yolov5s 网络结构
Parameter name: epoch Parameter value:-1 Parameter name: best_fitness Parameter value:[ 0.65574] Parameter name: training_results Parameter value:None Parameter name: model Parameter value:Model( (model): Sequential( (0): Focus( (conv): Conv( (conv): Conv2d(12, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (1): Conv( (conv): Conv2d(32, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (2): C3( (cv1): Conv( (conv): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(64, 32, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(32, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (3): Conv( (conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (4): C3( (cv1): Conv( (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (1): Bottleneck( (cv1): Conv( (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (2): Bottleneck( (cv1): Conv( (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (5): Conv( (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (6): C3( (cv1): Conv( (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (1): Bottleneck( (cv1): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) (2): Bottleneck( (cv1): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (7): Conv( (conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (8): SPP( (cv1): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): 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) ) ) (9): C3( (cv1): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (10): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (11): Upsample(scale_factor=2.0, mode='nearest') (12): Concat() (13): C3( (cv1): Conv( (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (14): Conv( (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (15): Upsample(scale_factor=2.0, mode='nearest') (16): Concat() (17): C3( (cv1): Conv( (conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(64, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (18): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (19): Concat() (20): C3( (cv1): Conv( (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(128, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (21): Conv( (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (22): Concat() (23): C3( (cv1): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv3): Conv( (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(512, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (m): Sequential( (0): Bottleneck( (cv1): Conv( (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) (cv2): Conv( (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False) (bn): BatchNorm2d(256, eps=0.001, momentum=0.03, affine=True, track_running_stats=True) (act): SiLU() ) ) ) ) (24): Detect( (m): ModuleList( (0): Conv2d(128, 21, kernel_size=(1, 1), stride=(1, 1)) (1): Conv2d(256, 21, kernel_size=(1, 1), stride=(1, 1)) (2): Conv2d(512, 21, kernel_size=(1, 1), stride=(1, 1)) ) ) ) ) Parameter name: ema Parameter value:None Parameter name: updates Parameter value:None Parameter name: optimizer Parameter value:None Parameter name: wandb_id Parameter value:None
各层参数的形状
model.0.conv.conv.weight torch.Size([32, 12, 3, 3]) model.0.conv.conv.bias torch.Size([32]) model.1.conv.weight torch.Size([64, 32, 3, 3]) model.1.conv.bias torch.Size([64]) model.2.cv1.conv.weight torch.Size([32, 64, 1, 1]) model.2.cv1.conv.bias torch.Size([32]) model.2.cv2.conv.weight torch.Size([32, 64, 1, 1]) model.2.cv2.conv.bias torch.Size([32]) model.2.cv3.conv.weight torch.Size([64, 64, 1, 1]) model.2.cv3.conv.bias torch.Size([64]) model.2.m.0.cv1.conv.weight torch.Size([32, 32, 1, 1]) model.2.m.0.cv1.conv.bias torch.Size([32]) model.2.m.0.cv2.conv.weight torch.Size([32, 32, 3, 3]) model.2.m.0.cv2.conv.bias torch.Size([32]) model.3.conv.weight torch.Size([128, 64, 3, 3]) model.3.conv.bias torch.Size([128]) model.4.cv1.conv.weight torch.Size([64, 128, 1, 1]) model.4.cv1.conv.bias torch.Size([64]) model.4.cv2.conv.weight torch.Size([64, 128, 1, 1]) model.4.cv2.conv.bias torch.Size([64]) model.4.cv3.conv.weight torch.Size([128, 128, 1, 1]) model.4.cv3.conv.bias torch.Size([128]) model.4.m.0.cv1.conv.weight torch.Size([64, 64, 1, 1]) model.4.m.0.cv1.conv.bias torch.Size([64]) model.4.m.0.cv2.conv.weight torch.Size([64, 64, 3, 3]) model.4.m.0.cv2.conv.bias torch.Size([64]) model.4.m.1.cv1.conv.weight torch.Size([64, 64, 1, 1]) model.4.m.1.cv1.conv.bias torch.Size([64]) model.4.m.1.cv2.conv.weight torch.Size([64, 64, 3, 3]) model.4.m.1.cv2.conv.bias torch.Size([64]) model.4.m.2.cv1.conv.weight torch.Size([64, 64, 1, 1]) model.4.m.2.cv1.conv.bias torch.Size([64]) model.4.m.2.cv2.conv.weight torch.Size([64, 64, 3, 3]) model.4.m.2.cv2.conv.bias torch.Size([64]) model.5.conv.weight torch.Size([256, 128, 3, 3]) model.5.conv.bias torch.Size([256]) model.6.cv1.conv.weight torch.Size([128, 256, 1, 1]) model.6.cv1.conv.bias torch.Size([128]) model.6.cv2.conv.weight torch.Size([128, 256, 1, 1]) model.6.cv2.conv.bias torch.Size([128]) model.6.cv3.conv.weight torch.Size([256, 256, 1, 1]) model.6.cv3.conv.bias torch.Size([256]) model.6.m.0.cv1.conv.weight torch.Size([128, 128, 1, 1]) model.6.m.0.cv1.conv.bias torch.Size([128]) model.6.m.0.cv2.conv.weight torch.Size([128, 128, 3, 3]) model.6.m.0.cv2.conv.bias torch.Size([128]) model.6.m.1.cv1.conv.weight torch.Size([128, 128, 1, 1]) model.6.m.1.cv1.conv.bias torch.Size([128]) model.6.m.1.cv2.conv.weight torch.Size([128, 128, 3, 3]) model.6.m.1.cv2.conv.bias torch.Size([128]) model.6.m.2.cv1.conv.weight torch.Size([128, 128, 1, 1]) model.6.m.2.cv1.conv.bias torch.Size([128]) model.6.m.2.cv2.conv.weight torch.Size([128, 128, 3, 3]) model.6.m.2.cv2.conv.bias torch.Size([128]) model.7.conv.weight torch.Size([512, 256, 3, 3]) model.7.conv.bias torch.Size([512]) model.8.cv1.conv.weight torch.Size([256, 512, 1, 1]) model.8.cv1.conv.bias torch.Size([256]) model.8.cv2.conv.weight torch.Size([512, 1024, 1, 1]) model.8.cv2.conv.bias torch.Size([512]) model.9.cv1.conv.weight torch.Size([256, 512, 1, 1]) model.9.cv1.conv.bias torch.Size([256]) model.9.cv2.conv.weight torch.Size([256, 512, 1, 1]) model.9.cv2.conv.bias torch.Size([256]) model.9.cv3.conv.weight torch.Size([512, 512, 1, 1]) model.9.cv3.conv.bias torch.Size([512]) model.9.m.0.cv1.conv.weight torch.Size([256, 256, 1, 1]) model.9.m.0.cv1.conv.bias torch.Size([256]) model.9.m.0.cv2.conv.weight torch.Size([256, 256, 3, 3]) model.9.m.0.cv2.conv.bias torch.Size([256]) model.10.conv.weight torch.Size([256, 512, 1, 1]) model.10.conv.bias torch.Size([256]) model.13.cv1.conv.weight torch.Size([128, 512, 1, 1]) model.13.cv1.conv.bias torch.Size([128]) model.13.cv2.conv.weight torch.Size([128, 512, 1, 1]) model.13.cv2.conv.bias torch.Size([128]) model.13.cv3.conv.weight torch.Size([256, 256, 1, 1]) model.13.cv3.conv.bias torch.Size([256]) model.13.m.0.cv1.conv.weight torch.Size([128, 128, 1, 1]) model.13.m.0.cv1.conv.bias torch.Size([128]) model.13.m.0.cv2.conv.weight torch.Size([128, 128, 3, 3]) model.13.m.0.cv2.conv.bias torch.Size([128]) model.14.conv.weight torch.Size([128, 256, 1, 1]) model.14.conv.bias torch.Size([128]) model.17.cv1.conv.weight torch.Size([64, 256, 1, 1]) model.17.cv1.conv.bias torch.Size([64]) model.17.cv2.conv.weight torch.Size([64, 256, 1, 1]) model.17.cv2.conv.bias torch.Size([64]) model.17.cv3.conv.weight torch.Size([128, 128, 1, 1]) model.17.cv3.conv.bias torch.Size([128]) model.17.m.0.cv1.conv.weight torch.Size([64, 64, 1, 1]) model.17.m.0.cv1.conv.bias torch.Size([64]) model.17.m.0.cv2.conv.weight torch.Size([64, 64, 3, 3]) model.17.m.0.cv2.conv.bias torch.Size([64]) model.18.conv.weight torch.Size([128, 128, 3, 3]) model.18.conv.bias torch.Size([128]) model.20.cv1.conv.weight torch.Size([128, 256, 1, 1]) model.20.cv1.conv.bias torch.Size([128]) model.20.cv2.conv.weight torch.Size([128, 256, 1, 1]) model.20.cv2.conv.bias torch.Size([128]) model.20.cv3.conv.weight torch.Size([256, 256, 1, 1]) model.20.cv3.conv.bias torch.Size([256]) model.20.m.0.cv1.conv.weight torch.Size([128, 128, 1, 1]) model.20.m.0.cv1.conv.bias torch.Size([128]) model.20.m.0.cv2.conv.weight torch.Size([128, 128, 3, 3]) model.20.m.0.cv2.conv.bias torch.Size([128]) model.21.conv.weight torch.Size([256, 256, 3, 3]) model.21.conv.bias torch.Size([256]) model.23.cv1.conv.weight torch.Size([256, 512, 1, 1]) model.23.cv1.conv.bias torch.Size([256]) model.23.cv2.conv.weight torch.Size([256, 512, 1, 1]) model.23.cv2.conv.bias torch.Size([256]) model.23.cv3.conv.weight torch.Size([512, 512, 1, 1]) model.23.cv3.conv.bias torch.Size([512]) model.23.m.0.cv1.conv.weight torch.Size([256, 256, 1, 1]) model.23.m.0.cv1.conv.bias torch.Size([256]) model.23.m.0.cv2.conv.weight torch.Size([256, 256, 3, 3]) model.23.m.0.cv2.conv.bias torch.Size([256]) model.24.m.0.weight torch.Size([21, 128, 1, 1]) model.24.m.0.bias torch.Size([21]) model.24.m.1.weight torch.Size([21, 256, 1, 1]) model.24.m.1.bias torch.Size([21]) model.24.m.2.weight torch.Size([21, 512, 1, 1]) model.24.m.2.bias torch.Size([21])
输入为640* 640* 3时，各层输出形状
Layer (type:depth-idx) Output Shape Param
Model [1, 25200, 7] -- ├─Sequential: 1-1 -- -- │ └─Focus: 2-1 [1, 32, 320, 320] -- │ │ └─Conv: 3-1 [1, 32, 320, 320] (3,488) │ └─Conv: 2-2 [1, 64, 160, 160] -- │ │ └─Conv2d: 3-2 [1, 64, 160, 160] (18,496) │ │ └─SiLU: 3-3 [1, 64, 160, 160] -- │ └─C3: 2-3 [1, 64, 160, 160] -- │ │ └─Conv: 3-4 [1, 32, 160, 160] (2,080) │ │ └─Sequential: 3-5 [1, 32, 160, 160] (10,304) │ │ └─Conv: 3-6 [1, 32, 160, 160] (2,080) │ │ └─Conv: 3-7 [1, 64, 160, 160] (4,160) │ └─Conv: 2-4 [1, 128, 80, 80] -- │ │ └─Conv2d: 3-8 [1, 128, 80, 80] (73,856) │ │ └─SiLU: 3-9 [1, 128, 80, 80] -- │ └─C3: 2-5 [1, 128, 80, 80] -- │ │ └─Conv: 3-10 [1, 64, 80, 80] (8,256) │ │ └─Sequential: 3-11 [1, 64, 80, 80] (123,264) │ │ └─Conv: 3-12 [1, 64, 80, 80] (8,256) │ │ └─Conv: 3-13 [1, 128, 80, 80] (16,512) │ └─Conv: 2-6 [1, 256, 40, 40] -- │ │ └─Conv2d: 3-14 [1, 256, 40, 40] (295,168) │ │ └─SiLU: 3-15 [1, 256, 40, 40] -- │ └─C3: 2-7 [1, 256, 40, 40] -- │ │ └─Conv: 3-16 [1, 128, 40, 40] (32,896) │ │ └─Sequential: 3-17 [1, 128, 40, 40] (492,288) │ │ └─Conv: 3-18 [1, 128, 40, 40] (32,896) │ │ └─Conv: 3-19 [1, 256, 40, 40] (65,792) │ └─Conv: 2-8 [1, 512, 20, 20] -- │ │ └─Conv2d: 3-20 [1, 512, 20, 20] (1,180,160) │ │ └─SiLU: 3-21 [1, 512, 20, 20] -- │ └─SPP: 2-9 [1, 512, 20, 20] -- │ │ └─Conv: 3-22 [1, 256, 20, 20] (131,328) │ │ └─ModuleList: 3-23 -- -- │ │ └─Conv: 3-24 [1, 512, 20, 20] (524,800) │ └─C3: 2-10 [1, 512, 20, 20] -- │ │ └─Conv: 3-25 [1, 256, 20, 20] (131,328) │ │ └─Sequential: 3-26 [1, 256, 20, 20] (655,872) │ │ └─Conv: 3-27 [1, 256, 20, 20] (131,328) │ │ └─Conv: 3-28 [1, 512, 20, 20] (262,656) │ └─Conv: 2-11 [1, 256, 20, 20] -- │ │ └─Conv2d: 3-29 [1, 256, 20, 20] (131,328) │ │ └─SiLU: 3-30 [1, 256, 20, 20] -- │ └─Upsample: 2-12 [1, 256, 40, 40] -- │ └─Concat: 2-13 [1, 512, 40, 40] -- │ └─C3: 2-14 [1, 256, 40, 40] -- │ │ └─Conv: 3-31 [1, 128, 40, 40] (65,664) │ │ └─Sequential: 3-32 [1, 128, 40, 40] (164,096) │ │ └─Conv: 3-33 [1, 128, 40, 40] (65,664) │ │ └─Conv: 3-34 [1, 256, 40, 40] (65,792) │ └─Conv: 2-15 [1, 128, 40, 40] -- │ │ └─Conv2d: 3-35 [1, 128, 40, 40] (32,896) │ │ └─SiLU: 3-36 [1, 128, 40, 40] -- │ └─Upsample: 2-16 [1, 128, 80, 80] -- │ └─Concat: 2-17 [1, 256, 80, 80] -- │ └─C3: 2-18 [1, 128, 80, 80] -- │ │ └─Conv: 3-37 [1, 64, 80, 80] (16,448) │ │ └─Sequential: 3-38 [1, 64, 80, 80] (41,088) │ │ └─Conv: 3-39 [1, 64, 80, 80] (16,448) │ │ └─Conv: 3-40 [1, 128, 80, 80] (16,512) │ └─Conv: 2-19 [1, 128, 40, 40] -- │ │ └─Conv2d: 3-41 [1, 128, 40, 40] (147,584) │ │ └─SiLU: 3-42 [1, 128, 40, 40] -- │ └─Concat: 2-20 [1, 256, 40, 40] -- │ └─C3: 2-21 [1, 256, 40, 40] -- │ │ └─Conv: 3-43 [1, 128, 40, 40] (32,896) │ │ └─Sequential: 3-44 [1, 128, 40, 40] (164,096) │ │ └─Conv: 3-45 [1, 128, 40, 40] (32,896) │ │ └─Conv: 3-46 [1, 256, 40, 40] (65,792) │ └─Conv: 2-22 [1, 256, 20, 20] -- │ │ └─Conv2d: 3-47 [1, 256, 20, 20] (590,080) │ │ └─SiLU: 3-48 [1, 256, 20, 20] -- │ └─Concat: 2-23 [1, 512, 20, 20] -- │ └─C3: 2-24 [1, 512, 20, 20] -- │ │ └─Conv: 3-49 [1, 256, 20, 20] (131,328) │ │ └─Sequential: 3-50 [1, 256, 20, 20] (655,872) │ │ └─Conv: 3-51 [1, 256, 20, 20] (131,328) │ │ └─Conv: 3-52 [1, 512, 20, 20] (262,656) │ └─Detect: 2-25 [1, 25200, 7] --
