YOLOv5-7.0添加BottleNet transformer

YOLOv5主干特征提取网络为CNN网络，CNN具有平移不变性和局部性，缺乏全局建模长距离建模的能力，引入自然语言领域的Transformer可以形成CNN+TransFormer架构，充分结合两者的优点，提高目标检测效果。

1. BoTNet

论文地址：https://arxiv.org/abs/2101.11605
BoTNet是一种简单但功能强大的主干网络，该架构将注意力模块纳入了包括图像分类，目标检测，实例分割等多种计算机视觉任务。通过仅将ResNet的最后三个的卷积层替换成MHSA层，并不进行其他改变，在墓边检测方面显著改善了极限，同时减少了参数两，从而使得延迟最小化。
Transformer中的MHSA和BoTNet中的MHSA的区别：

• 归一化，Transformer使用 Layer Normalization，而BoTNet使用 Batch Normalization。
• 非线性激活，Transformer仅仅使用一个非线性激活在FPN block模块中，BoTNet使用了3个非线性激活。
• 输出投影，Transformer中的MHSA包含一个输出投影，BoTNet则没有。
• 优化器，Transformer使用Adam优化器训练，BoTNet使用sgd+ momentum
  

2.YOLOv5融合Bottleneck Transformers

2.1 修改common.py，添加CTR3

class MHSA(nn.Module):
    def __init__(self, n_dims, width=14, height=14, heads=4, pos_emb=False):
        super(MHSA, self).__init__()

        self.heads = heads
        self.query = nn.Conv2d(n_dims, n_dims, kernel_size=1)
        self.key = nn.Conv2d(n_dims, n_dims, kernel_size=1)
        self.value = nn.Conv2d(n_dims, n_dims, kernel_size=1)
        self.pos = pos_emb
        if self.pos:
            self.rel_h = nn.Parameter(torch.randn([1, heads, (n_dims) // heads, 1, int(height)]), requires_grad=True)
            self.rel_w = nn.Parameter(torch.randn([1, heads, (n_dims) // heads, int(width), 1]), requires_grad=True)
        self.softmax = nn.Softmax(dim=-1)

    def forward(self, x):
        n_batch, C, width, height = x.size()
        q = self.query(x).view(n_batch, self.heads, C // self.heads, -1)
        k = self.key(x).view(n_batch, self.heads, C // self.heads, -1)
        v = self.value(x).view(n_batch, self.heads, C // self.heads, -1)
        # print('q shape:{},k shape:{},v shape:{}'.format(q.shape,k.shape,v.shape))  #1,4,64,256
        content_content = torch.matmul(q.permute(0, 1, 3, 2), k)  # 1,C,h*w,h*w
        # print("qkT=",content_content.shape)
        c1, c2, c3, c4 = content_content.size()
        if self.pos:
            # print("old content_content shape",content_content.shape) #1,4,256,256
            content_position = (self.rel_h + self.rel_w).view(1, self.heads, C // self.heads, -1).permute(0, 1, 3,
                                                                                                          2)  # 1,4,1024,64

            content_position = torch.matmul(content_position, q)  # ([1, 4, 1024, 256])
            content_position = content_position if (
                        content_content.shape == content_position.shape) else content_position[:, :, :c3, ]
            assert (content_content.shape == content_position.shape)
            # print('new pos222-> shape:',content_position.shape)
            # print('new content222-> shape:',content_content.shape)
            energy = content_content + content_position
        else:
            energy = content_content
        attention = self.softmax(energy)
        out = torch.matmul(v, attention.permute(0, 1, 3, 2))  # 1,4,256,64
        out = out.view(n_batch, C, width, height)
        return out


class BottleneckTransformer(nn.Module):
    # Transformer bottleneck
    # expansion = 1

    def __init__(self, c1, c2, stride=1, heads=4, mhsa=True, resolution=None, expansion=1):
        super(BottleneckTransformer, self).__init__()
        c_ = int(c2 * expansion)
        self.cv1 = Conv(c1, c_, 1, 1)
        # self.bn1 = nn.BatchNorm2d(c2)
        if not mhsa:
            self.cv2 = Conv(c_, c2, 3, 1)
        else:
            self.cv2 = nn.ModuleList()
            self.cv2.append(MHSA(c2, width=int(resolution[0]), height=int(resolution[1]), heads=heads))
            if stride == 2:
                self.cv2.append(nn.AvgPool2d(2, 2))
            self.cv2 = nn.Sequential(*self.cv2)
        # self.bn2 = nn.BatchNorm2d(planes)
        # self.cv3 = nn.Conv2d(planes, expansion * planes, kernel_size=1, bias=False)
        # self.bn3 = nn.BatchNorm2d(expansion * planes)
        # self.shortcut = nn.Sequential()
        self.shortcut = c1 == c2
        if stride != 1 or c1 != expansion * c2:
            self.shortcut = nn.Sequential(
                nn.Conv2d(c1, expansion * c2, kernel_size=1, stride=stride),
                nn.BatchNorm2d(expansion * c2)
            )
        self.fc1 = nn.Linear(c2, c2)

    def forward(self, x):
        # print("transforme  input bottleck shape:",x.shape)
        # out = F.relu(self.bn1(self.conv1(x)))
        # out = F.relu(self.bn2(self.conv2(out)))
        # out = self.bn3(self.conv3(out))
        # out += self.shortcut(x)
        # out = F.relu(out)
        out = x + self.cv2(self.cv1(x)) if self.shortcut else self.cv2(self.cv1(x))
        return out


class CTR3(nn.Module):
    # CSP Bottleneck with 3 convolutions
    def __init__(self, c1, c2, n=1, e=0.5, e2=1, w=20, h=20):  # ch_in, ch_out, number, , expansion,w,h
        super(CTR3, self).__init__()
        c_ = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c1, c_, 1, 1)
        self.cv3 = Conv(2 * c_, c2, 1)  # act=FReLU(c2)
        self.m = nn.Sequential(
            *[BottleneckTransformer(c_, c_, stride=1, heads=4, mhsa=True, resolution=(w, h), expansion=e2) for _ in
              range(n)])
        # self.m = nn.Sequential(*[CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)])

    def forward(self, x):
        # print("CTR3-INPUT:",x.shape)
        # return self.cv3
        return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))

2.2修改yolo.py，在parse_model注册CTR3

 if m in {
                Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv, CTR3,
                BottleneckCSP, C3, C3TR, C3SPP, C3Ghost, nn.ConvTranspose2d, DWConvTranspose2d, C3x}:

2.3 修改yaml文件

# YOLOv5 🚀 by Ultralytics, GPL-3.0 license

# Parameters
nc: 8  # 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 v7.0 backbone
backbone:
  # [from, number, module, args]
  [[-1, 1, Conv, [64, 6, 2, 2]],  # 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, CTR3, [1024]],
   [-1, 1, SPPF, [1024, 5]],  # 9
  ]

# YOLOv5 v7.0 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)
   [ [ 17, 20, 23 ], 1, Detect, [ nc, anchors ] ],         # Detect(P3, P4, P5)

  ]