这篇博客【魔改YOLOv5-6.x(中)】:加入ACON激活函数、CBAM和CA注意力机制、加权双向特征金字塔BiFPN简要介绍了BiFPN的原理,以及YOLOv5作者如何结合BiFPN。
之前尝试过设置可学习的权重参数,将不同的分支进行Add操作,具体可以参考这篇博客:【YOLOv5-6.x】设置可学习权重结合BiFPN(Add操作)。
本文将尝试直接进行Concat操作来结合BiFPN。
本文以yolov5s.yaml为例进行修改,原作者在YOLOv5-6.1版本中已经给出了使用Concat操作结合BiFPN的模型文件,但是没有设置可学习参数,因此本文直接在其基础上进行修改:将Concat
全部换成BiFPN_Concat
# 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, 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, C3, [1024]],
[-1, 1, SPPF, [1024, 5]], # 9
]
# YOLOv5 v6.0 BiFPN head
head:
[[-1, 1, Conv, [512, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 6], 1, BiFPN_Concat2, [1]], # cat backbone P4 <--- BiFPN change
[-1, 3, C3, [512, False]], # 13
[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 4], 1, BiFPN_Concat2, [1]], # cat backbone P3 <--- BiFPN change
[-1, 3, C3, [256, False]], # 17 (P3/8-small)
[-1, 1, Conv, [256, 3, 2]],
[[-1, 14, 6], 1, BiFPN_Concat3, [1]], # cat P4 <--- BiFPN change
[-1, 3, C3, [512, False]], # 20 (P4/16-medium)
[-1, 1, Conv, [512, 3, 2]],
[[-1, 10], 1, BiFPN_Concat2, [1]], # cat head P5 <--- BiFPN change
[-1, 3, C3, [1024, False]], # 23 (P5/32-large)
[[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
]
可以参考这篇博客:【YOLOv5-6.x】模型参数及detect层输出测试(自用),进行模型配置文件测试并查看输出结果:
from n params module arguments
0 -1 1 3520 models.common.Conv [3, 32, 6, 2, 2]
1 -1 1 18560 models.common.Conv [32, 64, 3, 2]
2 -1 1 18816 models.common.C3 [64, 64, 1]
3 -1 1 73984 models.common.Conv [64, 128, 3, 2]
4 -1 2 115712 models.common.C3 [128, 128, 2]
5 -1 1 295424 models.common.Conv [128, 256, 3, 2]
6 -1 3 625152 models.common.C3 [256, 256, 3]
7 -1 1 1180672 models.common.Conv [256, 512, 3, 2]
8 -1 1 1182720 models.common.C3 [512, 512, 1]
9 -1 1 656896 models.common.SPPF [512, 512, 5]
10 -1 1 131584 models.common.Conv [512, 256, 1, 1]
11 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
12 [-1, 6] 1 2 models.common.BiFPN_Concat2 [1]
13 -1 1 361984 models.common.C3 [512, 256, 1, False]
14 -1 1 33024 models.common.Conv [256, 128, 1, 1]
15 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
16 [-1, 4] 1 2 models.common.BiFPN_Concat2 [1]
17 -1 1 90880 models.common.C3 [256, 128, 1, False]
18 -1 1 147712 models.common.Conv [128, 128, 3, 2]
19 [-1, 14, 6] 1 3 models.common.BiFPN_Concat3 [1]
20 -1 1 361984 models.common.C3 [512, 256, 1, False]
21 -1 1 590336 models.common.Conv [256, 256, 3, 2]
22 [-1, 10] 1 2 models.common.BiFPN_Concat2 [1]
23 -1 1 1182720 models.common.C3 [512, 512, 1, False]
24 [17, 20, 23] 1 229245 models.yolo.Detect [80, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
Model Summary: 270 layers, 7300934 parameters, 7300934 gradients, 16.7 GFLOPs
# 结合BiFPN 设置可学习参数 学习不同分支的权重
# 两个分支concat操作
class BiFPN_Concat2(nn.Module):
def __init__(self, dimension=1):
super(BiFPN_Concat2, self).__init__()
self.d = dimension
self.w = nn.Parameter(torch.ones(2, dtype=torch.float32), requires_grad=True)
self.epsilon = 0.0001
def forward(self, x):
w = self.w
weight = w / (torch.sum(w, dim=0) + self.epsilon) # 将权重进行归一化
# Fast normalized fusion
x = [weight[0] * x[0], weight[1] * x[1]]
return torch.cat(x, self.d)
# 三个分支concat操作
class BiFPN_Concat3(nn.Module):
def __init__(self, dimension=1):
super(BiFPN_Concat3, self).__init__()
self.d = dimension
# 设置可学习参数 nn.Parameter的作用是:将一个不可训练的类型Tensor转换成可以训练的类型parameter
# 并且会向宿主模型注册该参数 成为其一部分 即model.parameters()会包含这个parameter
# 从而在参数优化的时候可以自动一起优化
self.w = nn.Parameter(torch.ones(3, dtype=torch.float32), requires_grad=True)
self.epsilon = 0.0001
def forward(self, x):
w = self.w
weight = w / (torch.sum(w, dim=0) + self.epsilon) # 将权重进行归一化
# Fast normalized fusion
x = [weight[0] * x[0], weight[1] * x[1], weight[2] * x[2]]
return torch.cat(x, self.d)
parse_model
函数中找到elif m is Concat:
语句,将其改为:# 添加bifpn_concat结构
elif m in [Concat, BiFPN_Concat2, BiFPN_Concat3]:
c2 = sum(ch[x] for x in f)
BiFPN_Concat2
和BiFPN_Concat3
函数中定义的w
参数,加入g1
g0, g1, g2 = [], [], [] # optimizer parameter groups
for v in model.modules():
# hasattr: 测试指定的对象是否具有给定的属性,返回一个布尔值
if hasattr(v, 'bias') and isinstance(v.bias, nn.Parameter): # bias
g2.append(v.bias) # biases
if isinstance(v, nn.BatchNorm2d): # weight (no decay)
g0.append(v.weight)
elif hasattr(v, 'weight') and isinstance(v.weight, nn.Parameter): # weight (with decay)
g1.append(v.weight)
# BiFPN_Concat
elif isinstance(v, BiFPN_Concat2) and hasattr(v, 'w') and isinstance(v.w, nn.Parameter):
g1.append(v.w)
elif isinstance(v, BiFPN_Concat3) and hasattr(v, 'w') and isinstance(v.w, nn.Parameter):
g1.append(v.w)
想要查看BiFPN_Concat层的参数更新情况,可以参考这篇博客【Pytorch】查看模型某一层的参数数值(自用),直接定位到w
参数,随着模型训练输出对应的值。