PyTorch - 如何在评估模式下停用 dropout

Question

这是我定义的模型，它是一个具有 2 个完全连接层的简单 lstm。

 import copy
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim

class mylstm(nn.Module):
    def __init__(self,input_dim, output_dim, hidden_dim,linear_dim):
        super(mylstm, self).__init__()
        self.hidden_dim=hidden_dim
        self.lstm=nn.LSTMCell(input_dim,self.hidden_dim)
        self.linear1=nn.Linear(hidden_dim,linear_dim)
        self.linear2=nn.Linear(linear_dim,output_dim)
    def forward(self, input):
        out,_=self.lstm(input)
        out=nn.Dropout(p=0.3)(out)
        out=self.linear1(out)
        out=nn.Dropout(p=0.3)(out)
        out=self.linear2(out)
        return out

x_train and x_val are float dataframe with shape (4478,30) , while y_train and y_val are float df with shape (4478,10)

     x_train.head()
Out[271]:
       0       1       2       3    ...        26      27      28      29
0  1.6110  1.6100  1.6293  1.6370   ...    1.6870  1.6925  1.6950  1.6905
1  1.6100  1.6293  1.6370  1.6530   ...    1.6925  1.6950  1.6905  1.6960
2  1.6293  1.6370  1.6530  1.6537   ...    1.6950  1.6905  1.6960  1.6930
3  1.6370  1.6530  1.6537  1.6620   ...    1.6905  1.6960  1.6930  1.6955
4  1.6530  1.6537  1.6620  1.6568   ...    1.6960  1.6930  1.6955  1.7040

[5 rows x 30 columns]

x_train.shape
Out[272]: (4478, 30)

定义变量并做一次 bp，我可以发现验证损失是 1.4941

 model=mylstm(30,10,200,100).double()
from torch import optim
optimizer=optim.RMSprop(model.parameters(), lr=0.001, alpha=0.9)
criterion=nn.L1Loss()
input_=torch.autograd.Variable(torch.from_numpy(np.array(x_train)))
target=torch.autograd.Variable(torch.from_numpy(np.array(y_train)))
input2_=torch.autograd.Variable(torch.from_numpy(np.array(x_val)))
target2=torch.autograd.Variable(torch.from_numpy(np.array(y_val)))
optimizer.zero_grad()
output=model(input_)
loss=criterion(output,target)
loss.backward()
optimizer.step()
moniter=criterion(model(input2_),target2)

moniter
Out[274]: tensor(1.4941, dtype=torch.float64, grad_fn=<L1LossBackward>)

但是我再次调用了 forward 函数，由于 dropout 的随机性，我得到了一个不同的数字

moniter=criterion(model(input2_),target2)
moniter
Out[275]: tensor(1.4943, dtype=torch.float64, grad_fn=<L1LossBackward>)

我应该怎么做才能消除预测短语中的所有丢失？

我试过 eval() ：

 moniter=criterion(model.eval()(input2_),target2)
moniter
Out[282]: tensor(1.4942, dtype=torch.float64, grad_fn=<L1LossBackward>)

moniter=criterion(model.eval()(input2_),target2)
moniter
Out[283]: tensor(1.4945, dtype=torch.float64, grad_fn=<L1LossBackward>)

并传递一个附加参数 p 来控制 dropout：

 import copy
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
class mylstm(nn.Module):
    def __init__(self,input_dim, output_dim, hidden_dim,linear_dim,p):
        super(mylstm, self).__init__()
        self.hidden_dim=hidden_dim
        self.lstm=nn.LSTMCell(input_dim,self.hidden_dim)
        self.linear1=nn.Linear(hidden_dim,linear_dim)
        self.linear2=nn.Linear(linear_dim,output_dim)
    def forward(self, input,p):
        out,_=self.lstm(input)
        out=nn.Dropout(p=p)(out)
        out=self.linear1(out)
        out=nn.Dropout(p=p)(out)
        out=self.linear2(out)
        return out

model=mylstm(30,10,200,100,0.3).double()

output=model(input_)
loss=criterion(output,target)
loss.backward()
optimizer.step()
moniter=criterion(model(input2_,0),target2)
Traceback (most recent call last):

  File "<ipython-input-286-e49b6fac918b>", line 1, in <module>
    output=model(input_)

  File "D:\Users\shan xu\Anaconda3\lib\site-packages\torch\nn\modules\module.py", line 489, in __call__
    result = self.forward(*input, **kwargs)

TypeError: forward() missing 1 required positional argument: 'p'

但他们都没有工作。

原文由 Tommy Yu 发布，翻译遵循 CC BY-SA 4.0 许可协议

Answer 1

您必须在 --- 中定义 --- __init__ nn.Dropout 层，并将其分配给您的模型以响应调用 eval() 。

所以像这样改变你的模型应该适合你：

 class mylstm(nn.Module):
    def __init__(self,input_dim, output_dim, hidden_dim,linear_dim,p):
        super(mylstm, self).__init__()
        self.hidden_dim=hidden_dim
        self.lstm=nn.LSTMCell(input_dim,self.hidden_dim)
        self.linear1=nn.Linear(hidden_dim,linear_dim)
        self.linear2=nn.Linear(linear_dim,output_dim)

        # define dropout layer in __init__
        self.drop_layer = nn.Dropout(p=p)
    def forward(self, input):
        out,_= self.lstm(input)

        # apply model dropout, responsive to eval()
        out= self.drop_layer(out)
        out= self.linear1(out)

        # apply model dropout, responsive to eval()
        out= self.drop_layer(out)
        out= self.linear2(out)
        return out

如果你像这样改变它，一旦你调用 eval() ，这个 dropout 就会失效。

注意：如果您想在之后继续训练，您需要在您的模型上调用 train() 以退出评估模式。

---

您还可以在此处找到用于评估模式的 eval() 的小型工作示例： nn.Dropout vs. F.dropout pyTorch

原文由 MBT 发布，翻译遵循 CC BY-SA 4.0 许可协议

Answer 2

我添加这个答案只是因为我现在面临着同样的问题，同时试图通过辍学分歧重现深度贝叶斯主动学习。如果您需要保持 dropout 处于活动状态（例如，为相同的测试实例引导一组不同的预测），您只需让模型处于训练模式，无需定义自己的 dropout 层。

由于在 pytorch 中你需要定义自己的预测函数，你可以像这样向它添加一个参数：

 def predict_class(model, test_instance, active_dropout=False):
    if active_dropout:
        model.train()
    else:
        model.eval()

原文由 Edoardo Guerriero 发布，翻译遵循 CC BY-SA 4.0 许可协议