GAN | 代码简单实现生成对抗网络（GAN）（PyTorch）

2014年GAN发表，直到最近大火的AI生成全部有GAN的踪迹，快来简单实现它！！！

GAN通过计算图和博弈论的创新组合，他们表明，如果有足够的建模能力，相互竞争的两个模型将能够通过普通的旧反向传播进行共同训练。

这些模型扮演着两种不同的（字面意思是对抗的）角色。给定一些真实的数据集R，G是生成器，试图创建看起来像真实数据的假数据，而D是鉴别器，从真实集或G获取数据并标记差异。 G就像一造假机器，通过多次画画练习，使得画出来的话像真图一样。而D是试图区分的侦探团队。（除了在这种情况下，伪造者G永远看不到原始数据——只能看到D的判断。他们就像盲人摸象的探索伪造的人。

Sourse

GAN实现代码

#!/usr/bin/env pythonimport numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
from torch.autograd import Variablematplotlib_is_available = True
try:from matplotlib import pyplot as plt
except ImportError:print("Will skip plotting; matplotlib is not available.")matplotlib_is_available = False# Data params
data_mean = 4
data_stddev = 1.25# ### Uncomment only one of these to define what data is actually sent to the Discriminator
#(name, preprocess, d_input_func) = ("Raw data", lambda data: data, lambda x: x)
#(name, preprocess, d_input_func) = ("Data and variances", lambda data: decorate_with_diffs(data, 2.0), lambda x: x * 2)
#(name, preprocess, d_input_func) = ("Data and diffs", lambda data: decorate_with_diffs(data, 1.0), lambda x: x * 2)
(name, preprocess, d_input_func) = ("Only 4 moments", lambda data: get_moments(data), lambda x: 4)print("Using data [%s]" % (name))# ##### DATA: Target data and generator input datadef get_distribution_sampler(mu, sigma):return lambda n: torch.Tensor(np.random.normal(mu, sigma, (1, n)))  # Gaussiandef get_generator_input_sampler():return lambda m, n: torch.rand(m, n)  # Uniform-dist data into generator, _NOT_ Gaussian# ##### MODELS: Generator model and discriminator modelclass Generator(nn.Module):def __init__(self, input_size, hidden_size, output_size, f):super(Generator, self).__init__()self.map1 = nn.Linear(input_size, hidden_size)self.map2 = nn.Linear(hidden_size, hidden_size)self.map3 = nn.Linear(hidden_size, output_size)self.f = fdef forward(self, x):x = self.map1(x)x = self.f(x)x = self.map2(x)x = self.f(x)x = self.map3(x)return xclass Discriminator(nn.Module):def __init__(self, input_size, hidden_size, output_size, f):super(Discriminator, self).__init__()self.map1 = nn.Linear(input_size, hidden_size)self.map2 = nn.Linear(hidden_size, hidden_size)self.map3 = nn.Linear(hidden_size, output_size)self.f = fdef forward(self, x):x = self.f(self.map1(x))x = self.f(self.map2(x))return self.f(self.map3(x))def extract(v):return v.data.storage().tolist()def stats(d):return [np.mean(d), np.std(d)]def get_moments(d):# Return the first 4 moments of the data providedmean = torch.mean(d)diffs = d - meanvar = torch.mean(torch.pow(diffs, 2.0))std = torch.pow(var, 0.5)zscores = diffs / stdskews = torch.mean(torch.pow(zscores, 3.0))kurtoses = torch.mean(torch.pow(zscores, 4.0)) - 3.0  # excess kurtosis, should be 0 for Gaussianfinal = torch.cat((mean.reshape(1,), std.reshape(1,), skews.reshape(1,), kurtoses.reshape(1,)))return finaldef decorate_with_diffs(data, exponent, remove_raw_data=False):mean = torch.mean(data.data, 1, keepdim=True)mean_broadcast = torch.mul(torch.ones(data.size()), mean.tolist()[0][0])diffs = torch.pow(data - Variable(mean_broadcast), exponent)if remove_raw_data:return torch.cat([diffs], 1)else:return torch.cat([data, diffs], 1)def train():# Model parametersg_input_size = 1      # Random noise dimension coming into generator, per output vectorg_hidden_size = 5     # Generator complexityg_output_size = 1     # Size of generated output vectord_input_size = 500    # Minibatch size - cardinality of distributionsd_hidden_size = 10    # Discriminator complexityd_output_size = 1     # Single dimension for 'real' vs. 'fake' classificationminibatch_size = d_input_sized_learning_rate = 1e-3g_learning_rate = 1e-3sgd_momentum = 0.9num_epochs = 5000print_interval = 100d_steps = 20g_steps = 20dfe, dre, ge = 0, 0, 0d_real_data, d_fake_data, g_fake_data = None, None, Nonediscriminator_activation_function = torch.sigmoidgenerator_activation_function = torch.tanhd_sampler = get_distribution_sampler(data_mean, data_stddev)gi_sampler = get_generator_input_sampler()G = Generator(input_size=g_input_size,hidden_size=g_hidden_size,output_size=g_output_size,f=generator_activation_function)D = Discriminator(input_size=d_input_func(d_input_size),hidden_size=d_hidden_size,output_size=d_output_size,f=discriminator_activation_function)criterion = nn.BCELoss()  # Binary cross entropy: http://pytorch.org/docs/nn.html#bcelossd_optimizer = optim.SGD(D.parameters(), lr=d_learning_rate, momentum=sgd_momentum)g_optimizer = optim.SGD(G.parameters(), lr=g_learning_rate, momentum=sgd_momentum)for epoch in range(num_epochs):for d_index in range(d_steps):# 1. Train D on real+fakeD.zero_grad()#  1A: Train D on reald_real_data = Variable(d_sampler(d_input_size))d_real_decision = D(preprocess(d_real_data))d_real_error = criterion(d_real_decision, Variable(torch.ones([1])))  # ones = trued_real_error.backward() # compute/store gradients, but don't change params#  1B: Train D on faked_gen_input = Variable(gi_sampler(minibatch_size, g_input_size))d_fake_data = G(d_gen_input).detach()  # detach to avoid training G on these labelsd_fake_decision = D(preprocess(d_fake_data.t()))d_fake_error = criterion(d_fake_decision, Variable(torch.zeros([1])))  # zeros = faked_fake_error.backward()d_optimizer.step()     # Only optimizes D's parameters; changes based on stored gradients from backward()dre, dfe = extract(d_real_error)[0], extract(d_fake_error)[0]for g_index in range(g_steps):# 2. Train G on D's response (but DO NOT train D on these labels)G.zero_grad()gen_input = Variable(gi_sampler(minibatch_size, g_input_size))g_fake_data = G(gen_input)dg_fake_decision = D(preprocess(g_fake_data.t()))g_error = criterion(dg_fake_decision, Variable(torch.ones([1])))  # Train G to pretend it's genuineg_error.backward()g_optimizer.step()  # Only optimizes G's parametersge = extract(g_error)[0]if epoch % print_interval == 0:print("Epoch %s: D (%s real_err, %s fake_err) G (%s err); Real Dist (%s),  Fake Dist (%s) " %(epoch, dre, dfe, ge, stats(extract(d_real_data)), stats(extract(d_fake_data))))if matplotlib_is_available:print("Plotting the generated distribution...")values = extract(g_fake_data)print(" Values: %s" % (str(values)))plt.hist(values, bins=50)plt.xlabel('Value')plt.ylabel('Count')plt.title('Histogram of Generated Distribution')plt.grid(True)plt.show()train()

代码输出结果

个人总结

GAN从编程的角度来看（纯个人理解，不对可指正）

• 利用numpy的random方法，随机生成多维的噪音向量

• 创建一个G网络用来生成

• 创建一个D网络用来判断

• 俩个网络在训练时分别进行优化

• 先训练D网络去判断真假：如果训练D为真时，进行传播；如果训练D为假时，进行传播，投入优化器（1为真，0为假）

• 在D的基础上训练G。

*因为是随机生成，所以每次生成结果不同