目录
1、YOLOv5的模型图。
2、BackBone简单介绍。
3、YOLOv5的Backbone文件。
4、YOLOv5Backbone的code部分
5、完整的code部分
6、结果展示
（1）Adam优化器
（2）SGD优化器
- 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
- 🍖 原作者：K同学啊 | 接辅导、项目定制
这周的这篇博客，主要是使用YOLOv5里面的Backbone模块，搭建网络。
我们先来看下YOLOv5的Backbone部分。
1、YOLOv5的模型图。
（图片来源于网络）
上图中，最左侧的事YOLOv5里面的BackBone，上篇博文，主要讲的是YOLOv5里面的BackBone里面的C3模块。
这篇文章，我们主要看YOLOv5里面的BackBone部分，即：YOLOv5的骨干网络。
2、BackBone简单介绍。
YOLOv5 模型主要由 Backbone、Neck 和Head 三部分组成，网络模型（如上图所示）。
①Backbone 主要负责对输入图像进行特征提取。
②Neck 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。
③Head 进行最终的回归预测。
YOLOv5的骨干网络（Backbone）是其整体架构中至关重要的部分，主要负责从输入图像中提取丰富的特征信息。以下是关于YOLOv5骨干网络的详细说明：
骨干网络采用自下而上的路径从原始图像中提取特征。输入图像经过一系列的卷积、批量归一化和激活函数处理后，得到不同尺度的特征图。这些特征图不仅包含了丰富的细节信息，还具备了一定的语义信息，为后续的目标检测任务提供了有力的支持。
YOLOv5的骨干网络采用了BottleNeckCSP结构，这是一种特殊设计的残差网络结构。
（1）该结构由Focus结构、三组CBL+CSP1_x 和 CBL+SPP串行搭建而成。（如下图）
（2）其中，（如下图）。CBL（Convolutional Block）由卷积（Conv）、批量归一化（BN）和激活函数（Leaky ReLU）组成。（也可叫CBS，因为S的话取得是Silu激活函数，L的话，直接取激活函数Leaky，这俩都是可以的，）
（3）（如下图）CSP1_x则借鉴了CSPNet网络结构，由CBL/CBS模块、Res unit模块以及卷积层Concat组成，其中x表示有x个CSP1模块。
（4）随着网络层数的加深，特征图的分辨率逐渐降低。浅层特征图感受野比较小，尺寸比较大，包含更多的细节和位置信息；而深层特征图尺寸比较小，感受野比较大，反映了图像的全局和抽象特征，包含更丰富的语义信息。
3、YOLOv5的Backbone文件。
4、YOLOv5Backbone的code部分
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 xdef autopad(k, p=None):
if p is None:
p = k // 2 if isinstance(k, int) else [x // 2 for x in k]
return pclass Conv(nn.Module):
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):
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):
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):
super().__init__()
c_ = int(c2 * e)
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):
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))
5、完整的code部分
import copy
import pathlib
import warnings
import matplotlib.pyplot as plt
import torch
from torch import nn
from torchvision import datasets
from torchvision.transforms import transforms
import matplotlib as mpl
mpl.use('Agg') # 在服务器上运行的时候，打开注释
'''
实现YOLOv5的 Backbone模块
'''
# 设置准备
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
print(device)
# 导入数据
data_dir = './weather_photos'
data_dir = pathlib.Path(data_dir)
data_paths = list(data_dir.glob('*'))
classNames = [str(path).split('/')[1] for path in data_paths]
print(classNames)
# 图像预处理
train_transforms = transforms.Compose([
transforms.Resize([224, 224]),
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]
)
])
test_transforms = transforms.Compose([
transforms.Resize([224, 224]),
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]
)
])
total_data = datasets.ImageFolder('./weather_photos', transform=train_transforms)
# 划分数据集
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_size, test_size) # 900 225
# 加载数据集
batch_size = 4
train_dl = torch.utils.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
test_dl = torch.utils.data.DataLoader(test_dataset, batch_size=batch_size, shuffle=True)
for X, y in test_dl:
print('Shape of X[N, C, H, W]', X.shape) # torch.Size([4, 3, 224, 224])
print('Shape of y', y.shape, y.dtype) # torch.Size([4]) torch.int64
break
# 搭建网络模型 （包含Backbone模块的模型）
def autopad(k, p=None):
if p is None:
p = k // 2 if isinstance(k, int) else [x // 2 for x in k]
return p
class Conv(nn.Module):
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):
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):
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):
super().__init__()
c_ = int(c2 * e)
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):
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
model = YOLOv5_backbone().to(device)
print(model)
# 编写训练函数
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集大小
num_batches = len(dataloader) # 批次数目，size/batch_size 向上取整
train_loss, train_acc = 0, 0
for X, y in dataloader:
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model(X)
loss = loss_fn(pred, y)
# 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
# 编写测试函数
def test(dataloader, model, loss_fn):
size = len(dataloader.dataset)
num_batches = len(dataloader)
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
# 正式训练
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)
# 获得当前学习率
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'
torch.save(best_model.state_dict(), PATH)
print('Done!!')
# 结果可视化
warnings.filterwarnings('ignore')
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="Train Accuracy")
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Acc')
plt.subplot(1,2,2)
plt.plot(epochs_range, train_loss, label="Train Loss")
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.savefig("/data/jupyter/deep_demo/p09_v5_backbone/resultImg.jpg") # 保存图片在服务器的位置
plt.show()
# 模型评估
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(f'epoch_test_acc:{epoch_test_acc}, epoch_test_loss:{epoch_test_loss}')
6、结果展示
（1）Adam优化器
（2）SGD优化器
7、总结
使用不同优化器，对于模型的精确度也是不一样的。
看YOLOv5源码的时候，可以先看.yaml文件，看下Backbone部分→Head部分。
看下总体的模型提，再去细看里面的代码。
收获还是很多的。
