一、本文介绍
本文记录的是利用ExtraDW
优化YOLOv10
中的RepNCSPELAN4
,详细说明了优化原因,注意事项等。ExtraDW
是MobileNetv4
模型中提出的新模块,允许以低成本增加网络深度和感受野,具有ConvNext和IB的组合优势。可以在提高模型精度的同时降低一定量的模型参数。
专栏目录:****YOLOv10改进目录一览 | 涉及卷积层、轻量化、注意力、损失函数、Backbone、SPPF、Neck、检测头等全方位改进
专栏地址:****YOLOv10改进专栏------以发表论文的角度,快速准确的找到有效涨点的创新点!
文章目录
- 一、本文介绍
- 二、UIB介绍
-
- [2.1 UIB结构设计](#2.1 UIB结构设计)
- [2.2 ExtraDW结构组成](#2.2 ExtraDW结构组成)
- [2.3 ExtraDW特点](#2.3 ExtraDW特点)
- 三、ExtraDW的实现代码
- 四、添加步骤
-
- [4.1 改进点1](#4.1 改进点1)
- [4.2 改进点2⭐](#4.2 改进点2⭐)
- 五、添加步骤
-
- [5.1 修改一](#5.1 修改一)
- [5.2 修改二](#5.2 修改二)
- [5.3 修改三](#5.3 修改三)
- 六、yaml模型文件
-
- [6.1 模型改进版本一](#6.1 模型改进版本一)
- [6.2 模型改进版本二⭐](#6.2 模型改进版本二⭐)
- 七、成功运行结果
二、UIB介绍
Universal Inverted Bottleneck(UIB)
通用反向瓶颈结构。
2.1 UIB结构设计
-
基于
MobileNetV4
UIB
建立在MobileNetV4
之上,即采用深度可分离卷积
和逐点
扩展及投影的反向瓶颈
结构。- 在
反向瓶颈块(IB)
中引入两个可选的深度可分离卷积
,一个在扩展层之前,另一个在扩展层和投影层之间。
-
UIB有四种可能的实例化形式:
- Inverted Bottleneck (IB):对扩展后的特征激活进行空间混合,以增加成本为代价提供更大的模型容量。
- ConvNext:通过在扩展之前进行空间混合,使用更大的核尺寸实现更便宜的空间混合。
- ExtraDW :文中引入的新变体,允许以低成本增加网络深度和感受野,具有
ConvNext
和IB
的组合优势。 - FFN :由两个
1x1逐点卷积(PW)
组成的栈,中间有激活和归一化层。
2.2 ExtraDW结构组成
结构组成:
- 在
IB块
中加入两个可选的深度可分离卷积
,一个在扩展层之前,另一个在扩展层和投影层之间。
2.3 ExtraDW特点
-
灵活性:
- 在每个网络阶段,可以灵活地进行空间和通道混合的权衡调整,根据需要扩大感受野,并最大化计算利用率,增强模型对输入特征的感知能力。
-
效率提升:
- 提供了一种廉价增加网络深度和感受野的方式。相比其他结构,它在增加网络深度和感受野的同时,不会带来过高的计算成本。
- 在论文中,与其他注意力机制结合时,能有效提高模型的运算强度,减少内存访问需求,从而提高模型效率。
论文:http://arxiv.org/abs/2404.10518
源码:https://github.com/tensorflow/models/blob/master/official/vision/modeling/backbones/mobilenet.py
三、ExtraDW的实现代码
ExtraDW模块
的实现代码如下:参考代码
python
import torch
import torch.nn as nn
from typing import Optional
def make_divisible(
value: float,
divisor: int,
min_value: Optional[float] = None,
round_down_protect: bool = True,
) -> int:
"""
This function is copied from here
"https://github.com/tensorflow/models/blob/master/official/vision/modeling/layers/nn_layers.py"
This is to ensure that all layers have channels that are divisible by 8.
Args:
value: A `float` of original value.
divisor: An `int` of the divisor that need to be checked upon.
min_value: A `float` of minimum value threshold.
round_down_protect: A `bool` indicating whether round down more than 10%
will be allowed.
Returns:
The adjusted value in `int` that is divisible against divisor.
"""
if min_value is None:
min_value = divisor
new_value = max(min_value, int(value + divisor / 2) // divisor * divisor)
# Make sure that round down does not go down by more than 10%.
if round_down_protect and new_value < 0.9 * value:
new_value += divisor
return int(new_value)
def conv2d(in_channels, out_channels, kernel_size=3, stride=1, groups=1, bias=False, norm=True, act=True):
conv = nn.Sequential()
padding = (kernel_size - 1) // 2
conv.append(nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, bias=bias, groups=groups))
if norm:
conv.append(nn.BatchNorm2d(out_channels))
if act:
conv.append(nn.ReLU6())
return conv
class UniversalInvertedBottleneckBlock(nn.Module):
def __init__(self, in_channels, out_channels, start_dw_kernel_size, middle_dw_kernel_size, middle_dw_downsample,
stride, expand_ratio):
"""An inverted bottleneck block with optional depthwises.
Referenced from here https://github.com/tensorflow/models/blob/master/official/vision/modeling/layers/nn_blocks.py
"""
super(UniversalInvertedBottleneckBlock, self).__init__()
# starting depthwise conv
self.start_dw_kernel_size = start_dw_kernel_size
if self.start_dw_kernel_size:
stride_ = stride if not middle_dw_downsample else 1
self._start_dw_ = conv2d(in_channels, in_channels, kernel_size=start_dw_kernel_size, stride=stride_, groups=in_channels, act=False)
# expansion with 1x1 convs
expand_filters = make_divisible(in_channels * expand_ratio, 8)
self._expand_conv = conv2d(in_channels, expand_filters, kernel_size=1)
# middle depthwise conv
self.middle_dw_kernel_size = middle_dw_kernel_size
if self.middle_dw_kernel_size:
stride_ = stride if middle_dw_downsample else 1
self._middle_dw = conv2d(expand_filters, expand_filters, kernel_size=middle_dw_kernel_size, stride=stride_, groups=expand_filters)
# projection with 1x1 convs
self._proj_conv = conv2d(expand_filters, out_channels, kernel_size=1, stride=1, act=False)
# expand depthwise conv (not used)
# _end_dw_kernel_size = 0
# self._end_dw = conv2d(out_channels, out_channels, kernel_size=_end_dw_kernel_size, stride=stride, groups=in_channels, act=False)
def forward(self, x):
if self.start_dw_kernel_size:
x = self._start_dw_(x)
# print("_start_dw_", x.shape)
x = self._expand_conv(x)
# print("_expand_conv", x.shape)
if self.middle_dw_kernel_size:
x = self._middle_dw(x)
# print("_middle_dw", x.shape)
x = self._proj_conv(x)
# print("_proj_conv", x.shape)
return x
def autopad(k, p=None, d=1): # kernel, padding, dilation
"""Pad to 'same' shape outputs."""
if d > 1:
k = d * (k - 1) + 1 if isinstance(k, int) else [d * (x - 1) + 1 for x in k] # actual kernel-size
if p is None:
p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad
return p
class Conv(nn.Module):
"""Standard convolution with args(ch_in, ch_out, kernel, stride, padding, groups, dilation, activation)."""
default_act = nn.SiLU() # default activation
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, d=1, act=True):
"""Initialize Conv layer with given arguments including activation."""
super().__init__()
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p, d), groups=g, dilation=d, bias=False)
self.bn = nn.BatchNorm2d(c2)
self.act = self.default_act if act is True else act if isinstance(act, nn.Module) else nn.Identity()
def forward(self, x):
"""Apply convolution, batch normalization and activation to input tensor."""
return self.act(self.bn(self.conv(x)))
def forward_fuse(self, x):
"""Perform transposed convolution of 2D data."""
return self.act(self.conv(x))
class Bottleneck(nn.Module):
"""Standard bottleneck."""
def __init__(self, c1, c2, shortcut=True, g=1, k=(3, 3), e=0.5):
"""Initializes a standard bottleneck module with optional shortcut connection and configurable parameters."""
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, k[0], 1)
self.cv2 = Conv(c_, c2, k[1], 1, g=g)
self.add = shortcut and c1 == c2
def forward(self, x):
"""Applies the YOLO FPN to input data."""
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
class C2f(nn.Module):
"""Faster Implementation of CSP Bottleneck with 2 convolutions."""
def __init__(self, c1, c2, n=1, shortcut=False, g=1, e=0.5):
"""Initializes a CSP bottleneck with 2 convolutions and n Bottleneck blocks for faster processing."""
super().__init__()
self.c = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, 2 * self.c, 1, 1)
self.cv2 = Conv((2 + n) * self.c, c2, 1) # optional act=FReLU(c2)
self.m = nn.ModuleList(Bottleneck(self.c, self.c, shortcut, g, k=((3, 3), (3, 3)), e=1.0) for _ in range(n))
def forward(self, x):
"""Forward pass through C2f layer."""
y = list(self.cv1(x).chunk(2, 1))
y.extend(m(y[-1]) for m in self.m)
return self.cv2(torch.cat(y, 1))
def forward_split(self, x):
"""Forward pass using split() instead of chunk()."""
y = list(self.cv1(x).split((self.c, self.c), 1))
y.extend(m(y[-1]) for m in self.m)
return self.cv2(torch.cat(y, 1))
class RepVGGDW(torch.nn.Module):
def __init__(self, ed) -> None:
super().__init__()
self.conv = Conv(ed, ed, 7, 1, 3, g=ed, act=False)
self.conv1 = Conv(ed, ed, 3, 1, 1, g=ed, act=False)
self.dim = ed
self.act = nn.SiLU()
def forward(self, x):
return self.act(self.conv(x) + self.conv1(x))
def forward_fuse(self, x):
return self.act(self.conv(x))
@torch.no_grad()
def fuse(self):
conv = fuse_conv_and_bn(self.conv.conv, self.conv.bn)
conv1 = fuse_conv_and_bn(self.conv1.conv, self.conv1.bn)
conv_w = conv.weight
conv_b = conv.bias
conv1_w = conv1.weight
conv1_b = conv1.bias
conv1_w = torch.nn.functional.pad(conv1_w, [2,2,2,2])
final_conv_w = conv_w + conv1_w
final_conv_b = conv_b + conv1_b
conv.weight.data.copy_(final_conv_w)
conv.bias.data.copy_(final_conv_b)
self.conv = conv
del self.conv1
class CIB(nn.Module):
"""Standard bottleneck."""
def __init__(self, c1, c2, shortcut=True, e=0.5, lk=False):
"""Initializes a bottleneck module with given input/output channels, shortcut option, group, kernels, and
expansion.
"""
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = nn.Sequential(
Conv(c1, c1, 3, g=c1),
Conv(c1, 2 * c_, 1),
Conv(2 * c_, 2 * c_, 3, g=2 * c_) if not lk else RepVGGDW(2 * c_),
Conv(2 * c_, c2, 1),
Conv(c2, c2, 3, g=c2),
UniversalInvertedBottleneckBlock(c2, c2, 5, 3, True, 1, 4)
)
self.add = shortcut and c1 == c2
def forward(self, x):
"""'forward()' applies the YOLO FPN to input data."""
return x + self.cv1(x) if self.add else self.cv1(x)
class C2fCIB_UIB(C2f):
"""Faster Implementation of CSP Bottleneck with 2 convolutions."""
def __init__(self, c1, c2, n=1, shortcut=False, lk=False, g=1, e=0.5):
"""Initialize CSP bottleneck layer with two convolutions with arguments ch_in, ch_out, number, shortcut, groups,
expansion.
"""
super().__init__(c1, c2, n, shortcut, g, e)
self.m = nn.ModuleList(CIB(self.c, self.c, shortcut, e=1.0, lk=lk) for _ in range(n))
四、添加步骤
4.1 改进点1
模块改进方法 1️⃣:直接加入UniversalInvertedBottleneckBlock模块
。
UniversalInvertedBottleneckBlock模块
添加后如下:
注意❗:在5.2和5.3小节
中需要声明的模块名称为:UniversalInvertedBottleneckBlock
。
4.2 改进点2⭐
模块改进方法 2️⃣:基于UniversalInvertedBottleneckBlock模块
的C2fCIB
。
第二种改进方法是对YOLOv10
中的C2fCIB模块
进行改进。UIB
中的ExtraDW
模块与C2fCIB
结合后,可以为YOLOv10
提供更丰富的特征表示,更好地调整特征的空间分布和通道信息,使得模型能够更有效地聚焦于目标相关的特征,减少无关信息的干扰,进而提高检测精度。
改进代码如下:
首先添加UniversalInvertedBottleneckBlock
模块改进CIB
模块。
python
class CIB(nn.Module):
"""Standard bottleneck."""
def __init__(self, c1, c2, shortcut=True, e=0.5, lk=False):
"""Initializes a bottleneck module with given input/output channels, shortcut option, group, kernels, and
expansion.
"""
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = nn.Sequential(
Conv(c1, c1, 3, g=c1),
Conv(c1, 2 * c_, 1),
Conv(2 * c_, 2 * c_, 3, g=2 * c_) if not lk else RepVGGDW(2 * c_),
Conv(2 * c_, c2, 1),
Conv(c2, c2, 3, g=c2),
UniversalInvertedBottleneckBlock(c2, c2, 5, 3, True, 1, 4)
)
self.add = shortcut and c1 == c2
def forward(self, x):
"""'forward()' applies the YOLO FPN to input data."""
return x + self.cv1(x) if self.add else self.cv1(x)
再添加如下代码将C2fCIB
重命名为C2fCIB_UIB
python
class C2fCIB_UIB(C2f):
"""Faster Implementation of CSP Bottleneck with 2 convolutions."""
def __init__(self, c1, c2, n=1, shortcut=False, lk=False, g=1, e=0.5):
"""Initialize CSP bottleneck layer with two convolutions with arguments ch_in, ch_out, number, shortcut, groups,
expansion.
"""
super().__init__(c1, c2, n, shortcut, g, e)
self.m = nn.ModuleList(CIB(self.c, self.c, shortcut, e=1.0, lk=lk) for _ in range(n))
注意❗:在5.2和5.3小节
中需要声明的模块名称为:C2fCIB_UIB
。
五、添加步骤
5.1 修改一
① 在ultralytics/nn/
目录下新建AddModules
文件夹用于存放模块代码
② 在AddModules
文件夹下新建UIB.py
,将第三节中的代码粘贴到此处
5.2 修改二
在AddModules
文件夹下新建__init__.py
(已有则不用新建),在文件内导入模块:from .UIB import *
5.3 修改三
在ultralytics/nn/modules/tasks.py
文件中,需要在两处位置添加各模块类名称。
首先:导入模块
其次:在parse_model函数
中注册UniversalInvertedBottleneckBlock
和C2fCIB_UIB
模块
六、yaml模型文件
6.1 模型改进版本一
在代码配置完成后,配置模型的YAML文件。
此处以ultralytics/cfg/models/v10/yolov10m.yaml
为例,在同目录下创建一个用于自己数据集训练的模型文件yolov10m-UIB.yaml
。
将yolov10m.yaml
中的内容复制到yolov10m-UIB.yaml
文件下,修改nc
数量等于自己数据中目标的数量。
在骨干网络中添加UniversalInvertedBottleneckBlock模块
,。
python
# Parameters
nc: 1 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
# [depth, width, max_channels]
m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients, 79.3 GFLOPs
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, UniversalInvertedBottleneckBlock, [128, 0, 3, True, 1, 2]]
- [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
- [-1, 6, UniversalInvertedBottleneckBlock, [256, 0, 3, True, 1, 2]]
- [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
- [-1, 6, UniversalInvertedBottleneckBlock, [512, 5, 3, True, 1, 4]]
- [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
- [-1, 3, UniversalInvertedBottleneckBlock, [1024, 5, 3, True, 1, 4]]
- [-1, 1, SPPF, [1024, 5]] # 9
- [-1, 1, PSA, [1024]] # 10
# YOLOv8.0n head
head:
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 6], 1, Concat, [1]] # cat backbone P4
- [-1, 3, C2f, [512]] # 13
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 4], 1, Concat, [1]] # cat backbone P3
- [-1, 3, C2f, [256]] # 16 (P3/8-small)
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 13], 1, Concat, [1]] # cat head P4
- [-1, 3, C2f, [512]] # 19 (P4/16-medium)
- [-1, 1, SCDown, [512, 3, 2]]
- [[-1, 10], 1, Concat, [1]] # cat head P5
- [-1, 3, C2fCIB, [1024, True, True]] # 22 (P5/32-large)
- [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)
6.2 模型改进版本二⭐
此处同样以ultralytics/cfg/models/v10/yolov11m.yaml
为例,在同目录下创建一个用于自己数据集训练的模型文件yolov10m-C2fCIB_UIB.yaml
。
将yolov10m.yaml
中的内容复制到yolov10m-C2fCIB_UIB.yaml
文件下,修改nc
数量等于自己数据中目标的数量。
📌 模型的修改方法是将骨干网络 中的所有C2fCIB模块
替换成C2fCIB_UIB模块
。
python
# Parameters
nc: 1 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
# [depth, width, max_channels]
m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients, 79.3 GFLOPs
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, SCDown, [512, 3, 2]] # 5-P4/16
- [-1, 6, C2f, [512, True]]
- [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
- [-1, 3, C2fCIB_UIB, [1024, True, True]]
- [-1, 1, SPPF, [1024, 5]] # 9
- [-1, 1, PSA, [1024]] # 10
# YOLOv8.0n head
head:
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 6], 1, Concat, [1]] # cat backbone P4
- [-1, 3, C2f, [512]] # 13
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 4], 1, Concat, [1]] # cat backbone P3
- [-1, 3, C2f, [256]] # 16 (P3/8-small)
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 13], 1, Concat, [1]] # cat head P4
- [-1, 3, C2f, [512]] # 19 (P4/16-medium)
- [-1, 1, SCDown, [512, 3, 2]]
- [[-1, 10], 1, Concat, [1]] # cat head P5
- [-1, 3, C2fCIB, [1024, True, True]] # 22 (P5/32-large)
- [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)
七、成功运行结果
分别打印网络模型可以看到UniversalInvertedBottleneckBlock
和C2fCIB_UIB
已经加入到模型中,并可以进行训练了。
YOLOv10m-UIB:
YOLOv10m-UIB summary: 522 layers, 22,630,438 parameters, 22,630,422 gradients, 71.3 GFLOPs
python
from n params module arguments
0 -1 1 1392 ultralytics.nn.modules.conv.Conv [3, 48, 3, 2]
1 -1 1 41664 ultralytics.nn.modules.conv.Conv [48, 96, 3, 2]
2 -1 2 79104 ultralytics.nn.modules.block.UniversalInvertedBottleneckBlock[96, 96, 0, 3, True, 1, 2]
3 -1 1 166272 ultralytics.nn.modules.conv.Conv [96, 192, 3, 2]
4 -1 4 611328 ultralytics.nn.modules.block.UniversalInvertedBottleneckBlock[192, 192, 0, 3, True, 1, 2]
5 -1 1 78720 ultralytics.nn.modules.block.SCDown [192, 384, 3, 2]
6 -1 4 4843008 ultralytics.nn.modules.block.UniversalInvertedBottleneckBlock[384, 384, 5, 3, True, 1, 4]
7 -1 1 228672 ultralytics.nn.modules.block.SCDown [384, 576, 3, 2]
8 -1 2 5401728 ultralytics.nn.modules.block.UniversalInvertedBottleneckBlock[576, 576, 5, 3, True, 1, 4]
9 -1 1 831168 ultralytics.nn.modules.block.SPPF [576, 576, 5]
10 -1 1 1253088 ultralytics.nn.modules.block.PSA [576, 576]
11 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
12 [-1, 6] 1 0 ultralytics.nn.modules.conv.Concat [1]
13 -1 2 1993728 ultralytics.nn.modules.block.C2f [960, 384, 2]
14 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
15 [-1, 4] 1 0 ultralytics.nn.modules.conv.Concat [1]
16 -1 2 517632 ultralytics.nn.modules.block.C2f [576, 192, 2]
17 -1 1 332160 ultralytics.nn.modules.conv.Conv [192, 192, 3, 2]
18 [-1, 13] 1 0 ultralytics.nn.modules.conv.Concat [1]
19 -1 2 1846272 ultralytics.nn.modules.block.C2f [576, 384, 2]
20 -1 1 152448 ultralytics.nn.modules.block.SCDown [384, 384, 3, 2]
21 [-1, 10] 1 0 ultralytics.nn.modules.conv.Concat [1]
22 -1 2 1969920 ultralytics.nn.modules.block.C2fCIB [960, 576, 2, True, True]
23 [16, 19, 22] 1 2282134 ultralytics.nn.modules.head.v10Detect [1, [192, 384, 576]]
YOLOv10m-UIB summary: 522 layers, 22,630,438 parameters, 22,630,422 gradients, 71.3 GFLOPs
YOLOv10m-C2fCIB_UIB:
YOLOv10m-C2fCIB_UIB summary: 532 layers, 19,598,182 parameters, 19,598,166 gradients, 68.9 GFLOPs
python
from n params module arguments
0 -1 1 1392 ultralytics.nn.modules.conv.Conv [3, 48, 3, 2]
1 -1 1 41664 ultralytics.nn.modules.conv.Conv [48, 96, 3, 2]
2 -1 2 111360 ultralytics.nn.modules.block.C2f [96, 96, 2, True]
3 -1 1 166272 ultralytics.nn.modules.conv.Conv [96, 192, 3, 2]
4 -1 4 813312 ultralytics.nn.modules.block.C2f [192, 192, 4, True]
5 -1 1 78720 ultralytics.nn.modules.block.SCDown [192, 384, 3, 2]
6 -1 4 3248640 ultralytics.nn.modules.block.C2f [384, 384, 4, True]
7 -1 1 228672 ultralytics.nn.modules.block.SCDown [384, 576, 3, 2]
8 -1 2 3729600 ultralytics.nn.AddModules.UIB.C2fCIB_UIB [576, 576, True, True]
9 -1 1 831168 ultralytics.nn.modules.block.SPPF [576, 576, 5]
10 -1 1 1253088 ultralytics.nn.modules.block.PSA [576, 576]
11 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
12 [-1, 6] 1 0 ultralytics.nn.modules.conv.Concat [1]
13 -1 2 1993728 ultralytics.nn.modules.block.C2f [960, 384, 2]
14 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
15 [-1, 4] 1 0 ultralytics.nn.modules.conv.Concat [1]
16 -1 2 517632 ultralytics.nn.modules.block.C2f [576, 192, 2]
17 -1 1 332160 ultralytics.nn.modules.conv.Conv [192, 192, 3, 2]
18 [-1, 13] 1 0 ultralytics.nn.modules.conv.Concat [1]
19 -1 2 1846272 ultralytics.nn.modules.block.C2f [576, 384, 2]
20 -1 1 152448 ultralytics.nn.modules.block.SCDown [384, 384, 3, 2]
21 [-1, 10] 1 0 ultralytics.nn.modules.conv.Concat [1]
22 -1 2 1969920 ultralytics.nn.modules.block.C2fCIB [960, 576, 2, True, True]
23 [16, 19, 22] 1 2282134 ultralytics.nn.modules.head.v10Detect [1, [192, 384, 576]]
YOLOv10m-C2fCIB_UIB summary: 532 layers, 19,598,182 parameters, 19,598,166 gradients, 68.9 GFLOPs