pytorch nn.functional.dropout的坑

作者：雷杰
链接：https://www.zhihu.com/question/67209417/answer/302434279
来源：知乎
著作权归作者所有。商业转载请联系作者获得授权，非商业转载请注明出处。

刚踩的坑, 差点就哭出来了TT. --- 我明明加了一百个dropout, 为什么结果一点都没变

使用F.dropout ( nn.functional.dropout )的时候需要设置它的training这个状态参数与模型整体的一致.

比如:

Class DropoutFC(nn.Module):
    def __init__(self):
        super(DropoutFC, self).__init__()
        self.fc = nn.Linear(100,20)

    def forward(self, input):
        out = self.fc(input)
        out = F.dropout(out, p=0.5)
        return out

Net = DropoutFC()
Net.train()

# train the Net

这段代码中的F.dropout实际上是没有任何用的, 因为它的training状态一直是默认值False. 由于F.dropout只是相当于引用的一个外部函数, 模型整体的training状态变化也不会引起F.dropout这个函数的training状态发生变化. 所以, 此处的out = F.dropout(out) 就是 out = out. Ref: https://github.com/pytorch/pytorch/blob/master/torch/nn/functional.py#L535

正确的使用方法如下, 将模型整体的training状态参数传入dropout函数

Class DropoutFC(nn.Module):
   def __init__(self):
       super(DropoutFC, self).__init__()
       self.fc = nn.Linear(100,20)

   def forward(self, input):
       out = self.fc(input)
       out = F.dropout(out, p=0.5, training=self.training)
       return out

Net = DropoutFC()
Net.train()

# train the Net

或者直接使用nn.Dropout() (nn.Dropout()实际上是对F.dropout的一个包装, 也将self.training传入了) Ref: https://github.com/pytorch/pytorch/blob/master/torch/nn/modules/dropout.py#L46

Class DropoutFC(nn.Module):
  def __init__(self):
      super(DropoutFC, self).__init__()
      self.fc = nn.Linear(100,20)
      self.dropout = nn.Dropout(p=0.5)

  def forward(self, input):
      out = self.fc(input)
      out = self.dropout(out)
      return out
Net = DropoutFC()
Net.train()

# train the Net