PyTorch学习（15）——生成对抗网络（GAN）

Gan的全称是Generative Adveratial Nets，生成对抗网络。

Generator采用随机数生成有意义的数据，Discriminator学习判定哪些是真实数据哪些是生成数据，并反向传递到Generator。

生成对抗网络接收一些信息，生成有意义的物体。

下面是示例代码：

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt
from torch.autograd import Variable

# 超参数
BATCH_SIZE = 64
LR_G = 0.0001  # learning rate for generator
LR_D = 0.0001  # learning rate for discriminator
N_IDEAS = 5  # think of this as number of ideas for generating an art work(Generator)
ART_COMPONETS = 15  # it could be total point G can draw in the canvas
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONETS) for _ in range(BATCH_SIZE)])

# show paiting range
# plt.plot(PAINT_POINTS[0], 2 * np.power(PAINT_POINTS[0], 2) + 1, c='#74BCFF', lw=3, label='upper bound')
# plt.plot(PAINT_POINTS[0], 1 * np.power(PAINT_POINTS[0], 2) + 0, c='#FF9359', lw=3, label='lower bound')
# plt.legend(loc='upper right')
# plt.show()


def artist_works():
    a = np.random.uniform(1, 2, size=BATCH_SIZE)[:, np.newaxis]
    painting = a * np.power(PAINT_POINTS, 2) + (a - 1)
    painting = torch.from_numpy(painting).float()
    return Variable(painting)

G = nn.Sequential(
    nn.Linear(N_IDEAS, 128),
    nn.ReLU(),
    nn.Linear(128, ART_COMPONETS)
)

D = nn.Sequential(
    nn.Linear(ART_COMPONETS, 128),
    nn.ReLU(),
    nn.Linear(128, 1),
    nn.Sigmoid(),
)


opt_D = torch.optim.Adam(D.parameters(), lr=LR_D)
opt_G = torch.optim.Adam(G.parameters(), lr=LR_G)

plt.ion()

for step in range(10000):
    artist_paintings = artist_works()
    G_ideas = Variable(torch.randn(BATCH_SIZE, N_IDEAS))
    G_paintings = G(G_ideas)

    prob_artist0 = D(artist_paintings)
    prob_artist1 = D(G_paintings)

    D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1 - prob_artist1))

    G_Loss = torch.mean(torch.log(1-prob_artist1))

    opt_D.zero_grad()
    D_loss.backward(retain_variables=True)
    opt_D.step()

    opt_G.zero_grad()
    G_Loss.backward()
    opt_G.step()
    if step % 50 == 0:  # plotting
        plt.cla()
        plt.plot(PAINT_POINTS[0], G_paintings.data.numpy()[0], c='#4AD631', lw=3, label='Generated painting', )
        plt.plot(PAINT_POINTS[0], 2 * np.power(PAINT_POINTS[0], 2) + 1, c='#74BCFF', lw=3, label='upper bound')
        plt.plot(PAINT_POINTS[0], 1 * np.power(PAINT_POINTS[0], 2) + 0, c='#FF9359', lw=3, label='lower bound')
        plt.text(-.5, 2.3, 'D accuracy=%.2f (0.5 for D to converge)' % prob_artist0.data.numpy().mean(),
                 fontdict={'size': 13})
        plt.text(-.5, 2, 'D score= %.2f (-1.38 for G to converge)' % -D_loss.data.numpy(), fontdict={'size': 13})
        plt.ylim((0, 3));
        plt.legend(loc='upper right', fontsize=10);
        plt.draw();
        plt.pause(0.01)
plt.ioff()
plt.show()

数据：

结果：

PyTorch学习（15）——生成对抗网络（GAN）

猜你喜欢