目录
1、动机
- 一个好的模型需要对输入数据的扰动鲁棒
- 使用有噪音的数据等价于Tikhonov正则
- 丢弃法:在层之间加入噪音
2、无偏差的加入噪音
对 x 加入噪音得到
,我们希望
E[
]=x
丢弃法对每个元素进行如下扰动
3、使用丢弃法
- 通常将丢弃法作用在隐藏全连接层的输出上
4、预测中的丢弃法
-
正则项只在训练中使用:他们影响模型参数的更新
-
在预测过程中,丢弃法直接返回输入
-
这样能保证确定性的输出
5、总结
- 丢弃法将一些输出项随机置0来控制模型复杂度
- 常作用在多层感知机的隐藏层输出上
- 丢弃概率是控制模型复杂度的超参数
附:代码实现
从零开始实现
import torch
from torch import nn
from d2l import torch as d2l
# Define the dropout_layer function
def dropout_layer(X, dropout):
assert 0 <= dropout <= 1
# p = 1,Discard all elements
if dropout == 1:
return torch.zeros_like(X)
# p = 0,Keep all elements
if dropout == 0:
return X
mask = (torch.rand(X.shape) > dropout).float()
return mask * X / (1.0 - dropout)
# Test the dropout_layer function
X= torch.arange(16, dtype = torch.float32).reshape((2, 8))
print(X)
print(dropout_layer(X, 0.))
print(dropout_layer(X, 0.5))
print(dropout_layer(X, 1.))
# Define model parameters
num_inputs, num_outputs, num_hiddens1, num_hiddens2 = 784, 10, 256, 256
# Definition model
dropout1, dropout2 = 0.2, 0.5
class Net(nn.Module):
def __init__(self, num_inputs, num_outputs, num_hiddens1, num_hiddens2,
is_training = True):
super(Net, self).__init__()
self.num_inputs = num_inputs
self.training = is_training
self.lin1 = nn.Linear(num_inputs, num_hiddens1)
self.lin2 = nn.Linear(num_hiddens1, num_hiddens2)
self.lin3 = nn.Linear(num_hiddens2, num_outputs)
self.relu = nn.ReLU()
def forward(self, X):
H1 = self.relu(self.lin1(X.reshape((-1, self.num_inputs))))
# 只有在训练模型时才使用dropout
if self.training == True:
# 在第一个全连接层之后添加一个dropout层
H1 = dropout_layer(H1, dropout1)
H2 = self.relu(self.lin2(H1))
if self.training == True:
# 在第二个全连接层之后添加一个dropout层
H2 = dropout_layer(H2, dropout2)
out = self.lin3(H2)
return out
net = Net(num_inputs, num_outputs, num_hiddens1, num_hiddens2)
# Training and testing
num_epochs, lr, batch_size = 10, 0.5, 256
loss = nn.CrossEntropyLoss(reduction='none')
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
trainer = torch.optim.SGD(net.parameters(), lr=lr)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, trainer)
简洁实现
net = nn.Sequential(nn.Flatten(),
nn.Linear(784, 256),
nn.ReLU(),
# 在第一个全连接层之后添加一个dropout层
nn.Dropout(dropout1),
nn.Linear(256, 256),
nn.ReLU(),
# 在第二个全连接层之后添加一个dropout层
nn.Dropout(dropout2),
nn.Linear(256, 10))
def init_weights(m):
if type(m) == nn.Linear:
nn.init.normal_(m.weight, std=0.01)
net.apply(init_weights);
trainer = torch.optim.SGD(net.parameters(), lr=lr)
d2l.train_ch3(net, train_iter, test_iter, loss, num_epochs, trainer)