求教pytorch這段代碼的意思learning_rate = 1e-6

pytorch中這段代碼的意思是把學習率learning_rate設為0.000001

但是設置學習率不是給learning_rate賦值就可以完成的,

在pytorch中設置learning_rate有六種方法(這裏的LR就是LearningRate的縮寫)

1等步長間隔調整學習率

optim.lr_scheduler.StepLR(optimizer, step_size, gamma=0.1, last_epoch=-1)

2cosine學習率

optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max, eta_min=0)

3指數衰減學習率

optim.lr_scheduler.ExponentialLR(optimizer, gamma, last_epoch=-1)

4自適應調整學習率

optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='min', factor=0.1, patience=10,verbose=False, threshold=1e-4, threshold_mode='rel',cooldown=0, min_lr=0, eps=1e-8)

5非等間隔調整學習率

optim.lr_scheduler.MultiStepLR(optimizer, milestones, gamma=0.1, last_epoch=-1)

6自定義網絡層隔學習率

optim.lr_scheduler.LambdaLR( optimizer, lr_lambda, last_epoch=-1)

我給妳壹個等步長間隔調整學習率的例子,妳看看吧

import torch

import torch.optim as optim

from torch.optim import lr_scheduler

from torchvision.models import AlexNet

model = AlexNet(num_classes=2)

optimizer = optim.SGD(params = model.parameters(), lr=0.05)

#每10次叠代，lr = lr * gamma

scheduler = lr_scheduler.StepLR(optimizer, step_size=10, gamma=0.1)

for epoch in range(40):

scheduler.step()

lr = scheduler.get_lr()

print(epoch, scheduler.get_lr()[0])

源代碼(註意源代碼的縮進)

上一篇:Java SE 6 新特性: Java DB&n

下一篇:阿爾法基金是什麽意思？