yolov8添加注意力机制-MHSA

ROCKY__________

已于 2023-07-30 17:04:55 修改

阅读量2.9k

点赞数 7

文章标签： YOLO python 深度学习

于 2023-07-19 12:53:01 首次发布

本文链接：https://blog.csdn.net/ROCKY__________/article/details/131804291

版权

文章介绍了如何在Ultralytics的神经网络库中创建一个新的MHSA类，该类实现了多头自注意力机制。代码展示了MHSA模块的定义，包括查询、键、值的卷积层，以及位置嵌入（如果启用）。接着，文章说明了如何在任务解析函数中导入MHSA，并在YOLOv8的配置文件中应用该模块，用于模型的头部。最后，给出了配置文件的示例，展示了如何将MHSA集成到YOLOv8n的检测头部。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

一在ultralytics/nn中新建MHSA.py

添加代码

import torch
import torch.nn as nn

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_weight = nn.Parameter(torch.randn([1, heads, (n_dims) // heads, 1, int(height)]),
                                             requires_grad=True)
            self.rel_w_weight = 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)
        content_content = torch.matmul(q.permute(0, 1, 3, 2), k)  # 1,C,h*w,h*w
        c1, c2, c3, c4 = content_content.size()
        if self.pos:
            content_position = (self.rel_h_weight + self.rel_w_weight).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)
            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

if __name__ == '__main__':
    input = torch.randn(50, 512, 7, 7)
    mhsa = MHSA(n_dims=512)
    output = mhsa(input)
    print(output.shape)

二在 ultralytics/nn/tasks.py中导入

from ultralytics.nn.MHSA import MHSA

找到parse_model函数，加入

        elif m in {MHSA}:
            args=[ch[f],*args]

三在配置文件中修改

在ultralytics/models/v8中添加配置文件yolov8n_MHSA.yaml

# Ultralytics YOLO 🚀, GPL-3.0 license

# Parameters
nc: 2 # number of classes
depth_multiple: 0.33  # scales module repeats
width_multiple: 0.25  # scales convolution channels

# 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
  - [-1, 1, MHSA, [14,14,4]]  #10
# YOLOv8.0n head
head:
  - [-1, 1, nn.Upsample, [None, 2, 'nearest']]
  - [[-1, 6], 1, Concat, [1]]  # cat backbone P4
  - [-1, 3, C2f, [512]]  # 12

  - [-1, 1, nn.Upsample, [None, 2, 'nearest']]
  - [[-1, 4], 1, Concat, [1]]  # cat backbone P3
  - [-1, 3, C2f, [256]]  # 15 (P3/8-small)

  - [-1, 1, Conv, [256, 3, 2]]
  - [[-1, 13], 1, Concat, [1]]  # cat head P4
  - [-1, 3, C2f, [512]]  # 18 (P4/16-medium)

  - [-1, 1, Conv, [512, 3, 2]]
  - [[-1, 10], 1, Concat, [1]]  # cat head P5
  - [-1, 3, C2f, [1024]]  # 21 (P5/32-large)

  - [[16, 19, 22], 1, Detect, [nc]]  # Detect(P3, P4, P5)

运行main.py（上一篇博客）