修改YOLOv5替换Focus层为ENet后遇张量类型不匹配RuntimeError求助

问题详情 #

修改YOLOv5网络结构，将原有Focus层替换为ENet网络后，运行时触发如下错误：

RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.cuda.HalfTensor) should be the same

待解决疑问 #

• 不清楚错误产生的根源，网上查询到的解决方案是将输入数据转换为对应设备类型，示例代码如下：

  device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
  inputs = inputs.to(device)
  

  但项目文件数量较多，不知道该将这段代码添加到何处。

• 错误起始于Bottleneck类forward方法中main=x的赋值操作，想了解Python中这种赋值操作是否属于不允许的行为？相关Bottleneck类代码如下：

  class Bottleneck(nn.Module):
      def __init__(self, channels, internal_ratio=4, kernel_size=3, padding=0, dilation=1,
                   asymmetric=False, dropout_prob=0, bias=False, relu=True):
          super(Bottleneck, self).__init__()
          internal_channels=channels//internal_ratio
  
          self.ext_conv1=nn.Sequential(
                  nn.Conv2d(channels,internal_channels,kernel_size=1,stride=1,bias=bias),
                  nn.BatchNorm2d(internal_channels),
                  activation())
          self.ext_conv2=nn.Sequential(
                      nn.Conv2d(internal_channels, internal_channels, kernel_size=kernel_size,
                              stride=1, padding=padding, dilation=dilation, bias=bias),
                      nn.BatchNorm2d(internal_channels),
                      activation())
          self.ext_conv3=nn.Sequential(
                  nn.Conv2d(internal_channels,channels,kernel_size=1,stride=1,bias=bias),
                  nn.BatchNorm2d(channels),
                  activation())
          self.ext_regul=nn.Dropout2d(p=dropout_prob)
          self.out_activation=activation()
                  
      def forward(self,x):
          main=x
          #print(type(x))
          #print("==========")
          ext=self.ext_conv1(x)
          ext=self.ext_conv2(ext)
          ext=self.ext_conv3(ext)
          ext=self.ext_regul(ext)
          out=main+ext
          return self.out_activation(out)
  

内容的提问来源于stack exchange，提问作者qianqiu