首页 文章详情

【深度学习】常见优化器的PyTorch实现

机器学习初学者 | 1004 2020-07-05 23:25 0 0 0
UniSMS (合一短信)


这里主要讲不同常见优化器代码的实现,以及在一个小数据集上做一个简单的比较。

备注:pytorch需要升级到最新版本

其中,SGD和SGDM,还有Adam是pytorch自带的优化器,而RAdam是最近提出的一个说是Adam更强的优化器,但是一般情况下真正的大佬还在用SGDM来做优化器

导入必要库:

import torchimport torch.nn as nnimport torch.nn.functional as Fimport torch.optim as optimimport matplotlib.pyplot as pltimport torch.utils.data as Datafrom torch.optim.optimizer import Optimizerimport math

主程序部分:

LR = 0.01BATCH_SIZE = 32EPOCH = 12
# fake datasetx = torch.unsqueeze(torch.linspace(-1, 1, 300), 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, num_workers=2)

class Net(nn.Module): def __init__(self): super(Net, self).__init__() self.hidden = nn.Linear(1, 20) self.prediction = nn.Linear(20, 1)
def forward(self, x): x = F.relu(self.hidden(x)) x = self.prediction(x) return x

def main(): net_SGD = Net() net_Momentum = Net() net_Adam = Net() net_RAdam = Net() nets = [net_SGD, net_Momentum, net_Adam, net_RAdam] opt_SGD = optim.SGD(net_SGD.parameters(), lr=LR) opt_Momentum = optim.SGD(net_Momentum.parameters(), lr=LR, momentum=0.9) opt_Adam = optim.Adam(net_Adam.parameters(), lr=LR, betas=(0.9, 0.99)) opt_RAdam = RAdam(net_RAdam.parameters(),lr=LR,weight_decay=0) optimizers = [opt_SGD, opt_Momentum, opt_Adam, opt_RAdam] loss_func = nn.MSELoss() losses_his = [[], [], [], []] # training for epoch in range(EPOCH): print('EPOCH:', epoch) for step, (batch_x, batch_y) in enumerate(loader): b_x = batch_x b_y = batch_y for net, opt, l_his in zip(nets, optimizers, losses_his): out = net(b_x) loss = loss_func(out, b_y) opt.zero_grad() loss.backward() opt.step() l_his.append(loss.item()) labels = ['SGD', 'Momentum', 'Adam','RAdam'] 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()

if __name__ == '__main__': main()

下图是优化器的对比:

ba391ec0b462752f35f8dabf0dc8f2fa.webp

可以看出来,Adam的效果可以说是非常好的。然后SGDM其次,SGDM是大佬们经常会使用的,所以在这里虽然看起来SGDM效果不如Adam,但是依然推荐在项目中,尝试一下SGDM的效果。

14dd3bcbc955bb31c17336a25e07d25b.webp


往期精彩回顾





获取一折本站知识星球优惠券,复制链接直接打开:

https://t.zsxq.com/yFQV7am

本站qq群1003271085。

加入微信群请扫码进群:

e728ca2bfc29afe330d3c2b6c2b2b311.webp

good-icon 0
favorite-icon 0
收藏
回复数量: 0
    暂无评论~~
    Ctrl+Enter