Some interesting modules

Share a module I wrote , I am here VOC There are no rising points in the data set , But many people have increased their data sets , Then I will simply say the principle

At first my idea was simple , Just feel the original $6\times 6$ The convolution parameters are too large , Then I want to use a $3\times 3$ And a $5\times 5$ Convolution substitution of , The main purpose is to save some parameters （ Although it's trivial ）, Later, some students said that they could try empty convolution , Then I put $5\times 5$ Instead of the normal convolution of, I'm going to replace it with $3\times 3$ The void convolution of , This can keep the receptive field unchanged and further reduce the parameters

The schematic diagram is as follows ,（a） For the original $6\times 6$ Convolution （b） Improved convolution for myself , I didn't add BN and SiLU, I tested , It is lower to add points than not ,

After the improvement, the parameters of the first layer are indeed reduced by more than half


Source code ：

class Inception_Conv(nn.Module):
    # Standard convolution
    def __init__(self, c1, c2, k=3, s=2, g=1, p=None):  # ch_in, ch_out, kernel, stride, padding, groups
        super().__init__()

        self.conv1 = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
        self.conv2 = nn.Conv2d(c1, c2, k , s, autopad(k+2 , p),dilation=2, groups=g, bias=False)

    def forward(self, x):
        x1 = self.conv1(x)
        x2 = self.conv2(x)
        x = torch.add(x1, x2)
        return x

The configuration file ：

# YOLOv5  by Ultralytics, GPL-3.0 license

# Parameters
nc: 80  # number of classes
depth_multiple: 0.33  # model depth multiple
width_multiple: 0.50  # layer channel multiple
anchors:
  - [10,13, 16,30, 33,23]  # P3/8
  - [30,61, 62,45, 59,119]  # P4/16
  - [116,90, 156,198, 373,326]  # P5/32

# YOLOv5 v6.0 backbone
backbone:
  # [from, number, module, args]
  [[-1, 1, Inception_Conv, [64, 3]], # 0-P1/2
   [-1, 1, Conv, [128, 3, 2]],  # 1-P2/4
   [-1, 3, C3, [128]],
   [-1, 1, Conv, [256, 3, 2]],  # 3-P3/8
   [-1, 6, C3, [256]],
   [-1, 1, Conv, [512, 3, 2]],  # 5-P4/16
   [-1, 9, C3, [512]],
   [-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32
   [-1, 3, C3, [1024]],
   [-1, 1, SPPF, [1024, 5]],  # 9
  ]


# YOLOv5 v6.0 head
head:
  [[-1, 1, Conv, [512, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
   [-1, 3, C3, [512, False]],  # 13

   [-1, 1, Conv, [256, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
   [-1, 3, C3, [256, False]],  # 17 (P3/8-small)

   [-1, 1, Conv, [256, 3, 2]],
   [[-1, 14], 1, Concat, [1]],  # cat head P4
   [-1, 3, C3, [512, False]],  # 20 (P4/16-medium)

   [-1, 1, Conv, [512, 3, 2]],
   [[-1, 10], 1, Concat, [1]],  # cat head P5
   [-1, 3, C3, [1024, False]],  # 23 (P5/32-large)

   [[17, 20, 23], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)
  ]