【6】pytorch中常用的优化器

【1】优化器介绍

(1)SGD, (2)Momentum, (3)RMSprop, (4)Adam

SGD 是最普通的优化器, 也可以说没有加速效果, 而 Momentum 是 SGD 的改良版, 它加入了动量原则. 后面的 RMSprop 又是 Momentum 的升级版. 而 Adam 又是 RMSprop 的升级版. 不过从这个结果中我们看到, Adam 的效果似乎比 RMSprop 要差一点. 所以说并不是越先进的优化器, 结果越佳. 我们在自己的试验中可以尝试不同的优化器, 找到那个最适合你数据/网络的优化器.

【2】代码

# -*- coding: utf-8 -*-#

#-------------------------------------------------------------------------------
# Name:         OptimizerYOUhuaqi
# Description:  
# Author:       Administrator
# Date:         2020/12/1
#-------------------------------------------------------------------------------
import torch
import torch.utils.data as Data
import torch.nn.functional as F
import matplotlib.pyplot as plt

torch.manual_seed(1)

LR=0.01
BATCH_SIZE=32
EPOCH=12

#FAKE DATASER
x=torch.unsqueeze(torch.linspace(-1,1,500),dim=1)
y=x.pow(2)+0.1*torch.normal(torch.zeros(*x.size()))


#plot dataset
plt.scatter(x.numpy(),y.numpy())
plt.show()

#load data
torch_dataset=Data.TensorDataset(x,y)
loader=Data.DataLoader(dataset=torch_dataset,batch_size=BATCH_SIZE,shuffle=True)


#network
class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.hidden = torch.nn.Linear(1, 20)   # hidden layer
        self.predict = torch.nn.Linear(20, 1)   # output layer


    def forward(self,x):
        x=F.relu(self.hidden(x))
        x=self.predict(x)
        return x

#每一个优化器创建一个Net
net_SGD      =Net()
net_Momentum =Net()
net_Rmsprop  =Net()
net_Adam     =Net()
nets=[net_Adam,net_SGD,net_Momentum,net_Rmsprop]

#创建不同的优化器
opt_SGD     =torch.optim.SGD(net_SGD.parameters(),lr=LR)
opt_Momentum=torch.optim.SGD(net_Momentum.parameters(),lr=LR,momentum=0.8)
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_Adam,opt_Momentum,opt_RMSprop,opt_SGD]

loss_func=torch.nn.MSELoss()
losses_his=[[],[],[],[]]

for epoch in range(EPOCH):
    print('Epoch: ', epoch)
    for step, (b_x, b_y) in enumerate(loader):
        # 对每个优化器, 优化属于他的神经网络
        for net, opt, l_his in zip(nets, optimizers, losses_his):
            output = net(b_x)              # get output for every net
            loss = loss_func(output, b_y)  # compute loss for every net
            opt.zero_grad()                # clear gradients for next train
            loss.backward()                # backpropagation, compute gradients
            opt.step()                     # apply gradients
            l_his.append(loss.data.numpy())     # loss recoder


    labels = ['Adam', 'Momentum', 'RMSprop', 'SGD']
    for i, l_his in enumerate(losses_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()

【3】效果图