PyTorch学习GAN（对抗神经网咯）

Posted 2021-06-07 My heart will go ~~

本节依据莫烦的画家的实例来介绍GAN的工作流程，代码如下。

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           # 两个神经网络generator,新手画家，生成器
LR_D = 0.0001           # discriminator，新手鉴赏家，判别器
N_IDEAS = 5             # 随机灵感
ART_COMPONENTS = 15     # 生成的15个线段
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONENTS) for _ in range(BATCH_SIZE)])



# 产生著名画家的数据（要学习的目的）
def artist_works():   
    # 生成-1到1的曲线，包含15个点，每个点为一个二次函数
    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

# 学习，产生类似的数据
G = nn.Sequential(                      # 
    nn.Linear(N_IDEAS, 128),  # 输入随机灵感来作画(输入了5个灵感)
    nn.ReLU(),
    nn.Linear(128, ART_COMPONENTS),     # 生成的15个y
)

#新手鉴赏家，接受画
D = nn.Sequential(                      # 
    nn.Linear(ART_COMPONENTS, 128),     # 接受创作的画（15个点）
    nn.ReLU(),
    nn.Linear(128, 1),#判别，是从哪接受的画
    nn.Sigmoid(),# 转换百分比
)

#优化两个
opt_D = torch.optim.Adam(D.parameters(), lr=LR_D)
opt_G = torch.optim.Adam(G.parameters(), lr=LR_G)

plt.ion()   # something about continuous plotting

#学习
for step in range(10000):
    artist_paintings = artist_works()  # 著名画家的画，均匀的二次函数
    G_ideas = torch.randn(BATCH_SIZE, N_IDEAS, requires_grad=True)  # 新手随机的idea
    G_paintings = G(G_ideas)  #根据idea生成画
                 
    prob_artist1 = D(G_paintings)               # D try to reduce this prob
    G_loss = torch.mean(torch.log(1. - prob_artist1))  
    opt_G.zero_grad()
    G_loss.backward()
    opt_G.step()
     
    prob_artist0 = D(artist_paintings) # 多大概率来自著名画家
    prob_artist1 = D(G_paintings.detach())  #多大概率来自知名画家
    #计算误差，用于反向传输
    D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
    opt_D.zero_grad()
    D_loss.backward(retain_graph=True)      # 保留之前网络的参数给下一次
    opt_D.step()

    if step % 50 == 0:  # 可视化
        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()