YOLOv8重要模块解读

• 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
• 🍖 原作者：K同学啊 | 接辅导、项目定制

yolov8的模块代码主要在./ultralytics/nn/modules/下

conv.py

在conv.py文件的头部，__all__中声明了当前模块的所有暴露的模型，如果需要自定义的话，需要将自定义的模块写到这个里面

__all__ = ('Conv', 'Conv2', 'LightConv', 'DWConv', 'DWConvTranspose2d', 'ConvTranspose', 'Focus', 'GhostConv',
           'ChannelAttention', 'SpatialAttention', 'CBAM', 'Concat', 'RepConv')

autopad

autopad的功能是返回padding的大小，使padding后输出张量的大小不变

参数k: 卷积核（kernel）的大小，类型可能是一个int也可能是一个序列
参数p: 填充（padding）的大小。默认为None
参数d: 扩张率（dilation rate）的大小，默认为1，普通卷积的扩张率为1，空洞卷积的扩张率大于1

def autopad(k, p=None, d=1):  # kernel, padding, dilation
    """Pad to 'same' shape outputs."""
    if d > 1:
        k = d * (k - 1) + 1 if isinstance(k, int) else [d * (x - 1) + 1 for x in k]  # actual kernel-size
    if p is None:
        p = k // 2 if isinstance(k, int) else [x // 2 for x in k]  # auto-pad
    return p

Conv

conv的功能是一个标准卷积模块，它继承自nn.Module。
参数c1: 输入通道数
参数c2: 输出通道数
参数k：卷积核大小，默认是1
参数s: 步长，默认是1
参数p: 填充，默认为None
参数g：组，默认为1
参数d：扩张率，默认为1
参数act：是否采用激活函数，默认为True，且采用SiLU为激活函数

class Conv(nn.Module):
    """Standard convolution with args(ch_in, ch_out, kernel, stride, padding, groups, dilation, activation)."""
    default_act = nn.SiLU()  # default activation

    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, d=1, act=True):
        """Initialize Conv layer with given arguments including activation."""
        super().__init__()
        self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p, d), groups=g, dilation=d, bias=False)
        self.bn = nn.BatchNorm2d(c2)
        self.act = self.default_act if act is True else act if isinstance(act, nn.Module) else nn.Identity()

    def forward(self, x):
        """Apply convolution, batch normalization and activation to input tensor."""
        return self.act(self.bn(self.conv(x)))

    def forward_fuse(self, x):
        """Perform transposed convolution of 2D data."""
        return self.act(self.conv(x))

Focus

Focus模块是YOLO作者自己设计出来的，为了减少浮点数和提高速度，而不是增加featuremap。本质是将图像进行切片，类似于下采样取值，将原图像的宽高信息切分，聚合到通道中。其结构如下：

class Focus(nn.Module):
    """Focus wh information into c-space."""

    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):
        """Initializes Focus object with user defined channel, convolution, padding, group and activation values."""
        super().__init__()
        self.conv = Conv(c1 * 4, c2, k, s, p, g, act=act)
        # self.contract = Contract(gain=2)

    def forward(self, x):
        """
        Applies convolution to concatenated tensor and returns the output.

        Input shape is (b,c,w,h) and output shape is (b,4c,w/2,h/2).
        """
        # 将输入沿着通道维度拼接4次得到新的输入，然后经过卷积层self.conv处理
        return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1))
        # 如果启用降维模块
        # return self.conv(self.contract(x))

block.py

c2f

c2f结构如图所示：

class C2f(nn.Module):
    """Faster Implementation of CSP Bottleneck with 2 convolutions."""
    # 有两个CSP Bottleneck

    def __init__(self, c1, c2, n=1, shortcut=False, g=1, e=0.5):
        """Initialize CSP bottleneck layer with two convolutions with arguments ch_in, ch_out, number, shortcut, groups,
        expansion.
        """
        super().__init__()
        self.c = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, 2 * self.c, 1, 1)
        self.cv2 = Conv((2 + n) * self.c, c2, 1)  # optional act=FReLU(c2)
        self.m = nn.ModuleList(Bottleneck(self.c, self.c, shortcut, g, k=((3, 3), (3, 3)), e=1.0) for _ in range(n))

    def forward(self, x):
        """Forward pass through C2f layer."""
        # 通过c2f层进行向前传播
        y = list(self.cv1(x).chunk(2, 1))
        y.extend(m(y[-1]) for m in self.m)
        return self.cv2(torch.cat(y, 1))

    def forward_split(self, x):
        """Forward pass using split() instead of chunk()."""
        # 通过split()而不是chunk()进行前向传播
        y = list(self.cv1(x).split((self.c, self.c), 1))
        y.extend(m(y[-1]) for m in self.m)
        return self.cv2(torch.cat(y, 1))