我对这个是否正确感到非常不确定。遗憾的是，我找不到很多关于如何参数化神经网络的好例子。
你觉得在这两个类中使用这种dropout方式如何？首先，我写出原始类：
class NeuralNet(nn.Module): def __init__(self, input_size, hidden_size, num_classes, p = dropout): super(NeuralNet, self).__init__() self.fc1 = nn.Linear(input_size, hidden_size) self.fc2 = nn.Linear(hidden_size, hidden_size) self.fc3 = nn.Linear(hidden_size, num_classes) def forward(self, x): out = F.relu(self.fc1(x)) out = F.relu(self.fc2(out)) out = self.fc3(out) return out然后在这里，我发现了两种不同的写法，我不知道如何区分。第一种使用：
self.drop_layer = nn.Dropout(p=p) 而第二种：
self.dropout = nn.Dropout(p) 这是我的结果：
class NeuralNet(nn.Module): def __init__(self, input_size, hidden_size, num_classes, p = dropout): super(NeuralNet, self).__init__() self.fc1 = nn.Linear(input_size, hidden_size) self.fc2 = nn.Linear(hidden_size, hidden_size) self.fc3 = nn.Linear(hidden_size, num_classes) self.drop_layer = nn.Dropout(p=p) def forward(self, x): out = F.relu(self.fc1(x)) out = F.relu(self.fc2(out)) out = self.fc3(out) return out class NeuralNet(nn.Module): def __init__(self, input_size, hidden_size, num_classes, p = dropout): super(NeuralNet, self).__init__() self.fc1 = nn.Linear(input_size, hidden_size) self.fc2 = nn.Linear(hidden_size, hidden_size) self.fc3 = nn.Linear(hidden_size, num_classes) self.dropout = nn.Dropout(p) def forward(self, x): out = F.relu(self.fc1(x)) out = F.relu(self.fc2(out)) out = self.fc3(out) return out这是否可行，如果不行，如何改进？它是否能给我预期的结果，即创建一个可以dropout某些神经元的神经网络。 重要细节，我只想对神经网络的第二层进行dropout，其余部分不做任何更改！
回答：
你提供的两个例子是完全相同的。self.drop_layer = nn.Dropout(p=p) 和 self.dropout = nn.Dropout(p) 唯一的区别是作者给层赋予了不同的变量名。dropout层通常在.__init__()方法中定义，并在.forward()中调用。像这样：
class NeuralNet(nn.Module): def __init__(self, input_size, hidden_size, num_classes, p = dropout): super(NeuralNet, self).__init__() self.fc1 = nn.Linear(input_size, hidden_size) self.fc2 = nn.Linear(hidden_size, hidden_size) self.fc3 = nn.Linear(hidden_size, num_classes) self.dropout = nn.Dropout(p) def forward(self, x): out = F.relu(self.fc1(x)) out = F.relu(self.fc2(out)) out = self.dropout(self.fc3(out)) return out你可以进行测试：
tensor(5440) # 非零值的总和tensor(2656) # dropout后非零值的总和让我们可视化它：
tensor([[ 1.1404, 0.2102, -0.1237, 0.4240, 0.0174], [-2.0872, 1.2790, 0.7804, -0.0962, -0.9730], [ 0.4788, -1.3408, 0.0483, 2.4125, -1.2463], [ 1.5761, 0.3592, 0.2302, 1.3980, 0.0154], [-0.4308, 0.2484, 0.8584, 0.1689, -1.3607]])现在，让我们应用dropout：
m = nn.Dropout(p=0.5)output = m(input)print(output)tensor([[ 0.0000, 0.0000, -0.0000, 0.8481, 0.0000], [-0.0000, 0.0000, 1.5608, -0.0000, -1.9459], [ 0.0000, -0.0000, 0.0000, 0.0000, -0.0000], [ 0.0000, 0.7184, 0.4604, 2.7959, 0.0308], [-0.0000, 0.0000, 0.0000, 0.0000, -0.0000]])大约一半的神经元被置为零，因为我们设定了神经元被置为零的概率为p=0.5！
