Desnet模型详解

模型介绍

DenseNet的主要思想是密集连接,它在卷积神经网络(CNN)中引入了密集块(Dense Block),在这些块中,每个层都与前面所有层直接连接。这种设计可以让信息更快速地传播,有助于解决梯度消失的问题,同时也能够增加网络的参数共享,减少参数量,提高模型的效率和性能。

Desnet原理

DenseNet 的原理可以总结为以下几个关键点:

  1. 密集连接的块: DenseNet 将网络分成多个密集块(Dense Block)。在每个密集块内,每一层都连接到前面所有的层,不仅仅是前一层。这种连接方式使得信息能够更加快速地传播,允许网络在更早的阶段融合不同层的特征。

  2. 跳跃连接: 每一层都从前面所有的层接收特征作为输入。这些输入通过堆叠而来,从而构成了一个密集的特征图。这种跳跃连接有助于解决梯度消失问题,因为每一层都可以直接访问之前层的梯度信息,使得训练更加稳定。

  3. 特征重用性: 由于每一层都与前面所有层连接,网络可以自动地学习到更加丰富和复杂的特征表示。这样的特征重用性有助于提高网络的性能,同时减少了需要训练的参数数量。

  4. 过渡层: 在密集块之间,通常会使用过渡层(Transition Layer)来控制特征图的大小。过渡层包括一个卷积层和一个池化层,用于减小特征图的尺寸,从而减少计算量。

Desnet模型详解_第1张图片

Desnet的结构

关于 DenseNet 的结构时,我们主要关注网络中的三个主要组成部分:密集块(Dense Block)、过渡层(Transition Layer)以及全局平均池化层。

密集块

密集块是 DenseNet 最核心的部分,由若干层组成。在密集块中,每一层都与前面所有层直接连接。这种密集连接的方式使得信息可以更充分地传递和重用。每一层的输出都是前面所有层输出的连结,这也意味着每一层的输入包括了前面所有层的特征。这种连接方式通过堆叠层的方式,构建了一个密集的特征图。

过渡层

在密集块之间,可以使用过渡层来控制特征图的大小,从而减少计算成本。过渡层由一个卷积层和一个池化层组成。卷积层用于减小通道数,从而降低特征图的维度。池化层(通常是平均池化)用于减小特征图的尺寸。这些操作有助于在保持网络性能的同时降低计算需求。

全局平均池化层

在整个 DenseNet 结构的末尾,通常会添加一个全局平均池化层。这一层的作用是将最终的特征图转换为全局汇总的特征,这对于分类任务是非常有用的。全局平均池化层计算每个通道上的平均值,将每个通道转换为一个标量,从而形成最终的预测。

DenseNet 结构的特点不仅在每个密集块内进行特征的密集连接,还在不同密集块之间使用过渡层来控制网络的尺寸和复杂度。这使得 DenseNet 能够在高度复杂的任务中表现出色,同时保持相对较少的参数。

这些在论文当中也有体现:

Desnet模型详解_第2张图片

Desnet的优缺点比较

优点

  • 密集连接促进信息传递和特征重用,提升了网络性能。

  • 跳跃连接减少了梯度消失,有助于训练深层网络。

  • 密集连接减少参数数量,提高了模型效率。

  • 早期融合多尺度特征,增强了表征能力。

  • 在小样本情况下表现更佳,充分利用有限数据。

缺点

  • 密集连接可能导致内存需求增大。

  • 连接多导致计算量增加,训练和推理时间较长。

  • 可能因复杂性导致过拟合,需考虑正则化。

其实综合考虑,Desnet在图像识别和计算机视觉任务中仍然是一个好的选择。

Pytorch实现Desnet

import torch
import torchvision
import torch.nn as nn
import torchsummary
import torch.nn.functional as F
from torch.hub import load_state_dict_from_url
from collections import OrderedDict
from torchvision.utils import _log_api_usage_once
import torch.utils.checkpoint as cp

model_urls = {
    "densenet121":"https://download.pytorch.org/models/densenet121-a639ec97.pth",
    "densenet161":"https://download.pytorch.org/models/densenet161-8d451a50.pth",
    "densenet169":"https://download.pytorch.org/models/densenet169-b2777c0a.pth",
    "densenet201":"https://download.pytorch.org/models/densenet201-c1103571.pth",
}
cfgs = {
    "densenet121":(6, 12, 24, 16),
    "densenet161":(6, 12, 36, 24),
    "densenet169":(6, 12, 32, 32),
    "densenet201":(6, 12, 48, 32),
}


class DenseLayer(nn.Module):
    def __init__(self, num_input_features, growth_rate, bn_size, drop_rate, memory_efficient = False):
        super(DenseLayer,self).__init__()
        self.norm1 = nn.BatchNorm2d(num_input_features)
        self.relu1 = nn.ReLU(inplace=True)
        self.conv1 = nn.Conv2d(num_input_features, bn_size * growth_rate, kernel_size=1, stride=1, bias=False)

        self.norm2 = nn.BatchNorm2d(bn_size * growth_rate)
        self.relu2 = nn.ReLU(inplace=True)
        self.conv2 = nn.Conv2d(bn_size * growth_rate, growth_rate, kernel_size=3, stride=1, padding=1, bias=False)

        self.drop_rate = float(drop_rate)
        self.memory_efficient = memory_efficient

    def bn_function(self, inputs):
        concated_features = torch.cat(inputs, 1)
        bottleneck_output = self.conv1(self.relu1(self.norm1(concated_features)))
        return bottleneck_output

    def any_requires_grad(self, input):
        for tensor in input:
            if tensor.requires_grad:
                return True
        return False

    @torch.jit.unused
    def call_checkpoint_bottleneck(self, input):
        def closure(*inputs):
            return self.bn_function(inputs)

        return cp.checkpoint(closure, *input)

    def forward(self, input):
        if isinstance(input, torch.Tensor):
            prev_features = [input]
        else:
            prev_features = input

        if self.memory_efficient and self.any_requires_grad(prev_features):
            if torch.jit.is_scripting():
                raise Exception("Memory Efficient not supported in JIT")

            bottleneck_output = self.call_checkpoint_bottleneck(prev_features)
        else:
            bottleneck_output = self.bn_function(prev_features)

        new_features = self.conv2(self.relu2(self.norm2(bottleneck_output)))
        if self.drop_rate > 0:
            new_features = F.dropout(new_features, p=self.drop_rate, training=self.training)
        return new_features


class DenseBlock(nn.ModuleDict):
    def __init__(self,num_layers,num_input_features,bn_size,growth_rate,
                 drop_rate,memory_efficient = False,):
        super(DenseBlock,self).__init__()
        for i in range(num_layers):
            layer = DenseLayer(
                num_input_features + i * growth_rate,
                growth_rate=growth_rate,
                bn_size=bn_size,
                drop_rate=drop_rate,
                memory_efficient=memory_efficient,
            )
            self.add_module("denselayer%d" % (i + 1), layer)

    def forward(self, init_features):
        features = [init_features]
        for name, layer in self.items():
            new_features = layer(features)
            features.append(new_features)
        return torch.cat(features, 1)


class Transition(nn.Sequential):
    """
    Densenet Transition Layer:
        1 × 1 conv
        2 × 2 average pool, stride 2
    """
    def __init__(self, num_input_features, num_output_features):
        super(Transition,self).__init__()
        self.norm = nn.BatchNorm2d(num_input_features)
        self.relu = nn.ReLU(inplace=True)
        self.conv = nn.Conv2d(num_input_features, num_output_features, kernel_size=1, stride=1, bias=False)
        self.pool = nn.AvgPool2d(kernel_size=2, stride=2)


class DenseNet(nn.Module):
    def __init__(self,growth_rate = 32,num_init_features = 64,block_config = None,num_classes = 1000,
                 bn_size = 4,drop_rate = 0.,memory_efficient = False,):

        super(DenseNet,self).__init__()
        _log_api_usage_once(self)

        # First convolution
        self.features = nn.Sequential(
            OrderedDict(
                [
                    ("conv0", nn.Conv2d(3, num_init_features, kernel_size=7, stride=2, padding=3, bias=False)),
                    ("norm0", nn.BatchNorm2d(num_init_features)),
                    ("relu0", nn.ReLU(inplace=True)),
                    ("pool0", nn.MaxPool2d(kernel_size=3, stride=2, padding=1)),
                ]
            )
        )

        # Each denseblock
        num_features = num_init_features
        for i, num_layers in enumerate(block_config):
            block = DenseBlock(
                num_layers=num_layers,
                num_input_features=num_features,
                bn_size=bn_size,
                growth_rate=growth_rate,
                drop_rate=drop_rate,
                memory_efficient=memory_efficient,
            )
            self.features.add_module("denseblock%d" % (i + 1), block)
            num_features = num_features + num_layers * growth_rate
            if i != len(block_config) - 1:
                trans = Transition(num_input_features=num_features, num_output_features=num_features // 2)
                self.features.add_module("transition%d" % (i + 1), trans)
                num_features = num_features // 2

        # Final batch norm
        self.features.add_module("norm5", nn.BatchNorm2d(num_features))
        # Linear layer
        self.classifier = nn.Linear(num_features, num_classes)

        # Official init from torch repo.
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight)
            elif isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.Linear):
                nn.init.constant_(m.bias, 0)

    def forward(self, x):
        features = self.features(x)
        out = F.relu(features, inplace=True)
        out = F.adaptive_avg_pool2d(out, (1, 1))
        out = torch.flatten(out, 1)
        out = self.classifier(out)
        return out

def densenet(growth_rate=32,num_init_features=64,num_classes=1000,mode="densenet121",pretrained=False,**kwargs):
    import re
    pattern = re.compile(
        r"^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$"
    )
    if mode == "densenet161":
        growth_rate=48
        num_init_features=96
    model = DenseNet(growth_rate, num_init_features, cfgs[mode],num_classes=num_classes, **kwargs)
    if pretrained:
        state_dict = load_state_dict_from_url(model_urls[mode], model_dir='./model', progress=True)  # 预训练模型地址
        for key in list(state_dict.keys()):
            res = pattern.match(key)
            if res:
                new_key = res.group(1) + res.group(2)
                state_dict[new_key] = state_dict[key]
                del state_dict[key]
        if num_classes != 1000:
            num_new_classes = num_classes
            weight = state_dict['classifier.weight']
            bias = state_dict['classifier.bias']
            weight_new = weight[:num_new_classes, :]
            bias_new = bias[:num_new_classes]
            state_dict['classifier.weight'] = weight_new
            state_dict['classifier.bias'] = bias_new
        model.load_state_dict(state_dict)
    return model

from torchsummaryX import summary

if __name__ == "__main__":
    in_channels = 3
    num_classes = 10

    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
    model = densenet(growth_rate=32, num_init_features=64, num_classes=num_classes, mode="densenet121", pretrained=True)
    model = model.to(device)
    print(model)
    summary(model, torch.zeros((1, in_channels, 224, 224)).to(device))

你可能感兴趣的:(深度学习杂文,深度学习,人工智能)