深度学习优化器(pytorch实现)

import torch
import torch.nn.functional as F
import matplotlib.pyplot as plt
import torch.utils.data as Data
from torch.autograd import Variable

LR = 0.01
BATCH_SIZE = 32
EPOCH = 12

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 Net(torch.nn.Module):
    def __init__(self, n_feature, n_hidden, n_output):
        super(Net,self).__init__()
        self.hidden = torch.nn.Linear(n_feature, n_hidden)
        self.predict = torch.nn.Linear(n_hidden, n_output)
    def forward(self, x):
        x = F.relu(self.hidden(x))
        x = self.predict(x)
        return x

net_SGD = Net(1, 20, 1)
opt_SGD = torch.optim.SGD(net_SGD.parameters(), lr=LR)

net_Momentum = Net(1, 20, 1)
opt_Momentum = torch.optim.SGD(net_Momentum.parameters(), lr=LR, momentum=0.9)  # 通常取0.5,0.9,0.99

net_RMSprop = Net(1, 20, 1)
opt_RMSprop = torch.optim.RMSprop(net_RMSprop.parameters(), lr=LR, alpha=0.9)  # 取值0.9或者0.5

net_Adam = Net(1, 20, 1)
opt_Adam = torch.optim.Adam(net_Adam.parameters(), lr=LR, betas=(0.9, 0.99))  # 超参1一般取（0.9，0.99）
loss_func = torch.nn.MSELoss()

nets = [net_SGD, net_Momentum, net_RMSprop, net_Adam]  # 放到一个list里，一个一个提取训练
optimizers = [opt_SGD, opt_Momentum, opt_RMSprop, opt_Adam]
losses_his = [[], [], [], []]   # 记录训练时不同神经网络的 loss
for epoch in range(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):  # [0,1],[2,3],[4,5]zip变成[(0,2,4),(1,3,5)],分别训练四种网络
            output = net(b_x)
            loss = loss_func(output, b_y)
            opt.zero_grad()
            loss.backward()
            opt.step()
            l_his.append(loss.data.numpy())     # 记录每种网络的loss

labels = ['SGD', 'Momentum', 'RMSprop', 'Adam']
plt.figure(figsize=(10,5))
for i,l_his in enumerate(losses_his):
    plt.subplot(3,2,i+1)
    plt.plot(l_his, label=labels[i])
    plt.legend(loc='best')
    plt.xlabel('Steps')
    plt.ylabel('Loss')
    plt.ylim((0, 0.2))
plt.subplot(3,2,5)
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()