第J3周：DenseNet算法实战与解析

•    🍨 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

• 🚀 文章来源：K同学的学习圈子

目录 

我的环境

• 语言环境：python3.8.18
• 编译器：jupyter notebook
• 深度学习环境：torch==2.0.1+cu118，torchvision==0.15.2+cu118 

一、代码实现

1.配置GPU 

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)

cuda

2.导入数据

data_dir = "E:/pytorch练习/训练营/J1/数据集/第8天/bird_photos"
data_dir = pathlib.Path(data_dir)
 
data_paths  = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[7] for path in data_paths]
print(classeNames)

['Bananaquit', 'Black Skimmer', 'Black Throated Bushtiti', 'Cockatoo']

3.加载数据

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("E:/pytorch练习/训练营/J1/数据集/第8天/bird_photos",transform = train_transforms)
print(total_data)

4.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_dataset)
print(test_dataset)
 
batch_size = 8
 
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

<torch.utils.data.dataset.Subset object at 0x000001974B7AD610>
<torch.utils.data.dataset.Subset object at 0x000001974B7AD640>
Shape of X [N, C, H, W]:  torch.Size([8, 3, 224, 224])
Shape of y:  torch.Size([8]) torch.int64
​

5.构造模型

import torch
import torch.nn as nn
import torch.nn.functional as F
 
class DenseLayer(nn.Sequential):
    """Basic unit of DenseBlock (using bottleneck layer) """
    def __init__(self, num_input_features, growth_rate, bn_size, drop_rate):
        super(DenseLayer, self).__init__()
        self.bn1 = nn.BatchNorm2d(num_input_features)
        self.relu1 = nn.ReLU()
        self.conv1 = nn.Conv2d(num_input_features, bn_size*growth_rate,
                                           kernel_size=1, stride=1, bias=False)
        self.bn2 = nn.BatchNorm2d(bn_size*growth_rate)
        self.relu2 = nn.ReLU()
        self.conv2 = nn.Conv2d(bn_size*growth_rate, growth_rate,
                                           kernel_size=3, stride=1, padding=1, bias=False)
        self.drop_rate = drop_rate
 
    def forward(self, x):
        output = self.bn1(x)
        output = self.relu1(output)
        output = self.conv1(output)
 
        output = self.bn2(output)
        output = self.relu2(output)
        output = self.conv2(output)
 
        if self.drop_rate > 0:
            output = F.dropout(output, p=self.drop_rate)
        return torch.cat([x, output], 1)

class DenseBlock(nn.Sequential):
    def __init__(self, num_layers, num_input_features, bn_size, growth_rate, drop_rate):
        super(DenseBlock, self).__init__()
        for i in range(num_layers):
            if i == 0:
                self.layer = nn.Sequential(
                    DenseLayer(num_input_features+i*growth_rate, growth_rate, bn_size,drop_rate)
                )
            else:
                layer = DenseLayer(num_input_features+i*growth_rate, growth_rate, bn_size,drop_rate)
                self.layer.add_module("denselayer%d" % (i+1), layer)
    
    def forward(self,input):
        return self.layer(input)

class Transition(nn.Sequential):
    def __init__(self, num_input_feature, num_output_features):
        super(Transition, self).__init__()
        self.bn = nn.BatchNorm2d(num_input_feature)
        self.relu = nn.ReLU()
        self.conv = nn.Conv2d(num_input_feature, num_output_features,
                                          kernel_size=1, stride=1, bias=False)
        self.pool = nn.AvgPool2d(2, stride=2)
 
    def forward(self,input):
        output = self.bn(input)
        output = self.relu(output)
        output = self.conv(output)
        output = self.pool(output)
 
        return output

model = DenseNet(32,(2, 2, 4, 4),64,4,0.5,0,10)
model.to(device)

 6.定义训练和测试函数 

def train(dataloader,model,optimizer,loss_fn):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)
 
    train_acc,train_loss = 0,0
 
    for X,y in dataloader:
        X,y = X.to(device),y.to(device)
 
        pred = model(X)
        loss = loss_fn(pred,y)
 
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()
 
        train_loss += loss.item()
        train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
 
    train_loss /= num_batches
    train_acc /= size
 
    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

 7.定义一些超参数

loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-2
opt = torch.optim.SGD(model.parameters(),lr=learn_rate)
 
import copy
 
epochs = 10
 
train_loss=[]
train_acc=[]
test_loss=[]
test_acc=[]
best_acc = 0

8.开始训练

for epoch in range(epochs):
 
    model.train()
    epoch_train_acc,epoch_train_loss = train(train_dl,model,opt,loss_fn)
 
    model.eval()
    epoch_test_acc,epoch_test_loss = test(test_dl,model,loss_fn)
 
    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)
 
    lr = opt.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))
 

 
print('Done')

 9.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[-10:], label='Training Accuracy')
plt.plot(epochs_range, test_acc[-10:], 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[-10:], label='Training Loss')
plt.plot(epochs_range, test_loss[-10:], label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

 

二、个人总结 

这周学习了DenseNet算法模型，效果相较于上周的ResNet50V2有了显著的提升 。

DenseNet（密集连接卷积网络）是一种深度学习模型，其核心特点是网络中的每一层都与其前面的所有层直接相连，形成了密集的连接模式。这种设计促进了特征的重用，并鼓励梯度流动，有助于缓解深度学习中的梯度消失问题。下面是DenseNet结构的关键组成部分：

1. 初始卷积层：网络通常以一个标准的卷积层开始，用于初步提取输入图像的特征，并可能伴随有池化层来缩小输入尺寸。

2. Dense Blocks（密集块）：DenseNet的主要构建模块。每个密集块内，每新增一个层，都会将其输出特征图与之前所有层的输出特征图进行拼接（concatenation），作为下一个层的输入。这保证了信息流的高效传递和特征的复用。为了控制模型复杂度，每个层通过较小的增长率（growth rate）来增加特征图的数量，即每个层产生的新特征图数量。

3. Bottleneck Layers（瓶颈层）：为了减少计算成本，实际应用中的DenseNet常采用Bottleneck层设计。这些层首先使用1x1卷积来减少输入特征图的数量，然后是BN（Batch Normalization）和ReLU激活函数，接着是3x3卷积来提取特征。这样的设计保持了模型的效率，同时维持了特征的丰富性。

4. Transition Layers（过渡层）：位于Dense Blocks之间，用于过渡并控制模型的复杂度。过渡层通常包含1x1的卷积用于压缩特征图的通道数（使用压缩因子θ），以及可选的平均池化（Average Pooling）来进一步减小空间尺寸，帮助减少计算负担和过拟合风险。

5. 分类层：网络的尾部通常包括全局平均池化（Global Average Pooling）层，用于将每个特征图的 spatial 维度压缩为一个值，随后连接一个或多个全连接层用于最终的分类或回归任务。

DenseNet通过其独特的密集连接机制，不仅增强了特征传播，还允许特征的多尺度融合，提高了模型的性能和训练效率。