yolov8 修改颈部网络 替换为bifpn

1 添加融合模块代码

创建revcol.py 写入以下Fusion模块代码 revcol.py文件放置位置 ultralytics.nn.backbone文件夹下面

class Fusion(nn.Module):
    def __init__(self, level, channels, first_col) -> None:
        super().__init__()
        
        self.level = level
        self.first_col = first_col
        self.down = Conv(channels[level-1], channels[level], k=2, s=2, p=0, act=False) if level in [1, 2, 3] else nn.Identity()
        if not first_col:
            self.up = nn.Sequential(Conv(channels[level+1], channels[level]), nn.Upsample(scale_factor=2, mode='nearest')) if level in [0, 1, 2] else nn.Identity()            

    def forward(self, *args):

        c_down, c_up = args
        
        if self.first_col:
            x = self.down(c_down)
            return x
        
        if self.level == 3:
            x = self.down(c_down)
        else:
            x = self.up(c_up) + self.down(c_down)
        return x

2 在task.py文件中导入 在parse_model函数里添加

        elif m is Fusion:
            args[0] = d[args[0]]
            c1, c2 = [ch[x] for x in f], (sum([ch[x] for x in f]) if args[0] == 'concat' else ch[f[0]])

3 编写模型yaml文件

# Ultralytics YOLO 🚀, AGPL-3.0 license
# YOLOv8 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect

# Parameters
nc: 80  # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
  # [depth, width, max_channels]
  n: [0.33, 0.25, 1024]  # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
  s: [0.33, 0.50, 1024]  # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
  m: [0.67, 0.75, 768]   # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
  l: [1.00, 1.00, 512]   # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
  x: [1.00, 1.25, 512]   # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
fusion_mode: bifpn
node_mode: C2f
head_channel: 256

# YOLOv8.0n backbone
backbone:
  # [from, repeats, module, args]
  - [-1, 1, Conv, [64, 3, 2]]  # 0-P1/2
  - [-1, 1, Conv, [128, 3, 2]]  # 1-P2/4
  - [-1, 3, C2f, [128, True]]
  - [-1, 1, Conv, [256, 3, 2]]  # 3-P3/8
  - [-1, 6, C2f, [256, True]]
  - [-1, 1, Conv, [512, 3, 2]]  # 5-P4/16
  - [-1, 6, C2f, [512, True]]
  - [-1, 1, Conv, [1024, 3, 2]]  # 7-P5/32
  - [-1, 3, C2f, [1024, True]]
  - [-1, 1, SPPF, [1024, 5]]  # 9

# YOLOv8.0n head
head:
  - [4, 1, Conv, [head_channel]]  # 10-P3/8
  - [6, 1, Conv, [head_channel]]  # 11-P4/16
  - [9, 1, Conv, [head_channel]]  # 12-P5/32

  - [-1, 1, nn.Upsample, [None, 2, 'nearest']] # 13 P5->P4
  - [[-1, 11], 1, Fusion, [fusion_mode]] # 14
  - [-1, 3, node_mode, [head_channel]] # 15-P4/16
  
  - [-1, 1, nn.Upsample, [None, 2, 'nearest']] # 16 P4->P3
  - [[-1, 10], 1, Fusion, [fusion_mode]] # 17
  - [-1, 3, node_mode, [head_channel]] # 18-P3/8

  - [2, 1, Conv, [head_channel, 3, 2]] # 19 P2->P3
  - [[-1, 10, 18], 1, Fusion, [fusion_mode]] # 20
  - [-1, 3, node_mode, [head_channel]] # 21-P3/8

  - [-1, 1, Conv, [head_channel, 3, 2]] # 22 P3->P4
  - [[-1, 11, 15], 1, Fusion, [fusion_mode]] # 23
  - [-1, 3, node_mode, [head_channel]] # 24-P4/16

  - [-1, 1, Conv, [head_channel, 3, 2]] # 25 P4->P5
  - [[-1, 12], 1, Fusion, [fusion_mode]] # 26
  - [-1, 3, node_mode, [head_channel]] # 27-P5/32

  - [[21, 24, 27], 1, Detect, [nc]]  # Detect(P3, P4, P5)

4 创建 train.py 运行即可

import warnings
warnings.filterwarnings('ignore')
from ultralytics import YOLO


if __name__ == '__main__':

    model = YOLO(r'yolov8-bifpn.yaml')
    model.train(data=r'./datasetp/data.yaml',
                imgsz=640,
                epochs=100,
                resume=True,
                batch=8,
                workers=0,
                device='0',
                optimizer='SGD', # using SGD
                project='runs/train',
                name='exp',
                model_name="yolov8"
                )