当前位置：首页 > java >正文

稠密连接网络(DensoNet)

java 2025/5/13 14:12:21

简介

稠密连接网络（DenseNet）是由康奈尔大学的 Gao Huang 等人在 2017 年提出的深度卷积神经网络架构。它在 ResNet 的残差连接思想基础上进一步创新，通过密集跨层连接（Dense Connectivity）显著提升了特征复用效率，减少了参数量。

enseNet 的核心思想

每一层的输入不仅来自前一层的输出，还来自前面所有层的输出。

与ResNet的主要区别在于，DenseNet里模块的输出不是像ResNet那样和模块的输出相加，而是在通道维上连结。这样模块的输出可以直接传入模块后面的层。在这个设计里，模块直接跟模块后面的所有层连接在了一起。这也是它被称为“稠密连接”的原因。

DenseNet的主要构建模块是稠密块（dense block）和过渡层（transition layer）。前者定义了输入和输出是如何连结的，后者则用来控制通道数，使之不过大。

DenseNet 的关键组件

稠密块

由多个密集连接的层组成，层间通过拼接（而非相加）传递特征。
每层的输出通道数固定（如 growth_rate=32），但整体通道数线性增长。

Transition Layer

用于连接不同 Dense Block，降低特征图尺寸并压缩通道数：
- 1x1 卷积：减少通道数（通常压缩为一半）。
- 2x2 平均池化：下采样。

代码：

import torch
from torch import nn
from torch.utils.data import DataLoader
import torchvision
import torchvision.transforms as transforms
import time
import sys
from torch import nn, optim
import torch.nn.functional as F# 定义全局平均池化层
class GlobalAvgPool2d(nn.Module):# 全局平均池化层可通过将池化窗口形状设置成输入的高和宽实现def __init__(self):super(GlobalAvgPool2d, self).__init__()def forward(self, x):return F.avg_pool2d(x, kernel_size=x.size()[2:])# 定义卷积块
def conv_block(in_channels, out_channels):blk = nn.Sequential(nn.BatchNorm2d(in_channels),nn.ReLU(),nn.Conv2d(in_channels, out_channels,kernel_size=3, padding=1))return blk# 定义DenseBlock
class DenseBlock(nn.Module):def __init__(self, num_convs, in_channels, out_channels):super(DenseBlock, self).__init__()net = []for i in range(num_convs):in_c = in_channels + i*out_channelsnet.append(conv_block(in_c, out_channels))self.net = nn.ModuleList(net)# 计算输出通道数self.out_channels = in_channels + num_convs*out_channelsdef forward(self, X):for blk in self.net:Y = blk(X)# 在通道维上将输入和输出连结X = torch.cat((X, Y), dim=1)return X# blk = DenseBlock(2, 3, 10)
# X = torch.rand(4, 3, 8, 8)
# Y = blk(X)
# print(Y.shape)# 定义过渡块
def transition_block(in_channels, out_channels):blk = nn.Sequential(nn.BatchNorm2d(in_channels),nn.ReLU(),nn.Conv2d(in_channels, out_channels, kernel_size=1),nn.AvgPool2d(kernel_size=2, stride=2))return blk# blk1 = transition_block(23, 10)
# print(blk1(Y).shape)net = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),nn.BatchNorm2d(64),nn.ReLU(),nn.MaxPool2d(kernel_size=3, stride=2, padding=1))num_channels, growth_rate = 64, 32
num_convs_in_dense_blocks = [4, 4, 4, 4]
for i, num_convs in enumerate(num_convs_in_dense_blocks):DB = DenseBlock(num_convs, num_channels, growth_rate)net.add_module('DenseBlock_%d' % i, DB)# 上一个稠密块的输出通道数num_channels = DB.out_channels# 在稠密块之间加入通道数减半的过渡层if i != len(num_convs_in_dense_blocks)-1:net.add_module('transition_block_%d' % i, transition_block(num_channels, num_channels//2))num_channels = num_channels // 2net.add_module('BN', nn.BatchNorm2d(num_channels))
net.add_module('relu', nn.ReLU())
# GlobalAvgPool2d的输出: (Batch, num_channels, 1, 1)
net.add_module('global_avg_pool', GlobalAvgPool2d())
net.add_module('fc', nn.Sequential(nn.Flatten(),nn.Linear(num_channels, 10)))# X = torch.rand((1, 1, 96, 96))
# for name, layer in net.named_children():
#     X = layer(X)
#     print(name, ' output shape:\t', X.shape)def load_data_fashion_mnist(batch_size, resize=None):"""function：将fashion mnist数据集划分为小批量样本Parameters:batch_size - 小批量样本的大小(int)resize - 对图像的维度进行扩大Returns:train_iter - 训练集样本划分为最小批的结果test_iter - 测试集样本划分为最小批的结果Modify:2020-11-262020-12-10 添加图像维度变化"""# 存储图像处理流程trans = []if resize:trans.append(transforms.Resize(size=resize))trans.append(transforms.ToTensor())transform = transforms.Compose(trans)mnist_train = torchvision.datasets.FashionMNIST(root='data/FashionMNIST',train=True,download=True,transform=transform)mnist_test = torchvision.datasets.FashionMNIST(root='data/FashionMNIST',train=False,download=True,transform=transform)if sys.platform.startswith('win'):# 0表示不用额外的进程来加速读取数据num_workers = 0else:num_workers = 4train_iter = torch.utils.data.DataLoader(mnist_train,batch_size=batch_size,shuffle=True,num_workers=num_workers)test_iter = torch.utils.data.DataLoader(mnist_test,batch_size=batch_size,shuffle=False,num_workers=num_workers)return train_iter, test_iterdef evaluate_accuracy(data_iter, net, device=None):"""function:计算多分类模型预测结果的准确率Parameters:data_iter - 样本划分为最小批的结果net - 定义的网络device - 指定计算在GPU或者CPU上进行Returns:准确率计算结果Modify:2020-11-302020-12-03 增加模型训练模型和推理模式的判别2020-12-10 增加指定运行计算位置的方法"""if device is None and isinstance(net, torch.nn.Module):# 如果没指定device就使用net的devicedevice = next(net.parameters()).deviceacc_sum, n = 0.0, 0with torch.no_grad():for X, y in data_iter:if isinstance(net, torch.nn.Module):# 评估模式, 这会关闭dropoutnet.eval()# .cpu()保证可以进行数值加减acc_sum += (net(X.to(device)).argmax(dim=1) ==y.to(device)).float().sum().cpu().item()# 改回训练模式net.train()# 自定义的模型, 2.13节之后不会用到, 不考虑GPUelse:if ('is_training' in net.__code__.co_varnames):# 将is_training设置成Falseacc_sum += (net(X, is_training=False).argmax(dim=1) ==y).float().sum().item()else:acc_sum += (net(X).argmax(dim=1) == y).float().sum().item()n += y.shape[0]return acc_sum / ndef train_ch5(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs):"""function:利用softmax回归模型对图像进行分类识别Parameters:net - 定义的网络train_iter - 训练集样本划分为最小批的结果test_iter - 测试集样本划分为最小批的结果num_epochs - 迭代次数batch_size - 最小批大小optimizer - 优化器device - 指定计算在GPU或者CPU上进行Returns:Modify:2020-12-10"""# 将模型加载到指定运算器中net = net.to(device)print("training on ", device)loss = torch.nn.CrossEntropyLoss()for epoch in range(num_epochs):train_l_sum, train_acc_sum, n, batch_count = 0.0, 0.0, 0, 0start = time.time()for X, y in train_iter:X = X.to(device)y = y.to(device)y_hat = net(X)l = loss(y_hat, y)# 梯度清零optimizer.zero_grad()l.backward()optimizer.step()train_l_sum += l.cpu().item()train_acc_sum += (y_hat.argmax(dim=1) == y).sum().cpu().item()n += y.shape[0]batch_count += 1test_acc = evaluate_accuracy(test_iter, net, device=device)print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f, \time %.1f sec' % (epoch+1, train_l_sum/batch_count,train_acc_sum/n, test_acc,time.time()-start))device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
batch_size = 256
train_iter, test_iter = load_data_fashion_mnist(batch_size,
resize=96)
lr, num_epochs = 0.001, 2
optimizer = torch.optim.Adam(net.parameters(), lr=lr)
train_ch5(net, train_iter, test_iter, batch_size, optimizer,
device, num_epochs)torch.save(net.state_dict(), 'DensoNet.params')

查看全文

http://www.xdnf.cn/news/5750.html