当前位置:网站首页>PyTorch四种常用优化器测试
PyTorch四种常用优化器测试
2022-07-06 09:16:00 【想成为风筝】
PyTorch四种常用优化器测试SGD、SGD(Momentum)、RMSprop、Adam
import os
os.environ['KMP_DUPLICATE_LIB_OK'] = 'TRUE'
import torch
import torch.utils.data as Data
import torch.nn.functional as F
import matplotlib.pyplot as plt
#超参数
LR =0.001
Batch_Size = 32
Epochs = 12
#生成训练数据
x = torch.unsqueeze(torch.linspace(-1,1,1000),dim=1)
y = x.pow(2) + 0.1 * torch.normal(torch.zeros(*x.size()))
torch_dataset = Data.TensorDataset(x,y)
loader = Data.DataLoader(dataset=torch_dataset,batch_size=Batch_Size,shuffle=True)
class Net2(torch.nn.Module):
def __init__(self):
super(Net2,self).__init__()
self.hidden = torch.nn.Linear(1,20)
self.predict = torch.nn.Linear(20,1)
#前向传递
def forward(self,x):
x = F.relu(self.hidden(x))
x = self.predict(x)
return x
net_SGD = Net2()
net_Momentum =Net2()
net_RMSprop = Net2()
net_Adam = Net2()
nets = [net_SGD,net_Momentum,net_RMSprop,net_Adam]
opt_SGD = torch.optim.SGD(net_SGD.parameters(),lr=LR)
opt_Momentum = torch.optim.SGD(net_Momentum.parameters(),lr=LR,momentum=0.9)
opt_RMSProp = torch.optim.RMSprop(net_RMSprop.parameters(),lr=LR,alpha=0.9)
opt_Adam = torch.optim.Adam(net_Adam.parameters(),lr=LR,betas=(0.9,0.99))
optimizers = [opt_SGD,opt_Momentum,opt_RMSProp,opt_Adam]
loss_func = torch.nn.MSELoss()
loss_his = [[],[],[],[]]
for epoch in range(Epochs):
for step,(batch_x,batch_y) in enumerate(loader):
for net,opt,l_his in zip(nets,optimizers,loss_his):
output = net(batch_x)
loss = loss_func(output,batch_y)
opt.zero_grad()
loss.backward()
opt.step()
l_his.append(loss.data.numpy()) #loss recoder
labels = ['SGD','Momentum','RMsprop','Adam']
for i ,l_his in enumerate(loss_his):
plt.plot(l_his, label=labels[i])
plt.legend(loc='best')
plt.xlabel('Steps')
plt.ylabel('Loss')
plt.ylim((0, 0.2))
plt.show()
边栏推荐
- Kept VRRP script, preemptive delay, VIP unicast details
- L2-006 tree traversal (25 points)
- [Presto] Presto parameter configuration optimization
- mysql实现读写分离
- ES6 let 和 const 命令
- 【flink】flink学习
- Mall project -- day09 -- order module
- Mysql的索引实现之B树和B+树
- Codeforces Round #771 (Div. 2)
- [Blue Bridge Cup 2017 preliminary] buns make up
猜你喜欢
[Flink] Flink learning
PHP - whether the setting error displays -php xxx When PHP executes, there is no code exception prompt
Apprentissage automatique - - régression linéaire (sklearn)
error C4996: ‘strcpy‘: This function or variable may be unsafe. Consider using strcpy_ s instead
[CDH] cdh5.16 configuring the setting of yarn task centralized allocation does not take effect
Vs2019 first MFC Application
Word排版(小計)
MongoDB
ToggleButton实现一个开关灯的效果
Mall project -- day09 -- order module
随机推荐
SQL时间注入
[Flink] cdh/cdp Flink on Yan log configuration
【Flink】CDH/CDP Flink on Yarn 日志配置
[CDH] cdh5.16 configuring the setting of yarn task centralized allocation does not take effect
Codeforces Round #753 (Div. 3)
Vert. x: A simple login access demo (simple use of router)
Machine learning notes week02 convolutional neural network
Composition des mots (sous - total)
ImportError: libmysqlclient. so. 20: Cannot open shared object file: no such file or directory solution
L2-006 tree traversal (25 points)
Common regular expression collation
第4阶段 Mysql数据库
{one week summary} take you into the ocean of JS knowledge
[CDH] modify the default port 7180 of cloudera manager in cdh/cdp environment
error C4996: ‘strcpy‘: This function or variable may be unsafe. Consider using strcpy_ s instead
Codeforces Round #753 (Div. 3)
express框架详解
yarn安装与使用
[Bluebridge cup 2020 preliminary] horizontal segmentation
JS array + array method reconstruction