DenseNet.zip

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 """ growth_rate:（int）DenseLayer中使用的增长率和最终输出通道数/过滤器数量(论文中的k) block_config：（4个int组成的列表）每个DenseBlock中的层数 num_init_features：（int）第一个Conv2d中的过滤器数量（输出通道数） bn_size：（int）瓶颈层中使用的因子 compression_rate：（float）过渡层中使用的压缩率 drop_rate：（float）每个DenseLayer之后的丢弃率 num_classes：（int）分类的类数 """ class DenseNet(nn.Module): def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16), num_init_features=64,bn_size=4, compression_rate=0.5, drop_rate=0, num_classes=1000): super(DenseNet, self).__init__() # 前部 self.features = nn.Sequential( #第一层 nn.Conv2d(3, num_init_features, kernel_size=7, stride=2, padding=3, bias=False), nn.BatchNorm2d(num_init_features), nn.ReLU(), #第二层 nn.MaxPool2d(3, stride=2, padding=1) ) # DenseBlock num_features = num_init_features for i, num_layers in enumerate(block_config): block = DenseBlock(num_layers, num_features, bn_size, growth_rate,drop_rate) if i == 0: self.block_tran = nn.Sequential( block ) else: self.block_tran.add_module("denseblock%d" % (i + 1), block)#添加一个block num_features += num_layers*growth_rate#更新通道数 if i != len(block_config) - 1:#除去最后一层不需要加Transition来连接两个相邻的DenseBlock transition = Transition(num_features, int(num_features*compression_rate)) self.block_tran.add_module("transition%d" % (i + 1), transition)#添加Transition num_features = int(num_features * compression_rate)#更新通道数 # 后部 bn+ReLU self.tail = nn.Sequential( nn.BatchNorm2d(num_features), nn.ReLU() ) # classification layer self.classifier = nn.Linear(num_features, num_classes) # params initialization for m in self.modules(): if isinstance(m, nn.Conv2d):#如果是卷积层，参数kaiming分布处理 nn.init.kaiming_normal_(m.weight) elif isinstance(m, nn.BatchNorm2d):#如果是批量归一化则伸缩参数为1，偏移为0 nn.init.constant_(m.bias, 0) nn.init.constant_(m.weight, 1) elif isinstance(m, nn.Linear):#如果是线性层偏移为0 nn.init.constant_(m.bias, 0) def forward(self, x): features = self.features(x) block_output = self.block_tran(features) tail_output = self.tail(block_output) out = F.avg_pool2d(tail_output, 7, stride=1).view(tail_output.size(0), -1)#平均池化 out = self.classifier(out) return out