GAN 生成对抗网络

GAN网络可以说是近来最火的神经网络模型，其变种包括WGAN，WCGAN，WGAN-l，circleGAN等，被广泛运用于计算机视觉中。

1. 准备数据和超参数

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt

# torch.manual_seed(1)    # reproducible
# np.random.seed(1)

# Hyper Parameters
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_COMPONENTS = 15     # it could be total point G can draw in the canvas
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONENTS) for _ in range(BATCH_SIZE)])

# show our beautiful painting 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()

2. 准备训练图像

def artist_works():     # painting from the famous artist (real target)
    a = np.random.uniform(1, 2, size=BATCH_SIZE)[:, np.newaxis]
    paintings = a * np.power(PAINT_POINTS, 2) + (a-1)
    paintings = torch.from_numpy(paintings).float()
    return paintings

3. 构建训练模型

G = nn.Sequential(                      # Generator
    nn.Linear(N_IDEAS, 128),            # random ideas (could from normal distribution)
    nn.ReLU(),
    nn.Linear(128, ART_COMPONENTS),     # making a painting from these random ideas
)

D = nn.Sequential(                      # Discriminator
    nn.Linear(ART_COMPONENTS, 128),     
# receive art work either from the famous artist or a newbie like G
    nn.ReLU(),
    nn.Linear(128, 1),
    nn.Sigmoid(),                       
# tell the probability that the art work is made by artist
)

注：

1. GAN的核心在于同时构建Generator和Discriminator，两个神经网络相互对抗以生成有效的结果
2. GAN的 G 和 D 使用普通全连接网络，但也可使用DCNN网络以获得更好的结果
3. GAN模型中常用Relu作为激活函数

4. 选择优化器

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

5. 训练过程

for step in range(10000):
    artist_paintings = artist_works()           # real painting from artist
    G_ideas = torch.randn(BATCH_SIZE, N_IDEAS)  # random ideas
    G_paintings = G(G_ideas)                    # fake painting from G (random ideas)

    prob_artist0 = D(artist_paintings)          # D try to increase this prob
    prob_artist1 = D(G_paintings)               # D try to reduce this prob

    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_graph=True)      # reusing computational graph
    opt_D.step()

    opt_G.zero_grad()
    G_loss.backward()
    opt_G.step()

注：
1.Pytorch中无GAN的损失函数，所以我们通常需要自定义损失函数
2.判别器的损失函数为：D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
3.生成器的损失函数为： G_loss = torch.mean(torch.log(1. - prob_artist1))

6. 可视化训练过程和结果

plt.ion()   # something about continuous plotting
    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()