv13): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn13): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish13): Mish()
)
(17): EmptyModule()
(18): Sequential(
(conv14): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn14): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish14): Mish()
)
(19): Sequential(
(conv15): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn15): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish15): Mish()
)
(20): EmptyModule()
(21): Sequential(
(conv16): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn16): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish16): Mish()
)
(22): EmptyModule()
(23): Sequential(
(conv17): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn17): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish17): Mish()
)
(24): Sequential(
(conv18): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn18): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish18): Mish()
)
(25): Sequential(
(conv19): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn19): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish19): Mish()
)
(26): EmptyModule()
(27): Sequential(
(conv20): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn20): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish20): Mish()
)
(28): Sequential(
(conv21): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn21): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish21): Mish()
)
(29): Sequential(
(conv22): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn22): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish22): Mish()
)
(30): EmptyModule()
(31): Sequential(
(conv23): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn23): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish23): Mish()
)
(32): Sequential(
(conv24): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn24): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish24): Mish()
)
(33): EmptyModule()
(34): Sequential(
(conv25): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn25): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish25): Mish()
)
(35): Sequential(
(conv26): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn26): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish26): Mish()
)
(36): EmptyModule()
(37): Sequential(
(conv27): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn27): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish27): Mish()
)
(38): Sequential(
(conv28): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn28): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish28): Mish()
)
(39): EmptyModule()
(40): Sequential(
(conv29): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn29): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish29): Mish()
)
(41): Sequential(
(conv30): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn30): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish30): Mish()
)
(42): EmptyModule()
(43): Sequential(
(conv31): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn31): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish31): Mish()
)
(44): Sequential(
(conv32): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn32): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish32): Mish()
)
(45): EmptyModule()
(46): Sequential(
(conv33): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn33): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish33): Mish()
)
(47): Sequential(
(conv34): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn34): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish34): Mish()
)
(48): EmptyModule()
(49): Sequential(
(conv35): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn35): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish35): Mish()
)
(50): Sequential(
(conv36): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn36): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish36): Mish()
)
(51): EmptyModule()
(52): Sequential(
(conv37): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn37): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish37): Mish()
)
(53): EmptyModule()
(54): Sequential(
(conv38): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn38): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish38): Mish()
)
(55): Sequential(
(conv39): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn39): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish39): Mish()
)
(56): Sequential(
(conv40): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn40): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish40): Mish()
)
(57): EmptyModule()
(58): Sequential(
(conv41): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn41): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish41): Mish()
)
(59): Sequential(
(conv42): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn42): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish42): Mish()
)
(60): Sequential(
(conv43): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn43): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish43): Mish()
)
(61): EmptyModule()
(62): Sequential(
(conv44): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn44): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish44): Mish()
)
(63): Sequential(
(conv45): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn45): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish45): Mish()
)
(64): EmptyModule()
(65): Sequential(
(conv46): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn46): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish46): Mish()
)
(66): Sequential(
(conv47): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn47): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish47): Mish()
)
(67): EmptyModule()
(68): Sequential(
(conv48): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn48): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish48): Mish()
)
(69): Sequential(
(conv49): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn49): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish49): Mish()
)
(70): EmptyModule()
(71): Sequential(
(conv50): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn50): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish50): Mish()
)
(72): Sequential(
(conv51): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn51): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish51): Mish()
)
(73): EmptyModule()
(74): Sequential(
(conv52): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn52): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish52): Mish()
)
(75): Sequential(
(conv53): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn53): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish53): Mish()
)
(76): EmptyModule()
(77): Sequential(
(conv54): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn54): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish54): Mish()
)
(78): Sequential(
(conv55): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn55): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish55): Mish()
)
(79): EmptyModule()
(80): Sequential(
(conv56): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn56): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish56): Mish()
)
(81): Sequential(
(conv57): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn57): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish57): Mish()
)
(82): EmptyModule()
(83): Sequential(
(conv58): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn58): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish58): Mish()
)
(84): EmptyModule()
(85): Sequential(
(conv59): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn59): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish59): Mish()
)
(86): Sequential(
(conv60): Conv2d(512, 1024, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn60): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish60): Mish()
)
(87): Sequential(
(conv61): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn61): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish61): Mish()
)
(88): EmptyModule()
(89): Sequential(
(conv62): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn62): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish62): Mish()
)
(90): Sequential(
(conv63): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn63): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish63): Mish()
)
(91): Sequential(
(conv64): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn64): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish64): Mish()
)
(92): EmptyModule()
(93): Sequential(
(conv65): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn65): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish65): Mish()
)
(94): Sequential(
(conv66): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn66): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish66): Mish()
)
(95): EmptyModule()
(96): Sequential(
(conv67): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn67): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish67): Mish()
)
(97): Sequential(
(conv68): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn68): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish68): Mish()
)
(98): EmptyModule()
(99): Sequential(
(conv69): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn69): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish69): Mish()
)
(100): Sequential(
(conv70): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn70): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish70): Mish()
)
(101): EmptyModule()
(102): Sequential(
(conv71): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn71): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish71): Mish()
)
(103): EmptyModule()
(104): Sequential(
(conv72): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn72): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(mish72): Mish()
)
(105): Sequential(
(conv73): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn73): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky73): LeakyReLU(negative_slope=0.1, inplace=True)
)
(106): Sequential(
(conv74): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn74): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky74): LeakyReLU(negative_slope=0.1, inplace=True)
)
(107): Sequential(
(conv75): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn75): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky75): LeakyReLU(negative_slope=0.1, inplace=True)
)
(108): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False)
(109): EmptyModule()
(110): MaxPool2d(kernel_size=9, stride=1, padding=4, dilation=1, ceil_mode=False)
(111): EmptyModule()
(112): MaxPool2d(kernel_size=13, stride=1, padding=6, dilation=1, ceil_mode=False)
(113): EmptyModule()
(114): Sequential(
(conv76): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn76): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky76): LeakyReLU(negative_slope=0.1, inplace=True)
)
(115): Sequential(
(conv77): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn77): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky77): LeakyReLU(negative_slope=0.1, inplace=True)
)
(116): Sequential(
(conv78): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn78): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky78): LeakyReLU(negative_slope=0.1, inplace=True)
)
(117): Sequential(
(conv79): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn79): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky79): LeakyReLU(negative_slope=0.1, inplace=True)
)
(118): Upsample_expand()
(119): EmptyModule()
(120): Sequential(
(conv80): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn80): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky80): LeakyReLU(negative_slope=0.1, inplace=True)
)
(121): EmptyModule()
(122): Sequential(
(conv81): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn81): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky81): LeakyReLU(negative_slope=0.1, inplace=True)
)
(123): Sequential(
(conv82): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn82): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky82): LeakyReLU(negative_slope=0.1, inplace=True)
)
(124): Sequential(
(conv83): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn83): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky83): LeakyReLU(negative_slope=0.1, inplace=True)
)
(125): Sequential(
(conv84): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn84): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky84): LeakyReLU(negative_slope=0.1, inplace=True)
)
(126): Sequential(
(conv85): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn85): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky85): LeakyReLU(negative_slope=0.1, inplace=True)
)
(127): Sequential(
(conv86): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn86): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky86): LeakyReLU(negative_slope=0.1, inplace=True)
)
(128): Upsample_expand()
(129): EmptyModule()
(130): Sequential(
(conv87): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn87): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky87): LeakyReLU(negative_slope=0.1, inplace=True)
)
(131): EmptyModule()
(132): Sequential(
(conv88): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn88): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky88): LeakyReLU(negative_slope=0.1, inplace=True)
)
(133): Sequential(
(conv89): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn89): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky89): LeakyReLU(negative_slope=0.1, inplace=True)
)
(134): Sequential(
(conv90): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn90): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky90): LeakyReLU(negative_slope=0.1, inplace=True)
)
(135): Sequential(
(conv91): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn91): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky91): LeakyReLU(negative_slope=0.1, inplace=True)
)
(136): Sequential(
(conv92): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn92): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky92): LeakyReLU(negative_slope=0.1, inplace=True)
)
(137): Sequential(
(conv93): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn93): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky93): LeakyReLU(negative_slope=0.1, inplace=True)
)
(138): Sequential(
(conv94): Conv2d(256, 255, kernel_size=(1, 1), stride=(1, 1))
)
(139): YoloLayer()
(140): EmptyModule()
(141): Sequential(
(conv95): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn95): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky95): LeakyReLU(negative_slope=0.1, inplace=True)
)
(142): EmptyModule()
(143): Sequential(
(conv96): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn96): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky96): LeakyReLU(negative_slope=0.1, inplace=True)
)
(144): Sequential(
(conv97): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn97): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky97): LeakyReLU(negative_slope=0.1, inplace=True)
)
(145): Sequential(
(conv98): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn98): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky98): LeakyReLU(negative_slope=0.1, inplace=True)
)
(146): Sequential(
(conv99): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn99): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky99): LeakyReLU(negative_slope=0.1, inplace=True)
)
(147): Sequential(
(conv100): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn100): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky100): LeakyReLU(negative_slope=0.1, inplace=True)
)
(148): Sequential(
(conv101): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn101): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky101): LeakyReLU(negative_slope=0.1, inplace=True)
)
(149): Sequential(
(conv102): Conv2d(512, 255, kernel_size=(1, 1), stride=(1, 1))
)
(150): YoloLayer()
(151): EmptyModule()
(152): Sequential(
(conv103): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn103): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky103): LeakyReLU(negative_slope=0.1, inplace=True)
)
(153): EmptyModule()
(154): Sequential(
(conv104): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn104): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky104): LeakyReLU(negative_slope=0.1, inplace=True)
)
(155): Sequential(
(conv105): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn105): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky105): LeakyReLU(negative_slope=0.1, inplace=True)
)
(156): Sequential(
(conv106): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn106): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky106): LeakyReLU(negative_slope=0.1, inplace=True)
)
(157): Sequential(
(conv107): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn107): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky107): LeakyReLU(negative_slope=0.1, inplace=True)
)
(158): Sequential(
(conv108): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn108): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky108): LeakyReLU(negative_slope=0.1, inplace=True)
)
(159): Sequential(
(conv109): Conv2d(512, 1024, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn109): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(leaky109): LeakyReLU(negative_slope=0.1, inplace=True)
)
(160): Sequential(
(conv110): Conv2d(1024, 255, kernel_size=(1, 1), stride=(1, 1))
)
(161): YoloLayer()
)
