根據不同的需求,在PyTorch中有時需要爲模型的可學習參數施加自定義的約束或正則項(regular term),下面具體介紹在PyTorch中爲可學習參數施加約束或正則項的方法,先看一下爲損失函數(Loss function)施加正則項的具體形式,如下爲L2正則項:
在上式中,是訓練誤差關於可學習參數w的函數,右邊的第二項表示L2正則項。在PyTorch中L2正則項是默認內置實現的,其中的weight_decay就表示L2正則項的超參數。具體如下:
optimizer = optim.SGD(net.parameters(), lr=0.01, weight_decay=0.01)
根據不同的需求,怎樣自定義自己的正則項函數呢?具體示例如下:
import torch
torch.manual_seed(1)
N, D_in, H, D_out = 10, 5, 5, 1
x = torch.randn(N, D_in)
y = torch.randn(N, D_out)
model = torch.nn.Sequential(
torch.nn.Linear(D_in, H),
torch.nn.ReLU(),
torch.nn.Linear(H, D_out),
)
criterion = torch.nn.MSELoss()
lr = 1e-4
weight_decay = 0 # for torch.optim.SGD
lmbd = 0.9 # for custom L2 regularization
optimizer = torch.optim.SGD(model.parameters(), lr=lr, weight_decay=weight_decay)
for t in range(100):
y_pred = model(x)
# Compute and print loss.
loss = criterion(y_pred, y)
optimizer.zero_grad()
reg_loss = None
for param in model.parameters():
if reg_loss is None:
reg_loss = 0.5 * torch.sum(param**2)
else:
reg_loss = reg_loss + 0.5 * param.norm(2)**2
loss += lmbd * reg_loss
loss.backward()
optimizer.step()
for name, param in model.named_parameters():
print(name, param)
在上述代碼中,如下部分可根據自己的需求,自定義自己的正則項約束:
reg_loss = None
for param in model.parameters():
if reg_loss is None:
reg_loss = 0.5 * torch.sum(param**2)
else:
reg_loss = reg_loss + 0.5 * param.norm(2)**2
參考:
1. How does one implement Weight regularization (l1 or l2) manually without optimum?
2. torch.norm