当前位置:网站首页>YOLOv5s-ShuffleNetV2
YOLOv5s-ShuffleNetV2
2022-07-04 19:56:00 【Master Ma】
Yes YOLOV5 Lightweight :
One 、backbone part
yaml The configuration file :
backbone:
# [from, number, module, args]
[[-1, 1, conv_bn_relu_maxpool, [32]], # 0-P2/4
[-1, 1, Shuffle_Block, [116, 2]], # 1-P3/8
[-1, 3, Shuffle_Block, [116, 1]], # 2
[-1, 1, Shuffle_Block, [232, 2]], # 3-P4/16
[-1, 7, Shuffle_Block, [232, 1]], # 4
[-1, 1, Shuffle_Block, [464, 2]], # 5-P5/32
[-1, 1, Shuffle_Block, [464, 1]], # 6
]
1.1、Focus Replace
The original YOLOv5s-5.0 Of stem It's a Focus Slicing operation , and v6 It's a 6x6Conv, Here is a copy v6 Yes Focus Improvement , Change it to 1 individual 3x3 Convolution ( Because my task itself is not complicated , Change it to 3x3 The parameters can be reduced after )
class conv_bn_relu_maxpool(nn.Module):
def __init__(self, c1, c2): # ch_in, ch_out
super(conv_bn_relu_maxpool, self).__init__()
self.conv = nn.Sequential(
nn.Conv2d(c1, c2, kernel_size=3, stride=2, padding=1, bias=False),
nn.BatchNorm2d(c2),
nn.ReLU(inplace=True),
)
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
def forward(self, x):
return self.maxpool(self.conv(x))
1.2、 all Conv+C3 Replace with Shuffle_Block
def channel_shuffle(x, groups):
batchsize, num_channels, height, width = x.data.size() # bs c h w
channels_per_group = num_channels // groups
# reshape
x = x.view(batchsize, groups, channels_per_group, height, width) # [bs,c,h,w] to [bs,group,channels_per_group,h,w]
x = torch.transpose(x, 1, 2).contiguous() # channel shuffle [bs,channels_per_group,group,h,w]
# flatten
x = x.view(batchsize, -1, height, width) # [bs,c,h,w]
return x
class Shuffle_Block(nn.Module):
def __init__(self, inp, oup, stride):
super(Shuffle_Block, self).__init__()
if not (1 <= stride <= 3):
raise ValueError('illegal stride value')
self.stride = stride
branch_features = oup // 2 # channel split to 2 feature map
assert (self.stride != 1) or (inp == branch_features << 1)
# stride=2 chart d The left branch =3x3DW Conv + 1x1Conv
if self.stride > 1:
self.branch1 = nn.Sequential(
self.depthwise_conv(inp, inp, kernel_size=3, stride=self.stride, padding=1),
nn.BatchNorm2d(inp),
nn.Conv2d(inp, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
nn.BatchNorm2d(branch_features),
nn.ReLU(inplace=True),
)
# Right branch =1x1Conv + 3x3DW Conv + 1x1Conv
self.branch2 = nn.Sequential(
nn.Conv2d(inp if (self.stride > 1) else branch_features,
branch_features, kernel_size=1, stride=1, padding=0, bias=False),
nn.BatchNorm2d(branch_features),
nn.ReLU(inplace=True),
self.depthwise_conv(branch_features, branch_features, kernel_size=3, stride=self.stride, padding=1),
nn.BatchNorm2d(branch_features),
nn.Conv2d(branch_features, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
nn.BatchNorm2d(branch_features),
nn.ReLU(inplace=True),
)
@staticmethod
def depthwise_conv(i, o, kernel_size, stride=1, padding=0, bias=False):
return nn.Conv2d(i, o, kernel_size, stride, padding, bias=bias, groups=i)
def forward(self, x):
# x/out: [bs, c, h, w]
if self.stride == 1:
x1, x2 = x.chunk(2, dim=1) # channel split to 2 feature map
out = torch.cat((x1, self.branch2(x2)), dim=1)
else:
out = torch.cat((self.branch1(x), self.branch2(x)), dim=1)
out = channel_shuffle(out, 2)
return out
1.3、 Cut off SPP
Cut off SPP Structure and the one behind C3 structure , because SPP Parallel operation will affect the speed .
Two 、head part
head:
[[-1, 1, Conv, [96, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[ -1, 4 ], 1, Concat, [1]], # cat backbone P4
[-1, 1, DWConvblock, [96, 3, 1]], # 10
[-1, 1, Conv, [96, 1, 1 ]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 2], 1, Concat, [1]], # cat backbone P3
[-1, 1, DWConvblock, [96, 3, 1]], # 14 (P3/8-small)
[-1, 1, DWConvblock, [96, 3, 2]],
[[-1, 11], 1, ADD, [1]], # cat head P4
[-1, 1, DWConvblock, [96, 3, 1]], # 17 (P4/16-medium)
[-1, 1, DWConvblock, [ 96, 3, 2]],
[[-1, 7], 1, ADD, [1]], # cat head P5
[-1, 1, DWConvblock, [96, 3, 1]], # 20 (P5/32-large)
[[14, 17, 20], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
]
2.1、 All layer structure inputs and outputs channel equal
2.2、 all C3 Replace all structures with DWConv
2.3、PAN Of the two Concat Change it to ADD
3、 ... and 、、 summary
ShuffleNeckV2 Four criteria for designing lightweight networks are proposed :
G1、 Input characteristics of convolution layer channel And output characteristics channel Try to be equal ;
G2、 Try not to use group convolution , Or convolution group g As small as possible ;
G3、 Network branches should be as few as possible , Avoid parallel structures ;
G4、 Element-Wise The operation should be as few as possible , Such as :ReLU、ADD、 Point by point convolution, etc ;
YOLOv5s-ShuffleNetV2 Summary of improvement points :
backbone Of Focus Replace with a 3x3Conv(c=32), because v5-6.0 Just replace it with a 6x6Conv, Here, in order to further reduce the amount of parameters , Replace with 3x3Conv;
backbone all Conv and C3 Replace with Shuffle Block;
Cut off SPP And the one behind C3 structure ,SPP There are too many parallel operations (G3)
head Input and output of all layers channel=96(G1)
head all C3 Change it to DWConv
PAN Of the two Concat Change it to ADD(channel Too big , Too much computation , Although it violates G4, But the amount of calculation is smaller )
Four 、 experimental result
GFLOPs= value /10^9
Parameter quantity (M)= value *4/1024/1024
Parameter quantity 、 Amount of computation 、 The weight file size is compressed to YOLOv5s Of 1/10, precision [email protected] It fell off 1% about (96.7%->95.5%),[email protected]~0.95 It fell off 5 A p.m. (88.5%->84%).
reference :https://blog.csdn.net/qq_38253797/article/details/124803531
边栏推荐
- Upgrade the smart switch, how much is the difference between the "zero fire version" and "single fire" wiring methods?
- 线上数据库迁移的几种方法
- TCP两次挥手,你见过吗?那四次握手呢?
- 1007 maximum subsequence sum (25 points) (PAT class a)
- 92. (cesium chapter) cesium building layering
- How to use async Awati asynchronous task processing instead of backgroundworker?
- YOLOv5s-ShuffleNetV2
- 1005 Spell It Right(20 分)(PAT甲级)
- 1002. A+B for Polynomials (25)(PAT甲级)
- 牛客小白月赛7 I 新建 Microsoft Office Word 文档
猜你喜欢
PointNeXt:通过改进的模型训练和缩放策略审视PointNet++
Multi table operation inner join query
Pointnet/Pointnet++点云数据集处理并训练
Write it down once Net analysis of thread burst height of an industrial control data acquisition platform
Pythagorean number law (any three numbers can meet the conditions of Pythagorean theorem)
Introduction to polyfit software
Online text line fixed length fill tool
TCP两次挥手,你见过吗?那四次握手呢?
做社交媒体营销应该注意些什么?Shopline卖家的成功秘笈在这里!
公司要上监控,Zabbix 和 Prometheus 怎么选?这么选准没错!
随机推荐
HDU 1097 A hard puzzle
Explore the contour drawing function drawcontours() of OpenCV in detail with practical examples
数据集划分
The page element is vertically and horizontally centered, realizing the vertical and horizontal centering of known or unknown width.
New wizard effect used by BCG
What should we pay attention to when doing social media marketing? Here is the success secret of shopline sellers!
C # use stopwatch to measure the running time of the program
Educational Codeforces Round 22 E. Army Creation
Data set division
Jetpack Compose 教程
解密函数计算异步任务能力之「任务的状态及生命周期管理」
项目中遇到的线上数据迁移方案1---总体思路整理和技术梳理
92.(cesium篇)cesium楼栋分层
牛客小白月赛7 F题
Oracle with as ora-00903: invalid table name multi report error
Crystal optoelectronics: ar-hud products of Chang'an dark blue sl03 are supplied by the company
Niuke Xiaobai month race 7 e applese's super ability
Introduction to polyfit software
1006 Sign In and Sign Out(25 分)(PAT甲级)
HDU 6440 2018 Chinese college student program design network competition