Dropout 缓解过拟合

Dropout 缓解过拟合

搭建两个神经网络, 一个没有 dropout, 一个有 dropout. 没有 dropout 的容易出现 过拟合, 那我们就命名为 net_overfitting,
另一个就是 net_dropped. torch.nn.Dropout(0.5) 这里的 0.5 指的是随机有 50% 的神经元会被关闭/丢弃.
这两个神经网络分开训练. 训练的环境都一样.

import torch
import matplotlib.pyplot as plt

torch.manual_seed(1)    # reproducible

N_SAMPLES = 20#数据量20
N_HIDDEN = 300#300个神经元拟合它

# training data
x = torch.unsqueeze(torch.linspace(-1, 1, N_SAMPLES), 1)
y = x + 0.3*torch.normal(torch.zeros(N_SAMPLES, 1), torch.ones(N_SAMPLES, 1))

# test data
test_x = torch.unsqueeze(torch.linspace(-1, 1, N_SAMPLES), 1)
test_y = test_x + 0.3*torch.normal(torch.zeros(N_SAMPLES, 1), torch.ones(N_SAMPLES, 1))

# show data
# plt.scatter(x.data.numpy(), y.data.numpy(), c='magenta', s=50, alpha=0.5, label='train')
# plt.scatter(test_x.data.numpy(), test_y.data.numpy(), c='cyan', s=50, alpha=0.5, label='test')
# plt.legend(loc='upper left')
# plt.ylim((-2.5, 2.5))
# plt.show()
#搭建两个神经网络, 一个没有 dropout, 一个有 dropout. 没有 dropout 的容易出现 过拟合
net_overfitting = torch.nn.Sequential(
    torch.nn.Linear(1, N_HIDDEN),
    torch.nn.ReLU(),
    torch.nn.Linear(N_HIDDEN, N_HIDDEN),
    torch.nn.ReLU(),
    torch.nn.Linear(N_HIDDEN, 1),
)

net_dropped = torch.nn.Sequential(
    torch.nn.Linear(1, N_HIDDEN),
    torch.nn.Dropout(0.5),  # 随机抽取50%节点数连接屏蔽掉
    torch.nn.ReLU(),
    torch.nn.Linear(N_HIDDEN, N_HIDDEN),
    torch.nn.Dropout(0.5),  #同上
    torch.nn.ReLU(),
    torch.nn.Linear(N_HIDDEN, 1),
)
#训练
optimizer_ofit = torch.optim.Adam(net_overfitting.parameters(), lr=0.01)
optimizer_drop = torch.optim.Adam(net_dropped.parameters(), lr=0.01)
loss_func = torch.nn.MSELoss()

for t in range(500):
    pred_ofit = net_overfitting(x)
    pred_drop = net_dropped(x)

    loss_ofit = loss_func(pred_ofit, y)
    loss_drop = loss_func(pred_drop, y)

    optimizer_ofit.zero_grad()
    optimizer_drop.zero_grad()
    loss_ofit.backward()
    loss_drop.backward()
    optimizer_ofit.step()
    optimizer_drop.step()
    #预测的时候不需要屏蔽50% 训练的时候屏蔽50%
    if t % 10 == 0:  # 每 10 步画一次图
        # 将神经网络转换成预测形式, 画好图之后改回 训练形式
        net_overfitting.eval()#先进行预测
        net_dropped.eval()  # 因为 drop 网络在 train 的时候和 test 的时候参数不一样.

        # plotting

        plt.cla()


        test_pred_ofit = net_overfitting(test_x)#预测结果
        test_pred_drop = net_dropped(test_x)#训练结果

        plt.scatter(x.data.numpy(), y.data.numpy(), c='magenta', s=50, alpha=0.3, label='train')

        plt.scatter(test_x.data.numpy(), test_y.data.numpy(), c='cyan', s=50, alpha=0.3, label='test')

        plt.plot(test_x.data.numpy(), test_pred_ofit.data.numpy(), 'r-', lw=3, label='overfitting')

        plt.plot(test_x.data.numpy(), test_pred_drop.data.numpy(), 'b--', lw=3, label='dropout(50%)')

        plt.text(0, -1.2, 'overfitting loss=%.4f' % loss_func(test_pred_ofit, test_y).data.numpy(),
                 fontdict={'size': 20, 'color': 'red'})

        plt.text(0, -1.5, 'dropout loss=%.4f' % loss_func(test_pred_drop, test_y).data.numpy(),
                 fontdict={'size': 20, 'color': 'blue'})

        plt.legend(loc='upper left');
        plt.ylim((-2.5, 2.5));
        plt.pause(0.1)

        # 将两个网络改回 训练形式
        net_overfitting.train()
        net_dropped.train()

plt.ioff()

plt.show()