yolov5 的网络结构
yolov5 的网络结构的配置文件在models文件夹下，有yolov5n.yaml, yolov5s.yaml, yolov5m.yaml等等。几个网络结构其实都一样，通过depth_multiple和width_multiple参数来控制网络结构的深度和宽度。
主要理解 head和backbone部分，根据需求对模型做修改时主要也是改动backbone部分。
# 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 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)
]
最新版本的yolov5 v6.0的backbone已经不用Focus了，直接使用卷积核为6，stride为2的卷积层。
from：输入来自那一层，-1代表前面一层，1代表第1层，3代表第3层，[-1, 6]则表示来自上一层和第6层的输入维度相加(concat)
number：模块的数量，最终数量需要乘width，然后四舍五入取整，如果小于1，取1。
module：子模块名称
args：模块参数，有 out_channel，kernel_size，stride，padding，bias等
自定义网络结构
自定义网络结构需要三个步骤：
- 在model文件夹下创建新的模块，比如mobilenetv3.py，编写你要加的结构，再在yolo.py中导入
- 更改yaml配置文件，将backbone或者head中需要改的层换成自己的
- 更改yolo.py 中的parse_model解析函数
第一步，编写mobilenetv3.py
# MobileNetV3
import torch.nn as nn
class h_sigmoid(nn.Module):
def __init__(self, inplace=True):
super(h_sigmoid, self).__init__()
self.relu = nn.ReLU6(inplace=inplace)
def forward(self, x):
return self.relu(x + 3) / 6
class h_swish(nn.Module):
def __init__(self, inplace=True):
super(h_swish, self).__init__()
self.sigmoid = h_sigmoid(inplace=inplace)
def forward(self, x):
y = self.sigmoid(x)
return x * y
class SELayer(nn.Module):
def __init__(self, channel, reduction=4):
super(SELayer, self).__init__()
self.avg_pool = nn.AdaptiveAvgPool2d(1)
self.fc = nn.Sequential(
nn.Linear(channel, channel // reduction),
nn.ReLU(inplace=True),
nn.Linear(channel // reduction, channel),
h_sigmoid()
)
def forward(self, x):
b, c, _, _ = x.size()
y = self.avg_pool(x)
y = y.view(b, c)
y = self.fc(y).view(b, c, 1, 1)
return x * y
class conv_bn_hswish(nn.Module):
"""
This equals to
def conv_3x3_bn(inp, oup, stride):
return nn.Sequential(
nn.Conv2d(inp, oup, 3, stride, 1, bias=False),
nn.BatchNorm2d(oup),
h_swish()
)
"""
def __init__(self, c1, c2, stride):
super(conv_bn_hswish, self).__init__()
self.conv = nn.Conv2d(c1, c2, 3, stride, 1, bias=False)
self.bn = nn.BatchNorm2d(c2)
self.act = h_swish()
def forward(self, x):
return self.act(self.bn(self.conv(x)))
def fuseforward(self, x):
return self.act(self.conv(x))
class MobileNetV3_InvertedResidual(nn.Module):
def __init__(self, inp, oup, hidden_dim, kernel_size, stride, use_se, use_hs):
super(MobileNetV3_InvertedResidual, self).__init__()
assert stride in [1, 2]
self.identity = stride == 1 and inp == oup
if inp == hidden_dim:
self.conv = nn.Sequential(
# dw
nn.Conv2d(hidden_dim, hidden_dim, kernel_size, stride, (kernel_size - 1) // 2, groups=hidden_dim,
bias=False),
nn.BatchNorm2d(hidden_dim),
h_swish() if use_hs else nn.ReLU(inplace=True),
# Squeeze-and-Excite
SELayer(hidden_dim) if use_se else nn.Sequential(),
# pw-linear
nn.Conv2d(hidden_dim, oup, 1, 1, 0, bias=False),
nn.BatchNorm2d(oup),
)
else:
self.conv = nn.Sequential(
# pw
nn.Conv2d(inp, hidden_dim, 1, 1, 0, bias=False),
nn.BatchNorm2d(hidden_dim),
h_swish() if use_hs else nn.ReLU(inplace=True),
# dw
nn.Conv2d(hidden_dim, hidden_dim, kernel_size, stride, (kernel_size - 1) // 2, groups=hidden_dim,
bias=False),
nn.BatchNorm2d(hidden_dim),
# Squeeze-and-Excite
SELayer(hidden_dim) if use_se else nn.Sequential(),
h_swish() if use_hs else nn.ReLU(inplace=True),
# pw-linear
nn.Conv2d(hidden_dim, oup, 1, 1, 0, bias=False),
nn.BatchNorm2d(oup),
)
def forward(self, x):
y = self.conv(x)
if self.identity:
return x + y
else:
return y
在yolo.py中加上：
from models.mobilenetv3 import *
第二步，新建mobilenetv3small.yaml，将yolov5中的backbone的卷积模块全部替换成MobileNetV3_InvertedResidual
# parameters
nc: 80 # number of classes
depth_multiple: 1.0
width_multiple: 1.0
# anchors
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
# custom backbone
backbone:
# MobileNetV3-small
# [from, number, module, args]
[[-1, 1, conv_bn_hswish, [16, 2]], # 0-p1/2
[-1, 1, MobileNetV3_InvertedResidual, [16, 16, 3, 2, 1, 0]], # 1-p2/4
[-1, 1, MobileNetV3_InvertedResidual, [24, 72, 3, 2, 0, 0]], # 2-p3/8
[-1, 1, MobileNetV3_InvertedResidual, [24, 88, 3, 1, 0, 0]], # 3-p3/8
[-1, 1, MobileNetV3_InvertedResidual, [40, 96, 5, 2, 1, 1]], # 4-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [40, 240, 5, 1, 1, 1]], # 5-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [40, 240, 5, 1, 1, 1]], # 6-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [48, 120, 5, 1, 1, 1]], # 7-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [48, 144, 5, 1, 1, 1]], # 8-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [96, 288, 5, 2, 1, 1]], # 9-p5/32
[-1, 1, MobileNetV3_InvertedResidual, [96, 576, 5, 1, 1, 1]], # 10-p5/32
[-1, 1, MobileNetV3_InvertedResidual, [96, 576, 5, 1, 1, 1]], # 11-p5/32
]
head:
[[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 8], 1, Concat, [1]], # cat backbone P4
[-1, 1, C3, [256, False]], # 15
[-1, 1, Conv, [128, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 3], 1, Concat, [1]], # cat backbone P3
[-1, 1, C3, [128, False]], # 19 (P3/8-small)
[-1, 1, Conv, [128, 3, 2]],
[[-1, 16], 1, Concat, [1]], # cat head P4
[-1, 1, C3, [256, False]], # 22 (P4/16-medium)
[-1, 1, Conv, [256, 3, 2]],
[[-1, 12], 1, Concat, [1]], # cat head P5
[-1, 1, C3, [512, False]], # 25 (P5/32-large)
[[19, 22, 25], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
]
第三步，更改解析模块，models/yolo.py中的parse_model函数
if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv,
BottleneckCSP, C3, C3TR, C3SPP, C3Ghost, conv_bn_hswish, MobileNetV3_InvertedResidual]:
开始训练
python train.py --data coco.yaml --img 640 --batch 64 --cfg mobilenetv3small.yaml --weights ''
型号参数
from n params module arguments
0 -1 1 464 models.mobilenetv3.conv_bn_hswish [3, 16, 2]
1 -1 1 612 models.mobilenetv3.MobileNetV3_InvertedResidual[16, 16, 16, 3, 2, 1, 0]
2 -1 1 3864 models.mobilenetv3.MobileNetV3_InvertedResidual[16, 24, 72, 3, 2, 0, 0]
3 -1 1 5416 models.mobilenetv3.MobileNetV3_InvertedResidual[24, 24, 88, 3, 1, 0, 0]
4 -1 1 13736 models.mobilenetv3.MobileNetV3_InvertedResidual[24, 40, 96, 5, 2, 1, 1]
5 -1 1 55340 models.mobilenetv3.MobileNetV3_InvertedResidual[40, 40, 240, 5, 1, 1, 1]
6 -1 1 55340 models.mobilenetv3.MobileNetV3_InvertedResidual[40, 40, 240, 5, 1, 1, 1]
7 -1 1 21486 models.mobilenetv3.MobileNetV3_InvertedResidual[40, 48, 120, 5, 1, 1, 1]
8 -1 1 28644 models.mobilenetv3.MobileNetV3_InvertedResidual[48, 48, 144, 5, 1, 1, 1]
9 -1 1 91848 models.mobilenetv3.MobileNetV3_InvertedResidual[48, 96, 288, 5, 2, 1, 1]
10 -1 1 294096 models.mobilenetv3.MobileNetV3_InvertedResidual[96, 96, 576, 5, 1, 1, 1]
11 -1 1 294096 models.mobilenetv3.MobileNetV3_InvertedResidual[96, 96, 576, 5, 1, 1, 1]
12 -1 1 25088 models.common.Conv [96, 256, 1, 1]
13 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
14 [-1, 8] 1 0 models.common.Concat [1]
15 -1 1 308736 models.common.C3 [304, 256, 1, False]
16 -1 1 33024 models.common.Conv [256, 128, 1, 1]
17 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
18 [-1, 3] 1 0 models.common.Concat [1]
19 -1 1 77568 models.common.C3 [152, 128, 1, False]
20 -1 1 147712 models.common.Conv [128, 128, 3, 2]
21 [-1, 16] 1 0 models.common.Concat [1]
22 -1 1 296448 models.common.C3 [256, 256, 1, False]
23 -1 1 590336 models.common.Conv [256, 256, 3, 2]
24 [-1, 12] 1 0 models.common.Concat [1]
25 -1 1 1182720 models.common.C3 [512, 512, 1, False]
26 [19, 22, 25] 1 18879 models.yolo.Detect [2, [[10, 13, 16, 30, 33, 23], [30, 61, 62, 45, 59, 119], [116, 90, 156, 198, 373, 326]], [128, 256, 512]]
Model Summary: 340 layers, 3545453 parameters, 3545453 gradients, 6.3 GFLOPs
yaml 文件中的这些参数[16, 16, 3, 2, 1, 0]]也是自定义的
[-1, 1, MobileNetV3_InvertedResidual, [16, 16, 3, 2, 1, 0]], # 1-p2/4
[-1, 1, MobileNetV3_InvertedResidual, [24, 72, 3, 2, 0, 0]], # 2-p3/8
[-1, 1, MobileNetV3_InvertedResidual, [24, 88, 3, 1, 0, 0]], # 3-p3/8
[-1, 1, MobileNetV3_InvertedResidual, [40, 96, 5, 2, 1, 1]], # 4-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [40, 240, 5, 1, 1, 1]], # 5-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [40, 240, 5, 1, 1, 1]], # 6-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [48, 120, 5, 1, 1, 1]], # 7-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [48, 144, 5, 1, 1, 1]], # 8-p4/16
[-1, 1, MobileNetV3_InvertedResidual, [96, 288, 5, 2, 1, 1]], # 9-p5/32
[-1, 1, MobileNetV3_InvertedResidual, [96, 576, 5, 1, 1, 1]], # 10-p5/32
[-1, 1, MobileNetV3_InvertedResidual, [96, 576, 5, 1, 1, 1]], # 11-p5/32
在mobilenetv3.py中都有定义，除了inp(input_channel)以外，其他一一对应
class MobileNetV3_InvertedResidual(nn.Module):
def __init__(self, inp, oup, hidden_dim, kernel_size, stride, use_se, use_hs):
以[16, 16, 3, 2, 1, 0]]为例：
oup=16,
hidden_dim=16,
kernel_size=3,
stride=2,
use_se=1，表示是否使用SELayer
use_hs=0，表示使用h_swish还是ReLU
use_se表示是否使用SELayer
参考
- 目标检测 YOLOv5 自定义网络结构
文章出处登录后可见！
已经登录？立即刷新
