參考連接: 注意力機制
參考連接: 深度學習卷積神經網絡重要結構之通道注意力和空間注意力模塊
參考連接: 用於卷積神經網絡的注意力機制(Attention)----CBAM: Convolutional Block Attention Module
參考連接: moskomule/senet.pytorch
參考連接: Squeeze-and-Excitation Networks
參考連接: CBAM: Convolutional Block Attention Modulepython
空間注意力機制:
代碼實驗展現:git
import torch from torch import nn class SpatialAttention(nn.Module): def __init__(self, kernel_size=7): super(SpatialAttention, self).__init__() assert kernel_size in (3, 7), 'kernel size must be 3 or 7' padding = 3 if kernel_size == 7 else 1 self.conv1 = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False) # 7,3 3,1 self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = torch.mean(x, dim=1, keepdim=True) max_out, _ = torch.max(x, dim=1, keepdim=True) x = torch.cat([avg_out, max_out], dim=1) x = self.conv1(x) return self.sigmoid(x) if __name__ == '__main__': SA = SpatialAttention(7) data_in = torch.randn(8,32,300,300) data_out = SA(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 1, 300, 300])
控制檯結果輸出展現:github
Windows PowerShell 版權全部 (C) Microsoft Corporation。保留全部權利。 嘗試新的跨平臺 PowerShell https://aka.ms/pscore6 加載我的及系統配置文件用了 1003 毫秒。 (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> & 'D:\Anaconda3\python.exe' 'c:\Users\chenxuqi\.vscode\extensions\ms-python.python-2021.1.502429796\pythonFiles\lib\python\debugpy\launcher' '55088' '--' 'c:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM\ 空間注意力機制.py' torch.Size([8, 32, 300, 300]) torch.Size([8, 1, 300, 300]) (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> conda activate base (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM>
通道注意力機制:web
代碼實驗展現:網絡
import torch from torch import nn class ChannelAttention(nn.Module): def __init__(self, in_planes, ratio=16): super(ChannelAttention, self).__init__() self.avg_pool = nn.AdaptiveAvgPool2d(1) self.max_pool = nn.AdaptiveMaxPool2d(1) self.fc1 = nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False) self.relu1 = nn.ReLU() self.fc2 = nn.Conv2d(in_planes // ratio, in_planes, 1, bias=False) self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = self.fc2(self.relu1(self.fc1(self.avg_pool(x)))) max_out = self.fc2(self.relu1(self.fc1(self.max_pool(x)))) out = avg_out + max_out return self.sigmoid(out) if __name__ == '__main__': CA = ChannelAttention(32) data_in = torch.randn(8,32,300,300) data_out = CA(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 32, 1, 1])
控制檯結果輸出展現:svg
Windows PowerShell 版權全部 (C) Microsoft Corporation。保留全部權利。 嘗試新的跨平臺 PowerShell https://aka.ms/pscore6 加載我的及系統配置文件用了 882 毫秒。 (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> conda activate base (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> & 'D:\Anaconda3\python.exe' 'c:\Users\chenxuqi\.vscode\extensions\ms-python.python-2021.1.502429796\pythonFiles\lib\python\debugpy\launcher' '55339' '--' 'c:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM\ 通道注意力機制.py' torch.Size([8, 32, 300, 300]) torch.Size([8, 32, 1, 1]) (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM>
CBAM注意力機制:
代碼實驗展現:學習
import torch from torch import nn class ChannelAttention(nn.Module): def __init__(self, in_planes, ratio=16): super(ChannelAttention, self).__init__() self.avg_pool = nn.AdaptiveAvgPool2d(1) self.max_pool = nn.AdaptiveMaxPool2d(1) self.fc1 = nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False) self.relu1 = nn.ReLU() self.fc2 = nn.Conv2d(in_planes // ratio, in_planes, 1, bias=False) self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = self.fc2(self.relu1(self.fc1(self.avg_pool(x)))) max_out = self.fc2(self.relu1(self.fc1(self.max_pool(x)))) out = avg_out + max_out return self.sigmoid(out) class SpatialAttention(nn.Module): def __init__(self, kernel_size=7): super(SpatialAttention, self).__init__() assert kernel_size in (3, 7), 'kernel size must be 3 or 7' padding = 3 if kernel_size == 7 else 1 self.conv1 = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False) # 7,3 3,1 self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = torch.mean(x, dim=1, keepdim=True) max_out, _ = torch.max(x, dim=1, keepdim=True) x = torch.cat([avg_out, max_out], dim=1) x = self.conv1(x) return self.sigmoid(x) class CBAM(nn.Module): def __init__(self, in_planes, ratio=16, kernel_size=7): super(CBAM, self).__init__() self.ca = ChannelAttention(in_planes, ratio) self.sa = SpatialAttention(kernel_size) def forward(self, x): out = x * self.ca(x) result = out * self.sa(out) return result if __name__ == '__main__': print('testing ChannelAttention'.center(100,'-')) torch.manual_seed(seed=20200910) CA = ChannelAttention(32) data_in = torch.randn(8,32,300,300) data_out = CA(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 32, 1, 1]) if __name__ == '__main__': print('testing SpatialAttention'.center(100,'-')) torch.manual_seed(seed=20200910) SA = SpatialAttention(7) data_in = torch.randn(8,32,300,300) data_out = SA(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 1, 300, 300]) if __name__ == '__main__': print('testing CBAM'.center(100,'-')) torch.manual_seed(seed=20200910) cbam = CBAM(32, 16, 7) data_in = torch.randn(8,32,300,300) data_out = cbam(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 1, 300, 300])
控制檯結果輸出展現:測試
Windows PowerShell 版權全部 (C) Microsoft Corporation。保留全部權利。 嘗試新的跨平臺 PowerShell https://aka.ms/pscore6 加載我的及系統配置文件用了 1029 毫秒。 (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> & 'D:\Anaconda3\python.exe' 'c:\Users\chenxuqi\.vscode\extensions\ms-python.python-2021.1.502429796\pythonFiles\lib\python\debugpy\launcher' '55659' '--' 'c:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM\cbam注意力機制.py' --------------------------------------testing ChannelAttention-------------------------------------- torch.Size([8, 32, 300, 300]) torch.Size([8, 32, 1, 1]) --------------------------------------testing SpatialAttention-------------------------------------- torch.Size([8, 32, 300, 300]) torch.Size([8, 1, 300, 300]) --------------------------------------------testing CBAM-------------------------------------------- torch.Size([8, 32, 300, 300]) torch.Size([8, 32, 300, 300]) (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> conda activate base (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM> (base) PS C:\Users\chenxuqi\Desktop\News4cxq\測試注意力機制CBAM>
SE注意力機制:ui
代碼實驗展現:url
from torch import nn import torch class SELayer(nn.Module): def __init__(self, channel, reduction=16): super(SELayer, self).__init__() self.avg_pool = nn.AdaptiveAvgPool2d(1) self.fc = nn.Sequential( nn.Linear(channel, channel // reduction, bias=False), nn.ReLU(inplace=True), nn.Linear(channel // reduction, channel, bias=False), nn.Sigmoid() ) def forward(self, x): b, c, _, _ = x.size() y = self.avg_pool(x).view(b, c) y = self.fc(y).view(b, c, 1, 1) return x * y.expand_as(x) # return x * y if __name__ == '__main__': torch.manual_seed(seed=20200910) data_in = torch.randn(8,32,300,300) SE = SELayer(32) data_out = SE(data_in) print(data_in.shape) # torch.Size([8, 32, 300, 300]) print(data_out.shape) # torch.Size([8, 32, 300, 300])
控制檯輸出結果展現:
Windows PowerShell 版權全部 (C) Microsoft Corporation。保留全部權利。 嘗試新的跨平臺 PowerShell https://aka.ms/pscore6 加載我的及系統配置文件用了 979 毫秒。 (base) PS F:\Iris_SSD_small\senet.pytorch-master> & 'D:\Anaconda3\envs\pytorch_1.7.1_cu102\python.exe' 'c:\Users\chenxuqi\.vscode\extensions\ms-python.python-2021.1.502429796\pythonFiles\lib\python\debugpy\launcher' '54904' '--' 'f:\Iris_SSD_small\senet.pytorch-master\senet\se_module.py' torch.Size([8, 32, 300, 300]) torch.Size([8, 32, 300, 300]) (base) PS F:\Iris_SSD_small\senet.pytorch-master> conda activate pytorch_1.7.1_cu102 (pytorch_1.7.1_cu102) PS F:\Iris_SSD_small\senet.pytorch-master>
本文同步分享在 博客「敲代碼的小風」（CSDN）。
若有侵權，請聯繫 support@oschina.cn 刪除。
本文參與「OSC源創計劃」，歡迎正在閱讀的你也加入，一塊兒分享。