第P9周：YOLOv5-Backbone模块实现

最新推荐文章于 2024-06-17 17:14:52 发布
最新推荐文章于 2024-06-17 17:14:52 发布
阅读量902 收藏 14
点赞数 16
文章标签： YOLO
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/ck1487921366/article/details/135702601
版权

目录

代码及运行结果：

一、 前期准备

二、搭建包含Backbone模块的模型

三、 训练模型

四、 结果可视化

YOLOv5-Backbone模块分析与总结：

Conv模块

C3模块

SPP模块

代码及运行结果：

一、 前期准备

import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os,PIL,pathlib,warnings

warnings.filterwarnings("ignore")             #忽略警告信息

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(device)

import os,PIL,random,pathlib

data_dir = './9-data/'
data_dir = pathlib.Path(data_dir)

data_paths  = list(data_dir.glob('*'))
classeNames = [str(path).split("/")[1] for path in data_paths]
print(classeNames)

# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
train_transforms = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    # transforms.RandomHorizontalFlip(), # 随机水平翻转
    transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
    transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])

test_transform = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
    transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])

total_data = datasets.ImageFolder("./9-data/",transform=train_transforms)
print(total_data)

print(total_data.class_to_idx)

train_size = int(0.8 * len(total_data))
test_size  = len(total_data) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(total_data, [train_size, test_size])
print(train_dataset, test_dataset)

batch_size = 4

train_dl = torch.utils.data.DataLoader(train_dataset,
                                           batch_size=batch_size,
                                           shuffle=True,
                                           num_workers=1)
test_dl = torch.utils.data.DataLoader(test_dataset,
                                          batch_size=batch_size,
                                          shuffle=True,
                                          num_workers=1)

for X, y in test_dl:
    print("Shape of X [N, C, H, W]: ", X.shape)
    print("Shape of y: ", y.shape, y.dtype)
    break
cuda
['cloudy', 'rain', 'shine', 'sunrise']
Dataset ImageFolder
    Number of datapoints: 1125
    Root location: ./9-data/
    StandardTransform
Transform: Compose(
               Resize(size=[224, 224], interpolation=bilinear)
               ToTensor()
               Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
           )
{'cloudy': 0, 'rain': 1, 'shine': 2, 'sunrise': 3}
<torch.utils.data.dataset.Subset object at 0x7fb9a46d3160> <torch.utils.data.dataset.Subset object at 0x7fb86c86a550>
Shape of X [N, C, H, W]:  torch.Size([4, 3, 224, 224])
Shape of y:  torch.Size([4]) torch.int64

二、搭建包含Backbone模块的模型

import torch.nn.functional as F

def autopad(k, p=None):  # kernel, padding
    # Pad to 'same'
    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
    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):  # ch_in, ch_out, kernel, stride, padding, groups
        super().__init__()
        self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
        self.bn = nn.BatchNorm2d(c2)
        self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())

    def forward(self, x):
        return self.act(self.bn(self.conv(x)))

class Bottleneck(nn.Module):
    # Standard bottleneck
    def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, shortcut, groups, expansion
        super().__init__()
        c_ = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c_, c2, 3, 1, g=g)
        self.add = shortcut and c1 == c2

    def forward(self, x):
        return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))

class C3(nn.Module):
    # CSP Bottleneck with 3 convolutions
    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, number, shortcut, groups, expansion
        super().__init__()
        c_ = int(c2 * e)  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c1, c_, 1, 1)
        self.cv3 = Conv(2 * c_, c2, 1)  # act=FReLU(c2)
        self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))

    def forward(self, x):
        return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))
    
class SPPF(nn.Module):
    # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
    def __init__(self, c1, c2, k=5):  # equivalent to SPP(k=(5, 9, 13))
        super().__init__()
        c_ = c1 // 2  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c_ * 4, c2, 1, 1)
        self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2)

    def forward(self, x):
        x = self.cv1(x)
        with warnings.catch_warnings():
            warnings.simplefilter('ignore')  # suppress torch 1.9.0 max_pool2d() warning
            y1 = self.m(x)
            y2 = self.m(y1)
            return self.cv2(torch.cat([x, y1, y2, self.m(y2)], 1))

class YOLOv5_backbone(nn.Module):
    def __init__(self):
        super(YOLOv5_backbone, self).__init__()
        
        self.Conv_1 = Conv(3, 64, 3, 2, 2) 
        self.Conv_2 = Conv(64, 128, 3, 2) 
        self.C3_3   = C3(128,128)
        self.Conv_4 = Conv(128, 256, 3, 2) 
        self.C3_5   = C3(256,256)
        self.Conv_6 = Conv(256, 512, 3, 2) 
        self.C3_7   = C3(512,512)
        self.Conv_8 = Conv(512, 1024, 3, 2) 
        self.C3_9   = C3(1024, 1024)
        self.SPPF   = SPPF(1024, 1024, 5)
        
        # 全连接网络层，用于分类
        self.classifier = nn.Sequential(
            nn.Linear(in_features=65536, out_features=100),
            nn.ReLU(),
            nn.Linear(in_features=100, out_features=4)
        )
        
    def forward(self, x):
        x = self.Conv_1(x)
        x = self.Conv_2(x)
        x = self.C3_3(x)
        x = self.Conv_4(x)
        x = self.C3_5(x)
        x = self.Conv_6(x)
        x = self.C3_7(x)
        x = self.Conv_8(x)
        x = self.C3_9(x)
        x = self.SPPF(x)
        
        x = torch.flatten(x, start_dim=1)
        x = self.classifier(x)

        return x

device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
    
model = YOLOv5_backbone().to(device)
print(model)

# 统计模型参数量以及其他指标
import torchsummary as summary
summary.summary(model, (3, 224, 224))

Using cuda device
YOLOv5_backbone(
  (Conv_1): Conv(
    (conv): Conv2d(3, 64, kernel_size=(3, 3), stride=(2, 2), padding=(2, 2), bias=False)
    (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (Conv_2): Conv(
    (conv): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_3): C3(
    (cv1): Conv(
      (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(128, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_4): Conv(
    (conv): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_5): C3(
    (cv1): Conv(
      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(128, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_6): Conv(
    (conv): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_7): C3(
    (cv1): Conv(
      (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(256, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (Conv_8): Conv(
    (conv): Conv2d(512, 1024, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
    (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (act): SiLU()
  )
  (C3_9): C3(
    (cv1): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv3): Conv(
      (conv): Conv2d(1024, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): Sequential(
      (0): Bottleneck(
        (cv1): Conv(
          (conv): Conv2d(512, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
        (cv2): Conv(
          (conv): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (act): SiLU()
        )
      )
    )
  )
  (SPPF): SPPF(
    (cv1): Conv(
      (conv): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (cv2): Conv(
      (conv): Conv2d(2048, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (act): SiLU()
    )
    (m): MaxPool2d(kernel_size=5, stride=1, padding=2, dilation=1, ceil_mode=False)
  )
  (classifier): Sequential(
    (0): Linear(in_features=65536, out_features=100, bias=True)
    (1): ReLU()
    (2): Linear(in_features=100, out_features=4, bias=True)
  )
)
----------------------------------------------------------------
        Layer (type)               Output Shape         Param #
================================================================
            Conv2d-1         [-1, 64, 113, 113]           1,728
       BatchNorm2d-2         [-1, 64, 113, 113]             128
              SiLU-3         [-1, 64, 113, 113]               0
              Conv-4         [-1, 64, 113, 113]               0
            Conv2d-5          [-1, 128, 57, 57]          73,728
       BatchNorm2d-6          [-1, 128, 57, 57]             256
              SiLU-7          [-1, 128, 57, 57]               0
              Conv-8          [-1, 128, 57, 57]               0
            Conv2d-9           [-1, 64, 57, 57]           8,192
      BatchNorm2d-10           [-1, 64, 57, 57]             128
             SiLU-11           [-1, 64, 57, 57]               0
             Conv-12           [-1, 64, 57, 57]               0
           Conv2d-13           [-1, 64, 57, 57]           4,096
      BatchNorm2d-14           [-1, 64, 57, 57]             128
             SiLU-15           [-1, 64, 57, 57]               0
             Conv-16           [-1, 64, 57, 57]               0
           Conv2d-17           [-1, 64, 57, 57]          36,864
      BatchNorm2d-18           [-1, 64, 57, 57]             128
             SiLU-19           [-1, 64, 57, 57]               0
             Conv-20           [-1, 64, 57, 57]               0
       Bottleneck-21           [-1, 64, 57, 57]               0
           Conv2d-22           [-1, 64, 57, 57]           8,192
      BatchNorm2d-23           [-1, 64, 57, 57]             128
             SiLU-24           [-1, 64, 57, 57]               0
             Conv-25           [-1, 64, 57, 57]               0
           Conv2d-26          [-1, 128, 57, 57]          16,384
      BatchNorm2d-27          [-1, 128, 57, 57]             256
             SiLU-28          [-1, 128, 57, 57]               0
             Conv-29          [-1, 128, 57, 57]               0
               C3-30          [-1, 128, 57, 57]               0
           Conv2d-31          [-1, 256, 29, 29]         294,912
      BatchNorm2d-32          [-1, 256, 29, 29]             512
             SiLU-33          [-1, 256, 29, 29]               0
             Conv-34          [-1, 256, 29, 29]               0
           Conv2d-35          [-1, 128, 29, 29]          32,768
      BatchNorm2d-36          [-1, 128, 29, 29]             256
             SiLU-37          [-1, 128, 29, 29]               0
             Conv-38          [-1, 128, 29, 29]               0
           Conv2d-39          [-1, 128, 29, 29]          16,384
      BatchNorm2d-40          [-1, 128, 29, 29]             256
             SiLU-41          [-1, 128, 29, 29]               0
             Conv-42          [-1, 128, 29, 29]               0
           Conv2d-43          [-1, 128, 29, 29]         147,456
      BatchNorm2d-44          [-1, 128, 29, 29]             256
             SiLU-45          [-1, 128, 29, 29]               0
             Conv-46          [-1, 128, 29, 29]               0
       Bottleneck-47          [-1, 128, 29, 29]               0
           Conv2d-48          [-1, 128, 29, 29]          32,768
      BatchNorm2d-49          [-1, 128, 29, 29]             256
             SiLU-50          [-1, 128, 29, 29]               0
             Conv-51          [-1, 128, 29, 29]               0
           Conv2d-52          [-1, 256, 29, 29]          65,536
      BatchNorm2d-53          [-1, 256, 29, 29]             512
             SiLU-54          [-1, 256, 29, 29]               0
             Conv-55          [-1, 256, 29, 29]               0
               C3-56          [-1, 256, 29, 29]               0
           Conv2d-57          [-1, 512, 15, 15]       1,179,648
      BatchNorm2d-58          [-1, 512, 15, 15]           1,024
             SiLU-59          [-1, 512, 15, 15]               0
             Conv-60          [-1, 512, 15, 15]               0
           Conv2d-61          [-1, 256, 15, 15]         131,072
      BatchNorm2d-62          [-1, 256, 15, 15]             512
             SiLU-63          [-1, 256, 15, 15]               0
             Conv-64          [-1, 256, 15, 15]               0
           Conv2d-65          [-1, 256, 15, 15]          65,536
      BatchNorm2d-66          [-1, 256, 15, 15]             512
             SiLU-67          [-1, 256, 15, 15]               0
             Conv-68          [-1, 256, 15, 15]               0
           Conv2d-69          [-1, 256, 15, 15]         589,824
      BatchNorm2d-70          [-1, 256, 15, 15]             512
             SiLU-71          [-1, 256, 15, 15]               0
             Conv-72          [-1, 256, 15, 15]               0
       Bottleneck-73          [-1, 256, 15, 15]               0
           Conv2d-74          [-1, 256, 15, 15]         131,072
      BatchNorm2d-75          [-1, 256, 15, 15]             512
             SiLU-76          [-1, 256, 15, 15]               0
             Conv-77          [-1, 256, 15, 15]               0
           Conv2d-78          [-1, 512, 15, 15]         262,144
      BatchNorm2d-79          [-1, 512, 15, 15]           1,024
             SiLU-80          [-1, 512, 15, 15]               0
             Conv-81          [-1, 512, 15, 15]               0
               C3-82          [-1, 512, 15, 15]               0
           Conv2d-83           [-1, 1024, 8, 8]       4,718,592
      BatchNorm2d-84           [-1, 1024, 8, 8]           2,048
             SiLU-85           [-1, 1024, 8, 8]               0
             Conv-86           [-1, 1024, 8, 8]               0
           Conv2d-87            [-1, 512, 8, 8]         524,288
      BatchNorm2d-88            [-1, 512, 8, 8]           1,024
             SiLU-89            [-1, 512, 8, 8]               0
             Conv-90            [-1, 512, 8, 8]               0
           Conv2d-91            [-1, 512, 8, 8]         262,144
      BatchNorm2d-92            [-1, 512, 8, 8]           1,024
             SiLU-93            [-1, 512, 8, 8]               0
             Conv-94            [-1, 512, 8, 8]               0
           Conv2d-95            [-1, 512, 8, 8]       2,359,296
      BatchNorm2d-96            [-1, 512, 8, 8]           1,024
             SiLU-97            [-1, 512, 8, 8]               0
             Conv-98            [-1, 512, 8, 8]               0
       Bottleneck-99            [-1, 512, 8, 8]               0
          Conv2d-100            [-1, 512, 8, 8]         524,288
     BatchNorm2d-101            [-1, 512, 8, 8]           1,024
            SiLU-102            [-1, 512, 8, 8]               0
            Conv-103            [-1, 512, 8, 8]               0
          Conv2d-104           [-1, 1024, 8, 8]       1,048,576
     BatchNorm2d-105           [-1, 1024, 8, 8]           2,048
            SiLU-106           [-1, 1024, 8, 8]               0
            Conv-107           [-1, 1024, 8, 8]               0
              C3-108           [-1, 1024, 8, 8]               0
          Conv2d-109            [-1, 512, 8, 8]         524,288
     BatchNorm2d-110            [-1, 512, 8, 8]           1,024
            SiLU-111            [-1, 512, 8, 8]               0
            Conv-112            [-1, 512, 8, 8]               0
       MaxPool2d-113            [-1, 512, 8, 8]               0
       MaxPool2d-114            [-1, 512, 8, 8]               0
       MaxPool2d-115            [-1, 512, 8, 8]               0
          Conv2d-116           [-1, 1024, 8, 8]       2,097,152
     BatchNorm2d-117           [-1, 1024, 8, 8]           2,048
            SiLU-118           [-1, 1024, 8, 8]               0
            Conv-119           [-1, 1024, 8, 8]               0
            SPPF-120           [-1, 1024, 8, 8]               0
          Linear-121                  [-1, 100]       6,553,700
            ReLU-122                  [-1, 100]               0
          Linear-123                    [-1, 4]             404
================================================================
Total params: 21,729,592
Trainable params: 21,729,592
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 137.59
Params size (MB): 82.89
Estimated Total Size (MB): 221.06
----------------------------------------------------------------

三、 训练模型

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)  # 训练集的大小
    num_batches = len(dataloader)   # 批次数目, (size/batch_size，向上取整)

    train_loss, train_acc = 0, 0  # 初始化训练损失和正确率
    
    for X, y in dataloader:  # 获取图片及其标签
        X, y = X.to(device), y.to(device)
        
        # 计算预测误差
        pred = model(X)          # 网络输出
        loss = loss_fn(pred, y)  # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
        
        # 反向传播
        optimizer.zero_grad()  # grad属性归零
        loss.backward()        # 反向传播
        optimizer.step()       # 每一步自动更新
        
        # 记录acc与loss
        train_acc  += (pred.argmax(1) == y).type(torch.float).sum().item()
        train_loss += loss.item()
            
    train_acc  /= size
    train_loss /= num_batches

    return train_acc, train_loss

def test (dataloader, model, loss_fn):
    size        = len(dataloader.dataset)  # 测试集的大小
    num_batches = len(dataloader)          # 批次数目, (size/batch_size，向上取整)
    test_loss, test_acc = 0, 0
    
    # 当不进行训练时，停止梯度更新，节省计算内存消耗
    with torch.no_grad():
        for imgs, target in dataloader:
            imgs, target = imgs.to(device), target.to(device)
            
            # 计算loss
            target_pred = model(imgs)
            loss        = loss_fn(target_pred, target)
            
            test_loss += loss.item()
            test_acc  += (target_pred.argmax(1) == target).type(torch.float).sum().item()

    test_acc  /= size
    test_loss /= num_batches

    return test_acc, test_loss

import copy

optimizer  = torch.optim.Adam(model.parameters(), lr= 1e-4)
loss_fn    = nn.CrossEntropyLoss() # 创建损失函数

epochs     = 60

train_loss = []
train_acc  = []
test_loss  = []
test_acc   = []

best_acc = 0    # 设置一个最佳准确率，作为最佳模型的判别指标

for epoch in range(epochs):
    
    model.train()
    epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
    
    model.eval()
    epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
    
    # 保存最佳模型到 best_model
    if epoch_test_acc > best_acc:
        best_acc   = epoch_test_acc
        best_model = copy.deepcopy(model)
    
    train_acc.append(epoch_train_acc)
    train_loss.append(epoch_train_loss)
    test_acc.append(epoch_test_acc)
    test_loss.append(epoch_test_loss)
    
    # 获取当前的学习率
    lr = optimizer.state_dict()['param_groups'][0]['lr']
    
    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
    print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, 
                          epoch_test_acc*100, epoch_test_loss, lr))
    
# 保存最佳模型到文件中
PATH = './best_model.pth'  # 保存的参数文件名
torch.save(best_model.state_dict(), PATH)

print('Done')

Epoch: 1, Train_acc:52.7%, Train_loss:1.244, Test_acc:50.7%, Test_loss:1.547, Lr:1.00E-04
Epoch: 2, Train_acc:60.3%, Train_loss:0.995, Test_acc:72.4%, Test_loss:0.809, Lr:1.00E-04
Epoch: 3, Train_acc:70.0%, Train_loss:0.827, Test_acc:69.3%, Test_loss:0.769, Lr:1.00E-04
Epoch: 4, Train_acc:77.3%, Train_loss:0.626, Test_acc:85.3%, Test_loss:0.357, Lr:1.00E-04
Epoch: 5, Train_acc:77.7%, Train_loss:0.602, Test_acc:78.2%, Test_loss:0.554, Lr:1.00E-04
Epoch: 6, Train_acc:80.2%, Train_loss:0.512, Test_acc:73.3%, Test_loss:0.596, Lr:1.00E-04
Epoch: 7, Train_acc:81.8%, Train_loss:0.477, Test_acc:87.1%, Test_loss:0.331, Lr:1.00E-04
Epoch: 8, Train_acc:84.2%, Train_loss:0.424, Test_acc:84.4%, Test_loss:0.347, Lr:1.00E-04
Epoch: 9, Train_acc:86.6%, Train_loss:0.351, Test_acc:84.9%, Test_loss:0.352, Lr:1.00E-04
Epoch:10, Train_acc:85.6%, Train_loss:0.370, Test_acc:87.6%, Test_loss:0.410, Lr:1.00E-04
Epoch:11, Train_acc:90.1%, Train_loss:0.252, Test_acc:89.8%, Test_loss:0.304, Lr:1.00E-04
Epoch:12, Train_acc:88.8%, Train_loss:0.315, Test_acc:88.4%, Test_loss:0.292, Lr:1.00E-04
Epoch:13, Train_acc:92.1%, Train_loss:0.226, Test_acc:85.8%, Test_loss:0.349, Lr:1.00E-04
Epoch:14, Train_acc:91.3%, Train_loss:0.213, Test_acc:90.7%, Test_loss:0.278, Lr:1.00E-04
Epoch:15, Train_acc:94.6%, Train_loss:0.180, Test_acc:88.9%, Test_loss:0.260, Lr:1.00E-04
Epoch:16, Train_acc:90.8%, Train_loss:0.279, Test_acc:84.9%, Test_loss:0.470, Lr:1.00E-04
Epoch:17, Train_acc:95.1%, Train_loss:0.137, Test_acc:89.8%, Test_loss:0.322, Lr:1.00E-04
Epoch:18, Train_acc:94.4%, Train_loss:0.179, Test_acc:88.4%, Test_loss:0.350, Lr:1.00E-04
Epoch:19, Train_acc:96.3%, Train_loss:0.109, Test_acc:92.0%, Test_loss:0.317, Lr:1.00E-04
Epoch:20, Train_acc:95.6%, Train_loss:0.145, Test_acc:88.0%, Test_loss:0.352, Lr:1.00E-04
Epoch:21, Train_acc:94.6%, Train_loss:0.159, Test_acc:91.1%, Test_loss:0.280, Lr:1.00E-04
Epoch:22, Train_acc:96.7%, Train_loss:0.082, Test_acc:84.4%, Test_loss:0.478, Lr:1.00E-04
Epoch:23, Train_acc:95.8%, Train_loss:0.127, Test_acc:90.7%, Test_loss:0.267, Lr:1.00E-04
Epoch:24, Train_acc:96.2%, Train_loss:0.104, Test_acc:91.6%, Test_loss:0.259, Lr:1.00E-04
Epoch:25, Train_acc:98.0%, Train_loss:0.080, Test_acc:89.8%, Test_loss:0.262, Lr:1.00E-04
Epoch:26, Train_acc:96.2%, Train_loss:0.104, Test_acc:91.1%, Test_loss:0.308, Lr:1.00E-04
Epoch:27, Train_acc:96.8%, Train_loss:0.107, Test_acc:92.4%, Test_loss:0.257, Lr:1.00E-04
Epoch:28, Train_acc:96.2%, Train_loss:0.088, Test_acc:88.9%, Test_loss:0.311, Lr:1.00E-04
Epoch:29, Train_acc:96.9%, Train_loss:0.099, Test_acc:91.1%, Test_loss:0.242, Lr:1.00E-04
Epoch:30, Train_acc:98.8%, Train_loss:0.046, Test_acc:91.6%, Test_loss:0.273, Lr:1.00E-04
Epoch:31, Train_acc:99.0%, Train_loss:0.028, Test_acc:91.1%, Test_loss:0.293, Lr:1.00E-04
Epoch:32, Train_acc:98.0%, Train_loss:0.053, Test_acc:89.8%, Test_loss:0.360, Lr:1.00E-04
Epoch:33, Train_acc:99.0%, Train_loss:0.040, Test_acc:85.3%, Test_loss:0.659, Lr:1.00E-04
Epoch:34, Train_acc:96.8%, Train_loss:0.097, Test_acc:89.8%, Test_loss:0.353, Lr:1.00E-04
Epoch:35, Train_acc:98.2%, Train_loss:0.071, Test_acc:91.1%, Test_loss:0.371, Lr:1.00E-04
Epoch:36, Train_acc:97.4%, Train_loss:0.086, Test_acc:87.6%, Test_loss:0.412, Lr:1.00E-04
Epoch:37, Train_acc:97.8%, Train_loss:0.061, Test_acc:90.2%, Test_loss:0.447, Lr:1.00E-04
Epoch:38, Train_acc:97.9%, Train_loss:0.048, Test_acc:89.8%, Test_loss:0.400, Lr:1.00E-04
Epoch:39, Train_acc:98.4%, Train_loss:0.057, Test_acc:88.4%, Test_loss:0.402, Lr:1.00E-04
Epoch:40, Train_acc:99.0%, Train_loss:0.025, Test_acc:88.9%, Test_loss:0.539, Lr:1.00E-04
Epoch:41, Train_acc:98.0%, Train_loss:0.068, Test_acc:91.1%, Test_loss:0.362, Lr:1.00E-04
Epoch:42, Train_acc:97.2%, Train_loss:0.066, Test_acc:90.2%, Test_loss:0.350, Lr:1.00E-04
Epoch:43, Train_acc:98.8%, Train_loss:0.037, Test_acc:93.3%, Test_loss:0.379, Lr:1.00E-04
Epoch:44, Train_acc:99.0%, Train_loss:0.034, Test_acc:90.2%, Test_loss:0.425, Lr:1.00E-04
Epoch:45, Train_acc:98.3%, Train_loss:0.045, Test_acc:91.1%, Test_loss:0.374, Lr:1.00E-04
Epoch:46, Train_acc:97.6%, Train_loss:0.108, Test_acc:90.7%, Test_loss:0.488, Lr:1.00E-04
Epoch:47, Train_acc:99.4%, Train_loss:0.016, Test_acc:90.7%, Test_loss:0.380, Lr:1.00E-04
Epoch:48, Train_acc:98.9%, Train_loss:0.032, Test_acc:87.1%, Test_loss:0.620, Lr:1.00E-04
Epoch:49, Train_acc:98.7%, Train_loss:0.037, Test_acc:90.7%, Test_loss:0.430, Lr:1.00E-04
Epoch:50, Train_acc:99.2%, Train_loss:0.018, Test_acc:90.2%, Test_loss:0.398, Lr:1.00E-04
Epoch:51, Train_acc:96.6%, Train_loss:0.090, Test_acc:88.9%, Test_loss:0.474, Lr:1.00E-04
Epoch:52, Train_acc:98.6%, Train_loss:0.042, Test_acc:88.4%, Test_loss:0.457, Lr:1.00E-04
Epoch:53, Train_acc:99.0%, Train_loss:0.034, Test_acc:88.9%, Test_loss:0.476, Lr:1.00E-04
Epoch:54, Train_acc:99.0%, Train_loss:0.034, Test_acc:89.3%, Test_loss:0.337, Lr:1.00E-04
Epoch:55, Train_acc:99.6%, Train_loss:0.011, Test_acc:94.2%, Test_loss:0.530, Lr:1.00E-04
Epoch:56, Train_acc:99.3%, Train_loss:0.010, Test_acc:89.8%, Test_loss:0.502, Lr:1.00E-04
Epoch:57, Train_acc:98.7%, Train_loss:0.036, Test_acc:88.4%, Test_loss:0.441, Lr:1.00E-04
Epoch:58, Train_acc:99.7%, Train_loss:0.009, Test_acc:89.8%, Test_loss:0.419, Lr:1.00E-04
Epoch:59, Train_acc:99.8%, Train_loss:0.008, Test_acc:92.0%, Test_loss:0.512, Lr:1.00E-04
Epoch:60, Train_acc:98.1%, Train_loss:0.064, Test_acc:90.2%, Test_loss:0.316, Lr:1.00E-04
Done

四、 结果可视化

import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore")               #忽略警告信息
plt.rcParams['font.sans-serif']    = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False      # 用来正常显示负号
plt.rcParams['figure.dpi']         = 100        #分辨率

epochs_range = range(epochs)

plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)

plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

# 将参数加载到model当中
best_model.load_state_dict(torch.load(PATH, map_location=device))
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)

print(epoch_test_acc, epoch_test_loss)

# 查看是否与我们记录的最高准确率一致
print(epoch_test_acc)

0.9422222222222222 0.35362279524906826
0.9422222222222222

YOLOv5-Backbone模块分析与总结：

YOLOv5s 模型主要由 Backbone、Neck 和Head 三部分组成：

  • Backbone 主要负责对输入图像进行特征提取。
  • Neck 负责对特征图进行多尺度特征融合，并把这些特征传递给预测层。
  • Head 进行最终的回归预测。

下面是Backbone骨干网络的分析：
        骨干网络是指用来提取图像特征的网络，它的主要作用是将原始的输入图像转化为多层特征图，以便后续的目标检测任务使用。

        在Yolov5中，使用的是CSPDarknet53或ResNet骨干网络，这两个网络都是相对轻量级的，能够在保证较高检测精度的同时，尽可能地减少计算量和内存占用。
        Yolov5的Backbone中的主要结构有Conv模块、C3模块、SPPF模块。

Conv模块

        Conv模块是卷积神经网络中常用的一种基础模块，它主要由卷积层、BN层和激活函数组成。下面对这些组成部分进行详细解析。

  • 卷积层是卷积神经网络中最基础的层之一，用于提取输入特征中的局部空间信息。卷积操作可以看作是一个滑动窗口，窗口在输入特征上滑动，并将窗口内的特征值与卷积核进行卷积操作，从而得到输出特征。卷积层通常由多个卷积核组成，每个卷积核对应一个输出通道。卷积核的大小、步长、填充方式等超参数决定了卷积层的输出大小和感受野大小。卷积神经网络中，卷积层通常被用来构建特征提取器。
  • BN层是在卷积层之后加入的一种归一化层，用于规范化神经网络中的特征值分布。它可以加速训练过程，提高模型泛化能力，减轻模型对初始化的依赖性。BN层的输入为一个batch的特征图，它将每个通道上的特征进行均值和方差的计算，并对每个通道上的特征进行标准化处理。标准化后的特征再通过一个可学习的仿射变换（拉伸和偏移）进行还原，从而得到BN层的输出。
  • 激活函数是一种非线性函数，用于给神经网络引入非线性变换能力。常用的激活函数包括sigmoid、ReLU、LeakyReLU、ELU等。它们在输入值的不同范围内都有不同的输出表现，可以更好地适应不同类型的数据分布。

综上所述，Conv模块是卷积神经网络中常用的基础模块，它通过卷积操作提取局部空间信息，并通过BN层规范化特征值分布，最后通过激活函数引入非线性变换能力，从而实现对输入特征的转换和提取。

C3模块

        C3模块是YOLOv5网络中的一个重要组成部分，其主要作用是增加网络的深度和感受野，提高特征提取的能力。

        C3模块模块的具体内容以及在上一篇文章中有了详细说明，本篇就不再赘述。

SPP模块

        SPP模块是一种池化模块，通常应用于卷积神经网络中，旨在实现输入数据的空间不变性和位置不变性，以便于提高神经网络的识别能力。其主要思想是将不同大小的感受野应用于同一张图像，从而能够捕捉到不同尺度的特征信息。在SPP模块中，首先对输入特征图进行不同大小的池化操作，以得到一组不同大小的特征图。然后将这些特征图连接在一起，并通过全连接层进行降维，最终得到固定大小的特征向量。

SPP模块通常由三个步骤组成：

  • 池化：将输入特征图分别进行不同大小的池化操作，以获得一组不同大小的特征图。
  • 连接：将不同大小的特征图连接在一起。
  • 全连接：通过全连接层将连接后的特征向量降维，得到固定大小的特征向量。
轻风不语
关注
  • 16
    点赞
  • 踩
  • 14
    收藏
    觉得还不错? 一键收藏
  • 1
    评论
【YOLOv5】Backbone、Neck、Head各模块详解
qq_44878985的博客
03-01 6万+
详解Yolov5原理及backbone、neck、head三个模块的作用及结构。
YOLOv5-Backbone模块实现
weixin_73067446的博客
01-26 791
Backbone主要负责对输入图像进行特征提取。Neck负责对特征图进行多尺度特征融合，并把这些特征传递给预测层。Head进行最终的回归预测。
YOLOv5 backbone（一）
浅汐入梦的博客
09-18 5426
yolov5中Backbone部分
YOLOv5的Backbone详解
日拱一卒，功不唐捐
03-20 4万+
本文从yolov5s入手，细致分析了yolov5s的网络配置文件参数、backbone部分、neck部分和最终的head部分，并结合网络结构图和源码对各模块进行了详细介绍
yolov5深度剖析(2)—backbone
Ring__Rain的博客
08-11 4186
转自：https://blog.csdn.net/weixin_36714575/article/details/114211796?spm=1001.2014.3001.5501 前言 上一篇中我们简单回顾了yolov5的架构和搭建了开发环境。需要回顾的小伙伴可以点击下面传送门，基于第一篇的理解我们再进行源码解读就会有事半功倍的效果。 yolov5架构剖析和环境搭建传送门 这篇中我们开始解析yolov5的源码，类似yolov3我们同样从模型的构建开始。这部分核心的代码在yolo.py文件中。我
YOLOV5之backbone介绍与迁移使用
qq_45752541的博客
09-17 6872
yolov5的特征提取网络：将PAN与PFN深度融合，对不同尺度鲁棒性强，可以即插即用，后接不同的检测器。
yolov5的backbone学习
weixin_41399424的博客
11-15 7094
参考了: YOLOv5深度刨析之配置与初始化超参数 - 知乎 YOLOv5代码详解（common.py部分） - 灰信网（软件开发博客聚合） 理解bottleneck architecture_GL3_24的博客-CSDN博客_bottleneck层
YOLOV5解析
热门推荐
丝萝愿为乔木的博客
10-13 5万+
网络由三个主要组件组成： 1）Backbone：在不同图像细粒度上聚合并形成图像特征的卷积神经网络。 2）Neck：一系列混合和组合图像特征的网络层，并将图像特征传递到预测层。 3）Head：对图像特征进行预测，生成边界框和并预测类别。 对于YOLOV5，无论是V5s，V5m，V5l还是V5x其Backbone，Neck和Head一致。唯一的区别在与模型的深度和宽度设置。 总结构框架： 下面逐一解析： 1)Backbone 先代码，有个大概脉络： # YOLOv5 backbone backbone:
第P9周：YOLOv5-Backbone模块实现天气识别
qq_52386407的博客
05-31 624
输出：Shape of X [N, C, H, W]: torch.Size([4, 3, 224, 224])输出：{'cloudy': 0, 'rain': 1, 'shine': 2, 'sunrise': 3}输出：['cloudy', 'rain', 'shine', 'sunrise']学习环境：torch1.9.0+cu111。语言环境：Python3.9。
mobile-yolov5-pruning-distillation:mobilev2-yolov5s剪枝，蒸馏，支持ncnn，tensorRT部署。超轻但性能更好！
03-19
移动式yolov5修剪蒸馏mobilev2-yolov5s的通道修剪和蒸馏。超轻但性能更好！ TensorRT版本===> Android版本===> 背景yolov5s在640x640分辨率下的计算量和参数量分别为8.39G和7.07M。在速度上仍然有提升空间，通过替换...
yolov4-tiny.weights
06-29
yolov4 Tiny版本的预训练权重，提取了yolo v4的backbone层，可用作训练yolo-V4.
agile2go:Rails 4-Backbone.js-SemanticUI
05-14
$ foreman start -f Profile.development发展测试：RSpec，茉莉前端：Semantic-ui，Backbone.js，Handlebars 数据库：Mongodb，Firebase 服务器：彪马测验捆绑执行耙捆绑执行耙茉莉花贡献请参阅。 另请参见。维护者
YOLOv5：修改backbone为ConvNeXt
09-14
1、官方源项目地址 ：https://github.com/ultralytics/yolov5 2、在yolov5-6.1中修改backbone为ConvNeXt。 3、训练、测试、预测命令与官方版本一致。
快速编译安装tensorrt_yolo
FL1623863129的博客
06-15 379
通过 PyPI 安装模块，您只需执行以下命令即可：如果您希望获取最新的开发版本或者为项目做出贡献，可以按照以下步骤从 GitHub 克隆代码库并安装：git clone https://github.com/laugh12321/TensorRT-YOLO # 克隆代码库在以上步骤中，您可以先克隆代码库并进行本地构建，然后再使用pip安装生成的 Wheel 包，确保安装的是最新版本并具有最新的功能和改进。Deploy。
（1）图像识别yolov5—安装教程
qq_26043945的博客
06-12 66
首先，你需要在你的计算机上下载的文件包，。下载后对压缩文件进行解压。通常使用识别物体，需要安装必要的，如PyTorch、torchvision、OpenCV等。打开YOLOv5文件夹，在里面打开 “” 文件，里面就是必要的依赖项。然后打开命令行，进入当前文件目录（我这里保存到盘的中）。然后通过下面的指令，系统自动安装这些依赖库。
从0到1实现YOLOv3
最新发布
周陽讀書
06-17 46
介绍如何从配置文件创建网络架构、加载权重和设计输入/输出管道近年来，人们开发了许多用于目标检测的算法，其中一些包括YOLO、SSD、Mask RCNN和RetinaNet。致力于改进目标检测，学习目标检测的最好方法是自己从头开始实现算法。使用PyTorch实现一个基于YOLOv3的目标检测算法。代码置于这个Github存储库中：主要有5个部分：0.卷积神经网络如何工作，边界盒回归，IoU和非最大抑制1.理解YOLO工作原理2.创建网络架构的层3.实现网络的前向传递。
yolov10主要特点
yzf060109的专栏
06-14 694
以下是YOLOv10与其他模型的快速对比：YOLOv10-S：46.3 AP，2.49毫秒延迟YOLOv10-M：51.1 AP，4.74毫秒延迟YOLOv10-L：53.2 AP，7.28毫秒延迟YOLOv10-X：54.4 AP，10.70毫秒延迟这些改进使YOLOv10成为实时目标检测任务的首选，从自动驾驶到智能监控。与YOLOv9-C相比，YOLOv10-B（一个平衡的变体）延迟减少了46%，参数量减少了25%，同时保持了相同的性能水平。一对多分配：在训练过程中使用，以提供丰富的监督信号。
yolov5backbone的作用
03-27
yolov5backbone是YOLOv5算法中的核心组成部分，作为网络的骨干，主要负责提取图像特征。其作用是将输入的图像通过卷积神经网络进行多层次的特征提取，从低层次到高层次逐渐提取出图像的不同特征，如边缘、纹理、形状等，最终生成一个高维度的特征向量，用于目标检测和分类等任务。yolov5backbone采用了一种轻量化的网络结构，具有更快的运行速度和更高的精度，是目前主流的目标检测算法之一。

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值