对抗神经网络的说明
对抗神经网络:谁和谁进行对抗呢,两个神经网络之间进行相应的对抗,两个神经网络G和D像敌人一样,D会不断的根据已有的数据去模仿一个数据,而D会不断的进行纠错,看模仿的数据是否符合要求。随着训练次数的增加,G和D越来越精进,从而可以实现在训练结束之前达到模仿的目的。
卷积神经网络基于pytorch的实现手写数字识别
循环神经网络基于pytorch之手写数字识别
开项目
- 导入相关的包
import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt
2.定义宏
# 一批数据里面有64
BATCH_SIZE = 64
# 模仿者的学习率
LR_G = 0.0001 # learning rate for generator
# 找事的学习率
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
# 将生成的数据拼接成相应的array数组,维度为2维,大小为(64, 15)
PAINT_POINTS = np.vstack([np.linspace(-1, 1, ART_COMPONENTS) for _ in range(BATCH_SIZE)])
- 定义模仿G和找事D的神经网络
# product an image
G = nn.Sequential( # Generator
# 使用随机灵感去生成
nn.Linear(N_IDEAS, 128), # random ideas (could from normal distribution)
nn.ReLU(),
# 生成15个点
nn.Linear(128, ART_COMPONENTS), # making a painting from these random ideas
)
# Evaluation picture
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),
# 转换成相应百分比的形式
nn.Sigmoid(), # tell the probability that the art work is made by artist
)
定义优化函数
opt_D = torch.optim.Adam(D.parameters(), lr=LR_D)
opt_G = torch.optim.Adam(G.parameters(), lr=LR_G)
生成供G学习的模板
# 这些曲线是由点由组成的,最终返回的是一些点
def artist_works(): # painting from the famous artist (real target)
# 生成一个二维在1到2之间的随机数个数为64个且在列上加一个维度,为2维数据
a = np.random.uniform(1, 2, size=BATCH_SIZE)[:, np.newaxis]
# y = ax^2+a
paintings = a * np.power(PAINT_POINTS, 2) + (a - 1)
paintings = torch.from_numpy(paintings).float()
# paintings的shape是(64, 15)
return paintings
进行相应的训练
for step in range(10000):
artist_paintings = artist_works() # real painting from artist
# 生成正态分布构建的假数据
G_ideas = torch.randn(BATCH_SIZE, N_IDEAS, requires_grad=True) # random ideas\n
# G_painting变成一个(64, 15)的数据
G_paintings = G(G_ideas) # fake painting from G (random ideas)
# pro_artist1是网络产生图像为真图像的概率
prob_artist1 = D(G_paintings) # D try to reduce this prob
# G 尝试最小化D 预测为假的概率
G_loss = torch.mean(torch.log(1. - prob_artist1))
opt_G.zero_grad()
G_loss.backward()
opt_G.step()
prob_artist0 = D(artist_paintings) # D try to increase this prob
prob_artist1 = D(G_paintings.detach()) # D try to reduce this prob
# 得到其预测为假的概率,至于这个计算方法是论文规定,不需要考虑
# torch.log(prob_artist0)增加这个预测为真的概率
D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
opt_D.zero_grad()
D_loss.backward(retain_graph=True) # reusing computational graph
opt_D.step()
这篇比较详细
这篇可以帮助理解损失函数的意思
由于自己能力有限,这些东西有的不是很懂,
完整代码
"""
View more, visit my tutorial page: https://mofanpy.com/tutorials/
My Youtube Channel: https://www.youtube.com/user/MorvanZhou
Dependencies:
torch: 0.4
numpy
matplotlib
"""
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()
def artist_works(): # painting from the famous artist (real target)
a = np.random.uniform(1, 2, size=BATCH_SIZE)[:, np.newaxis]
print(a.shape)
paintings = a * np.power(PAINT_POINTS, 2) + (a - 1)
print(paintings.shape)
paintings = torch.from_numpy(paintings).float()
return paintings
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
)
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
for step in range(10000):
artist_paintings = artist_works() # real painting from artist
G_ideas = torch.randn(BATCH_SIZE, N_IDEAS, requires_grad=True) # random ideas\n
G_paintings = G(G_ideas) # fake painting from G (random ideas)
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) # D try to increase this prob
prob_artist1 = D(G_paintings.detach()) # D try to reduce this prob
D_loss = - torch.mean(torch.log(prob_artist0) + torch.log(1. - prob_artist1))
opt_D.zero_grad()
D_loss.backward(retain_graph=True) # reusing computational graph
opt_D.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()